From Matt.Peters at windriver.com Mon Oct 1 13:12:23 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Mon, 1 Oct 2018 13:12:23 +0000 Subject: [Starlingx-discuss] Analysis of patch e547a7e & ebd41fd for StartlingX upstreaming In-Reply-To: References: Message-ID: Hi Chenjie, I agree that this is still an issue upstream that needs to be fixed. Regards, Matt From: Xu, Chenjie [mailto:chenjie.xu at intel.com] Sent: Sunday, September 30, 2018 11:07 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Zhao, Forrest; Guo, Ruijing Subject: [Starlingx-discuss] Analysis of patch e547a7e & ebd41fd for StartlingX upstreaming Hi Matt, For patches e547a7e and ebd41fd, I think it's a bug in Neutron upstream and will submit the patch to fix it in the next step. This bug is caused by passing the wrong parameter after create or update router. You can find the analysis report within attached file. Could you please help review and comment! Best Regards, Xu, Chenjie -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ghada.Khalil at windriver.com Mon Oct 1 13:52:56 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Mon, 1 Oct 2018 13:52:56 +0000 Subject: [Starlingx-discuss] [multi-os] replace "/etc" in config files by a variable "sysconfdir" Message-ID: <151EE31B9FCCA54397A757BC674650F0BA459909@ALA-MBD.corp.ad.wrs.com> Hi Yong, This proposal will need to be reviewed with the multi-OS technical leads (Brent, Saul). Once reviewed and accepted, it will need to tracked via the appropriate story in storyboard. Regards, Ghada From: Hu, Yong [mailto:yong.hu at intel.com] Sent: Friday, September 28, 2018 8:24 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] [multi-os] replace "/etc" in config files by a variable "sysconfdir" As a part of initiatives, I would like to suggest to replace “/etc”, which is hardcoded in some StarlingX config files, by a variable something like “sysconfdir”, which would be “instantiated” a bit later upon the OS type (CentOS, ClearLinux or Ubuntu). Taking “stx-nfv/nfv/nfv-vim/nfv_vim/config.ini” as an example, “/etc” is coded in this “config.ini” directly to refer other config files: config_file=/etc/nfv/nfv_plugins/alarm_handlers/config.ini config_file=/etc/nfv/nfv_plugins/nfvi_plugins/config.ini My recommendation is to use a variable “sysconfdir” as instead, and to “define” this variable on later stage, like in “nfv.spec” for CentOS, in which “sysconfdir” is “/etc”. The benefit I see here is in other OSs, such as Clear Linux in which we can set different value, for example, “/usr/local/etc” for this variable “sysconfdir”. Should you have any questions or feedback, please let me know. If I don’t see objections with this kind of changes, I will make some patches accordingly afterward. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Don.Penney at windriver.com Mon Oct 1 14:22:58 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 14:22:58 +0000 Subject: [Starlingx-discuss] [multi-os] replace "/etc" in config files by a variable "sysconfdir" In-Reply-To: References: Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B85C3@ALA-MBD.corp.ad.wrs.com> There are a set of defined RPM macros already available: https://fedoraproject.org/wiki/Packaging:RPMMacros For example, /etc should be replaced with %{_sysconfdir} in the spec files. From: Hu, Yong [mailto:yong.hu at intel.com] Sent: Friday, September 28, 2018 8:24 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] [multi-os] replace "/etc" in config files by a variable "sysconfdir" As a part of initiatives, I would like to suggest to replace “/etc”, which is hardcoded in some StarlingX config files, by a variable something like “sysconfdir”, which would be “instantiated” a bit later upon the OS type (CentOS, ClearLinux or Ubuntu). Taking “stx-nfv/nfv/nfv-vim/nfv_vim/config.ini” as an example, “/etc” is coded in this “config.ini” directly to refer other config files: config_file=/etc/nfv/nfv_plugins/alarm_handlers/config.ini config_file=/etc/nfv/nfv_plugins/nfvi_plugins/config.ini My recommendation is to use a variable “sysconfdir” as instead, and to “define” this variable on later stage, like in “nfv.spec” for CentOS, in which “sysconfdir” is “/etc”. The benefit I see here is in other OSs, such as Clear Linux in which we can set different value, for example, “/usr/local/etc” for this variable “sysconfdir”. Should you have any questions or feedback, please let me know. If I don’t see objections with this kind of changes, I will make some patches accordingly afterward. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sgw at linux.intel.com Mon Oct 1 14:49:42 2018 From: sgw at linux.intel.com (Saul Wold) Date: Mon, 1 Oct 2018 07:49:42 -0700 Subject: [Starlingx-discuss] Final reviews needed for branching manifest Message-ID: Folks, Dean posted this last week, we need final reviews to get the manifest for branching merged Here is a gerrit link for the reviews that need to be merged: https://review.openstack.org/#/q/status:open+branch:master+topic:release-backport Thanks Sau! From Don.Penney at windriver.com Mon Oct 1 14:51:29 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 14:51:29 +0000 Subject: [Starlingx-discuss] Final reviews needed for branching manifest In-Reply-To: References: Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B860D@ALA-MBD.corp.ad.wrs.com> Sorry... I've been out of town since Thursday. I don't see any reviews from that query, though -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: Monday, October 01, 2018 10:50 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Final reviews needed for branching manifest Folks, Dean posted this last week, we need final reviews to get the manifest for branching merged Here is a gerrit link for the reviews that need to be merged: https://review.openstack.org/#/q/status:open+branch:master+topic:release-backport Thanks Sau! _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ken.Young at windriver.com Mon Oct 1 14:51:45 2018 From: Ken.Young at windriver.com (Young, Ken) Date: Mon, 1 Oct 2018 14:51:45 +0000 Subject: [Starlingx-discuss] Final reviews needed for branching manifest In-Reply-To: References: Message-ID: Saul, Clicking on the link, I get an empty list. Am I missing something? /KenY On 2018-10-01, 10:49 AM, "Saul Wold" wrote: Folks, Dean posted this last week, we need final reviews to get the manifest for branching merged Here is a gerrit link for the reviews that need to be merged: https://review.openstack.org/#/q/status:open+branch:master+topic:release-backport Thanks Sau! _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Don.Penney at windriver.com Mon Oct 1 14:53:38 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 14:53:38 +0000 Subject: [Starlingx-discuss] Final reviews needed for branching manifest In-Reply-To: References: Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8621@ALA-MBD.corp.ad.wrs.com> There are some reviews here: https://review.openstack.org/#/q/topic:create-r/2018.10+status:open -----Original Message----- From: Young, Ken [mailto:Ken.Young at windriver.com] Sent: Monday, October 01, 2018 10:52 AM To: Saul Wold; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Final reviews needed for branching manifest Saul, Clicking on the link, I get an empty list. Am I missing something? /KenY On 2018-10-01, 10:49 AM, "Saul Wold" wrote: Folks, Dean posted this last week, we need final reviews to get the manifest for branching merged Here is a gerrit link for the reviews that need to be merged: https://review.openstack.org/#/q/status:open+branch:master+topic:release-backport Thanks Sau! _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From sgw at linux.intel.com Mon Oct 1 14:55:19 2018 From: sgw at linux.intel.com (Saul Wold) Date: Mon, 1 Oct 2018 07:55:19 -0700 Subject: [Starlingx-discuss] [multi-os] replace "/etc" in config files by a variable "sysconfdir" In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B85C3@ALA-MBD.corp.ad.wrs.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B85C3@ALA-MBD.corp.ad.wrs.com> Message-ID: On 10/01/2018 07:22 AM, Penney, Don wrote: > There are a set of defined RPM macros already available: > > https://fedoraproject.org/wiki/Packaging:RPMMacros > > For example, /etc should be replaced with %{_sysconfdir} in the spec files. > This issue is not in the specfiles, but in other files that have /etc hardcoded in them (.ini or .conf files). We would need to replace the hardcoded /etc with a token (sysconfdir) that can be modified (by sed likely) during installation. This modification would use the %{_sysconfdir}, in RPM case to replace the token in the config file. This would be true of other hardcoded path variables we might find along the way. Sau! > *From:*Hu, Yong [mailto:yong.hu at intel.com] > *Sent:* Friday, September 28, 2018 8:24 PM > *To:* starlingx-discuss at lists.starlingx.io > *Subject:* [Starlingx-discuss] [multi-os] replace "/etc" in config files > by a variable "sysconfdir" > > As a part of initiatives, I would like to suggest to replace “/etc”, > which is hardcoded in some StarlingX config files, > > by a variable something like “sysconfdir”, which would be “instantiated” > a bit later upon the OS type (CentOS, ClearLinux or Ubuntu). > > Taking “stx-nfv/nfv/nfv-vim/nfv_vim/config.ini” as an example, > > “/etc” is coded in this “config.ini” directly to refer other config files: > > config_file=/etc/nfv/nfv_plugins/alarm_handlers/config.ini > > config_file=/etc/nfv/nfv_plugins/nfvi_plugins/config.ini > > My recommendation is to use a variable “sysconfdir” as instead, and to > “define” this variable on later stage, > > like in “nfv.spec” for CentOS, in which “sysconfdir” is “/etc”. > > The benefit I see here is in other OSs, such as Clear Linux in which we > can set different value, > > for example, “/usr/local/etc” for this variable “sysconfdir”. > > Should you have any questions or feedback, please let me know. > > If I don’t see objections with this kind of changes, I will make some > patches accordingly afterward. > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From sgw at linux.intel.com Mon Oct 1 14:58:45 2018 From: sgw at linux.intel.com (Saul Wold) Date: Mon, 1 Oct 2018 07:58:45 -0700 Subject: [Starlingx-discuss] Final reviews needed for branching manifest In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B860D@ALA-MBD.corp.ad.wrs.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B860D@ALA-MBD.corp.ad.wrs.com> Message-ID: <5151b799-e60b-bbfe-1357-6feb1611c2d8@linux.intel.com> On 10/01/2018 07:51 AM, Penney, Don wrote: > Sorry... I've been out of town since Thursday. > > I don't see any reviews from that query, though > Strange, it works for me, it's the query that Dean gave me, I wonder what is different. Can you find: https://review.openstack.org/#/c/606153/ And then click Topic link to get the list. Sau! > -----Original Message----- > From: Saul Wold [mailto:sgw at linux.intel.com] > Sent: Monday, October 01, 2018 10:50 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Final reviews needed for branching manifest > > > Folks, > > Dean posted this last week, we need final reviews to get the manifest > for branching merged > > Here is a gerrit link for the reviews that need to be merged: > > https://review.openstack.org/#/q/status:open+branch:master+topic:release-backport > > Thanks > Sau! > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From juan.carlos.alonso at intel.com Mon Oct 1 15:09:00 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Mon, 1 Oct 2018 15:09:00 +0000 Subject: [Starlingx-discuss] There are not network interfaces in controller-1 In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB28D30B@ALA-MBD.corp.ad.wrs.com> References: <8557B550001AFB46A43A0CCC314BF85144243ABB@FMSMSX108.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB28D30B@ALA-MBD.corp.ad.wrs.com> Message-ID: <8557B550001AFB46A43A0CCC314BF85144243D76@FMSMSX108.amr.corp.intel.com> Yes, [wrsroot at controller-0 ~(keystone_admin)]$ system host-list | grep "controller-1" | 4 | controller-1 | controller | locked | disabled | online | Tested against the last ISO too: stx-2018-10-01-156.iso Regards. Juan Carlos Alonso From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, September 29, 2018 9:40 AM To: Alonso, Juan Carlos ; starlingx-discuss at lists.starlingx.io Subject: RE: There are not network interfaces in controller-1 Did controller-1 go online ? Brent From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Friday, September 28, 2018 10:17 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] There are not network interfaces in controller-1 Hi, Was there a change on how to create or provision network interfaces for controllers? When Controller-1 was installed and try to provision it, there is not output for "system host-if-list -a controller-1", for controller-0 there is. So, if controller-1 is not full provisioned it cannot 'unlock' and cannot be used. I followed steps in Wiki for Virtual multinode controller storage. https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Controller_Storage Execute against ISO: stx-2018-09-28-148.iso Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From Don.Penney at windriver.com Mon Oct 1 15:14:48 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 15:14:48 +0000 Subject: [Starlingx-discuss] Final reviews needed for branching manifest In-Reply-To: <5151b799-e60b-bbfe-1357-6feb1611c2d8@linux.intel.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B860D@ALA-MBD.corp.ad.wrs.com> <5151b799-e60b-bbfe-1357-6feb1611c2d8@linux.intel.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B866F@ALA-MBD.corp.ad.wrs.com> Thanks Saul. They're all W+1'ed now. -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: Monday, October 01, 2018 10:59 AM To: Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Final reviews needed for branching manifest On 10/01/2018 07:51 AM, Penney, Don wrote: > Sorry... I've been out of town since Thursday. > > I don't see any reviews from that query, though > Strange, it works for me, it's the query that Dean gave me, I wonder what is different. Can you find: https://review.openstack.org/#/c/606153/ And then click Topic link to get the list. Sau! > -----Original Message----- > From: Saul Wold [mailto:sgw at linux.intel.com] > Sent: Monday, October 01, 2018 10:50 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Final reviews needed for branching manifest > > > Folks, > > Dean posted this last week, we need final reviews to get the manifest > for branching merged > > Here is a gerrit link for the reviews that need to be merged: > > https://review.openstack.org/#/q/status:open+branch:master+topic:release-backport > > Thanks > Sau! > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From juan.carlos.alonso at intel.com Mon Oct 1 16:47:08 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Mon, 1 Oct 2018 16:47:08 +0000 Subject: [Starlingx-discuss] There are not network interfaces in controller-1 References: <8557B550001AFB46A43A0CCC314BF85144243ABB@FMSMSX108.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB28D30B@ALA-MBD.corp.ad.wrs.com> Message-ID: <8557B550001AFB46A43A0CCC314BF85144243E69@FMSMSX108.amr.corp.intel.com> Hi, Bug in Launchpad was open: https://bugs.launchpad.net/starlingx/+bug/1795466 Regards. Juan Carlos Alonso From: Alonso, Juan Carlos Sent: Monday, October 1, 2018 10:09 AM To: 'Rowsell, Brent' ; starlingx-discuss at lists.starlingx.io Subject: RE: There are not network interfaces in controller-1 Yes, [wrsroot at controller-0 ~(keystone_admin)]$ system host-list | grep "controller-1" | 4 | controller-1 | controller | locked | disabled | online | Tested against the last ISO too: stx-2018-10-01-156.iso Regards. Juan Carlos Alonso From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, September 29, 2018 9:40 AM To: Alonso, Juan Carlos >; starlingx-discuss at lists.starlingx.io Subject: RE: There are not network interfaces in controller-1 Did controller-1 go online ? Brent From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Friday, September 28, 2018 10:17 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] There are not network interfaces in controller-1 Hi, Was there a change on how to create or provision network interfaces for controllers? When Controller-1 was installed and try to provision it, there is not output for "system host-if-list -a controller-1", for controller-0 there is. So, if controller-1 is not full provisioned it cannot 'unlock' and cannot be used. I followed steps in Wiki for Virtual multinode controller storage. https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Controller_Storage Execute against ISO: stx-2018-09-28-148.iso Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtroyer at gmail.com Mon Oct 1 17:01:48 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 1 Oct 2018 12:01:48 -0500 Subject: [Starlingx-discuss] r/2018.10 branch merges complete Message-ID: It looks like the last of the 'setup' reviews for the r/2018.10 branch merged a few hours ago, thank you to those who reviewed those. This means that a) master is open again for ongoing work b) we are ready to do release branch builds directly from Gerrit via stx-manifest (check out r/2018.10 and use that default.xml) Don't forge that bug fixes should merge in master before being backported to r/2018.10. dt -- Dean Troyer dtroyer at gmail.com From bruce.e.jones at intel.com Mon Oct 1 17:13:30 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Mon, 1 Oct 2018 17:13:30 +0000 Subject: [Starlingx-discuss] r/2018.10 branch merges complete In-Reply-To: References: Message-ID: <9A85D2917C58154C960D95352B22818BAB5A9549@fmsmsx117.amr.corp.intel.com> Dean, thank you! I'm looking forward to seeing posts to the mailing list of successful builds on the branch, hint hint :) brucej -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Monday, October 1, 2018 10:02 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] r/2018.10 branch merges complete It looks like the last of the 'setup' reviews for the r/2018.10 branch merged a few hours ago, thank you to those who reviewed those. This means that a) master is open again for ongoing work b) we are ready to do release branch builds directly from Gerrit via stx-manifest (check out r/2018.10 and use that default.xml) Don't forge that bug fixes should merge in master before being backported to r/2018.10. dt -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ghada.Khalil at windriver.com Mon Oct 1 17:49:12 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Mon, 1 Oct 2018 17:49:12 +0000 Subject: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 Message-ID: <151EE31B9FCCA54397A757BC674650F0BA459D8E@ALA-MBD.corp.ad.wrs.com> A critical issue was found during Wind River's sanity run today. This issue impacts successful configuration of multi-node systems in the virtual as well as baremetal environments. https://bugs.launchpad.net/starlingx/+bug/1795400 The issue appears to be related to the change of the software version to 18.10. Don Penney is investigating and is working on a fix. The fix will be sourced in both master and the stx.2018.10 release branch. Regards, Ghada -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ghada.Khalil at windriver.com Mon Oct 1 17:54:57 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Mon, 1 Oct 2018 17:54:57 +0000 Subject: [Starlingx-discuss] There are not network interfaces in controller-1 In-Reply-To: <8557B550001AFB46A43A0CCC314BF85144243E69@FMSMSX108.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF85144243ABB@FMSMSX108.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB28D30B@ALA-MBD.corp.ad.wrs.com> <8557B550001AFB46A43A0CCC314BF85144243E69@FMSMSX108.amr.corp.intel.com> Message-ID: <151EE31B9FCCA54397A757BC674650F0BA459DA2@ALA-MBD.corp.ad.wrs.com> Hi Juan Carlos, Can you please collect logs from your system and attach them to the Launchpad? Thanks, Ghada From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Monday, October 01, 2018 12:47 PM To: Rowsell, Brent; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] There are not network interfaces in controller-1 Hi, Bug in Launchpad was open: https://bugs.launchpad.net/starlingx/+bug/1795466 Regards. Juan Carlos Alonso From: Alonso, Juan Carlos Sent: Monday, October 1, 2018 10:09 AM To: 'Rowsell, Brent' >; starlingx-discuss at lists.starlingx.io Subject: RE: There are not network interfaces in controller-1 Yes, [wrsroot at controller-0 ~(keystone_admin)]$ system host-list | grep "controller-1" | 4 | controller-1 | controller | locked | disabled | online | Tested against the last ISO too: stx-2018-10-01-156.iso Regards. Juan Carlos Alonso From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, September 29, 2018 9:40 AM To: Alonso, Juan Carlos >; starlingx-discuss at lists.starlingx.io Subject: RE: There are not network interfaces in controller-1 Did controller-1 go online ? Brent From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Friday, September 28, 2018 10:17 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] There are not network interfaces in controller-1 Hi, Was there a change on how to create or provision network interfaces for controllers? When Controller-1 was installed and try to provision it, there is not output for "system host-if-list -a controller-1", for controller-0 there is. So, if controller-1 is not full provisioned it cannot 'unlock' and cannot be used. I followed steps in Wiki for Virtual multinode controller storage. https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Controller_Storage Execute against ISO: stx-2018-09-28-148.iso Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From erich.cordoba.malibran at intel.com Mon Oct 1 17:58:12 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Mon, 1 Oct 2018 17:58:12 +0000 Subject: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 In-Reply-To: <151EE31B9FCCA54397A757BC674650F0BA459D8E@ALA-MBD.corp.ad.wrs.com> References: <151EE31B9FCCA54397A757BC674650F0BA459D8E@ALA-MBD.corp.ad.wrs.com> Message-ID: <357f5ef3fd237329ec35aa47b7185fb6a4f0b718.camel@intel.com> Are network interfaces shown after installing controller-1 ? The behavior looks similar to the issue reported this weekend by Juan Carlos. https://bugs.launchpad.net/starlingx/+bug/1795466 On Mon, 2018-10-01 at 17:49 +0000, Khalil, Ghada wrote: > A critical issue was found during Wind River’s sanity run today. > This issue impacts successful configuration of multi-node systems in > the virtual as well as baremetal environments. > https://bugs.launchpad.net/starlingx/+bug/1795400 > > The issue appears to be related to the change of the software version > to 18.10. Don Penney is investigating and is working on a fix. The > fix will be sourced in both master and the stx.2018.10 release > branch. > > Regards, > Ghada > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Don.Penney at windriver.com Mon Oct 1 18:28:09 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 18:28:09 +0000 Subject: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 In-Reply-To: <357f5ef3fd237329ec35aa47b7185fb6a4f0b718.camel@intel.com> References: <151EE31B9FCCA54397A757BC674650F0BA459D8E@ALA-MBD.corp.ad.wrs.com> <357f5ef3fd237329ec35aa47b7185fb6a4f0b718.camel@intel.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8801@ALA-MBD.corp.ad.wrs.com> Hi Erich, It would be the same issue. Sysinv is unable to populate host data due to an issue with the handling of the version string: [root at controller-0 wrsroot(keystone_admin)]# system host-disk-list controller-1 [root at controller-0 wrsroot(keystone_admin)]# system host-if-list -a controller-1 [root at controller-0 wrsroot(keystone_admin)]# system host-port-list controller-1 -----Original Message----- From: Cordoba Malibran, Erich [mailto:erich.cordoba.malibran at intel.com] Sent: Monday, October 01, 2018 1:58 PM To: Khalil, Ghada; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 Are network interfaces shown after installing controller-1 ? The behavior looks similar to the issue reported this weekend by Juan Carlos. https://bugs.launchpad.net/starlingx/+bug/1795466 On Mon, 2018-10-01 at 17:49 +0000, Khalil, Ghada wrote: > A critical issue was found during Wind River’s sanity run today. > This issue impacts successful configuration of multi-node systems in > the virtual as well as baremetal environments. > https://bugs.launchpad.net/starlingx/+bug/1795400 > > The issue appears to be related to the change of the software version > to 18.10. Don Penney is investigating and is working on a fix. The > fix will be sourced in both master and the stx.2018.10 release > branch. > > Regards, > Ghada > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ghada.Khalil at windriver.com Mon Oct 1 19:02:57 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Mon, 1 Oct 2018 19:02:57 +0000 Subject: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8801@ALA-MBD.corp.ad.wrs.com> References: <151EE31B9FCCA54397A757BC674650F0BA459D8E@ALA-MBD.corp.ad.wrs.com> <357f5ef3fd237329ec35aa47b7185fb6a4f0b718.camel@intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8801@ALA-MBD.corp.ad.wrs.com> Message-ID: <151EE31B9FCCA54397A757BC674650F0BA459DDD@ALA-MBD.corp.ad.wrs.com> Thanks Don. I've marked https://bugs.launchpad.net/starlingx/+bug/1795466 as a duplicate of https://bugs.launchpad.net/starlingx/+bug/1795400 Ghada -----Original Message----- From: Penney, Don Sent: Monday, October 01, 2018 2:28 PM To: Cordoba Malibran, Erich; Khalil, Ghada; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 Hi Erich, It would be the same issue. Sysinv is unable to populate host data due to an issue with the handling of the version string: [root at controller-0 wrsroot(keystone_admin)]# system host-disk-list controller-1 [root at controller-0 wrsroot(keystone_admin)]# system host-if-list -a controller-1 [root at controller-0 wrsroot(keystone_admin)]# system host-port-list controller-1 -----Original Message----- From: Cordoba Malibran, Erich [mailto:erich.cordoba.malibran at intel.com] Sent: Monday, October 01, 2018 1:58 PM To: Khalil, Ghada; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Critical issue found in stx master as of 2018-09-28_12-39-00 Are network interfaces shown after installing controller-1 ? The behavior looks similar to the issue reported this weekend by Juan Carlos. https://bugs.launchpad.net/starlingx/+bug/1795466 On Mon, 2018-10-01 at 17:49 +0000, Khalil, Ghada wrote: > A critical issue was found during Wind River’s sanity run today. > This issue impacts successful configuration of multi-node systems in > the virtual as well as baremetal environments. > https://bugs.launchpad.net/starlingx/+bug/1795400 > > The issue appears to be related to the change of the software version > to 18.10. Don Penney is investigating and is working on a fix. The fix > will be sourced in both master and the stx.2018.10 release branch. > > Regards, > Ghada > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Don.Penney at windriver.com Mon Oct 1 20:21:10 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 20:21:10 +0000 Subject: [Starlingx-discuss] r/2018.10 branch merges complete In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5A9549@fmsmsx117.amr.corp.intel.com> References: <9A85D2917C58154C960D95352B22818BAB5A9549@fmsmsx117.amr.corp.intel.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8995@ALA-MBD.corp.ad.wrs.com> For backporting a fix to r/2018.10, I've got a repo setup on that branch and cherry-picked the change, and I'm building a load to test while waiting for Zuul to process the update to master. Maybe a stupid question, but do I need to change the "Change-Id" for the commit (ie. delete it and amend the commit) to avoid collision with the original when I create the new gerrit review? -----Original Message----- From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Monday, October 01, 2018 1:14 PM To: Dean Troyer; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] r/2018.10 branch merges complete Dean, thank you! I'm looking forward to seeing posts to the mailing list of successful builds on the branch, hint hint :) brucej -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Monday, October 1, 2018 10:02 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] r/2018.10 branch merges complete It looks like the last of the 'setup' reviews for the r/2018.10 branch merged a few hours ago, thank you to those who reviewed those. This means that a) master is open again for ongoing work b) we are ready to do release branch builds directly from Gerrit via stx-manifest (check out r/2018.10 and use that default.xml) Don't forge that bug fixes should merge in master before being backported to r/2018.10. dt -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From dtroyer at gmail.com Mon Oct 1 20:40:11 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 1 Oct 2018 15:40:11 -0500 Subject: [Starlingx-discuss] r/2018.10 branch merges complete In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8995@ALA-MBD.corp.ad.wrs.com> References: <9A85D2917C58154C960D95352B22818BAB5A9549@fmsmsx117.amr.corp.intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8995@ALA-MBD.corp.ad.wrs.com> Message-ID: On Mon, Oct 1, 2018 at 3:21 PM Penney, Don wrote: > For backporting a fix to r/2018.10, I've got a repo setup on that branch and cherry-picked the change, and I'm building a load to test while waiting for Zuul to process the update to master. Maybe a stupid question, but do I need to change the "Change-Id" for the commit (ie. delete it and amend the commit) to avoid collision with the original when I create the new gerrit review? No, in fact for cherry-picking you want the change-id to be the same. Gerrit understands this and will do it for you if you do the cherry-pick in the web UI. This simplifies matching the reviews. dt -- Dean Troyer dtroyer at gmail.com From Don.Penney at windriver.com Mon Oct 1 20:51:56 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Mon, 1 Oct 2018 20:51:56 +0000 Subject: [Starlingx-discuss] r/2018.10 branch merges complete In-Reply-To: References: <9A85D2917C58154C960D95352B22818BAB5A9549@fmsmsx117.amr.corp.intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8995@ALA-MBD.corp.ad.wrs.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B89E6@ALA-MBD.corp.ad.wrs.com> The process I've followed here for the backport, partly due to waiting for Zuul to process the original review, is: - Cloned r/2018.10 - Cherry-picked the original review, using the cmd provided by Gerrit: - Build a load (in progress) - Re-test In this particular case, the change is small enough (and the difference between master and r/2018.10 is minimal at this point) that I'd feel comfortable with a straight cherry-pick. For other changes, I think the build and retest steps would be important. I suppose I could have used the "cherry-pick" button on Gerrit and then downloaded that update into my repo for the testing, setting workflow to -1 while I complete testing, but I thought I should wait until the update had been merged, based on your original email. Side note: After 1h42, Zuul has passed the "check" and is running the "gate". So it will likely be merged very shortly. -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Monday, October 01, 2018 4:40 PM To: Penney, Don Cc: JONES, BRUCE; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] r/2018.10 branch merges complete On Mon, Oct 1, 2018 at 3:21 PM Penney, Don wrote: > For backporting a fix to r/2018.10, I've got a repo setup on that branch and cherry-picked the change, and I'm building a load to test while waiting for Zuul to process the update to master. Maybe a stupid question, but do I need to change the "Change-Id" for the commit (ie. delete it and amend the commit) to avoid collision with the original when I create the new gerrit review? No, in fact for cherry-picking you want the change-id to be the same. Gerrit understands this and will do it for you if you do the cherry-pick in the web UI. This simplifies matching the reviews. dt -- Dean Troyer dtroyer at gmail.com From dtroyer at gmail.com Mon Oct 1 21:12:05 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 1 Oct 2018 16:12:05 -0500 Subject: [Starlingx-discuss] r/2018.10 branch merges complete In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B89E6@ALA-MBD.corp.ad.wrs.com> References: <9A85D2917C58154C960D95352B22818BAB5A9549@fmsmsx117.amr.corp.intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B8995@ALA-MBD.corp.ad.wrs.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B89E6@ALA-MBD.corp.ad.wrs.com> Message-ID: [Grrr... gmail changed on me this weekend and the new quoting looks like crap...sorry] On Mon, Oct 1, 2018 at 3:52 PM Penney, Don wrote: > The process I've followed here for the backport, partly due to waiting for Zuul to process the original review, is: > - Cloned r/2018.10 > - Cherry-picked the original review, using the cmd provided by Gerrit: > - Build a load (in progress) > - Re-test This sounds fine. You could even run it through the check pipeline while waiting. The flow you mention below is also good (cp in the web ui). FWIW I usually do what you did above mostly because I want to run the tests locally first. > In this particular case, the change is small enough (and the difference between master and r/2018.10 is minimal at this point) that I'd feel comfortable with a straight cherry-pick. For other changes, I think the build and retest steps would be important. I suppose I could have used the "cherry-pick" button on Gerrit and then downloaded that update into my repo for the testing, setting workflow to -1 while I complete testing, but I thought I should wait until the update had been merged, based on your original email. The practice in OpenStack has been to wait for the review on master to merge. This is in part because it was a different group that ultimately approved stable backports, this is no longer necessarily true for many projects. We can decide how strict to be on that matter, I would suggest that we be reasonably sure that it will merge on master before +W on the backport. This case sounds pretty safe as you say but unless there is a reason to be in a hurry > Side note: After 1h42, Zuul has passed the "check" and is running the "gate". So it will likely be merged very shortly. Zuul does seem to be doing better so far this week. dt -- Dean Troyer dtroyer at gmail.com From bruce.e.jones at intel.com Mon Oct 1 21:17:51 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Mon, 1 Oct 2018 21:17:51 +0000 Subject: [Starlingx-discuss] agenda for multi-os sub-project call today 4PM Pacific Message-ID: <9A85D2917C58154C960D95352B22818BAB5A9C2C@fmsmsx117.amr.corp.intel.com> Zoom link: https://zoom.us/j/342730236 This file is https://etherpad.openstack.org/p/stx-multios This is the meeting notes and agenda file for the StarlingX Multi-OS project Agenda for the October 1st call: * Introduction / Welcome * TL / PL volunteers / assignments * Objectives and goals * Deliverables o Spec for the Clear Linux MVP o TBD... -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Mon Oct 1 23:01:31 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Mon, 1 Oct 2018 23:01:31 +0000 Subject: [Starlingx-discuss] [Build] Build report ww39 sept 23rd - sept29 2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB47105260DE@fmsmsx104.amr.corp.intel.com> This is the build report for StarlingX project for ww39 As of sept 21st we started to report "Build failed in stage 'build-pkgs --std'." issues in our Intel internal CI/CD infrastructure. There were multiple workarounds to issues discovered besides the initial failure like: - New packages being introduced as part of the CentOS 7.5 branch. - Issues with our local CI/CD environment. - Fuzzy patches. - "too many loops" errors. WindRiver started to report the same errors on their build system during this week and we proceeded as a team to root cause the issues. After several iterations and multiple fixes, the issue was root caused and solved "The root cause of the issue has been found. It turns out that the mock environment is recycled between build packages to save time. After the CentOS 7.5 upgrade the autoconf-archive package is built and installed in the mock environment. This package contains a set of autoconf macros which conflict with the build scripts for ceph. Therefore, ceph fails with the 'too many loops' error and all the dependant packages fails as well.." [1] After 26 failed attempted builds we got our first successful build on September 29th So the balance for the week: Failed builds - 26 Successful builds - 4 [1] https://bugs.launchpad.net/bugs/1794415 Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Mon Oct 1 23:55:36 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Mon, 1 Oct 2018 23:55:36 +0000 Subject: [Starlingx-discuss] agenda for multi-os sub-project call today 4PM Pacific In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5A9C2C@fmsmsx117.amr.corp.intel.com> References: <9A85D2917C58154C960D95352B22818BAB5A9C2C@fmsmsx117.amr.corp.intel.com> Message-ID: <9A85D2917C58154C960D95352B22818BAB5A9E1C@fmsmsx117.amr.corp.intel.com> We didn't have a quorum so we had no meeting. Saul will run the call next week and the week after in my absence. brucej From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Monday, October 1, 2018 2:18 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] agenda for multi-os sub-project call today 4PM Pacific Zoom link: https://zoom.us/j/342730236 This file is https://etherpad.openstack.org/p/stx-multios This is the meeting notes and agenda file for the StarlingX Multi-OS project Agenda for the October 1st call: * Introduction / Welcome * TL / PL volunteers / assignments * Objectives and goals * Deliverables o Spec for the Clear Linux MVP o TBD... -------------- next part -------------- An HTML attachment was scrubbed... URL: From glenn.seiler at windriver.com Tue Oct 2 00:54:57 2018 From: glenn.seiler at windriver.com (Seiler, Glenn) Date: Tue, 2 Oct 2018 00:54:57 +0000 Subject: [Starlingx-discuss] StarlingX Initial Governance In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> References: <0309b193-6b70-00d3-9ac4-053eb4f28fd4@openstack.org> <2A0287E6-6C87-49D8-A896-D0D663ECF8C7@openstack.org> <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> Message-ID: Has this been done yet (Governance doc in stx-governance repo)? The way I understand the timeline from the last call I attended, and from talking to various people is the following: August 2018-PTG Initial 4 people appointed Q4 2018 (around launch and Berlin) 3 open seats appointed 1H 2019 (time undetermined) 2 more open seats appointed August 2019 4 initial seats up for voting TBD- not clear.... Other seats up for voting I think it is necessary for this to be very clear with a timeline similar to, and as simple as, the above so that when we are doing our outreach to prospective new contributers we can be crystal clear about when they might expect to be appointed a chair, or voted into a chair. Regards, -glenn -----Original Message----- From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Thursday, September 27, 2018 9:15 AM To: Claire Massey; Thierry Carrez; Ildiko Vancsa Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] StarlingX Initial Governance Thank you for the feedback. We appreciate it. The way I'd like to suggest that we handle this is to get the governance document into source control. At the Documentation team meeting this week we agreed to create a stx-governance repo and place the document there. Then I will post changes to gerrit to address this feedback that can be reviewed and approved. Makes sense? @Ildiko, please let me know when the repo goes live. I'll be working on converting the current draft from wiki to .rst this week. brucej -----Original Message----- From: Claire Massey [mailto:claire at openstack.org] Sent: Thursday, September 27, 2018 8:21 AM To: Thierry Carrez Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] StarlingX Initial Governance Hi all, I agree with Thierry and think that in order to encourage growth for the project, the first election period should to be adjusted. The next several months will be a critical time for StarlingX to welcome diverse engagement. From what we’ve experienced while launching other projects, like Kata Containers for example, new companies that are interested in engaging with the project for the first time are also very interested in having the opportunity to be elected to the technical leadership body within a reasonable amount of time. The opportunity to contribute to the project and possibly earn an elected seat early on in the life of the project can be very motivating for new companies that want to get involved. Based on the current StarlingX election timeline shown in the governance documentation[1], September 2019 is the soonest date when a new person would be able to be elected to the TSC. I personally think that waiting an entire year is just too long and may discourage new contributors from getting involved in the short-term. I recommend that instead of waiting a full year to hold the first election, that it's held sooner, in *May 2019*, which is eight months from now. The planned StarlingX release cadence and schedule[2] shows that the next two releases are timed to happen in March and July 2019, so hosting the first election between those two releases, in May 2019, could work well. [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance#Technical_Steering_Committee [2] https://wiki.openstack.org/wiki/StarlingX/Release_Plan#Release_Cadence_and_Schedule Thanks, Claire > On Sep 27, 2018, at 10:13 AM, Thierry Carrez wrote: > > Hi everyone, > > I reread the Governance document[1] and found an inconsistency we'll likely want to address before the initial TSC formally blesses it: > > In "Technical Steering Committee": >> It will be comprised of 7 members who will be appointed at Launch but fully elected by the Contributors within the first year > > then: >> In April 2020, the remaining 4 seats are up for election > > "appointed at Launch" is a bit vague, I would use "initially appointed" (which would start the clock to when the TSC is fully formed). But more importantly, "fully elected by the Contributors within the first year" is at odds with still having appointed members around in March, 2020. The date or the period needs to be adjusted to match. > > We'll also likely want to clarify initial terms for the non-WR and non-Intel initial TSC members. One option is to appoint one of them with a short term, and the other two with long terms. Another option would be to randomly select the short one (like for the WR and Intel initially-appointed seats) > > Finally, I would suggest saying that if somone steps down from their TSC seat, their seat should automatically be up for election at the next round. So imagine someone that is initially appointed for a year, stepping down after 4 months: their seat would be up for reelection, which would make the first election determine 4 seats instead of just 3. That will facilitate keeping an active TSC with a limited number of stale members. > > That level of subtlety could be conveyed through a separate document that details the corner cases of the election process, if you find it too distracting for the Governance document. > > [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance > > -- > Thierry Carrez (ttx) > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Tue Oct 2 05:42:20 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Tue, 2 Oct 2018 05:42:20 +0000 Subject: [Starlingx-discuss] [Build] Build report ww39 sept 23rd - sept29 2018 In-Reply-To: <0B566C62EC792145B40E29EFEBF1AB47105260DE@fmsmsx104.amr.corp.intel.com> References: <0B566C62EC792145B40E29EFEBF1AB47105260DE@fmsmsx104.amr.corp.intel.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F2B38DAD6@SHSMSX104.ccr.corp.intel.com> Thanks Cesar and team root-causing this issue. Looking into the bug 1794415, we have two solutions (or workaround?) for the issue, are you continue to drive which one to adopt in the build system? Thanks. Thx. - cindy From: Lara, Cesar [mailto:cesar.lara at intel.com] Sent: Tuesday, October 2, 2018 7:02 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] [Build] Build report ww39 sept 23rd - sept29 2018 This is the build report for StarlingX project for ww39 As of sept 21st we started to report "Build failed in stage 'build-pkgs --std'." issues in our Intel internal CI/CD infrastructure. There were multiple workarounds to issues discovered besides the initial failure like: - New packages being introduced as part of the CentOS 7.5 branch. - Issues with our local CI/CD environment. - Fuzzy patches. - "too many loops" errors. WindRiver started to report the same errors on their build system during this week and we proceeded as a team to root cause the issues. After several iterations and multiple fixes, the issue was root caused and solved "The root cause of the issue has been found. It turns out that the mock environment is recycled between build packages to save time. After the CentOS 7.5 upgrade the autoconf-archive package is built and installed in the mock environment. This package contains a set of autoconf macros which conflict with the build scripts for ceph. Therefore, ceph fails with the 'too many loops' error and all the dependant packages fails as well.." [1] After 26 failed attempted builds we got our first successful build on September 29th So the balance for the week: Failed builds - 26 Successful builds - 4 [1] https://bugs.launchpad.net/bugs/1794415 Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From saul.wold at intel.com Tue Oct 2 13:12:43 2018 From: saul.wold at intel.com (Saul Wold) Date: Tue, 2 Oct 2018 06:12:43 -0700 Subject: [Starlingx-discuss] Release Meeting?? Message-ID: Morning Folks, Bruce and I attempted to attend the "Release meeting" this morning and it was just us. Given we just cut the branch and our release is later this month, I would have thought this meeting would occur as a status check for the release. I noted there is no associated etherpad either. Then I looked at the "Release Project" section and noted that release issues are discussed in the Wed Community meeting. I am going to change the Meeting Wiki to show that the Release Meeting is a placeholder and not a regular meeting. I gather that if a Release meeting is needed, it will be announced on the Mailing list? Thanks Sau! From sgw at linux.intel.com Tue Oct 2 13:29:59 2018 From: sgw at linux.intel.com (Saul Wold) Date: Tue, 2 Oct 2018 06:29:59 -0700 Subject: [Starlingx-discuss] build-pkgs cannot complete std build----aclocal: too many loops In-Reply-To: References: <2FD5DDB5A04D264C80D42CA35194914F2B38B7FE@SHSMSX104.ccr.corp.intel.com> Message-ID: Folks, What's the status of the review of these PRs for stx-ceph, this is still causing build breakage in the 2018.10 branch. I think this will require an update to the 2018.10 manifest also Sau! On 09/29/2018 06:47 AM, Sun, Austin wrote: > Hi Cindy and All: > I have generated two PR, one is for master and another is for r/2018.10. please help check if the PR process is correct . > > -[0] https://github.com/starlingx-staging/stx-ceph/pull/2 > -[1] https://github.com/starlingx-staging/stx-ceph/pull/3 > > > Thanks. > BR > Austin Sun. > > > -----Original Message----- > From: Xie, Cindy > Sent: Saturday, September 29, 2018 6:03 PM > To: Sun, Austin ; Saul Wold ; Scott Little ; starlingx-discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] build-pkgs cannot complete std build----aclocal: too many loops > > Austin, > Thanks for the finding - can we generate pull-requests for ax_require_define.m4 against StarlingX-staging? > > Thx. - cindy > > -----Original Message----- > From: Sun, Austin [mailto:austin.sun at intel.com] > Sent: Saturday, September 29, 2018 4:41 PM > To: Saul Wold ; Scott Little ; starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std build----aclocal: too many loops > > Hi Saul , Scott and Erich: > > I met same issue in my local env, I did some study aclocal script and some test. > > According your analysis , autoconf-archive-2017.03.21-1.el7.noarch.rpm was added , which included a lot of system m4 files used by aclocal, according [0] and [1], '--install' option will cp system macros (.m4) to local m4 folder , > > From error log: > > BUILDSTDERR: aclocal: installing /usr/share/aclocal/ax_require_defined.m4 to m4/ax_require_defined.m4 > BUILDSTDERR: aclocal: installing 'm4/ax_require_defined.m4' from '/usr/share/aclocal/ax_require_defined.m4' > BUILDSTDERR: aclocal: running: cp /usr/share/aclocal/ax_require_defined.m4 m4/ax_require_defined.m4 > BUILDSTDERR: aclocal: running aclocal anew, because some files were installed locally > BUILDSTDERR: aclocal: error: too many loops > > ax_require_defined.m4 cause this issue, If copied ax_require_defined.m4 to code base stx/git/ceph/m4/, ceph can be built successfully . > > since cannot upload change to starlingx-staging , please use [3] ax_require_defined.m4 file which is same as file in autoconf-archive-2017.03.21-1.el7.noarch.rpm package > > - About why ax_require_defined.m4 is a must for ceph build: > ax_cxx_compile_stdcxx_11.m4 is upgraded from local serial 4 to system serial 18 including in autoconf-archive rpm , which is using AX_REQUIRE_DEFINED defined in ax_require_defined.m4 > have tried force upgrade local m4/ax_cxx_compile_stdcxx_11.m4 to serial 19 , it can solve this issue too. > > -[0] https://www.gnu.org/software/automake/manual/html_node/Serials.html#Serials > -[1] https://www.gnu.org/software/automake/manual/html_node/aclocal-Invocation.html#aclocal-Invocation > -[2] http://git.savannah.gnu.org/gitweb/?p=autoconf-archive.git;a=blob_plain;f=m4/ax_require_defined.m4 > > Thanks. > BR > Austin Sun. > > -----Original Message----- > From: Saul Wold [mailto:sgw at linux.intel.com] > Sent: Saturday, September 29, 2018 8:20 AM > To: Scott Little ; starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std build > > > > On 09/28/2018 01:39 PM, Scott Little wrote: >> Ok, we've seen 3 ceph failures in our last 6 builds. >> >> The common factor:  tpm2-tools builds on 'b0' before ceph builds. >> >> Our theory.  The buildRequires of tpm2-tools causes autoconf-archive >> to be installed... which installs a bunch of .m4 files in >> /usr/share/aclocal ... which causes ceph grief when it calls aclocal. >> >> I don't really know automake or aclocal all that well.  I'm assuming >> /usr/share/aclocal is acting something like a cache, but it's a cache >> whos contents are incompatible with ceph. >> >> Do we have any autotools / aclocal / m4 experts in the house? >> >> Possible fixes: >> - ceph: can we tell it to not use the aclocal cache... explicitly (a >> flag to aclocal?)  ... or implicitly (update ceph's m4 files so they >> look 'newer' than the cache)? > > Not sure about that, I would have to dig deeper into aclocal, it's been a while since I dug into that. > >> - tpm2-tools: Can we remove the dependence on autoconf-archive? No >> other package we build seems to need it. >> > A quick scan show that the autoconf-archive was put in there for travis support, and goes away this past March upstream when they coverted to using a container for travis. If we could use a newer version of tpm2 that might solve this. > > Maybe Erich's solution can work > > Sau! > > >> Scott >> >> >> >> On 18-09-27 04:45 PM, Saul Wold wrote: >>> >>> And of course it worked the third time! >>> >>> So, I lost the good logs. >>> >>> Sau! >>> >>> >>> On 09/27/2018 12:56 PM, Scott Little wrote: >>>> On 18-09-27 03:53 PM, Scott Little wrote: >>>>> Our latest build, based on code synced at 2018-09-27T15:28:00  UTC, >>>>> built successfully. >>>>> >>>>> It took three attempts to get ceph built.  The first two passes >>>>> aborted quickly due to missing packages.  The final pass did not >>>>> exhibit the 'aclocal: too many loops'**issue. >>>>> >>>>> The only build I have that exhibited the too many loops error was a >>>>> snapshot on 2018-09-20T15:50:40 UTC >>>>> >>>>> I do have a designer with an older snapshot that seems to hit it >>>>> regularly, so I'll work with him and see if we can learn more. >>>>> >>>>> I think we need more data from the community >>>>> - Who's build is failing on ceph with *aclocal: too many loops?* >>>>> - Who is building successfully ? >>>>> - Who can build only intermittently? >>>>> >>>>> >>>>> >>>>> Info to collect for failed builds: >>>> - repo sync timestamp >>>>> - build command used? >>>>> - Was it a new workspace, a cleaned workspace, or a previously used >>>>> workspace? >>>> - $MY_WORKSPACE/CONTEXT >>>>> - $MY_WORKSPACE/build-std.log >>>>> - $MY_WORKSPACE/std/results/*/ceph-*/*.log >>>>> >>>>> For successful builds, same info. Rather than full build logs, I >>>>> can settle for: >>>>> - grep '\(Success building\|iteration\|building ceph\)' >>>>> $MY_WORKSPACE/build-std.log >>>>> - grep compute_resources: build-std.log >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> On 18-09-27 02:21 PM, Saul Wold wrote: >>>>>> On 09/26/2018 09:16 AM, Scott Little wrote: >>>>>>> aclocal 'too many loops' has been popping up sporadically for a >>>>>>> week or two now.  Possibly 7.5 related. >>>>>>> >>>>>>> I suspect that there is a build order and/or race condition >>>>>>> element to this.   It often goes away if you just run build-pkgs >>>>>>> a second time. >>>>>>> >>>>>> I am seeing this failure also, but it does not go away after a >>>>>> second rebuild.  I have the lastest stx-root (build-tools) with >>>>>> the recent patches. >>>>>> >>>>>> Is this directly related to the fuzz issue or is there something >>>>>> else we need to address in CEPH itself. >>>>>> >>>>>> This is blocking my local build. >>>>>> >>>>>> Sau! >>>>> >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> Starlingx-discuss mailing list >>>>> Starlingx-discuss at lists.starlingx.io >>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discus >>>>> s >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> Starlingx-discuss mailing list >>>> Starlingx-discuss at lists.starlingx.io >>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>>> >> > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From Ghada.Khalil at windriver.com Tue Oct 2 13:59:07 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Tue, 2 Oct 2018 13:59:07 +0000 Subject: [Starlingx-discuss] Release Meeting?? In-Reply-To: References: Message-ID: <151EE31B9FCCA54397A757BC674650F0BA45A093@ALA-MBD.corp.ad.wrs.com> Hi Saul, Sorry for the confusion. As you noted below, I did update the "Releases" wiki page to indicate that release topics are discussed on Wednesday (in an effort to reduce some of the meeting). I didn't realize that the same information was duplicated on another wiki: https://wiki.openstack.org/wiki/Starlingx/Meetings Ghada -----Original Message----- From: Saul Wold [mailto:saul.wold at intel.com] Sent: Tuesday, October 02, 2018 9:13 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Release Meeting?? Morning Folks, Bruce and I attempted to attend the "Release meeting" this morning and it was just us. Given we just cut the branch and our release is later this month, I would have thought this meeting would occur as a status check for the release. I noted there is no associated etherpad either. Then I looked at the "Release Project" section and noted that release issues are discussed in the Wed Community meeting. I am going to change the Meeting Wiki to show that the Release Meeting is a placeholder and not a regular meeting. I gather that if a Release meeting is needed, it will be announced on the Mailing list? Thanks Sau! _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From bruce.e.jones at intel.com Tue Oct 2 16:57:38 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Tue, 2 Oct 2018 16:57:38 +0000 Subject: [Starlingx-discuss] StarlingX Initial Governance In-Reply-To: References: <0309b193-6b70-00d3-9ac4-053eb4f28fd4@openstack.org> <2A0287E6-6C87-49D8-A896-D0D663ECF8C7@openstack.org> <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> Message-ID: <9A85D2917C58154C960D95352B22818BAB5AA132@fmsmsx117.amr.corp.intel.com> The governance repo create request is waiting for approvals from the OpenStack infra team [0]. If it gets approved & created before I leave for vacation, I'll post the document and some updates to it to the StarlingX gerrit for review and approval, to address previous feedback, which may cause some changes to the timeline you have below. [0] https://review.openstack.org/#/c/607117/ brucej -----Original Message----- From: Seiler, Glenn [mailto:glenn.seiler at windriver.com] Sent: Monday, October 1, 2018 5:55 PM To: Jones, Bruce E ; Claire Massey ; Thierry Carrez ; Ildiko Vancsa Cc: starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] StarlingX Initial Governance Has this been done yet (Governance doc in stx-governance repo)? The way I understand the timeline from the last call I attended, and from talking to various people is the following: August 2018-PTG Initial 4 people appointed Q4 2018 (around launch and Berlin) 3 open seats appointed 1H 2019 (time undetermined) 2 more open seats appointed August 2019 4 initial seats up for voting TBD- not clear.... Other seats up for voting I think it is necessary for this to be very clear with a timeline similar to, and as simple as, the above so that when we are doing our outreach to prospective new contributers we can be crystal clear about when they might expect to be appointed a chair, or voted into a chair. Regards, -glenn -----Original Message----- From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Thursday, September 27, 2018 9:15 AM To: Claire Massey; Thierry Carrez; Ildiko Vancsa Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] StarlingX Initial Governance Thank you for the feedback. We appreciate it. The way I'd like to suggest that we handle this is to get the governance document into source control. At the Documentation team meeting this week we agreed to create a stx-governance repo and place the document there. Then I will post changes to gerrit to address this feedback that can be reviewed and approved. Makes sense? @Ildiko, please let me know when the repo goes live. I'll be working on converting the current draft from wiki to .rst this week. brucej -----Original Message----- From: Claire Massey [mailto:claire at openstack.org] Sent: Thursday, September 27, 2018 8:21 AM To: Thierry Carrez Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] StarlingX Initial Governance Hi all, I agree with Thierry and think that in order to encourage growth for the project, the first election period should to be adjusted. The next several months will be a critical time for StarlingX to welcome diverse engagement. From what we’ve experienced while launching other projects, like Kata Containers for example, new companies that are interested in engaging with the project for the first time are also very interested in having the opportunity to be elected to the technical leadership body within a reasonable amount of time. The opportunity to contribute to the project and possibly earn an elected seat early on in the life of the project can be very motivating for new companies that want to get involved. Based on the current StarlingX election timeline shown in the governance documentation[1], September 2019 is the soonest date when a new person would be able to be elected to the TSC. I personally think that waiting an entire year is just too long and may discourage new contributors from getting involved in the short-term. I recommend that instead of waiting a full year to hold the first election, that it's held sooner, in *May 2019*, which is eight months from now. The planned StarlingX release cadence and schedule[2] shows that the next two releases are timed to happen in March and July 2019, so hosting the first election between those two releases, in May 2019, could work well. [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance#Technical_Steering_Committee [2] https://wiki.openstack.org/wiki/StarlingX/Release_Plan#Release_Cadence_and_Schedule Thanks, Claire > On Sep 27, 2018, at 10:13 AM, Thierry Carrez wrote: > > Hi everyone, > > I reread the Governance document[1] and found an inconsistency we'll likely want to address before the initial TSC formally blesses it: > > In "Technical Steering Committee": >> It will be comprised of 7 members who will be appointed at Launch but fully elected by the Contributors within the first year > > then: >> In April 2020, the remaining 4 seats are up for election > > "appointed at Launch" is a bit vague, I would use "initially appointed" (which would start the clock to when the TSC is fully formed). But more importantly, "fully elected by the Contributors within the first year" is at odds with still having appointed members around in March, 2020. The date or the period needs to be adjusted to match. > > We'll also likely want to clarify initial terms for the non-WR and non-Intel initial TSC members. One option is to appoint one of them with a short term, and the other two with long terms. Another option would be to randomly select the short one (like for the WR and Intel initially-appointed seats) > > Finally, I would suggest saying that if somone steps down from their TSC seat, their seat should automatically be up for election at the next round. So imagine someone that is initially appointed for a year, stepping down after 4 months: their seat would be up for reelection, which would make the first election determine 4 seats instead of just 3. That will facilitate keeping an active TSC with a limited number of stale members. > > That level of subtlety could be conveyed through a separate document that details the corner cases of the election process, if you find it too distracting for the Governance document. > > [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance > > -- > Thierry Carrez (ttx) > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From scott.little at windriver.com Tue Oct 2 17:19:26 2018 From: scott.little at windriver.com (Scott Little) Date: Tue, 2 Oct 2018 13:19:26 -0400 Subject: [Starlingx-discuss] build-pkgs cannot complete std build----aclocal: too many loops In-Reply-To: References: <2FD5DDB5A04D264C80D42CA35194914F2B38B7FE@SHSMSX104.ccr.corp.intel.com> Message-ID: <1af499f8-3fc8-9ce6-009e-9f17ceee9182@windriver.com> Merged on master.  Lets everyone give it a test build.  It's never been better than 50/50 to hit us, so we want as many test builds as possible to be sure this fixes it. Scott On 18-10-02 09:29 AM, Saul Wold wrote: > > Folks, > > What's the status of the review of these PRs for stx-ceph, this is > still causing build breakage in the 2018.10 branch. > > I think this will require an update to the 2018.10 manifest also > > Sau! > > > On 09/29/2018 06:47 AM, Sun, Austin wrote: >> Hi Cindy and All: >>      I have generated two PR, one is for master and another is for >> r/2018.10.  please help check if the PR process is correct . >> >> -[0] https://github.com/starlingx-staging/stx-ceph/pull/2 >> -[1] https://github.com/starlingx-staging/stx-ceph/pull/3 >> >> >> Thanks. >> BR >> Austin Sun. >> >> >> -----Original Message----- >> From: Xie, Cindy >> Sent: Saturday, September 29, 2018 6:03 PM >> To: Sun, Austin ; Saul Wold >> ; Scott Little ; >> starlingx-discuss at lists.starlingx.io >> Subject: RE: [Starlingx-discuss] build-pkgs cannot complete std >> build----aclocal: too many loops >> >> Austin, >> Thanks for the finding - can we generate pull-requests for >> ax_require_define.m4 against StarlingX-staging? >> >> Thx. - cindy >> >> -----Original Message----- >> From: Sun, Austin [mailto:austin.sun at intel.com] >> Sent: Saturday, September 29, 2018 4:41 PM >> To: Saul Wold ; Scott Little >> ; starlingx-discuss at lists.starlingx.io >> Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std >> build----aclocal: too many loops >> >> Hi Saul , Scott and Erich: >> >>      I met same issue in my local env,  I did some study aclocal >> script and some test. >>              According your analysis , >> autoconf-archive-2017.03.21-1.el7.noarch.rpm was added , which >> included a lot of system m4 files used by aclocal,  according [0] and >> [1],  '--install' option will cp system macros (.m4) to local m4 >> folder , >> >>  From error log: >> >> BUILDSTDERR: aclocal: installing >> /usr/share/aclocal/ax_require_defined.m4 to m4/ax_require_defined.m4 >> BUILDSTDERR: aclocal: installing 'm4/ax_require_defined.m4' from >> '/usr/share/aclocal/ax_require_defined.m4' >> BUILDSTDERR: aclocal: running: cp >> /usr/share/aclocal/ax_require_defined.m4 m4/ax_require_defined.m4 >> BUILDSTDERR: aclocal: running aclocal anew, because some files were >> installed locally >> BUILDSTDERR: aclocal: error: too many loops >> >> ax_require_defined.m4 cause this issue, If  copied >> ax_require_defined.m4 to code base stx/git/ceph/m4/,  ceph can be >> built successfully . >> >> since cannot upload change to starlingx-staging , please use [3] >> ax_require_defined.m4 file which is same as file in >> autoconf-archive-2017.03.21-1.el7.noarch.rpm package >> >> - About  why ax_require_defined.m4 is a must for ceph build: >>      ax_cxx_compile_stdcxx_11.m4 is upgraded from local serial 4 to >> system serial 18 including in autoconf-archive rpm , which is using >> AX_REQUIRE_DEFINED defined in ax_require_defined.m4 >>       have tried force upgrade local m4/ax_cxx_compile_stdcxx_11.m4 >> to serial 19 ,  it can solve this issue too. >> >> -[0] >> https://www.gnu.org/software/automake/manual/html_node/Serials.html#Serials >> -[1] >> https://www.gnu.org/software/automake/manual/html_node/aclocal-Invocation.html#aclocal-Invocation >> -[2] >> http://git.savannah.gnu.org/gitweb/?p=autoconf-archive.git;a=blob_plain;f=m4/ax_require_defined.m4 >> >> Thanks. >> BR >> Austin Sun. >> >> -----Original Message----- >> From: Saul Wold [mailto:sgw at linux.intel.com] >> Sent: Saturday, September 29, 2018 8:20 AM >> To: Scott Little ; >> starlingx-discuss at lists.starlingx.io >> Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std build >> >> >> >> On 09/28/2018 01:39 PM, Scott Little wrote: >>> Ok, we've seen 3 ceph failures in our last 6 builds. >>> >>> The common factor:  tpm2-tools builds on 'b0' before ceph builds. >>> >>> Our theory.  The buildRequires of tpm2-tools causes autoconf-archive >>> to be installed... which installs a bunch of .m4 files in >>> /usr/share/aclocal ... which causes ceph grief when it calls aclocal. >>> >>> I don't really know automake or aclocal all that well.  I'm assuming >>> /usr/share/aclocal is acting something like a cache, but it's a cache >>> whos contents are incompatible with ceph. >>> >>> Do we have any autotools / aclocal / m4 experts in the house? >>> >>> Possible fixes: >>> - ceph: can we tell it to not use the aclocal cache... explicitly (a >>> flag to aclocal?)  ... or implicitly (update ceph's m4 files so they >>> look 'newer' than the cache)? >> >> Not sure about that, I would have to dig deeper into aclocal, it's >> been a while since I dug into that. >> >>> - tpm2-tools: Can we remove the dependence on autoconf-archive? No >>> other package we build seems to need it. >>> >> A quick scan show that the autoconf-archive was put in there for >> travis support, and goes away this past March upstream when they >> coverted to using a container for travis.  If we could use a newer >> version of tpm2 that might solve this. >> >> Maybe Erich's solution can work >> >> Sau! >> >> >>> Scott >>> >>> >>> >>> On 18-09-27 04:45 PM, Saul Wold wrote: >>>> >>>> And of course it worked the third time! >>>> >>>> So, I lost the good logs. >>>> >>>> Sau! >>>> >>>> >>>> On 09/27/2018 12:56 PM, Scott Little wrote: >>>>> On 18-09-27 03:53 PM, Scott Little wrote: >>>>>> Our latest build, based on code synced at 2018-09-27T15:28:00  UTC, >>>>>> built successfully. >>>>>> >>>>>> It took three attempts to get ceph built.  The first two passes >>>>>> aborted quickly due to missing packages.  The final pass did not >>>>>> exhibit the 'aclocal: too many loops'**issue. >>>>>> >>>>>> The only build I have that exhibited the too many loops error was a >>>>>> snapshot on 2018-09-20T15:50:40 UTC >>>>>> >>>>>> I do have a designer with an older snapshot that seems to hit it >>>>>> regularly, so I'll work with him and see if we can learn more. >>>>>> >>>>>> I think we need more data from the community >>>>>> - Who's build is failing on ceph with *aclocal: too many loops?* >>>>>> - Who is building successfully ? >>>>>> - Who can build only intermittently? >>>>>> >>>>>> >>>>>> >>>>>> Info to collect for failed builds: >>>>> - repo sync timestamp >>>>>> - build command used? >>>>>> - Was it a new workspace, a cleaned workspace, or a previously used >>>>>> workspace? >>>>> - $MY_WORKSPACE/CONTEXT >>>>>> - $MY_WORKSPACE/build-std.log >>>>>> - $MY_WORKSPACE/std/results/*/ceph-*/*.log >>>>>> >>>>>> For successful builds, same info. Rather than full build logs, I >>>>>> can settle for: >>>>>> - grep '\(Success building\|iteration\|building ceph\)' >>>>>> $MY_WORKSPACE/build-std.log >>>>>> - grep compute_resources: build-std.log >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> On 18-09-27 02:21 PM, Saul Wold wrote: >>>>>>> On 09/26/2018 09:16 AM, Scott Little wrote: >>>>>>>> aclocal 'too many loops' has been popping up sporadically for a >>>>>>>> week or two now.  Possibly 7.5 related. >>>>>>>> >>>>>>>> I suspect that there is a build order and/or race condition >>>>>>>> element to this.   It often goes away if you just run build-pkgs >>>>>>>> a second time. >>>>>>>> >>>>>>> I am seeing this failure also, but it does not go away after a >>>>>>> second rebuild.  I have the lastest stx-root (build-tools) with >>>>>>> the recent patches. >>>>>>> >>>>>>> Is this directly related to the fuzz issue or is there something >>>>>>> else we need to address in CEPH itself. >>>>>>> >>>>>>> This is blocking my local build. >>>>>>> >>>>>>> Sau! >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Starlingx-discuss mailing list >>>>>> Starlingx-discuss at lists.starlingx.io >>>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discus >>>>>> s >>>>> >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> Starlingx-discuss mailing list >>>>> Starlingx-discuss at lists.starlingx.io >>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>>>> >>> >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> From sgw at linux.intel.com Tue Oct 2 17:23:53 2018 From: sgw at linux.intel.com (Saul Wold) Date: Tue, 2 Oct 2018 10:23:53 -0700 Subject: [Starlingx-discuss] build-pkgs cannot complete std build----aclocal: too many loops In-Reply-To: <1af499f8-3fc8-9ce6-009e-9f17ceee9182@windriver.com> References: <2FD5DDB5A04D264C80D42CA35194914F2B38B7FE@SHSMSX104.ccr.corp.intel.com> <1af499f8-3fc8-9ce6-009e-9f17ceee9182@windriver.com> Message-ID: On 10/02/2018 10:19 AM, Scott Little wrote: > Merged on master.  Lets everyone give it a test build.  It's never been > better than 50/50 to hit us, so we want as many test builds as possible > to be sure this fixes it. > Thanks Scott, I will start up a new build on master. Sau! > Scott > > > On 18-10-02 09:29 AM, Saul Wold wrote: >> >> Folks, >> >> What's the status of the review of these PRs for stx-ceph, this is >> still causing build breakage in the 2018.10 branch. >> >> I think this will require an update to the 2018.10 manifest also >> >> Sau! >> >> >> On 09/29/2018 06:47 AM, Sun, Austin wrote: >>> Hi Cindy and All: >>>      I have generated two PR, one is for master and another is for >>> r/2018.10.  please help check if the PR process is correct . >>> >>> -[0] https://github.com/starlingx-staging/stx-ceph/pull/2 >>> -[1] https://github.com/starlingx-staging/stx-ceph/pull/3 >>> >>> >>> Thanks. >>> BR >>> Austin Sun. >>> >>> >>> -----Original Message----- >>> From: Xie, Cindy >>> Sent: Saturday, September 29, 2018 6:03 PM >>> To: Sun, Austin ; Saul Wold >>> ; Scott Little ; >>> starlingx-discuss at lists.starlingx.io >>> Subject: RE: [Starlingx-discuss] build-pkgs cannot complete std >>> build----aclocal: too many loops >>> >>> Austin, >>> Thanks for the finding - can we generate pull-requests for >>> ax_require_define.m4 against StarlingX-staging? >>> >>> Thx. - cindy >>> >>> -----Original Message----- >>> From: Sun, Austin [mailto:austin.sun at intel.com] >>> Sent: Saturday, September 29, 2018 4:41 PM >>> To: Saul Wold ; Scott Little >>> ; starlingx-discuss at lists.starlingx.io >>> Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std >>> build----aclocal: too many loops >>> >>> Hi Saul , Scott and Erich: >>> >>>      I met same issue in my local env,  I did some study aclocal >>> script and some test. >>>              According your analysis , >>> autoconf-archive-2017.03.21-1.el7.noarch.rpm was added , which >>> included a lot of system m4 files used by aclocal,  according [0] and >>> [1],  '--install' option will cp system macros (.m4) to local m4 >>> folder , >>> >>>  From error log: >>> >>> BUILDSTDERR: aclocal: installing >>> /usr/share/aclocal/ax_require_defined.m4 to m4/ax_require_defined.m4 >>> BUILDSTDERR: aclocal: installing 'm4/ax_require_defined.m4' from >>> '/usr/share/aclocal/ax_require_defined.m4' >>> BUILDSTDERR: aclocal: running: cp >>> /usr/share/aclocal/ax_require_defined.m4 m4/ax_require_defined.m4 >>> BUILDSTDERR: aclocal: running aclocal anew, because some files were >>> installed locally >>> BUILDSTDERR: aclocal: error: too many loops >>> >>> ax_require_defined.m4 cause this issue, If  copied >>> ax_require_defined.m4 to code base stx/git/ceph/m4/,  ceph can be >>> built successfully . >>> >>> since cannot upload change to starlingx-staging , please use [3] >>> ax_require_defined.m4 file which is same as file in >>> autoconf-archive-2017.03.21-1.el7.noarch.rpm package >>> >>> - About  why ax_require_defined.m4 is a must for ceph build: >>>      ax_cxx_compile_stdcxx_11.m4 is upgraded from local serial 4 to >>> system serial 18 including in autoconf-archive rpm , which is using >>> AX_REQUIRE_DEFINED defined in ax_require_defined.m4 >>>       have tried force upgrade local m4/ax_cxx_compile_stdcxx_11.m4 >>> to serial 19 ,  it can solve this issue too. >>> >>> -[0] >>> https://www.gnu.org/software/automake/manual/html_node/Serials.html#Serials >>> >>> -[1] >>> https://www.gnu.org/software/automake/manual/html_node/aclocal-Invocation.html#aclocal-Invocation >>> >>> -[2] >>> http://git.savannah.gnu.org/gitweb/?p=autoconf-archive.git;a=blob_plain;f=m4/ax_require_defined.m4 >>> >>> >>> Thanks. >>> BR >>> Austin Sun. >>> >>> -----Original Message----- >>> From: Saul Wold [mailto:sgw at linux.intel.com] >>> Sent: Saturday, September 29, 2018 8:20 AM >>> To: Scott Little ; >>> starlingx-discuss at lists.starlingx.io >>> Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std build >>> >>> >>> >>> On 09/28/2018 01:39 PM, Scott Little wrote: >>>> Ok, we've seen 3 ceph failures in our last 6 builds. >>>> >>>> The common factor:  tpm2-tools builds on 'b0' before ceph builds. >>>> >>>> Our theory.  The buildRequires of tpm2-tools causes autoconf-archive >>>> to be installed... which installs a bunch of .m4 files in >>>> /usr/share/aclocal ... which causes ceph grief when it calls aclocal. >>>> >>>> I don't really know automake or aclocal all that well.  I'm assuming >>>> /usr/share/aclocal is acting something like a cache, but it's a cache >>>> whos contents are incompatible with ceph. >>>> >>>> Do we have any autotools / aclocal / m4 experts in the house? >>>> >>>> Possible fixes: >>>> - ceph: can we tell it to not use the aclocal cache... explicitly (a >>>> flag to aclocal?)  ... or implicitly (update ceph's m4 files so they >>>> look 'newer' than the cache)? >>> >>> Not sure about that, I would have to dig deeper into aclocal, it's >>> been a while since I dug into that. >>> >>>> - tpm2-tools: Can we remove the dependence on autoconf-archive? No >>>> other package we build seems to need it. >>>> >>> A quick scan show that the autoconf-archive was put in there for >>> travis support, and goes away this past March upstream when they >>> coverted to using a container for travis.  If we could use a newer >>> version of tpm2 that might solve this. >>> >>> Maybe Erich's solution can work >>> >>> Sau! >>> >>> >>>> Scott >>>> >>>> >>>> >>>> On 18-09-27 04:45 PM, Saul Wold wrote: >>>>> >>>>> And of course it worked the third time! >>>>> >>>>> So, I lost the good logs. >>>>> >>>>> Sau! >>>>> >>>>> >>>>> On 09/27/2018 12:56 PM, Scott Little wrote: >>>>>> On 18-09-27 03:53 PM, Scott Little wrote: >>>>>>> Our latest build, based on code synced at 2018-09-27T15:28:00  UTC, >>>>>>> built successfully. >>>>>>> >>>>>>> It took three attempts to get ceph built.  The first two passes >>>>>>> aborted quickly due to missing packages.  The final pass did not >>>>>>> exhibit the 'aclocal: too many loops'**issue. >>>>>>> >>>>>>> The only build I have that exhibited the too many loops error was a >>>>>>> snapshot on 2018-09-20T15:50:40 UTC >>>>>>> >>>>>>> I do have a designer with an older snapshot that seems to hit it >>>>>>> regularly, so I'll work with him and see if we can learn more. >>>>>>> >>>>>>> I think we need more data from the community >>>>>>> - Who's build is failing on ceph with *aclocal: too many loops?* >>>>>>> - Who is building successfully ? >>>>>>> - Who can build only intermittently? >>>>>>> >>>>>>> >>>>>>> >>>>>>> Info to collect for failed builds: >>>>>> - repo sync timestamp >>>>>>> - build command used? >>>>>>> - Was it a new workspace, a cleaned workspace, or a previously used >>>>>>> workspace? >>>>>> - $MY_WORKSPACE/CONTEXT >>>>>>> - $MY_WORKSPACE/build-std.log >>>>>>> - $MY_WORKSPACE/std/results/*/ceph-*/*.log >>>>>>> >>>>>>> For successful builds, same info. Rather than full build logs, I >>>>>>> can settle for: >>>>>>> - grep '\(Success building\|iteration\|building ceph\)' >>>>>>> $MY_WORKSPACE/build-std.log >>>>>>> - grep compute_resources: build-std.log >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> On 18-09-27 02:21 PM, Saul Wold wrote: >>>>>>>> On 09/26/2018 09:16 AM, Scott Little wrote: >>>>>>>>> aclocal 'too many loops' has been popping up sporadically for a >>>>>>>>> week or two now.  Possibly 7.5 related. >>>>>>>>> >>>>>>>>> I suspect that there is a build order and/or race condition >>>>>>>>> element to this.   It often goes away if you just run build-pkgs >>>>>>>>> a second time. >>>>>>>>> >>>>>>>> I am seeing this failure also, but it does not go away after a >>>>>>>> second rebuild.  I have the lastest stx-root (build-tools) with >>>>>>>> the recent patches. >>>>>>>> >>>>>>>> Is this directly related to the fuzz issue or is there something >>>>>>>> else we need to address in CEPH itself. >>>>>>>> >>>>>>>> This is blocking my local build. >>>>>>>> >>>>>>>> Sau! >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> Starlingx-discuss mailing list >>>>>>> Starlingx-discuss at lists.starlingx.io >>>>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discus >>>>>>> s >>>>>> >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Starlingx-discuss mailing list >>>>>> Starlingx-discuss at lists.starlingx.io >>>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>>>>> >>>> >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>> > From abraham.arce.moreno at intel.com Tue Oct 2 17:31:15 2018 From: abraham.arce.moreno at intel.com (Arce Moreno, Abraham) Date: Tue, 2 Oct 2018 17:31:15 +0000 Subject: [Starlingx-discuss] Stories for stx-tools deployment In-Reply-To: <282f2903-db5a-e98e-b884-f9d68af14dc5@windriver.com> References: <511d268b-c23c-2520-152c-fcf7e9286e09@windriver.com> <282f2903-db5a-e98e-b884-f9d68af14dc5@windriver.com> Message-ID: Thanks Michel, I am taking your email as a base to understand from your experience more on the details to work the stories below. > I created four stories: > > deployment/libvirt: sizing for reference configurations >   https://storyboard.openstack.org/#!/story/2003835 Currently this is our current libvirt configuration [0] [ All-In-One] - setup_allinone.sh - Controllers: 2 - vCPUs: 6 - Memory: 18 GB - Disks: - 0 600 GB - 1 200 GB - 2 200 GB - NICs: 4 [ Standard Controller ] - setup_standard_controller.sh - Controllers: 2 - vCPUs: 4 - Memory: ~ 17 GB - Disks: - 0 200 GB - 1 200 GB - 4 NICs - Computes: 2 - vCPUs: 4 - Memory: ~ 17 GB - Disks - 0 200 GB - 1 200 GB - 4 NICs Some questions: - Do we need any changes to the above configurations? - We are covered for Simplex, Duplex and Controller Storage configurations but Dedicated Storage is not identified as such, do we need to: 1. Create a specific setup_dedicated_controller.sh? 2. or should we reuse existing setup_standard_controller.sh passing a parameter to identify between Controller or Dedicated Storage? - Once we have these optimizations do we need to port them to our VirtualBox configurations? - In the story description, we mentioned: "Optimize cpu, memory, hardware emulation, etc for test coverage including..." Any more thoughts from your side on what the test coverage means? Any specific Things you want us to consider from a testing perspective? Does it involves things like tempest? > deployment/libvirt: virtual disk placement for reference configurations >   https://storyboard.openstack.org/#!/story/2003836 Can you please let us understand the use case for this story? > deployment/libvirt: tenant networking for reference configurations >   https://storyboard.openstack.org/#!/story/2003837 What do you think about this [1] reference configuration for tenant networking? [0] https://git.openstack.org/cgit/openstack/stx-tools/tree/deployment/libvirt [1] https://git.openstack.org/cgit/opnstack/stx-tools/tree/deployment/provision/simplex_stage_2.sh From michel.thebeau at windriver.com Tue Oct 2 19:08:04 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Tue, 2 Oct 2018 15:08:04 -0400 Subject: [Starlingx-discuss] Stories for stx-tools deployment In-Reply-To: References: <511d268b-c23c-2520-152c-fcf7e9286e09@windriver.com> <282f2903-db5a-e98e-b884-f9d68af14dc5@windriver.com> Message-ID: <02b60bc7-891d-7dce-58ec-1c718e365ef6@windriver.com> On 2018-10-02 01:31 PM, Arce Moreno, Abraham wrote: > Thanks Michel, Hi Abraham > I am taking your email as a base to understand from your experience more on the details > to work the stories below. > >> I created four stories: >> >> deployment/libvirt: sizing for reference configurations >>   https://storyboard.openstack.org/#!/story/2003835 > Currently this is our current libvirt configuration [0] > > [ All-In-One] > > - setup_allinone.sh > - Controllers: 2 > - vCPUs: 6 > - Memory: 18 GB > - Disks: > - 0 600 GB > - 1 200 GB > - 2 200 GB > - NICs: 4 > > [ Standard Controller ] > > - setup_standard_controller.sh > - Controllers: 2 > - vCPUs: 4 > - Memory: ~ 17 GB > - Disks: > - 0 200 GB > - 1 200 GB > - 4 NICs > - Computes: 2 > - vCPUs: 4 > - Memory: ~ 17 GB > - Disks > - 0 200 GB > - 1 200 GB > - 4 NICs > > Some questions: > > - Do we need any changes to the above configurations? There was a recent "stx.2018.10 test plan" email on the starlingx-discuss list, which I think would be useful for making this decision: do we need any changes to controller-storage, duplex or simplex configurations to support tests under virtualization.  I wrote in the notes for this story "Prepare a list of reference test cases..." and then soon after (same day) that test plan email came out - so, free alignment "yay!". I want to shrink the memory footprint down for the October release. I think we could fit simplex under 16G, duplex under 32G and controller-storage under 64G at the very least.  For example, previous experience (previous releases) put the controller-storage at 24G total for the four VMs. > - We are covered for Simplex, Duplex and Controller Storage configurations > but Dedicated Storage is not identified as such, do we need to: > 1. Create a specific setup_dedicated_controller.sh? > 2. or should we reuse existing setup_standard_controller.sh passing a parameter > to identify between Controller or Dedicated Storage? I do not have a strong opinion.  For example, currently you run setup_allinone.sh for both duplex and simplex.  One ignores the second controller when running simplex.  If you added storage xml to setup_dedicated_controller.sh I expect you'd be fine - one could ignore the storage VMs.  (The script name would not be right.) In practice for previous releases I would give the controller VM the cpu, memory, disks, networks needed to support either configuration. > - Once we have these optimizations do we need to port them to our VirtualBox > configurations? (I did not observe anyone supporting VirtualBox, but... ) "yes", I would expect the methods to be aligned. > - In the story description, we mentioned: > "Optimize cpu, memory, hardware emulation, etc for test coverage including..." > Any more thoughts from your side on what the test coverage means? Any specific > Things you want us to consider from a testing perspective? Does it involves things > like tempest? I would refer to the recent "stx.2018.10 test plan" email thread. Someone somewhere has started to think about which tests can run on hardware and which under virtualization.  I am thinking about what needs to be there for those tests. > >> deployment/libvirt: virtual disk placement for reference configurations >>   https://storyboard.openstack.org/#!/story/2003836 > Can you please let us understand the use case for this story? The host's root device can be overloaded with the host and 6 VMs running there. One needs to be careful to ensure the host's root disk is not overloaded when running automation, for example.  But, I'm really looking forward to running multiple clusters on the host using the stx-tools implementation (which is another good reason to bring the memory foot print down).  One requires more disks for that. It's the same story either way: configure where you want the virtual disks to be stored. > >> deployment/libvirt: tenant networking for reference configurations >>   https://storyboard.openstack.org/#!/story/2003837 > What do you think about this [1] reference configuration for tenant networking? The series of commands appears eerily familiar, almost as if it was copy/pasted from our internal wiki (lol).   Vlans 10 and 500 are outside of the 100-400 listed on the wikis, and providernet-b has no previous definition. I missed that script when I was testing the libvirt deployment method, otherwise I would have updated it as an example that matches the wiki, rather than carrying the attached script. M > [0] https://git.openstack.org/cgit/openstack/stx-tools/tree/deployment/libvirt > [1] https://git.openstack.org/cgit/opnstack/stx-tools/tree/deployment/provision/simplex_stage_2.sh > -------------- next part -------------- #PROVIDERNET=providernet-a #neutron providernet-create ${PROVIDERNET} --type=vlan #neutron providernet-range-create --name ${PROVIDERNET}-range1 --range 100-400 ${PROVIDERNET} #system host-if-modify -p ${PROVIDERNET} -nt data compute-0 eth1000 #system host-cpu-modify compute-0 -f vswitch -p0 1 ADMINID=`openstack project list | grep admin | awk '{print $2}'` neutron providernet-list PHYSNET0="${PROVIDERNET}" PUBLICNET='public-net0' PRIVATENET='private-net0' INTERNALNET='internal-net0' EXTERNALNET='external-net0' PUBLICSUBNET='public-subnet0' PRIVATESUBNET='private-subnet0' INTERNALSUBNET='internal-subnet0' EXTERNALSUBNET='external-subnet0' PUBLICROUTER='public-router0' PRIVATEROUTER='private-router0' neutron net-create --tenant-id ${ADMINID} --provider:network_type=vlan --provider:physical_network=${PHYSNET0} --provider:segmentation_id=100 --router:external ${EXTERNALNET} neutron net-create --tenant-id ${ADMINID} --provider:network_type=vlan --provider:physical_network=${PHYSNET0} --provider:segmentation_id=400 ${PUBLICNET} neutron net-create --tenant-id ${ADMINID} --provider:network_type=vlan --provider:physical_network=${PHYSNET0} --provider:segmentation_id=150 ${PRIVATENET} neutron net-create --tenant-id ${ADMINID} ${INTERNALNET} PUBLICNETID=`neutron net-list | grep ${PUBLICNET} | awk '{print $2}'` PRIVATENETID=`neutron net-list | grep ${PRIVATENET} | awk '{print $2}'` INTERNALNETID=`neutron net-list | grep ${INTERNALNET} | awk '{print $2}'` EXTERNALNETID=`neutron net-list | grep ${EXTERNALNET} | awk '{print $2}'` neutron subnet-create --tenant-id ${ADMINID} --name ${PUBLICSUBNET} ${PUBLICNET} 192.168.101.0/24 neutron subnet-create --tenant-id ${ADMINID} --name ${PRIVATESUBNET} ${PRIVATENET} 192.168.201.0/24 neutron subnet-create --tenant-id ${ADMINID} --name ${INTERNALSUBNET} --no-gateway ${INTERNALNET} 10.10.0.0/24 neutron subnet-create --tenant-id ${ADMINID} --name ${EXTERNALSUBNET} --gateway 192.168.1.1 --disable-dhcp ${EXTERNALNET} 192.168.1.0/24 neutron router-create ${PUBLICROUTER} neutron router-create ${PRIVATEROUTER} PRIVATEROUTERID=`neutron router-list | grep ${PRIVATEROUTER} | awk '{print $2}'` PUBLICROUTERID=`neutron router-list | grep ${PUBLICROUTER} | awk '{print $2}'` neutron router-gateway-set --disable-snat ${PUBLICROUTERID} ${EXTERNALNETID} neutron router-gateway-set --disable-snat ${PRIVATEROUTERID} ${EXTERNALNETID} neutron router-interface-add ${PUBLICROUTER} ${PUBLICSUBNET} neutron router-interface-add ${PRIVATEROUTER} ${PRIVATESUBNET} ### #PROVIDERNET=providernet-a #system host-if-modify -p ${PROVIDERNET} -nt data compute-1 \ #system host-cpu-modify compute-1 -f vswitch -p0 1 From Frank.Miller at windriver.com Tue Oct 2 19:27:22 2018 From: Frank.Miller at windriver.com (Miller, Frank) Date: Tue, 2 Oct 2018 19:27:22 +0000 Subject: [Starlingx-discuss] Containerizing the StarlingX Infrastructure Message-ID: Folks: I captured notes from this morning's presentation: * Brent presented this chart package for Containerization of StarlingX Infrastructure: https://wiki.openstack.org/wiki/Containerizing_StarlingX_Infrastructure * Some questions/answers that came up during the presentation: o Which configs will move to this containerized infrastructure? Answer: All configs will get the containerized configuration. o For 1 and 2 node configurations, CEPH will run on the controllers. o Where will RBD provisioner run? Answer: it will run in the cube space. o If trying to run an application, where will it run? Will it be possible to run apps in containers? Answer: Yes (Some limitations: eg: no high performance networking, multi-tenant limitation) o Containerized OpenStack services will require us to build our own docker images and those images will be CentOS based. o Can we generate these docker images using other O/S's? Answer: This will be a design consideration. o Will it be possible to put docker images on docker hub? Answer: Where to put the images will be looked at as part of one of the StoryBoards for this initiative. o Does the customer network share the same physical network with cluster? Answer: Networks may be physically the same but a different logical network will be used for the cluster. o Will software update change? Answer: Yes - we will be moving toward a k8s approach to update software/containers. o Can you provide some schedule info? Answer: Target is for the March release. Specs are expected to be out in the next 1-2 weeks. o If community folks are interested in participating please contact Brent or Frank. o In future, to containerize the flock will the decoupling of the flock need to be completed? Answer: No - there is no direct dependency on the decoupling work. Frank -----Original Appointment----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, September 29, 2018 10:38 AM To: Rowsell, Brent; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Containerizing the StarlingX Infrastructure When: Tuesday, October 02, 2018 10:00 AM-11:00 AM (UTC-05:00) Eastern Time (US & Canada). Where: https://zoom.us/j/342730236 Folks, Setting up this call to give the community an overview of the StarlingX infrastructure containerization initiative. Hope you can join. Thanks, Brent << File: ATT00001.txt >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Tue Oct 2 20:15:52 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Tue, 2 Oct 2018 20:15:52 +0000 Subject: [Starlingx-discuss] [build][meetings] build meeting minutes Message-ID: <0B566C62EC792145B40E29EFEBF1AB47105271D9@fmsmsx104.amr.corp.intel.com> 10/1/2018 Build team Meeting Minutes Attendees Saul, Brent, Ken, Bruce, Erich, Memo, Chuy, Mario, Marcela, Hayde, Abraham, Cesar Agenda - Build issues Findings and next steps Mock cleanup proposal General discussion - Preproposals for hosting with cengn - Brainstorming for next release - Re-scoping of stxb tool - Opens Notes Build issues - "Too many loops" Issue root caused, Ceph autotools identified, two solutions proposed, at the time of this meeting only one solution was tested. - One of the workarounds on both windriver and intel environments was to re-launch builds - Need still to test both patches and merge one on those Hosting with cengn - Brent camewith a proposal presentation for what and how a re we going to host on cengn premises - The idea is to replicate most of CentOS repos and built RPMs for people to ocnsume - We will aso host build artifacts and a finished ISO - The whole build process will have to happen on cengn premises - We need to mae our build tools availabe to them so they can replicate a mirror - We need to start separating the OS and host that as an artifact plus the packages that we ship with modifications STXB tool - We need to revisit use cases to start re-aquitechting the tools - Cesar to post first two relevant use cases to ML for dicussion - We want to start analysis focused onthese two use caess o New to StarlingX o Developer/contributor Opens - Release branch not completed yet - we need to start a testing cycle and there are still a few reviews to be merged - Saul to post a link on things that havent been merged to the IRC so we can get those going - No build created against release branch yet, so we need to get that out the door so we can start our validation cycle - Erich to create a job that will use that branch Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From marcela.a.rosales.jimenez at intel.com Tue Oct 2 20:58:12 2018 From: marcela.a.rosales.jimenez at intel.com (Rosales Jimenez, Marcela A) Date: Tue, 2 Oct 2018 20:58:12 +0000 Subject: [Starlingx-discuss] fix for lshell download bug Message-ID: <14BD8E0F-A242-447E-BC28-4A749A5AEACC@intel.com> Hi team, I was working on this bug https://bugs.launchpad.net/starlingx/+bug/1793615 The problem is that the lshell src rpm is not found by the yumdownloader tool, and this impacts master and r2018.10 branch. In the download_mirror.sh script, the rpms_centos3rdparty.lst are downloaded first, then the StarlingX_3rd*.repo config files are deleted, and then the rpms_centos.lst are downloaded. So when it is the turn for centos repo (and lshell) to be downloaded, 3rd party repos were already deleted. This is fixed by moving the lshell from repo_centos.lst to repo_centos3rdparty.lst, so it is downloaded before StarlingX_3rd* are deleted. I sent this two patches, for master and r2018.10, could you review them? https://review.openstack.org/#/c/607014/ https://review.openstack.org/#/c/607017/ Thanks. Marcela -------------- next part -------------- An HTML attachment was scrubbed... URL: From scott.w.doenecke at intel.com Tue Oct 2 21:29:34 2018 From: scott.w.doenecke at intel.com (Doenecke, Scott W) Date: Tue, 2 Oct 2018 21:29:34 +0000 Subject: [Starlingx-discuss] Updated: StarlingX Brochure for Berlin - New Draft for Comments Message-ID: <9C2E28407FB51546B9A92BD2EB7218A0013A73D094@ORSMSX103.amr.corp.intel.com> All, Since we worked through the one-pager together, I took the wording at the time and put it where I thought it made the most sense on the brochure. So the text completely changed from the last version. Here is the updated brochure with the consolidated feedback I’d gotten. Claire, if you don’t mind, we may want to review it in one of our meetings during October and I’ll tackle one more refresh before the end of the month so we can print for Berlin. Cheers, Scott _____________________________________________ From: Doenecke, Scott W Sent: Friday, September 21, 2018 9:03 PM To: claire at openstack.org; glenn.seiler at windriver.com; Allison Price ; Broughton, Travis V ; Ildiko Vancsa ; Jeff ; starlingx-discuss at lists.starlingx.io Subject: StarlingX Brochure for Berlin - Draft for comments Hello Claire et al, I’ve got a first cut at a possible StarlingX brochure for Berlin. Would you please take a look at it and see what you’d like to edit. Cheers, Scott << File: StarlingX Handout_v1.1.pdf >> -----Original Appointment----- From: claire at openstack.org [mailto:claire at openstack.org] Sent: Wednesday, August 22, 2018 10:08 AM To: claire at openstack.org; Doenecke, Scott W; glenn.seiler at windriver.com; Allison Price; Broughton, Travis V; Ildiko Vancsa; Jeff; starlingx-discuss at lists.starlingx.io Subject: Invitation: Bi-Weekly StarlingX Marketing/Community Content Work @ Every 2 weeks from 8am to 9am on Wednesday from Wed Sep 5 to Wed Nov 7 (PDT) (scott.w.doenecke at intel.com) When: Wednesday, September 19, 2018 10:00 AM-11:00 AM America/Chicago. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: StarlingX Handout_v1.3.pdf Type: application/pdf Size: 156023 bytes Desc: StarlingX Handout_v1.3.pdf URL: From bruce.e.jones at intel.com Tue Oct 2 22:08:23 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Tue, 2 Oct 2018 22:08:23 +0000 Subject: [Starlingx-discuss] [r/2018.10] StarlingX Release 2018.10 RC1 branch is complete In-Reply-To: References: Message-ID: <9A85D2917C58154C960D95352B22818BAB5AA8AD@fmsmsx117.amr.corp.intel.com> > We have not defined a formal release candidate process Is defining the formal process something that the Release team will do? brucej From dtroyer at gmail.com Wed Oct 3 02:37:06 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Tue, 2 Oct 2018 21:37:06 -0500 Subject: [Starlingx-discuss] [r/2018.10] StarlingX Release 2018.10 RC1 branch is complete In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5AA8AD@fmsmsx117.amr.corp.intel.com> References: <9A85D2917C58154C960D95352B22818BAB5AA8AD@fmsmsx117.amr.corp.intel.com> Message-ID: <5CD8780B-9438-4278-A6B0-7B885BCEF0F2@gmail.com> > On Oct 2, 2018, at 5:08 PM, Jones, Bruce E wrote: > > Is defining the formal process something that the Release team will do? If we want one, yes. I’m not sure we need one right now, but adding it will be straightforward as I do tag the branch point as rc1 (maybe it should be rc0?) even though it isn’t release-ready. Effectively our RC cycle is currently a single tag and nothing else. In OpenStack-land the release is tagged off the last rc, per project. dt — Dean Troyer From hai.tao.wang at intel.com Wed Oct 3 02:40:38 2018 From: hai.tao.wang at intel.com (Wang, Hai Tao) Date: Wed, 3 Oct 2018 02:40:38 +0000 Subject: [Starlingx-discuss] Containerizing the StarlingX Infrastructure In-Reply-To: References: Message-ID: <90D309A9E5805640B40D067D1B8EC8BF625C6C90@SHSMSX103.ccr.corp.intel.com> Hi Miller, Thanks for the minutes. I missed the sync and have a question as below. * If trying to run an application, where will it run? Will it be possible to run apps in containers? Answer: Yes (Some limitations: eg: no high performance networking, multi-tenant limitation) Does this limitation mean that we will not implement "privileged container", i.e., with full access to host, for nova-compute, libvirt, and Neutron agents being containerized? Thanks Haitao From: Miller, Frank [mailto:Frank.Miller at windriver.com] Sent: Wednesday, October 03, 2018 3:27 AM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Containerizing the StarlingX Infrastructure Folks: I captured notes from this morning's presentation: * Brent presented this chart package for Containerization of StarlingX Infrastructure: https://wiki.openstack.org/wiki/Containerizing_StarlingX_Infrastructure * Some questions/answers that came up during the presentation: * Which configs will move to this containerized infrastructure? Answer: All configs will get the containerized configuration. * For 1 and 2 node configurations, CEPH will run on the controllers. * Where will RBD provisioner run? Answer: it will run in the cube space. * If trying to run an application, where will it run? Will it be possible to run apps in containers? Answer: Yes (Some limitations: eg: no high performance networking, multi-tenant limitation) * Containerized OpenStack services will require us to build our own docker images and those images will be CentOS based. * Can we generate these docker images using other O/S's? Answer: This will be a design consideration. * Will it be possible to put docker images on docker hub? Answer: Where to put the images will be looked at as part of one of the StoryBoards for this initiative. * Does the customer network share the same physical network with cluster? Answer: Networks may be physically the same but a different logical network will be used for the cluster. * Will software update change? Answer: Yes - we will be moving toward a k8s approach to update software/containers. * Can you provide some schedule info? Answer: Target is for the March release. Specs are expected to be out in the next 1-2 weeks. * If community folks are interested in participating please contact Brent or Frank. * In future, to containerize the flock will the decoupling of the flock need to be completed? Answer: No - there is no direct dependency on the decoupling work. Frank -----Original Appointment----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, September 29, 2018 10:38 AM To: Rowsell, Brent; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Containerizing the StarlingX Infrastructure When: Tuesday, October 02, 2018 10:00 AM-11:00 AM (UTC-05:00) Eastern Time (US & Canada). Where: https://zoom.us/j/342730236 Folks, Setting up this call to give the community an overview of the StarlingX infrastructure containerization initiative. Hope you can join. Thanks, Brent << File: ATT00001.txt >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From ildiko at openstack.org Wed Oct 3 11:58:24 2018 From: ildiko at openstack.org (Ildiko Vancsa) Date: Wed, 3 Oct 2018 13:58:24 +0200 Subject: [Starlingx-discuss] StarlingX Initial Governance In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5AA132@fmsmsx117.amr.corp.intel.com> References: <0309b193-6b70-00d3-9ac4-053eb4f28fd4@openstack.org> <2A0287E6-6C87-49D8-A896-D0D663ECF8C7@openstack.org> <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> <9A85D2917C58154C960D95352B22818BAB5AA132@fmsmsx117.amr.corp.intel.com> Message-ID: <05117357-5A12-42E2-926D-41DD25B19B43@openstack.org> Hi Bruce, Thanks for uploading the document for review. I added a few comments to it before we merge and iterate on it further: https://review.openstack.org/#/c/607406/ Thanks, Ildikó > On 2018. Oct 2., at 18:57, Jones, Bruce E wrote: > > The governance repo create request is waiting for approvals from the OpenStack infra team [0]. If it gets approved & created before I leave for vacation, I'll post the document and some updates to it to the StarlingX gerrit for review and approval, to address previous feedback, which may cause some changes to the timeline you have below. > > [0] https://review.openstack.org/#/c/607117/ > > brucej > > -----Original Message----- > From: Seiler, Glenn [mailto:glenn.seiler at windriver.com] > Sent: Monday, October 1, 2018 5:55 PM > To: Jones, Bruce E ; Claire Massey ; Thierry Carrez ; Ildiko Vancsa > Cc: starlingx-discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] StarlingX Initial Governance > > Has this been done yet (Governance doc in stx-governance repo)? > > The way I understand the timeline from the last call I attended, and from talking to various people is the following: > > August 2018-PTG Initial 4 people appointed > Q4 2018 (around launch and Berlin) 3 open seats appointed > 1H 2019 (time undetermined) 2 more open seats appointed > August 2019 4 initial seats up for voting > TBD- not clear.... Other seats up for voting > > I think it is necessary for this to be very clear with a timeline similar to, and as simple as, the above so that when we are doing our outreach to prospective new contributers we can be crystal clear about when they might expect to be appointed a chair, or voted into a chair. > > Regards, > -glenn > > > -----Original Message----- > From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] > Sent: Thursday, September 27, 2018 9:15 AM > To: Claire Massey; Thierry Carrez; Ildiko Vancsa > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] StarlingX Initial Governance > > Thank you for the feedback. We appreciate it. > > The way I'd like to suggest that we handle this is to get the governance document into source control. > > At the Documentation team meeting this week we agreed to create a stx-governance repo and place the document there. Then I will post changes to gerrit to address this feedback that can be reviewed and approved. > > Makes sense? > > @Ildiko, please let me know when the repo goes live. I'll be working on converting the current draft from wiki to .rst this week. > > brucej > > -----Original Message----- > From: Claire Massey [mailto:claire at openstack.org] > Sent: Thursday, September 27, 2018 8:21 AM > To: Thierry Carrez > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] StarlingX Initial Governance > > Hi all, > > I agree with Thierry and think that in order to encourage growth for the project, the first election period should to be adjusted. > > The next several months will be a critical time for StarlingX to welcome diverse engagement. From what we’ve experienced while launching other projects, like Kata Containers for example, new companies that are interested in engaging with the project for the first time are also very interested in having the opportunity to be elected to the technical leadership body within a reasonable amount of time. The opportunity to contribute to the project and possibly earn an elected seat early on in the life of the project can be very motivating for new companies that want to get involved. > > Based on the current StarlingX election timeline shown in the governance documentation[1], September 2019 is the soonest date when a new person would be able to be elected to the TSC. I personally think that waiting an entire year is just too long and may discourage new contributors from getting involved in the short-term. I recommend that instead of waiting a full year to hold the first election, that it's held sooner, in *May 2019*, which is eight months from now. The planned StarlingX release cadence and schedule[2] shows that the next two releases are timed to happen in March and July 2019, so hosting the first election between those two releases, in May 2019, could work well. > > [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance#Technical_Steering_Committee > > [2] https://wiki.openstack.org/wiki/StarlingX/Release_Plan#Release_Cadence_and_Schedule > > Thanks, > Claire > >> On Sep 27, 2018, at 10:13 AM, Thierry Carrez wrote: >> >> Hi everyone, >> >> I reread the Governance document[1] and found an inconsistency we'll likely want to address before the initial TSC formally blesses it: >> >> In "Technical Steering Committee": >>> It will be comprised of 7 members who will be appointed at Launch but fully elected by the Contributors within the first year >> >> then: >>> In April 2020, the remaining 4 seats are up for election >> >> "appointed at Launch" is a bit vague, I would use "initially appointed" (which would start the clock to when the TSC is fully formed). But more importantly, "fully elected by the Contributors within the first year" is at odds with still having appointed members around in March, 2020. The date or the period needs to be adjusted to match. >> >> We'll also likely want to clarify initial terms for the non-WR and non-Intel initial TSC members. One option is to appoint one of them with a short term, and the other two with long terms. Another option would be to randomly select the short one (like for the WR and Intel initially-appointed seats) >> >> Finally, I would suggest saying that if somone steps down from their TSC seat, their seat should automatically be up for election at the next round. So imagine someone that is initially appointed for a year, stepping down after 4 months: their seat would be up for reelection, which would make the first election determine 4 seats instead of just 3. That will facilitate keeping an active TSC with a limited number of stale members. >> >> That level of subtlety could be conveyed through a separate document that details the corner cases of the election process, if you find it too distracting for the Governance document. >> >> [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance >> >> -- >> Thierry Carrez (ttx) >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From dtroyer at gmail.com Wed Oct 3 13:58:31 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Wed, 3 Oct 2018 08:58:31 -0500 Subject: [Starlingx-discuss] StarlingX Initial Governance In-Reply-To: <05117357-5A12-42E2-926D-41DD25B19B43@openstack.org> References: <0309b193-6b70-00d3-9ac4-053eb4f28fd4@openstack.org> <2A0287E6-6C87-49D8-A896-D0D663ECF8C7@openstack.org> <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> <9A85D2917C58154C960D95352B22818BAB5AA132@fmsmsx117.amr.corp.intel.com> <05117357-5A12-42E2-926D-41DD25B19B43@openstack.org> Message-ID: On Wed, Oct 3, 2018 at 6:59 AM Ildiko Vancsa wrote: > I added a few comments to it before we merge and iterate on it further: https://review.openstack.org/#/c/607406/ I've posted https://review.openstack.org/#/c/607396/ to set up the structure of the repo and get basic jobs running so we can merge anything. Ildiko I think you are the only one who can merge things at this point... dt -- Dean Troyer dtroyer at gmail.com From sgw at linux.intel.com Wed Oct 3 14:19:19 2018 From: sgw at linux.intel.com (Saul Wold) Date: Wed, 3 Oct 2018 07:19:19 -0700 Subject: [Starlingx-discuss] build-pkgs cannot complete std build----aclocal: too many loops In-Reply-To: References: <2FD5DDB5A04D264C80D42CA35194914F2B38B7FE@SHSMSX104.ccr.corp.intel.com> <1af499f8-3fc8-9ce6-009e-9f17ceee9182@windriver.com> Message-ID: On 10/02/2018 10:23 AM, Saul Wold wrote: > > > On 10/02/2018 10:19 AM, Scott Little wrote: >> Merged on master.  Lets everyone give it a test build.  It's never >> been better than 50/50 to hit us, so we want as many test builds as >> possible to be sure this fixes it. >> > > Thanks Scott, I will start up a new build on master. > Just a follow-up, my build of master dud work and I see that Scott has already merged this change to the 2018.10 branch. Thanks Sau! > Sau! > >> Scott >> >> >> On 18-10-02 09:29 AM, Saul Wold wrote: >>> >>> Folks, >>> >>> What's the status of the review of these PRs for stx-ceph, this is >>> still causing build breakage in the 2018.10 branch. >>> >>> I think this will require an update to the 2018.10 manifest also >>> >>> Sau! >>> >>> >>> On 09/29/2018 06:47 AM, Sun, Austin wrote: >>>> Hi Cindy and All: >>>>      I have generated two PR, one is for master and another is for >>>> r/2018.10.  please help check if the PR process is correct . >>>> >>>> -[0] https://github.com/starlingx-staging/stx-ceph/pull/2 >>>> -[1] https://github.com/starlingx-staging/stx-ceph/pull/3 >>>> >>>> >>>> Thanks. >>>> BR >>>> Austin Sun. >>>> >>>> >>>> -----Original Message----- >>>> From: Xie, Cindy >>>> Sent: Saturday, September 29, 2018 6:03 PM >>>> To: Sun, Austin ; Saul Wold >>>> ; Scott Little ; >>>> starlingx-discuss at lists.starlingx.io >>>> Subject: RE: [Starlingx-discuss] build-pkgs cannot complete std >>>> build----aclocal: too many loops >>>> >>>> Austin, >>>> Thanks for the finding - can we generate pull-requests for >>>> ax_require_define.m4 against StarlingX-staging? >>>> >>>> Thx. - cindy >>>> >>>> -----Original Message----- >>>> From: Sun, Austin [mailto:austin.sun at intel.com] >>>> Sent: Saturday, September 29, 2018 4:41 PM >>>> To: Saul Wold ; Scott Little >>>> ; starlingx-discuss at lists.starlingx.io >>>> Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std >>>> build----aclocal: too many loops >>>> >>>> Hi Saul , Scott and Erich: >>>> >>>>      I met same issue in my local env,  I did some study aclocal >>>> script and some test. >>>>              According your analysis , >>>> autoconf-archive-2017.03.21-1.el7.noarch.rpm was added , which >>>> included a lot of system m4 files used by aclocal,  according [0] >>>> and [1],  '--install' option will cp system macros (.m4) to local m4 >>>> folder , >>>> >>>>  From error log: >>>> >>>> BUILDSTDERR: aclocal: installing >>>> /usr/share/aclocal/ax_require_defined.m4 to m4/ax_require_defined.m4 >>>> BUILDSTDERR: aclocal: installing 'm4/ax_require_defined.m4' from >>>> '/usr/share/aclocal/ax_require_defined.m4' >>>> BUILDSTDERR: aclocal: running: cp >>>> /usr/share/aclocal/ax_require_defined.m4 m4/ax_require_defined.m4 >>>> BUILDSTDERR: aclocal: running aclocal anew, because some files were >>>> installed locally >>>> BUILDSTDERR: aclocal: error: too many loops >>>> >>>> ax_require_defined.m4 cause this issue, If  copied >>>> ax_require_defined.m4 to code base stx/git/ceph/m4/,  ceph can be >>>> built successfully . >>>> >>>> since cannot upload change to starlingx-staging , please use [3] >>>> ax_require_defined.m4 file which is same as file in >>>> autoconf-archive-2017.03.21-1.el7.noarch.rpm package >>>> >>>> - About  why ax_require_defined.m4 is a must for ceph build: >>>>      ax_cxx_compile_stdcxx_11.m4 is upgraded from local serial 4 to >>>> system serial 18 including in autoconf-archive rpm , which is using >>>> AX_REQUIRE_DEFINED defined in ax_require_defined.m4 >>>>       have tried force upgrade local m4/ax_cxx_compile_stdcxx_11.m4 >>>> to serial 19 ,  it can solve this issue too. >>>> >>>> -[0] >>>> https://www.gnu.org/software/automake/manual/html_node/Serials.html#Serials >>>> >>>> -[1] >>>> https://www.gnu.org/software/automake/manual/html_node/aclocal-Invocation.html#aclocal-Invocation >>>> >>>> -[2] >>>> http://git.savannah.gnu.org/gitweb/?p=autoconf-archive.git;a=blob_plain;f=m4/ax_require_defined.m4 >>>> >>>> >>>> Thanks. >>>> BR >>>> Austin Sun. >>>> >>>> -----Original Message----- >>>> From: Saul Wold [mailto:sgw at linux.intel.com] >>>> Sent: Saturday, September 29, 2018 8:20 AM >>>> To: Scott Little ; >>>> starlingx-discuss at lists.starlingx.io >>>> Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std build >>>> >>>> >>>> >>>> On 09/28/2018 01:39 PM, Scott Little wrote: >>>>> Ok, we've seen 3 ceph failures in our last 6 builds. >>>>> >>>>> The common factor:  tpm2-tools builds on 'b0' before ceph builds. >>>>> >>>>> Our theory.  The buildRequires of tpm2-tools causes autoconf-archive >>>>> to be installed... which installs a bunch of .m4 files in >>>>> /usr/share/aclocal ... which causes ceph grief when it calls aclocal. >>>>> >>>>> I don't really know automake or aclocal all that well.  I'm assuming >>>>> /usr/share/aclocal is acting something like a cache, but it's a cache >>>>> whos contents are incompatible with ceph. >>>>> >>>>> Do we have any autotools / aclocal / m4 experts in the house? >>>>> >>>>> Possible fixes: >>>>> - ceph: can we tell it to not use the aclocal cache... explicitly (a >>>>> flag to aclocal?)  ... or implicitly (update ceph's m4 files so they >>>>> look 'newer' than the cache)? >>>> >>>> Not sure about that, I would have to dig deeper into aclocal, it's >>>> been a while since I dug into that. >>>> >>>>> - tpm2-tools: Can we remove the dependence on autoconf-archive? No >>>>> other package we build seems to need it. >>>>> >>>> A quick scan show that the autoconf-archive was put in there for >>>> travis support, and goes away this past March upstream when they >>>> coverted to using a container for travis.  If we could use a newer >>>> version of tpm2 that might solve this. >>>> >>>> Maybe Erich's solution can work >>>> >>>> Sau! >>>> >>>> >>>>> Scott >>>>> >>>>> >>>>> >>>>> On 18-09-27 04:45 PM, Saul Wold wrote: >>>>>> >>>>>> And of course it worked the third time! >>>>>> >>>>>> So, I lost the good logs. >>>>>> >>>>>> Sau! >>>>>> >>>>>> >>>>>> On 09/27/2018 12:56 PM, Scott Little wrote: >>>>>>> On 18-09-27 03:53 PM, Scott Little wrote: >>>>>>>> Our latest build, based on code synced at 2018-09-27T15:28:00  UTC, >>>>>>>> built successfully. >>>>>>>> >>>>>>>> It took three attempts to get ceph built.  The first two passes >>>>>>>> aborted quickly due to missing packages.  The final pass did not >>>>>>>> exhibit the 'aclocal: too many loops'**issue. >>>>>>>> >>>>>>>> The only build I have that exhibited the too many loops error was a >>>>>>>> snapshot on 2018-09-20T15:50:40 UTC >>>>>>>> >>>>>>>> I do have a designer with an older snapshot that seems to hit it >>>>>>>> regularly, so I'll work with him and see if we can learn more. >>>>>>>> >>>>>>>> I think we need more data from the community >>>>>>>> - Who's build is failing on ceph with *aclocal: too many loops?* >>>>>>>> - Who is building successfully ? >>>>>>>> - Who can build only intermittently? >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> Info to collect for failed builds: >>>>>>> - repo sync timestamp >>>>>>>> - build command used? >>>>>>>> - Was it a new workspace, a cleaned workspace, or a previously used >>>>>>>> workspace? >>>>>>> - $MY_WORKSPACE/CONTEXT >>>>>>>> - $MY_WORKSPACE/build-std.log >>>>>>>> - $MY_WORKSPACE/std/results/*/ceph-*/*.log >>>>>>>> >>>>>>>> For successful builds, same info. Rather than full build logs, I >>>>>>>> can settle for: >>>>>>>> - grep '\(Success building\|iteration\|building ceph\)' >>>>>>>> $MY_WORKSPACE/build-std.log >>>>>>>> - grep compute_resources: build-std.log >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> On 18-09-27 02:21 PM, Saul Wold wrote: >>>>>>>>> On 09/26/2018 09:16 AM, Scott Little wrote: >>>>>>>>>> aclocal 'too many loops' has been popping up sporadically for a >>>>>>>>>> week or two now.  Possibly 7.5 related. >>>>>>>>>> >>>>>>>>>> I suspect that there is a build order and/or race condition >>>>>>>>>> element to this.   It often goes away if you just run build-pkgs >>>>>>>>>> a second time. >>>>>>>>>> >>>>>>>>> I am seeing this failure also, but it does not go away after a >>>>>>>>> second rebuild.  I have the lastest stx-root (build-tools) with >>>>>>>>> the recent patches. >>>>>>>>> >>>>>>>>> Is this directly related to the fuzz issue or is there something >>>>>>>>> else we need to address in CEPH itself. >>>>>>>>> >>>>>>>>> This is blocking my local build. >>>>>>>>> >>>>>>>>> Sau! >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> >>>>>>>> _______________________________________________ >>>>>>>> Starlingx-discuss mailing list >>>>>>>> Starlingx-discuss at lists.starlingx.io >>>>>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discus >>>>>>>> s >>>>>>> >>>>>>> >>>>>>> >>>>>>> >>>>>>> _______________________________________________ >>>>>>> Starlingx-discuss mailing list >>>>>>> Starlingx-discuss at lists.starlingx.io >>>>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>>>>>> >>>>> >>>> >>>> _______________________________________________ >>>> Starlingx-discuss mailing list >>>> Starlingx-discuss at lists.starlingx.io >>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>>> _______________________________________________ >>>> Starlingx-discuss mailing list >>>> Starlingx-discuss at lists.starlingx.io >>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >>>> >> > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From bruce.e.jones at intel.com Wed Oct 3 14:36:19 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Wed, 3 Oct 2018 14:36:19 +0000 Subject: [Starlingx-discuss] Notes from the project call Oct 3rd 2018 Message-ID: <9A85D2917C58154C960D95352B22818BAB5AAB0B@fmsmsx117.amr.corp.intel.com> Agenda and Notes for the Oct 3rd call * Release build and testing update o WR found some "download corruption" (1795739) and didn't get a successful build yesterday. The version number issue found by Intel and Wind River has been fixed in 1795540 on master and the release branch. WR will run sanity soonest. o Sanity tests are passing on master in WR. o Fix from Austin for Ceph builds 1795155 was merged into master and just merged into the release branch. o Saul had a successful build on master o Test teams to send out daily reports please to the email list! o We will push tags onto the branches as we get deeper into the test process and will tag the final release candidate o We reviewed release bug status ? https://bugs.launchpad.net/starlingx/+bugs?field.tag=stx.2018.10 o Documentation is covered for now, there are a few minor things to address. Docs team meeting today. docs.starlingx.io site has been created, but is empty. Infra team working on finding and fixing the issue. Should have the sphinx docs published soon. Release notes and API refs to follow. * PSA Reminder: Please join Freenode IRC Channel #starlingx -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtroyer at gmail.com Wed Oct 3 14:46:07 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Wed, 3 Oct 2018 09:46:07 -0500 Subject: [Starlingx-discuss] Notes from the project call Oct 3rd 2018 In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5AAB0B@fmsmsx117.amr.corp.intel.com> References: <9A85D2917C58154C960D95352B22818BAB5AAB0B@fmsmsx117.amr.corp.intel.com> Message-ID: On Wed, Oct 3, 2018 at 9:38 AM Jones, Bruce E wrote: > o Documentation is covered for now, there are a few minor things to address. Docs team meeting today. docs.starlingx.io site has been created, but is empty. Infra team working on finding and fixing the issue. Should have the sphinx docs published soon. Release notes and API refs to follow. One thing I didn't mention is that you can always look at a docs-related job output in the Gerrit review screen. On the right side of the page is a list of the jobs that have run with this review, clicking on a docs job name will display the html output of that job. If the job failed it will show you the log files like other jobs. dt -- Dean Troyer dtroyer at gmail.com From cesar.lara at intel.com Wed Oct 3 15:07:28 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 3 Oct 2018 15:07:28 +0000 Subject: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! Message-ID: <0B566C62EC792145B40E29EFEBF1AB4710527C90@fmsmsx104.amr.corp.intel.com> Hi team According to this wiki https://wiki.openstack.org/wiki/Starlingx/Meetings I will assume that what's not taken is available as far as time slots, so is there any preference on time and day? Last time we were planning this meeting Scott and Jason asked me not to use Tuesday. So my proposal will be Thursday's after TSC meeting 10am CST /11am EST/ 8am PST Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From sgw at linux.intel.com Wed Oct 3 16:44:17 2018 From: sgw at linux.intel.com (Saul Wold) Date: Wed, 3 Oct 2018 09:44:17 -0700 Subject: [Starlingx-discuss] What's the preferred patch directory structure Message-ID: Brent, I have noticed that we have a number of different patch directory layouts with in the stx-* repos. In some cases there are files/ or / directory at the top level such as sudo (files) or cluster-resource-agents (), some have files/ and patches/ at the top level such as net-snmp. Others have patches/ and/or files/ under the centos directory. Additional some of these directory contain patches with the same name. I am starting to do some clean-up to reduce some of the duplicate patches and want to look at making the directories more consistent also. Maybe this has been discussed and I lost it in the mix. Is there a spec / document showing what the correct layout should be? Thanks Sau! From scott.little at windriver.com Wed Oct 3 17:31:18 2018 From: scott.little at windriver.com (Scott Little) Date: Wed, 3 Oct 2018 13:31:18 -0400 Subject: [Starlingx-discuss] What's the preferred patch directory structure In-Reply-To: References: Message-ID: <3702ced5-5500-6fd3-81e7-bac765b051a3@windriver.com> The original thinking was .... 1) Patching a centos src.rpm a//centos/patches/ a//centos/meta-patches/ a//centos/files/ a//centos/srpm_path a//centos/build_srpm.data 2) Patching a downloaded tarball, we supply the packaging. b/downloads/ a//patches/ a//files/ a//centos/.spec a//centos/build_srpm.data Note: build_srpm.data provides SRC_DIR=$STX_BASE/downloads/ Note: This assumed the patches/files we add are distro agnostic. If our added content is centos specific, move patches/files under centos. 3) packaging our own content (small packages) a//    or possibly    ...//src/ a//centos/.spec a//centos/build_srpm.data Note: build_srpm.data provides SRC_DIR=$PKG_BASE/git/ 4a) packaging our own content, large package sored in a separate git. 4b) packaging 3rd party git b/git/ a//centos/.spec a//centos/build_srpm.data Note: build_srpm.data provides SRC_DIR=$STX_BASE/git/ Scott On 18-10-03 12:44 PM, Saul Wold wrote: > > Brent, > > I have noticed that we have a number of different patch directory > layouts with in the stx-* repos. > > In some cases there are files/ or / directory at the top > level such as sudo (files) or cluster-resource-agents (), > some have files/ and patches/ at the top level such as net-snmp.  > Others have patches/ and/or files/ under the centos directory. > Additional some of these directory contain patches with the same name. > > I am starting to do some clean-up to reduce some of the duplicate > patches and want to look at making the directories more consistent also. > > Maybe this has been discussed and I lost it in the mix. > Is there a spec / document showing what the correct layout should be? > > Thanks >    Sau! > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From bruce.e.jones at intel.com Wed Oct 3 17:36:47 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Wed, 3 Oct 2018 17:36:47 +0000 Subject: [Starlingx-discuss] StarlingX Initial Governance In-Reply-To: <05117357-5A12-42E2-926D-41DD25B19B43@openstack.org> References: <0309b193-6b70-00d3-9ac4-053eb4f28fd4@openstack.org> <2A0287E6-6C87-49D8-A896-D0D663ECF8C7@openstack.org> <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> <9A85D2917C58154C960D95352B22818BAB5AA132@fmsmsx117.amr.corp.intel.com> <05117357-5A12-42E2-926D-41DD25B19B43@openstack.org> Message-ID: <9A85D2917C58154C960D95352B22818BAB5AACF6@fmsmsx117.amr.corp.intel.com> Ildiko, thank you. I have uploaded a new version that addresses your comments. And I hand off whatever happens next to you. Please help as needed to get these changes merged and thank you! brucej -----Original Message----- From: Ildiko Vancsa [mailto:ildiko at openstack.org] Sent: Wednesday, October 3, 2018 4:58 AM To: Jones, Bruce E Cc: Seiler, Glenn ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] StarlingX Initial Governance Hi Bruce, Thanks for uploading the document for review. I added a few comments to it before we merge and iterate on it further: https://review.openstack.org/#/c/607406/ Thanks, Ildikó > On 2018. Oct 2., at 18:57, Jones, Bruce E wrote: > > The governance repo create request is waiting for approvals from the OpenStack infra team [0]. If it gets approved & created before I leave for vacation, I'll post the document and some updates to it to the StarlingX gerrit for review and approval, to address previous feedback, which may cause some changes to the timeline you have below. > > [0] https://review.openstack.org/#/c/607117/ > > brucej > > -----Original Message----- > From: Seiler, Glenn [mailto:glenn.seiler at windriver.com] > Sent: Monday, October 1, 2018 5:55 PM > To: Jones, Bruce E ; Claire Massey ; Thierry Carrez ; Ildiko Vancsa > Cc: starlingx-discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] StarlingX Initial Governance > > Has this been done yet (Governance doc in stx-governance repo)? > > The way I understand the timeline from the last call I attended, and from talking to various people is the following: > > August 2018-PTG Initial 4 people appointed > Q4 2018 (around launch and Berlin) 3 open seats appointed > 1H 2019 (time undetermined) 2 more open seats appointed > August 2019 4 initial seats up for voting > TBD- not clear.... Other seats up for voting > > I think it is necessary for this to be very clear with a timeline similar to, and as simple as, the above so that when we are doing our outreach to prospective new contributers we can be crystal clear about when they might expect to be appointed a chair, or voted into a chair. > > Regards, > -glenn > > > -----Original Message----- > From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] > Sent: Thursday, September 27, 2018 9:15 AM > To: Claire Massey; Thierry Carrez; Ildiko Vancsa > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] StarlingX Initial Governance > > Thank you for the feedback. We appreciate it. > > The way I'd like to suggest that we handle this is to get the governance document into source control. > > At the Documentation team meeting this week we agreed to create a stx-governance repo and place the document there. Then I will post changes to gerrit to address this feedback that can be reviewed and approved. > > Makes sense? > > @Ildiko, please let me know when the repo goes live. I'll be working on converting the current draft from wiki to .rst this week. > > brucej > > -----Original Message----- > From: Claire Massey [mailto:claire at openstack.org] > Sent: Thursday, September 27, 2018 8:21 AM > To: Thierry Carrez > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] StarlingX Initial Governance > > Hi all, > > I agree with Thierry and think that in order to encourage growth for the project, the first election period should to be adjusted. > > The next several months will be a critical time for StarlingX to welcome diverse engagement. From what we’ve experienced while launching other projects, like Kata Containers for example, new companies that are interested in engaging with the project for the first time are also very interested in having the opportunity to be elected to the technical leadership body within a reasonable amount of time. The opportunity to contribute to the project and possibly earn an elected seat early on in the life of the project can be very motivating for new companies that want to get involved. > > Based on the current StarlingX election timeline shown in the governance documentation[1], September 2019 is the soonest date when a new person would be able to be elected to the TSC. I personally think that waiting an entire year is just too long and may discourage new contributors from getting involved in the short-term. I recommend that instead of waiting a full year to hold the first election, that it's held sooner, in *May 2019*, which is eight months from now. The planned StarlingX release cadence and schedule[2] shows that the next two releases are timed to happen in March and July 2019, so hosting the first election between those two releases, in May 2019, could work well. > > [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance#Technical_Steering_Committee > > [2] https://wiki.openstack.org/wiki/StarlingX/Release_Plan#Release_Cadence_and_Schedule > > Thanks, > Claire > >> On Sep 27, 2018, at 10:13 AM, Thierry Carrez wrote: >> >> Hi everyone, >> >> I reread the Governance document[1] and found an inconsistency we'll likely want to address before the initial TSC formally blesses it: >> >> In "Technical Steering Committee": >>> It will be comprised of 7 members who will be appointed at Launch but fully elected by the Contributors within the first year >> >> then: >>> In April 2020, the remaining 4 seats are up for election >> >> "appointed at Launch" is a bit vague, I would use "initially appointed" (which would start the clock to when the TSC is fully formed). But more importantly, "fully elected by the Contributors within the first year" is at odds with still having appointed members around in March, 2020. The date or the period needs to be adjusted to match. >> >> We'll also likely want to clarify initial terms for the non-WR and non-Intel initial TSC members. One option is to appoint one of them with a short term, and the other two with long terms. Another option would be to randomly select the short one (like for the WR and Intel initially-appointed seats) >> >> Finally, I would suggest saying that if somone steps down from their TSC seat, their seat should automatically be up for election at the next round. So imagine someone that is initially appointed for a year, stepping down after 4 months: their seat would be up for reelection, which would make the first election determine 4 seats instead of just 3. That will facilitate keeping an active TSC with a limited number of stale members. >> >> That level of subtlety could be conveyed through a separate document that details the corner cases of the election process, if you find it too distracting for the Governance document. >> >> [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance >> >> -- >> Thierry Carrez (ttx) >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ildiko at openstack.org Wed Oct 3 18:01:53 2018 From: ildiko at openstack.org (Ildiko Vancsa) Date: Wed, 3 Oct 2018 20:01:53 +0200 Subject: [Starlingx-discuss] StarlingX Initial Governance In-Reply-To: <9A85D2917C58154C960D95352B22818BAB5AACF6@fmsmsx117.amr.corp.intel.com> References: <0309b193-6b70-00d3-9ac4-053eb4f28fd4@openstack.org> <2A0287E6-6C87-49D8-A896-D0D663ECF8C7@openstack.org> <9A85D2917C58154C960D95352B22818BAB5A725A@fmsmsx117.amr.corp.intel.com> <9A85D2917C58154C960D95352B22818BAB5AA132@fmsmsx117.amr.corp.intel.com> <05117357-5A12-42E2-926D-41DD25B19B43@openstack.org> <9A85D2917C58154C960D95352B22818BAB5AACF6@fmsmsx117.amr.corp.intel.com> Message-ID: Hi Bruce, Thank you for the quick turnaround on it, it looks great overall. I’m working with Dean now to get the repo ready for this and additional documents in the future. I will look into the folder structure and make sure that it lands where it’s supposed before merging the patch. I will take care of the above, thank you for getting the ball rolling. Enjoy your vacation! :) Thanks, Ildikó > On 2018. Oct 3., at 19:36, Jones, Bruce E wrote: > > Ildiko, thank you. I have uploaded a new version that addresses your comments. > > And I hand off whatever happens next to you. Please help as needed to get these changes merged and thank you! > > brucej > > -----Original Message----- > From: Ildiko Vancsa [mailto:ildiko at openstack.org] > Sent: Wednesday, October 3, 2018 4:58 AM > To: Jones, Bruce E > Cc: Seiler, Glenn ; starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] StarlingX Initial Governance > > Hi Bruce, > > Thanks for uploading the document for review. > > I added a few comments to it before we merge and iterate on it further: https://review.openstack.org/#/c/607406/ > > Thanks, > Ildikó > > >> On 2018. Oct 2., at 18:57, Jones, Bruce E wrote: >> >> The governance repo create request is waiting for approvals from the OpenStack infra team [0]. If it gets approved & created before I leave for vacation, I'll post the document and some updates to it to the StarlingX gerrit for review and approval, to address previous feedback, which may cause some changes to the timeline you have below. >> >> [0] https://review.openstack.org/#/c/607117/ >> >> brucej >> >> -----Original Message----- >> From: Seiler, Glenn [mailto:glenn.seiler at windriver.com] >> Sent: Monday, October 1, 2018 5:55 PM >> To: Jones, Bruce E ; Claire Massey ; Thierry Carrez ; Ildiko Vancsa >> Cc: starlingx-discuss at lists.starlingx.io >> Subject: RE: [Starlingx-discuss] StarlingX Initial Governance >> >> Has this been done yet (Governance doc in stx-governance repo)? >> >> The way I understand the timeline from the last call I attended, and from talking to various people is the following: >> >> August 2018-PTG Initial 4 people appointed >> Q4 2018 (around launch and Berlin) 3 open seats appointed >> 1H 2019 (time undetermined) 2 more open seats appointed >> August 2019 4 initial seats up for voting >> TBD- not clear.... Other seats up for voting >> >> I think it is necessary for this to be very clear with a timeline similar to, and as simple as, the above so that when we are doing our outreach to prospective new contributers we can be crystal clear about when they might expect to be appointed a chair, or voted into a chair. >> >> Regards, >> -glenn >> >> >> -----Original Message----- >> From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] >> Sent: Thursday, September 27, 2018 9:15 AM >> To: Claire Massey; Thierry Carrez; Ildiko Vancsa >> Cc: starlingx-discuss at lists.starlingx.io >> Subject: Re: [Starlingx-discuss] StarlingX Initial Governance >> >> Thank you for the feedback. We appreciate it. >> >> The way I'd like to suggest that we handle this is to get the governance document into source control. >> >> At the Documentation team meeting this week we agreed to create a stx-governance repo and place the document there. Then I will post changes to gerrit to address this feedback that can be reviewed and approved. >> >> Makes sense? >> >> @Ildiko, please let me know when the repo goes live. I'll be working on converting the current draft from wiki to .rst this week. >> >> brucej >> >> -----Original Message----- >> From: Claire Massey [mailto:claire at openstack.org] >> Sent: Thursday, September 27, 2018 8:21 AM >> To: Thierry Carrez >> Cc: starlingx-discuss at lists.starlingx.io >> Subject: Re: [Starlingx-discuss] StarlingX Initial Governance >> >> Hi all, >> >> I agree with Thierry and think that in order to encourage growth for the project, the first election period should to be adjusted. >> >> The next several months will be a critical time for StarlingX to welcome diverse engagement. From what we’ve experienced while launching other projects, like Kata Containers for example, new companies that are interested in engaging with the project for the first time are also very interested in having the opportunity to be elected to the technical leadership body within a reasonable amount of time. The opportunity to contribute to the project and possibly earn an elected seat early on in the life of the project can be very motivating for new companies that want to get involved. >> >> Based on the current StarlingX election timeline shown in the governance documentation[1], September 2019 is the soonest date when a new person would be able to be elected to the TSC. I personally think that waiting an entire year is just too long and may discourage new contributors from getting involved in the short-term. I recommend that instead of waiting a full year to hold the first election, that it's held sooner, in *May 2019*, which is eight months from now. The planned StarlingX release cadence and schedule[2] shows that the next two releases are timed to happen in March and July 2019, so hosting the first election between those two releases, in May 2019, could work well. >> >> [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance#Technical_Steering_Committee >> >> [2] https://wiki.openstack.org/wiki/StarlingX/Release_Plan#Release_Cadence_and_Schedule >> >> Thanks, >> Claire >> >>> On Sep 27, 2018, at 10:13 AM, Thierry Carrez wrote: >>> >>> Hi everyone, >>> >>> I reread the Governance document[1] and found an inconsistency we'll likely want to address before the initial TSC formally blesses it: >>> >>> In "Technical Steering Committee": >>>> It will be comprised of 7 members who will be appointed at Launch but fully elected by the Contributors within the first year >>> >>> then: >>>> In April 2020, the remaining 4 seats are up for election >>> >>> "appointed at Launch" is a bit vague, I would use "initially appointed" (which would start the clock to when the TSC is fully formed). But more importantly, "fully elected by the Contributors within the first year" is at odds with still having appointed members around in March, 2020. The date or the period needs to be adjusted to match. >>> >>> We'll also likely want to clarify initial terms for the non-WR and non-Intel initial TSC members. One option is to appoint one of them with a short term, and the other two with long terms. Another option would be to randomly select the short one (like for the WR and Intel initially-appointed seats) >>> >>> Finally, I would suggest saying that if somone steps down from their TSC seat, their seat should automatically be up for election at the next round. So imagine someone that is initially appointed for a year, stepping down after 4 months: their seat would be up for reelection, which would make the first election determine 4 seats instead of just 3. That will facilitate keeping an active TSC with a limited number of stale members. >>> >>> That level of subtlety could be conveyed through a separate document that details the corner cases of the election process, if you find it too distracting for the Governance document. >>> >>> [1] https://wiki.openstack.org/wiki/Starlingx/Initial_Governance >>> >>> -- >>> Thierry Carrez (ttx) >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From michael.l.tullis at intel.com Wed Oct 3 20:52:25 2018 From: michael.l.tullis at intel.com (Tullis, Michael L) Date: Wed, 3 Oct 2018 20:52:25 +0000 Subject: [Starlingx-discuss] [docs][meetings] docs team meeting minutes Message-ID: <3808363B39586544A6839C76CF81445EA1A3BB4C@FMSMSX151.amr.corp.intel.com> For notes and new action items from our meeting today, see our docs team etherpad: https://etherpad.openstack.org/p/stx-documentation Thx. -- Mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From sgw at linux.intel.com Wed Oct 3 21:12:08 2018 From: sgw at linux.intel.com (Saul Wold) Date: Wed, 3 Oct 2018 14:12:08 -0700 Subject: [Starlingx-discuss] What's the preferred patch directory structure In-Reply-To: <3702ced5-5500-6fd3-81e7-bac765b051a3@windriver.com> References: <3702ced5-5500-6fd3-81e7-bac765b051a3@windriver.com> Message-ID: <3ad6a503-e522-6773-9dd3-67e0771e6637@linux.intel.com> Ok, this is a good start, currently, I am not sure all the directories match this pattern, and I know that's OK for now. I would like to suggest some tweaks to this given we are thinking about supporting multi OS, there maybe "files" (configuration, inittialization, ...) and possibly patches that are common to mutliple OSes. Would you please propose a Spec (in stx-specs) for consideration, so we can better document this structure in Spec form, thanks. I think as we work on updating patches, we move toward a common pattern. On 10/03/2018 10:31 AM, Scott Little wrote: > The original thinking was .... > > 1) Patching a centos src.rpm > > a//centos/patches/ > a//centos/meta-patches/ > a//centos/files/ > a//centos/srpm_path > a//centos/build_srpm.data > For Ubuntu: a//debian/patches a//debian/source (See: https://www.debian.org/doc/manuals/debmake-doc/ch04.en.html) a//patches a//files I think this pattern more or less applies to the other examples you list below. Sau! > 2) Patching a downloaded tarball, we supply the packaging. > > b/downloads/ > a//patches/ > a//files/ > a//centos/.spec > a//centos/build_srpm.data > > Note: build_srpm.data provides SRC_DIR=$STX_BASE/downloads/ > Note: This assumed the patches/files we add are distro agnostic. If our > added content is centos specific, move patches/files under centos. > > 3) packaging our own content (small packages) > > a//    or possibly    ...//src/ > a//centos/.spec > a//centos/build_srpm.data > > Note: build_srpm.data provides SRC_DIR=$PKG_BASE/git/ > > 4a) packaging our own content, large package sored in a separate git. > 4b) packaging 3rd party git > > b/git/ > a//centos/.spec > a//centos/build_srpm.data > > Note: build_srpm.data provides SRC_DIR=$STX_BASE/git/ > > > > Scott > > > On 18-10-03 12:44 PM, Saul Wold wrote: >> >> Brent, >> >> I have noticed that we have a number of different patch directory >> layouts with in the stx-* repos. >> >> In some cases there are files/ or / directory at the top >> level such as sudo (files) or cluster-resource-agents (), >> some have files/ and patches/ at the top level such as net-snmp. >> Others have patches/ and/or files/ under the centos directory. >> Additional some of these directory contain patches with the same name. >> >> I am starting to do some clean-up to reduce some of the duplicate >> patches and want to look at making the directories more consistent also. >> >> Maybe this has been discussed and I lost it in the mix. >> Is there a spec / document showing what the correct layout should be? >> >> Thanks >>    Sau! >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Thu Oct 4 12:03:00 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Thu, 4 Oct 2018 12:03:00 +0000 Subject: [Starlingx-discuss] Validation Status Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E1951@fmsmsx101.amr.corp.intel.com> Hi All We have a successfully built an ISO for branch 2018.10 at Intel infrastructure, validation team started to setup the environments for virtual and bare-metal, test execution will start as soon as the those ones are up and running. Regards, José From Numan.Waheed at windriver.com Thu Oct 4 13:27:04 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Thu, 4 Oct 2018 13:27:04 +0000 Subject: [Starlingx-discuss] Sanity Results Updated on Test Status Page Message-ID: <3CAA827B7A79BA46B15B280EC82088FE481FF4F2@ALA-MBD.corp.ad.wrs.com> Hi, Test Status Page has been updated with sanity results from last night execution. https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_WR_TestStatus#Sanity_Results Regards, [WR logo signiture] Numan Waheed, Manager, Titanium Cloud Testing, Wind River Direct: 613.270.5762 * Mobile: 613.793.5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.jpg Type: image/jpeg Size: 1419 bytes Desc: image003.jpg URL: From scott.little at windriver.com Thu Oct 4 14:03:13 2018 From: scott.little at windriver.com (Scott Little) Date: Thu, 4 Oct 2018 10:03:13 -0400 Subject: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! In-Reply-To: <0B566C62EC792145B40E29EFEBF1AB4710527C90@fmsmsx104.amr.corp.intel.com> References: <0B566C62EC792145B40E29EFEBF1AB4710527C90@fmsmsx104.amr.corp.intel.com> Message-ID: I have two non-repeating meetings in that slot this morning.  Probably ok going forward. Scott On 18-10-03 11:07 AM, Lara, Cesar wrote: > > Hi team > > According to this wiki > > https://wiki.openstack.org/wiki/Starlingx/Meetings > > I will assume that what’s not taken is available as far as time slots, > so is there any preference on time and day? Last time we were planning > this meeting Scott and Jason asked me not to use Tuesday. > > So my proposal will be Thursday’s after TSC meeting 10am CST /11am > EST/ 8am PST > > Regards > > *//* > > */Cesar Lara/* > > Software Engineering Manager > > OpenSource Technology Center > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ken.Young at windriver.com Thu Oct 4 14:19:14 2018 From: Ken.Young at windriver.com (Young, Ken) Date: Thu, 4 Oct 2018 14:19:14 +0000 Subject: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! In-Reply-To: References: <0B566C62EC792145B40E29EFEBF1AB4710527C90@fmsmsx104.amr.corp.intel.com> Message-ID: <9967B6CE-79E8-4AAB-BEB8-F3E517B86E8B@windriver.com> Cesar, +1 from me as well. Perhaps we can have our first meeting here on Oct 11th? Please note that Oct 8th is Canadian Thanksgiving so Scoot, Jason and I will miss the meeting on Monday. Regards, Ken Y From: Scott Little Date: Thursday, October 4, 2018 at 10:05 AM To: "starlingx-discuss at lists.starlingx.io" Subject: Re: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! I have two non-repeating meetings in that slot this morning. Probably ok going forward. Scott On 18-10-03 11:07 AM, Lara, Cesar wrote: Hi team According to this wiki https://wiki.openstack.org/wiki/Starlingx/Meetings I will assume that what’s not taken is available as far as time slots, so is there any preference on time and day? Last time we were planning this meeting Scott and Jason asked me not to use Tuesday. So my proposal will be Thursday’s after TSC meeting 10am CST /11am EST/ 8am PST Regards Cesar Lara Software Engineering Manager OpenSource Technology Center _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Thu Oct 4 14:52:42 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Thu, 4 Oct 2018 14:52:42 +0000 Subject: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! In-Reply-To: <9967B6CE-79E8-4AAB-BEB8-F3E517B86E8B@windriver.com> References: <0B566C62EC792145B40E29EFEBF1AB4710527C90@fmsmsx104.amr.corp.intel.com> , <9967B6CE-79E8-4AAB-BEB8-F3E517B86E8B@windriver.com> Message-ID: Yes, the idea is to start this meeting up on October 11th, so I'll set that up Regards Cesar Lara Sent from my mobile phone ________________________________ From: "Young, Ken" Sent: Thursday, October 4, 2018 9:20 AM To: Little, Scott; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! Cesar, +1 from me as well. Perhaps we can have our first meeting here on Oct 11th? Please note that Oct 8th is Canadian Thanksgiving so Scoot, Jason and I will miss the meeting on Monday. Regards, Ken Y From: Scott Little Date: Thursday, October 4, 2018 at 10:05 AM To: "starlingx-discuss at lists.starlingx.io" Subject: Re: [Starlingx-discuss] [Build][Meetings] lets pick a new slot for our team meeting! I have two non-repeating meetings in that slot this morning. Probably ok going forward. Scott On 18-10-03 11:07 AM, Lara, Cesar wrote: Hi team According to this wiki https://wiki.openstack.org/wiki/Starlingx/Meetings I will assume that what’s not taken is available as far as time slots, so is there any preference on time and day? Last time we were planning this meeting Scott and Jason asked me not to use Tuesday. So my proposal will be Thursday’s after TSC meeting 10am CST /11am EST/ 8am PST Regards Cesar Lara Software Engineering Manager OpenSource Technology Center _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From jose.perez.carranza at intel.com Thu Oct 4 21:18:24 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Thu, 4 Oct 2018 21:18:24 +0000 Subject: [Starlingx-discuss] Duplex behavior question Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E3CC5@fmsmsx101.amr.corp.intel.com> Hi Folks What is the expected behavior of a "duplex - two node redundant configuration" when the controller-0 is powered-off (physically powerd off)? current behavior is that Starlingx stops working cannot access the horizon and cannot execute commands on controller-1 due is not active. Is this one an expected behavior ? Regards, José From Brent.Rowsell at windriver.com Thu Oct 4 21:23:55 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Thu, 4 Oct 2018 21:23:55 +0000 Subject: [Starlingx-discuss] Duplex behavior question In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E3CC5@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E3CC5@fmsmsx101.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB297288@ALA-MBD.corp.ad.wrs.com> Assuming the other controller was unlocked/enabled at the time then no. Please open a Launchpad and attach the logs Brent -----Original Message----- From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] Sent: Thursday, October 4, 2018 5:18 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Duplex behavior question Hi Folks What is the expected behavior of a "duplex - two node redundant configuration" when the controller-0 is powered-off (physically powerd off)? current behavior is that Starlingx stops working cannot access the horizon and cannot execute commands on controller-1 due is not active. Is this one an expected behavior ? Regards, José _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From sgw at linux.intel.com Thu Oct 4 22:01:57 2018 From: sgw at linux.intel.com (Saul Wold) Date: Thu, 4 Oct 2018 15:01:57 -0700 Subject: [Starlingx-discuss] DHCP Related Patches Message-ID: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> HI Brent, There are currently 6 DHCP related patches specifically to the client code. We have already determined that 2 of those patches the NSUPDATE protection could be removed, 2 seem to be StarlingX spceific but replaced with a enter-hook scripts provided in another patch. 1 is a backport dhclient-disable-NSUPDATE.patch support-disable-nsupdate.patch - Remove these 2 dhclient-handle-wrs-install-uuid.patch dhclient-dhcp6-wrs-install-uuid.patch - Can these be removed as the functionality is now added via the dhclient-enter-hooks script? dhclient-ipv6-bind-to-interface.patch - This is a backport, which will be available when CentOS updates to the newer version. dhclient-ipv6-conditionally-set-hostname.patch - Can this be moved to the dhclient-enter-hooks? dhclient-restrict-interfaces-to-command-line.patch - This seems to be adding functionality, in order to restrict the dhclient to only be active on one interface. There is no commit message for this one, or is there a way to refactor this? Is this a bug in dhclient? Was it ever filled or commented on in the DHCP mailing list? Thanks Sau! From Don.Penney at windriver.com Thu Oct 4 23:51:41 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Thu, 4 Oct 2018 23:51:41 +0000 Subject: [Starlingx-discuss] DHCP Related Patches In-Reply-To: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> References: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B9F4E@ALA-MBD.corp.ad.wrs.com> Here's the original commit message for dhclient-restrict-interfaces-to-command-line.patch. There was an issue with leases expiring, causing interfaces to lose their IP addresses, which in turn caused various issues due to the loss of comms: Author: Allain Legacy Date: Tue Jan 5 14:36:47 2016 -0500 dhclient: restrict interfaces to those on command line only By default, the dhclient process does not respect the list of interfaces supplied at the command line. It configures any interfaces found to be specified in the config file. Since we customize options for each interface in our config file and run a separate dhclient process for each interface we end up with multiple dhclient processes that each service all interfaces. This is undesirable because it is possible that a request is sent by process A but received by process B. This leads to lease expiry events even though a valid request packet was returned by the server. This change introduces a "--restrict-interfaces" option to the dhclient process to force it to ignore all interfaces in config files other than those specified at the command line. To activate this change our busybox version of ifup/ifdown has been modified to pass the "--restrict-interfaces" to dhclient as well as to request that each process use its own lease file to avoid file corruption. -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: Thursday, October 04, 2018 6:02 PM To: Rowsell, Brent; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] DHCP Related Patches HI Brent, There are currently 6 DHCP related patches specifically to the client code. We have already determined that 2 of those patches the NSUPDATE protection could be removed, 2 seem to be StarlingX spceific but replaced with a enter-hook scripts provided in another patch. 1 is a backport dhclient-disable-NSUPDATE.patch support-disable-nsupdate.patch - Remove these 2 dhclient-handle-wrs-install-uuid.patch dhclient-dhcp6-wrs-install-uuid.patch - Can these be removed as the functionality is now added via the dhclient-enter-hooks script? dhclient-ipv6-bind-to-interface.patch - This is a backport, which will be available when CentOS updates to the newer version. dhclient-ipv6-conditionally-set-hostname.patch - Can this be moved to the dhclient-enter-hooks? dhclient-restrict-interfaces-to-command-line.patch - This seems to be adding functionality, in order to restrict the dhclient to only be active on one interface. There is no commit message for this one, or is there a way to refactor this? Is this a bug in dhclient? Was it ever filled or commented on in the DHCP mailing list? Thanks Sau! _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From sgw at linux.intel.com Fri Oct 5 00:40:13 2018 From: sgw at linux.intel.com (Saul Wold) Date: Thu, 4 Oct 2018 17:40:13 -0700 Subject: [Starlingx-discuss] DHCP Related Patches In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B9F4E@ALA-MBD.corp.ad.wrs.com> References: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B9F4E@ALA-MBD.corp.ad.wrs.com> Message-ID: <6f68f38d-1838-bf99-7722-0cb0c1fec1af@linux.intel.com> On 10/04/2018 04:51 PM, Penney, Don wrote: > Here's the original commit message for dhclient-restrict-interfaces-to-command-line.patch. There was an issue with leases expiring, causing interfaces to lose their IP addresses, which in turn caused various issues due to the loss of comms: > Thanks Don, Was this reported anywhere? I am trying to ensure I follow any trails that have already been started. Thanks Sau! > Author: Allain Legacy > Date: Tue Jan 5 14:36:47 2016 -0500 > > dhclient: restrict interfaces to those on command line only > > By default, the dhclient process does not respect the list of interfaces > supplied at the command line. It configures any interfaces found to be > specified in the config file. Since we customize options for each interface in > our config file and run a separate dhclient process for each interface we end > up with multiple dhclient processes that each service all interfaces. This is > undesirable because it is possible that a request is sent by process A but > received by process B. This leads to lease expiry events even though a valid > request packet was returned by the server. > > This change introduces a "--restrict-interfaces" option to the dhclient process > to force it to ignore all interfaces in config files other than those specified > at the command line. > > To activate this change our busybox version of ifup/ifdown has been modified to > pass the "--restrict-interfaces" to dhclient as well as to request that each > process use its own lease file to avoid file corruption. > > > > -----Original Message----- > From: Saul Wold [mailto:sgw at linux.intel.com] > Sent: Thursday, October 04, 2018 6:02 PM > To: Rowsell, Brent; starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] DHCP Related Patches > > > HI Brent, > > There are currently 6 DHCP related patches specifically to the client code. > > We have already determined that 2 of those patches the NSUPDATE > protection could be removed, 2 seem to be StarlingX spceific but > replaced with a enter-hook scripts provided in another patch. 1 is a > backport > > > dhclient-disable-NSUPDATE.patch > support-disable-nsupdate.patch > - Remove these 2 > dhclient-handle-wrs-install-uuid.patch > dhclient-dhcp6-wrs-install-uuid.patch > - Can these be removed as the functionality is now added via the > dhclient-enter-hooks script? > > dhclient-ipv6-bind-to-interface.patch > - This is a backport, which will be available when CentOS updates to > the newer version. > > dhclient-ipv6-conditionally-set-hostname.patch > - Can this be moved to the dhclient-enter-hooks? > > dhclient-restrict-interfaces-to-command-line.patch > - This seems to be adding functionality, in order to restrict the > dhclient to only be active on one interface. There is no commit message > for this one, or is there a way to refactor this? Is this a bug in > dhclient? Was it ever filled or commented on in the DHCP mailing list? > > > Thanks > Sau! > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From sgw at linux.intel.com Fri Oct 5 00:42:01 2018 From: sgw at linux.intel.com (Saul Wold) Date: Thu, 4 Oct 2018 17:42:01 -0700 Subject: [Starlingx-discuss] DHCP Related Patches In-Reply-To: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> References: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> Message-ID: On 10/04/2018 03:01 PM, Saul Wold wrote: > > HI Brent, > > There are currently 6 DHCP related patches specifically to the client code. > > We have already determined that 2 of those patches the NSUPDATE > protection could be removed, 2 seem to be StarlingX spceific but > replaced with a enter-hook scripts provided in another patch. 1 is a > backport > > > dhclient-disable-NSUPDATE.patch > support-disable-nsupdate.patch >  - Remove these 2 > dhclient-handle-wrs-install-uuid.patch > dhclient-dhcp6-wrs-install-uuid.patch >  - Can these be removed as the functionality is now added via the > dhclient-enter-hooks script? > > dhclient-ipv6-bind-to-interface.patch >  - This is a backport, which will be available when CentOS updates to > the newer version. > MY bad, I was looking at the wrong code base, this is still a valid patch and seems to move where the bind to interface happens from after the bind to before the bind, again if this is a bug, was it reported anywhere to Redhat or Upstream on the DHCP list/bug system? Sau! > dhclient-ipv6-conditionally-set-hostname.patch >  - Can this be moved to the dhclient-enter-hooks? > > dhclient-restrict-interfaces-to-command-line.patch >  - This seems to be adding functionality, in order to restrict the > dhclient to only be active on one interface. There is no commit message > for this one, or is there a way to refactor this?  Is this a bug in > dhclient?  Was it ever filled or commented on in the DHCP mailing list? > > > Thanks >    Sau! > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ada.cabrales at intel.com Fri Oct 5 05:40:55 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Fri, 5 Oct 2018 05:40:55 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-04-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E716FC8C8@FMSMSX108.amr.corp.intel.com> Sanity completed successfully on stx-2018-10-03-11-r-2018.10.iso Duplex bare metal, duplex virtual, multinode with local storage, multinode with external storage (Ceph) virtual - installed and configured. Test plan execution began. Simplex bare metal and virtual, multinode local storage bare metal and virtual, multinode with external storage bare metal configuration WIP. On Duplex virtual, we experienced a problem where the nova-compute service on controller-0 was not enabled. After a while, the service enabled by itself. From jose.perez.carranza at intel.com Fri Oct 5 12:46:23 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Fri, 5 Oct 2018 12:46:23 +0000 Subject: [Starlingx-discuss] Duplex behavior question In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB297288@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E3CC5@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB297288@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E3D53@fmsmsx101.amr.corp.intel.com> Hi Brent My bad, I did not put the controller-1 as unlocked, I updated my steps and added that as a prerequisite. It is working now controller-1 is taking the StarlingX deployment up and running and I can access to Horizon and CLI without problems. Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Thursday, October 4, 2018 4:24 PM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Duplex behavior question > > Assuming the other controller was unlocked/enabled at the time then no. > Please open a Launchpad and attach the logs > > Brent > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Thursday, October 4, 2018 5:18 PM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Duplex behavior question > > Hi Folks > > What is the expected behavior of a "duplex - two node redundant > configuration" when the controller-0 is powered-off (physically powerd off)? > current behavior is that Starlingx stops working cannot access the horizon and > cannot execute commands on controller-1 due is not active. Is this one an > expected behavior ? > > Regards, > José > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Numan.Waheed at windriver.com Fri Oct 5 14:38:32 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Fri, 5 Oct 2018 14:38:32 +0000 Subject: [Starlingx-discuss] Sanity Results Updated on Test Status Page In-Reply-To: <3CAA827B7A79BA46B15B280EC82088FE481FF4F2@ALA-MBD.corp.ad.wrs.com> References: <3CAA827B7A79BA46B15B280EC82088FE481FF4F2@ALA-MBD.corp.ad.wrs.com> Message-ID: <3CAA827B7A79BA46B15B280EC82088FE481FFDC1@ALA-MBD.corp.ad.wrs.com> Sanity and Manual Regression results are updated on following page: https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_WR_TestStatus#Sanity_Results Regards, Numan. From: Waheed, Numan [mailto:Numan.Waheed at windriver.com] Sent: October-04-18 9:27 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Sanity Results Updated on Test Status Page Hi, Test Status Page has been updated with sanity results from last night execution. https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_WR_TestStatus#Sanity_Results Regards, [WR logo signiture] Numan Waheed, Manager, Titanium Cloud Testing, Wind River Direct: 613.270.5762 * Mobile: 613.793.5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.jpg Type: image/jpeg Size: 1419 bytes Desc: image001.jpg URL: From Numan.Waheed at windriver.com Fri Oct 5 19:08:00 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Fri, 5 Oct 2018 19:08:00 +0000 Subject: [Starlingx-discuss] Sanity Results Updated on Test Status Page In-Reply-To: <3CAA827B7A79BA46B15B280EC82088FE481FFDC1@ALA-MBD.corp.ad.wrs.com> References: <3CAA827B7A79BA46B15B280EC82088FE481FF4F2@ALA-MBD.corp.ad.wrs.com> <3CAA827B7A79BA46B15B280EC82088FE481FFDC1@ALA-MBD.corp.ad.wrs.com> Message-ID: <3CAA827B7A79BA46B15B280EC82088FE481FFF1A@ALA-MBD.corp.ad.wrs.com> Updated Results for stx.2018.10 release: Sanity: Sanity Results Config All-in-One Simplex Build ID 2018-10-04_01-52-00 Total 11 Pass 11 Fail 0 Bugs None Config All-in-One Duplex Build ID 2018-10-04_01-52-00 Total 15 Pass 15 Fail 0 Bugs None Config Multi Node Build ID 2018-10-03_09-47-10 Total 23 Pass 23 Fail 0 Bugs None Config Multi Node with CEPH Build ID 2018-10-04_01-52-00 Total 24 Pass 24 Fail 0 Bugs None Manual Regression: Domains Total Test Cases Passed Failed Total Executed Launchpad Backup and Restore 5 0 0 0 Distributed Cloud 30 8 0 8 Installation and Config 5 0 0 0 MTC 20 16 0 16 Networking 29 4 3 7 https://bugs.launchpad.net/starlingx/+bug/1796198 https://bugs.launchpad.net/starlingx/+bug/1790968 Nova 35 3 1 4 Security 17 0 0 0 Storage 11 1 2 3 https://bugs.launchpad.net/starlingx/+bug/1795962 System Inventory 8 6 0 6 System Test 11 2 0 2 Total 171 40 6 46 From: Waheed, Numan [mailto:Numan.Waheed at windriver.com] Sent: October-05-18 10:39 AM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Sanity Results Updated on Test Status Page Sanity and Manual Regression results are updated on following page: https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_WR_TestStatus#Sanity_Results Regards, Numan. From: Waheed, Numan [mailto:Numan.Waheed at windriver.com] Sent: October-04-18 9:27 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Sanity Results Updated on Test Status Page Hi, Test Status Page has been updated with sanity results from last night execution. https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_WR_TestStatus#Sanity_Results Regards, [WR logo signiture] Numan Waheed, Manager, Titanium Cloud Testing, Wind River Direct: 613.270.5762 * Mobile: 613.793.5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.jpg Type: image/jpeg Size: 1419 bytes Desc: image001.jpg URL: From sgw at linux.intel.com Fri Oct 5 21:40:10 2018 From: sgw at linux.intel.com (Saul Wold) Date: Fri, 5 Oct 2018 14:40:10 -0700 Subject: [Starlingx-discuss] DHCP Related Patches In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3B9F4E@ALA-MBD.corp.ad.wrs.com> References: <7edfe2b2-4b12-4dc2-7e4a-75e70fbda57f@linux.intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3B9F4E@ALA-MBD.corp.ad.wrs.com> Message-ID: On 10/04/2018 04:51 PM, Penney, Don wrote: > Here's the original commit message for dhclient-restrict-interfaces-to-command-line.patch. There was an issue with leases expiring, causing interfaces to lose their IP addresses, which in turn caused various issues due to the loss of comms: > > Author: Allain Legacy > Date: Tue Jan 5 14:36:47 2016 -0500 > > dhclient: restrict interfaces to those on command line only > > By default, the dhclient process does not respect the list of interfaces > supplied at the command line. It configures any interfaces found to be > specified in the config file. Since we customize options for each interface in > our config file and run a separate dhclient process for each interface we end > up with multiple dhclient processes that each service all interfaces. This is > undesirable because it is possible that a request is sent by process A but > received by process B. This leads to lease expiry events even though a valid > request packet was returned by the server. > So, a further question is why is the default behavior of having 1 process managing all the interfaces a bad thing? What problem were you trying to solve? Again, the WHY behind the original change not just that dhclient has a undesirable behavior. Why is there a dhclient/interface? > This change introduces a "--restrict-interfaces" option to the dhclient process > to force it to ignore all interfaces in config files other than those specified > at the command line. > > To activate this change our busybox version of ifup/ifdown has been modified to > pass the "--restrict-interfaces" to dhclient as well as to request that each > process use its own lease file to avoid file corruption. > I think this dates to the OE version, as Centos does not have busybox, but the changes are in initscripts. Thanks Sau! > > > -----Original Message----- > From: Saul Wold [mailto:sgw at linux.intel.com] > Sent: Thursday, October 04, 2018 6:02 PM > To: Rowsell, Brent; starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] DHCP Related Patches > > > HI Brent, > > There are currently 6 DHCP related patches specifically to the client code. > > We have already determined that 2 of those patches the NSUPDATE > protection could be removed, 2 seem to be StarlingX spceific but > replaced with a enter-hook scripts provided in another patch. 1 is a > backport > > > dhclient-disable-NSUPDATE.patch > support-disable-nsupdate.patch > - Remove these 2 > dhclient-handle-wrs-install-uuid.patch > dhclient-dhcp6-wrs-install-uuid.patch > - Can these be removed as the functionality is now added via the > dhclient-enter-hooks script? > > dhclient-ipv6-bind-to-interface.patch > - This is a backport, which will be available when CentOS updates to > the newer version. > > dhclient-ipv6-conditionally-set-hostname.patch > - Can this be moved to the dhclient-enter-hooks? > > dhclient-restrict-interfaces-to-command-line.patch > - This seems to be adding functionality, in order to restrict the > dhclient to only be active on one interface. There is no commit message > for this one, or is there a way to refactor this? Is this a bug in > dhclient? Was it ever filled or commented on in the DHCP mailing list? > > > Thanks > Sau! > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From sgw at linux.intel.com Fri Oct 5 23:22:06 2018 From: sgw at linux.intel.com (Saul Wold) Date: Fri, 5 Oct 2018 16:22:06 -0700 Subject: [Starlingx-discuss] dnsmasq patch for DHCP6 config Message-ID: <5f266bb8-cdd6-9680-a6ac-3a40c7d16175@linux.intel.com> Scott, Brent: You have a patch against dnsmasq that adds some code for handling IPv6 addrs in lease_update_from_configs(). Can you provide any more background, was this reported upstream? What exactly is this fixing? stx-integ/base/dnsmasq/centos/patches/dnsmasq-update-ipv6-leases-from-config.patch Thanks Sau! From ada.cabrales at intel.com Sat Oct 6 01:45:46 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Sat, 6 Oct 2018 01:45:46 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-05-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E716FCE35@FMSMSX108.amr.corp.intel.com> Sanity completed successfully on stx-2018-10-05-13-r-2018.10.iso - 31 TCs PASS / 0 TCs FAIL Simplex Bare metal: nova-compute service down. Launchpad - https://bugs.launchpad.net/starlingx/+bug/1796420 Simplex Virtual: 3/69 tests Passed Duplex Bare metal: 5/96 tests Passed Duplex Virtual: 14/101 tests Passed, 6 Failed - debug WIP Multinode local Storage Bare metal: 11/113 tests Passed Multinode local storage virtual: 6/109 tests Passed Multinode external Storage Bare metal: 3/116 tests Passed Multinode external storage virtual: 3/120 tests Passed Some updates proposed for the documentation of setting Multinode with external storage on virtual configuration. We are experiencing problems with our network switch configuration - nodes losing management network, causing reboots. Asked for help to Wind River - meeting scheduled for Mon 10/8 From chenjie.xu at intel.com Mon Oct 8 03:16:57 2018 From: chenjie.xu at intel.com (Xu, Chenjie) Date: Mon, 8 Oct 2018 03:16:57 +0000 Subject: [Starlingx-discuss] Analysis of patch e547a7e & ebd41fd for StartlingX upstreaming In-Reply-To: References: Message-ID: Hi Matt, Thank you for your comment! Sorry for the late reply! Last week is our national holiday. Best Regards, Xu, Chenjie From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 1, 2018 9:12 PM To: Xu, Chenjie Cc: starlingx-discuss at lists.starlingx.io; Zhao, Forrest ; Guo, Ruijing Subject: RE: [Starlingx-discuss] Analysis of patch e547a7e & ebd41fd for StartlingX upstreaming Hi Chenjie, I agree that this is still an issue upstream that needs to be fixed. Regards, Matt From: Xu, Chenjie [mailto:chenjie.xu at intel.com] Sent: Sunday, September 30, 2018 11:07 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Zhao, Forrest; Guo, Ruijing Subject: [Starlingx-discuss] Analysis of patch e547a7e & ebd41fd for StartlingX upstreaming Hi Matt, For patches e547a7e and ebd41fd, I think it's a bug in Neutron upstream and will submit the patch to fix it in the next step. This bug is caused by passing the wrong parameter after create or update router. You can find the analysis report within attached file. Could you please help review and comment! Best Regards, Xu, Chenjie -------------- next part -------------- An HTML attachment was scrubbed... URL: From austin.sun at intel.com Mon Oct 8 08:54:22 2018 From: austin.sun at intel.com (Sun, Austin) Date: Mon, 8 Oct 2018 08:54:22 +0000 Subject: [Starlingx-discuss] removing "make parallelize" meta patch Message-ID: Hi All: There are several meta patches to enable make in parallel , do we have any concern to remove it ? [0] ./base/openssh/centos/meta_patches/0001-Further-parallelize-openssh-build.patch [1] ./base/shadow-utils/centos/meta_patches/0001-Further-parallelize-shadow-utils-build.patch [2] ./base/sudo/centos/meta_patches/0003-Further-parallelize-sudo-build.patch Thanks. Austin Sun. From dtroyer at gmail.com Mon Oct 8 14:20:39 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 8 Oct 2018 09:20:39 -0500 Subject: [Starlingx-discuss] Retiring repos Message-ID: We have some repos in Gerrit that are no longer used (stx-gplv2, stx-gplv3, stx-utils) that would be good to get into a final disposition. I suggest we do not just remove them because they do contain history. The OpenStack practice[0] is to push one final commit into the repo that removes the content and leaves a note in README.rst explaining why it is empty. How does everyone feel about this? In this case the history is not long but it is the initial open sourced code so I would like to not remove the repos. I do think we should push an EOL tag to the last HEAD before removing the content (I thought OpenStack did this but I don't see it in their docs). dt [0] https://docs.openstack.org/infra/manual/drivers.html#retiring-a-project (note that parts of this process do not have equivalents in StarlingX, we are looking at only steps 2 and 3) -- Dean Troyer dtroyer at gmail.com From Brent.Rowsell at windriver.com Mon Oct 8 14:49:22 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Mon, 8 Oct 2018 14:49:22 +0000 Subject: [Starlingx-discuss] removing "make parallelize" meta patch In-Reply-To: References: Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A3B2F@ALA-MBD.corp.ad.wrs.com> Austin, No objections on removing these. Brent -----Original Message----- From: Sun, Austin [mailto:austin.sun at intel.com] Sent: Monday, October 8, 2018 4:54 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] removing "make parallelize" meta patch Hi All: There are several meta patches to enable make in parallel , do we have any concern to remove it ? [0] ./base/openssh/centos/meta_patches/0001-Further-parallelize-openssh-build.patch [1] ./base/shadow-utils/centos/meta_patches/0001-Further-parallelize-shadow-utils-build.patch [2] ./base/sudo/centos/meta_patches/0003-Further-parallelize-sudo-build.patch Thanks. Austin Sun. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Mon Oct 8 14:51:06 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Mon, 8 Oct 2018 14:51:06 +0000 Subject: [Starlingx-discuss] Retiring repos In-Reply-To: References: Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A3B46@ALA-MBD.corp.ad.wrs.com> Sounds good to me. Brent -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Monday, October 8, 2018 10:21 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Retiring repos We have some repos in Gerrit that are no longer used (stx-gplv2, stx-gplv3, stx-utils) that would be good to get into a final disposition. I suggest we do not just remove them because they do contain history. The OpenStack practice[0] is to push one final commit into the repo that removes the content and leaves a note in README.rst explaining why it is empty. How does everyone feel about this? In this case the history is not long but it is the initial open sourced code so I would like to not remove the repos. I do think we should push an EOL tag to the last HEAD before removing the content (I thought OpenStack did this but I don't see it in their docs). dt [0] https://docs.openstack.org/infra/manual/drivers.html#retiring-a-project (note that parts of this process do not have equivalents in StarlingX, we are looking at only steps 2 and 3) -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Mon Oct 8 15:26:45 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Mon, 8 Oct 2018 15:26:45 +0000 Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E427E@fmsmsx101.amr.corp.intel.com> Hi all I have a test scenario where requires a controller locked/disabled to be powered-off by command ' system host-power-off ' , currently on Duplex (redundant) configuration that is rejected with below message, is that normal behavior? Else I will open a Launchpad. ======= " Operation Rejected: Request rejected due to provisioning semantic check. . Please verify that the board management MAC or IP address being used is formatted correctly or not already provisioned against another host. " Regards, José From Brent.Rowsell at windriver.com Mon Oct 8 15:32:14 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Mon, 8 Oct 2018 15:32:14 +0000 Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E427E@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E427E@fmsmsx101.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A3BF3@ALA-MBD.corp.ad.wrs.com> Is the BMC configured ? What is the output of system host-show ? Brent -----Original Message----- From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] Sent: Monday, October 8, 2018 11:27 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario Hi all I have a test scenario where requires a controller locked/disabled to be powered-off by command ' system host-power-off ' , currently on Duplex (redundant) configuration that is rejected with below message, is that normal behavior? Else I will open a Launchpad. ======= " Operation Rejected: Request rejected due to provisioning semantic check. . Please verify that the board management MAC or IP address being used is formatted correctly or not already provisioned against another host. " Regards, José _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Mon Oct 8 15:37:38 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Mon, 8 Oct 2018 15:37:38 +0000 Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2A3BF3@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E427E@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2A3BF3@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E42B4@fmsmsx101.amr.corp.intel.com> Hi Brent Here is the output http://paste.openstack.org/show/731693/ Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Monday, October 8, 2018 10:32 AM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: Doubt on Duplex configuration scenario > > Is the BMC configured ? > What is the output of system host-show ? > > Brent > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Monday, October 8, 2018 11:27 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario > > Hi all > > I have a test scenario where requires a controller locked/disabled to be > powered-off by command ' system host-power-off ' , > currently on Duplex (redundant) configuration that is rejected with below > message, is that normal behavior? Else I will open a Launchpad. > > ======= > " Operation Rejected: Request rejected due to provisioning semantic check. . > Please verify that the board management MAC or IP address being used is > formatted correctly or not already provisioned against another host. " > > Regards, > José > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From erich.cordoba.malibran at intel.com Mon Oct 8 15:51:48 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Mon, 8 Oct 2018 15:51:48 +0000 Subject: [Starlingx-discuss] What's the process for new requirements? Message-ID: Hi, I would like to start to propose new features for the next release and I'm wondering what would be the lifecycle of feature requests. I see a stx-specs repository with some templates and examples and I can start from there, but once I have a written proposal what would be the next steps? Does the proposal should be discussed in mailing list or it will be in the gerrit change? What's the criteria to be accepted? Thanks! -Erich From Brent.Rowsell at windriver.com Mon Oct 8 16:01:58 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Mon, 8 Oct 2018 16:01:58 +0000 Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E42B4@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E427E@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2A3BF3@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E42B4@fmsmsx101.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A3E4E@ALA-MBD.corp.ad.wrs.com> Jose, The output show that there is no BMC configured on this server. | bm_ip | None | | bm_type | None | | bm_username | None The BMC needs to be on a network that is reachable from the OAM network and provisioned in system inventory in order for this test case to work. To provision the BMC: system host-update $HOST bm_password=${BM_PW} bm_username=${BM_UN} bm_ip=${BM_IP} bm_type=bmc Brent -----Original Message----- From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] Sent: Monday, October 8, 2018 11:38 AM To: Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: Doubt on Duplex configuration scenario Hi Brent Here is the output http://paste.openstack.org/show/731693/ Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Monday, October 8, 2018 10:32 AM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: Doubt on Duplex configuration scenario > > Is the BMC configured ? > What is the output of system host-show ? > > Brent > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Monday, October 8, 2018 11:27 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario > > Hi all > > I have a test scenario where requires a controller locked/disabled to > be powered-off by command ' system host-power-off name/id> ' , currently on Duplex (redundant) configuration that is > rejected with below message, is that normal behavior? Else I will open a Launchpad. > > ======= > " Operation Rejected: Request rejected due to provisioning semantic check. . > Please verify that the board management MAC or IP address being used > is formatted correctly or not already provisioned against another host. " > > Regards, > José > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Mon Oct 8 16:03:49 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Mon, 8 Oct 2018 16:03:49 +0000 Subject: [Starlingx-discuss] What's the process for new requirements? In-Reply-To: References: Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A3EA2@ALA-MBD.corp.ad.wrs.com> Erich, Please refer to the feature process on the wiki [0] [0] https://wiki.openstack.org/wiki/StarlingX/Feature_Development_Process Brent -----Original Message----- From: Cordoba Malibran, Erich [mailto:erich.cordoba.malibran at intel.com] Sent: Monday, October 8, 2018 11:52 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the process for new requirements? Hi, I would like to start to propose new features for the next release and I'm wondering what would be the lifecycle of feature requests. I see a stx-specs repository with some templates and examples and I can start from there, but once I have a written proposal what would be the next steps? Does the proposal should be discussed in mailing list or it will be in the gerrit change? What's the criteria to be accepted? Thanks! -Erich _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Mon Oct 8 16:06:09 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Mon, 8 Oct 2018 16:06:09 +0000 Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2A3E4E@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E427E@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2A3BF3@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E42B4@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2A3E4E@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E4404@fmsmsx101.amr.corp.intel.com> Thanks for your help Brent !! Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Monday, October 8, 2018 11:02 AM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: Doubt on Duplex configuration scenario > > Jose, > > The output show that there is no BMC configured on this server. > > | bm_ip | None | > | bm_type | None | > | bm_username | None > > The BMC needs to be on a network that is reachable from the OAM network > and provisioned in system inventory in order for this test case to work. > > To provision the BMC: > > system host-update $HOST bm_password=${BM_PW} > bm_username=${BM_UN} bm_ip=${BM_IP} bm_type=bmc > > > Brent > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Monday, October 8, 2018 11:38 AM > To: Rowsell, Brent ; starlingx- > discuss at lists.starlingx.io > Subject: RE: Doubt on Duplex configuration scenario > > Hi Brent > > Here is the output > > http://paste.openstack.org/show/731693/ > > Regards, > José > > > > -----Original Message----- > > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > > Sent: Monday, October 8, 2018 10:32 AM > > To: Perez Carranza, Jose ; starlingx- > > discuss at lists.starlingx.io > > Subject: RE: Doubt on Duplex configuration scenario > > > > Is the BMC configured ? > > What is the output of system host-show ? > > > > Brent > > > > -----Original Message----- > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > Sent: Monday, October 8, 2018 11:27 AM > > To: starlingx-discuss at lists.starlingx.io > > Subject: [Starlingx-discuss] Doubt on Duplex configuration scenario > > > > Hi all > > > > I have a test scenario where requires a controller locked/disabled to > > be powered-off by command ' system host-power-off > name/id> ' , currently on Duplex (redundant) configuration that is > > rejected with below message, is that normal behavior? Else I will open a > Launchpad. > > > > ======= > > " Operation Rejected: Request rejected due to provisioning semantic check. . > > Please verify that the board management MAC or IP address being used > > is formatted correctly or not already provisioned against another host. " > > > > Regards, > > José > > > > > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From sgw at linux.intel.com Mon Oct 8 16:13:44 2018 From: sgw at linux.intel.com (Saul Wold) Date: Mon, 8 Oct 2018 09:13:44 -0700 Subject: [Starlingx-discuss] removing "make parallelize" meta patch In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2A3B2F@ALA-MBD.corp.ad.wrs.com> References: <2588653EBDFFA34B982FAF00F1B4844EBB2A3B2F@ALA-MBD.corp.ad.wrs.com> Message-ID: <538bfb60-1e68-05bd-2499-a0d4a027360a@linux.intel.com> Austin, Yes please, I created a storyboard for the removal of those patches here: https://storyboard.openstack.org/#!/story/2003803 Thanks Sau! On 10/08/2018 07:49 AM, Rowsell, Brent wrote: > Austin, > > No objections on removing these. > > Brent > > -----Original Message----- > From: Sun, Austin [mailto:austin.sun at intel.com] > Sent: Monday, October 8, 2018 4:54 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] removing "make parallelize" meta patch > > Hi All: > There are several meta patches to enable make in parallel , do we have any concern to remove it ? > > [0] ./base/openssh/centos/meta_patches/0001-Further-parallelize-openssh-build.patch > [1] ./base/shadow-utils/centos/meta_patches/0001-Further-parallelize-shadow-utils-build.patch > [2] ./base/sudo/centos/meta_patches/0003-Further-parallelize-sudo-build.patch > > Thanks. > Austin Sun. > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From cesar.lara at intel.com Mon Oct 8 16:40:25 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Mon, 8 Oct 2018 16:40:25 +0000 Subject: [Starlingx-discuss] What's the process for new requirements? In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2A3EA2@ALA-MBD.corp.ad.wrs.com> References: <2588653EBDFFA34B982FAF00F1B4844EBB2A3EA2@ALA-MBD.corp.ad.wrs.com> Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052B2E1@fmsmsx104.amr.corp.intel.com> Somewhat in that tenure, do we ever finalized on the template for specs? Regards Cesar Lara -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Monday, October 8, 2018 11:04 AM To: Cordoba Malibran, Erich ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] What's the process for new requirements? Erich, Please refer to the feature process on the wiki [0] [0] https://wiki.openstack.org/wiki/StarlingX/Feature_Development_Process Brent -----Original Message----- From: Cordoba Malibran, Erich [mailto:erich.cordoba.malibran at intel.com] Sent: Monday, October 8, 2018 11:52 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the process for new requirements? Hi, I would like to start to propose new features for the next release and I'm wondering what would be the lifecycle of feature requests. I see a stx-specs repository with some templates and examples and I can start from there, but once I have a written proposal what would be the next steps? Does the proposal should be discussed in mailing list or it will be in the gerrit change? What's the criteria to be accepted? Thanks! -Erich _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Mon Oct 8 16:46:43 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Mon, 8 Oct 2018 16:46:43 +0000 Subject: [Starlingx-discuss] What's the process for new requirements? In-Reply-To: <0B566C62EC792145B40E29EFEBF1AB471052B2E1@fmsmsx104.amr.corp.intel.com> References: <2588653EBDFFA34B982FAF00F1B4844EBB2A3EA2@ALA-MBD.corp.ad.wrs.com> <0B566C62EC792145B40E29EFEBF1AB471052B2E1@fmsmsx104.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A3F9A@ALA-MBD.corp.ad.wrs.com> Yes, the spec template has been approved. https://git.openstack.org/cgit/openstack/stx-specs/commit/?id=476398fd15b2a0553dfe92e6a77b63bede3ac45a Brent -----Original Message----- From: Lara, Cesar [mailto:cesar.lara at intel.com] Sent: Monday, October 8, 2018 12:40 PM To: Rowsell, Brent ; Cordoba Malibran, Erich ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the process for new requirements? Somewhat in that tenure, do we ever finalized on the template for specs? Regards Cesar Lara -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Monday, October 8, 2018 11:04 AM To: Cordoba Malibran, Erich ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] What's the process for new requirements? Erich, Please refer to the feature process on the wiki [0] [0] https://wiki.openstack.org/wiki/StarlingX/Feature_Development_Process Brent -----Original Message----- From: Cordoba Malibran, Erich [mailto:erich.cordoba.malibran at intel.com] Sent: Monday, October 8, 2018 11:52 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the process for new requirements? Hi, I would like to start to propose new features for the next release and I'm wondering what would be the lifecycle of feature requests. I see a stx-specs repository with some templates and examples and I can start from there, but once I have a written proposal what would be the next steps? Does the proposal should be discussed in mailing list or it will be in the gerrit change? What's the criteria to be accepted? Thanks! -Erich _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From juan.carlos.alonso at intel.com Mon Oct 8 17:27:45 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Mon, 8 Oct 2018 17:27:45 +0000 Subject: [Starlingx-discuss] Image name is not showed in a Instance Message-ID: <8557B550001AFB46A43A0CCC314BF85144245F8D@FMSMSX108.amr.corp.intel.com> Hi, When a volume is created and then an Instance is launched, the field for "image" is empty in CLI and "Not Found" in Horizon. Is this behavior correct? After create a Volume and launch an Instance I created a Flavor, Image and Network. To create a Volume I used: $ openstack volume create --size 20 --type iscsi --image cirros --availability-zone nova volume-cirros To launch an Instance: $ openstack server create --volume volume-cirros --flavor m1.small --nic net-id=5886578b-1868-4c12-9742-9ae147b4daba vm-cirros-1 [wrsroot at controller-0 ~(keystone_admin)]$ openstack server show vm-cirros-1 | grep -w "image" | image The name of the image used to create the Volume to launch the Instances should be showed? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From juan.carlos.alonso at intel.com Mon Oct 8 17:30:09 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Mon, 8 Oct 2018 17:30:09 +0000 Subject: [Starlingx-discuss] Image name is not showed in a Instance In-Reply-To: <8557B550001AFB46A43A0CCC314BF85144245F8D@FMSMSX108.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF85144245F8D@FMSMSX108.amr.corp.intel.com> Message-ID: <8557B550001AFB46A43A0CCC314BF85144245FAF@FMSMSX108.amr.corp.intel.com> Fixed a mistake.. "Before" instead of "After" From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Monday, October 8, 2018 12:28 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Image name is not showed in a Instance Hi, When a volume is created and then an Instance is launched, the field for "image" is empty in CLI and "Not Found" in Horizon. Is this behavior correct? Before create a Volume and launch an Instance, I created a Flavor, Image and Network. To create a Volume I used: $ openstack volume create --size 20 --type iscsi --image cirros --availability-zone nova volume-cirros To launch an Instance: $ openstack server create --volume volume-cirros --flavor m1.small --nic net-id=5886578b-1868-4c12-9742-9ae147b4daba vm-cirros-1 [wrsroot at controller-0 ~(keystone_admin)]$ openstack server show vm-cirros-1 | grep -w "image" | image The name of the image used to create the Volume to launch the Instances should be showed? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Mon Oct 8 21:53:57 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Mon, 8 Oct 2018 21:53:57 +0000 Subject: [Starlingx-discuss] [Build] Build report ww40 sept 30 - Oct 7 2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052B6F1@fmsmsx104.amr.corp.intel.com> This is the build report for StarlingX project for ww40 We started to generate two different builds inside Intel infrastructure: one for Master branch and another one for Release branch r/2018.10 Last merged changes[0] projected for this release were integrated in the build for both branches as of October 4th during afternoon Build As part of the build process, we are testing automation to get release notes between each build (commits and branches that have changed since last build) Summary of builds Master branch - 5 successful 0 failed (builds # 155-160) starlingx-2018-10 branch - 5 successful 1 failed [1] (build# 9-15) [0] https://github.com/starlingx-staging/stx-ceph/pull/3 - "after adding autoconf-archive-2017.03.21-1.el7.noarch.rpm in system, building ceph package will meet "aclocal: error: too many loops" [1] build-rpm-parallel --std failed with rc=1, Failed to build packages: libvirt-4.7.0-1.tis.22.src.rpm Fixed Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Tue Oct 9 01:09:12 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Tue, 9 Oct 2018 01:09:12 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-08-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E716FDCA6@FMSMSX108.amr.corp.intel.com> Sanity Test on stx-2018-10-08-16-r-2018.10.iso- 31 TCs PASS, 0 TC FAIL Nova-compute disabled/down on Simplex bare metal is reproduced on Simplex virtual. Multinode local and external storage on bare metal configurations still experiencing continuous reboot on nodes. Wind River continues helping, checking our switch configuration. Simplex Bare metal - Setup failed - Launchpad link [0]. No progress on execution. Duplex Bare metal - 14 Passed, 1 Failed - Launchpad [1] Duplex Virtual - 23 Passed, 1 Failed (same as in Bare metal) Multinode Local storage bare metal - 14 Passed Multinode local storage virtual - 32 Passed, 1 fail (initial debug WIP) Multinode external storage bare metal - 3 Passed Multinode external storage virtual - 6 Passed [0] Simplex - https://bugs.launchpad.net/starlingx/+bug/1796420 [1] Duplex - https://bugs.launchpad.net/starlingx/+bug/1796751 From shuicheng.lin at intel.com Tue Oct 9 06:20:41 2018 From: shuicheng.lin at intel.com (Lin, Shuicheng) Date: Tue, 9 Oct 2018 06:20:41 +0000 Subject: [Starlingx-discuss] de-fuzz centos7.5 upgraded src rpm In-Reply-To: <9700A18779F35F49AF027300A49E7C7655383752@SHSMSX101.ccr.corp.intel.com> References: <9700A18779F35F49AF027300A49E7C76553830F7@SHSMSX101.ccr.corp.intel.com> <33b41902-d812-709a-c77a-dc902b5715d7@windriver.com> <9700A18779F35F49AF027300A49E7C76553834D4@SHSMSX101.ccr.corp.intel.com> <9c43419b-b546-a797-b2ab-8ffc349b5da6@windriver.com> <2588653EBDFFA34B982FAF00F1B4844EBB28C91F@ALA-MBD.corp.ad.wrs.com> <9700A18779F35F49AF027300A49E7C7655383752@SHSMSX101.ccr.corp.intel.com> Message-ID: <9700A18779F35F49AF027300A49E7C765FE0D3A2@SHSMSX101.ccr.corp.intel.com> Hi all, Here is the latest status of the two story. You could get detail info in the story. [Feature] Upgrade rpm to 4.14.0 version This story is abandoned due to there is required package version conflict between rpm4.14 and ima-evm-utils. [Enhancement] De-fuzz fuzzy patch All fuzzy patch during build-srpm stage are eliminated. And most fuzzy patch from StarlingX during build-rpm stage are eliminated also. Most of the remained fuzz are from src rpm itself. Please help review below patches: https://review.openstack.org/#/q/topic:patch_fuzz+(status:open) Best Regards Shuicheng From: Lin, Shuicheng [mailto:shuicheng.lin at intel.com] Sent: Saturday, September 29, 2018 10:26 AM To: Rowsell, Brent ; Little, Scott ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] de-fuzz centos7.5 upgraded src rpm Hi all, Two story are created to track rpm upgrade and fuzzy patch elimination. [Feature] Upgrade rpm to 4.14.0 version [Enhancement] De-fuzz fuzzy patch Best Regards Shuicheng From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, September 29, 2018 7:39 AM To: Little, Scott >; Lin, Shuicheng >; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] de-fuzz centos7.5 upgraded src rpm Agree with Scott Brent From: Scott Little [mailto:scott.little at windriver.com] Sent: Friday, September 28, 2018 9:54 AM To: Lin, Shuicheng >; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] de-fuzz centos7.5 upgraded src rpm Certainly that will prevent creation of .orig files, so the risk of build failures in the packaging phase will be eliminated. However I have strong distrust of fuzzy patches. It suggests to me that folks are upgrading packages without doing the due diligence to make sure the patches are still needed, and are being applied correctly. Also I have seen twice in my career, a fuzzy patch get improperly applied to a neighboring bit of code that happened to look similar to the proper target. It's rare, but it can occur. On the plus side, moving to 4.14 would give us a consistent version of rpm throughout the build. That I like. I'm thinking we should do it, but we need a no fuzzy patches policy, and perhaps we create an audit that looks for fuzzy patches. Scott On 18-09-28 02:27 AM, Lin, Shuicheng wrote: Hi Scott, What do you think about upgrade rpm in mirror lst to 4.14 version? So we could get rid of the fuzz issue. To do this upgrade, rpm will be moved from centos repo lst to tarball lst. Here is the release notes of rpm-4.13 to fix this fuzz issue: Enable –no-backup-if-mismatch by default in %patch macro (RhBug:884755) Best Regards Shuicheng From: Scott Little [mailto:scott.little at windriver.com] Sent: Thursday, September 27, 2018 11:24 PM To: Lin, Shuicheng ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] de-fuzz centos7.5 upgraded src rpm Yes, I've seen some upstream packages that ship from CentOS or EL7 with fuzzy patches. Can't do much about that. My list might include packages where the only fuzzy patches are those coming from CentOS. We'll have to sort through them by hand I think. Scott On 18-09-27 10:24 AM, Lin, Shuicheng wrote: Hi all, Here is the patch list to de-fuzz the patches in CentOS7.5 upgrade task. https://review.openstack.org/#/q/topic:patch_fuzz+(status:open+OR+status:merged) I notice there are some other src rpms also have the fuzzy issue. Will handle them later. Also, I find some SRC RPM itself has the fuzzy issue, such as openldap/rsync/netpbm/openssh etc. So we cannot eliminate all fuzzy issue just by rebase ourselves patches. Best Regards Shuicheng From: Scott Little [mailto:scott.little at windriver.com] Sent: Thursday, September 27, 2018 5:08 AM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] build-pkgs cannot complete std build I have also been investigating another intermittent build error affecting initscripts. https://bugs.launchpad.net/starlingx/+bug/1794611 So any given package might be built by rpm-4.11.3, or by rpm-4.14.0. It's a race with many inputs. Ideally it shouldn't matter, but it does. build-pkgs --serial might reduce the odds, but likely won't solve it entirely. In this case it is a new option that rpm passes to the patch command, --no-backup-if-mismatch, first introduced by rpm-4.13. This option has the effect of suppressing the creation of '.orig' files. Creation of .orig files are a consequence of applying patches that are not clean. Unclean patches are ones that require 'fuzzing', i.e. treat the patch line numbers as approximate, rather than a strict requirement, just so long as the before/after context seems to be correct. Prior to StarlingX, my policy for rebasing patches was that no fuzz is tolerated in our patches. All the work to upgrade to 7.5 has created a lot of fuzzy patches. audit-2.8.1-3.el7.tis.2 bash-4.2.46-30.el7.tis.3 dhcp-4.2.5-68.el7.centos.1.tis.8 dnsmasq-2.76-5.el7.tis.6 drbd-8.4.3-0.tis.6 facter-2.4.4-4.el7.tis.4 haproxy-1.5.18-7.el7.tis.7 initscripts-9.49.41-1.el7.tis.16 iptables-1.4.21-24.1.el7_5.tis.3 kubernetes-1.10.0-1.tis.1 libevent-2.0.21-4.el7.tis.2 lighttpd-1.4.50-1.el7.tis.6 logrotate-3.8.6-15.el7.tis.3 netpbm-10.79.00-7.el7.tis.2 net-snmp-5.7.2-33.el7_5.2.tis.10 net-tools-2.0-0.22.20131004git.el7.tis.2 nfs-utils-1.3.0-0.54.el7.tis.4 nss-pam-ldapd-0.8.13-16.el7.tis.4 ntp-4.2.6p5-28.el7.centos.tis.3 openldap-2.4.44-15.el7_5.tis.8 openssh-7.4p1-16.el7_4.tis.9 pam-1.1.8-22.el7.tis.4 puppet-4.8.2-1.el7.tis.2 puppet-ceph-2.4.1-1.el7.tis.4 puppet-horizon-11.5.0-1.el7.tis.1 python-2.7.5-69.el7_5.tis.3 python-keyring-5.7.1-1.tis.2 python-wsme-0.9.2-1.el7.tis.3 resource-agents-3.9.5-124.el7.tis.12 rsync-3.1.2-4.el7.tis.2 shadow-utils-4.1.5.1-24.el7.tis.4 sudo-1.8.19p2-14.el7_5.tis.3 watchdog-5.13-11.el7.tis.2 So any of these packages might or might not produce unwanted .orig files. The .orig files might or might not be packaged, or break packaging. So there are a number of threads to pull at here. 1) Are fuzzy patches tolerated. I vote no. Sooner or later a fuzzy patch will be mis-applied and cause us problems. 2) Can we force rpmbuild within mock to use a consistent policy with respect to creation of orig files? 3) Can we pre-build rpm, such that all packages build against the same rpm version? 4) Are we patching any other low level build tools that have similar issues? Possibly explaining ceph? TBD On 18-09-26 12:16 PM, Scott Little wrote: aclocal 'too many loops' has been popping up sporadically for a week or two now. Possibly 7.5 related. I suspect that there is a build order and/or race condition element to this. It often goes away if you just run build-pkgs a second time. The second possible element is that build-pkgs is using flags that preserve the mock environment between packages. The goal was to avoid reinstalling required packages that are often common across the packages we build. It was a build time speedup that has been in use for 2-3 years now without incident. Google shows that other folks have hit 'aclocal too many loops' as well, and there are suggestions that it might be fixed in a recent update to automake. Scanning the changelog for the latest and greatest from gnu.org doesn't show any obvious fixes addressing this issue. Scott On 18-09-26 11:32 AM, Cordoba Malibran, Erich wrote: BUILDSTDERR: aclocal: error: too many loops BUILDSTDERR: aclocal: Please contact . BUILDSTDERR: at /usr/share/automake-1.13/Automake/Channels.pm line 662. BUILDSTDERR: Automake::Channels::msg('automake', '', 'too many loops') called at /usr/share/automake-1.13/Automake/ChannelDefs.pm line 212 BUILDSTDERR: Automake::ChannelDefs::prog_error('too many loops') called at /usr/bin/aclocal line 1187 BUILDSTDERR: error: Bad exit status from /var/tmp/rpm-tmp.Sj0E7c (%build) BUILDSTDERR: Macro expanded in comment on line 214: %global _libexecdir %{_exec_prefix}/lib BUILDSTDERR: Bad exit status from /var/tmp/rpm-tmp.Sj0E7c (%build) _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From cindy.xie at intel.com Tue Oct 9 06:43:55 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Tue, 9 Oct 2018 06:43:55 +0000 Subject: [Starlingx-discuss] Weekly StarlingX non-OpenStack Distro meeting In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F2B32565E@SHSMSX104.ccr.corp.intel.com> References: <2FD5DDB5A04D264C80D42CA35194914F2B32565E@SHSMSX104.ccr.corp.intel.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D61F8C@SHSMSX104.ccr.corp.intel.com> Agenda for 10/10 meeting: - Bugs for 2018.10 review (Ghada) - non-OpenStack patch analysis status (Saul/Zhipeng) - Fuzz patch guidline and follow-up actions (Brent/Shuicheng) - Kernel driver upgrade status (Shuicheng) - Ceph upgrade status (Vivian/Dehao) - Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other From austin.sun at intel.com Tue Oct 9 07:51:13 2018 From: austin.sun at intel.com (Sun, Austin) Date: Tue, 9 Oct 2018 07:51:13 +0000 Subject: [Starlingx-discuss] removing "make parallelize" meta patch In-Reply-To: <538bfb60-1e68-05bd-2499-a0d4a027360a@linux.intel.com> References: <2588653EBDFFA34B982FAF00F1B4844EBB2A3B2F@ALA-MBD.corp.ad.wrs.com> <538bfb60-1e68-05bd-2499-a0d4a027360a@linux.intel.com> Message-ID: Hi Saul and Brent: There are the gerrits. https://review.openstack.org/608870 https://review.openstack.org/608871 https://review.openstack.org/608872 Thanks. BR Austin Sun. -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: Tuesday, October 9, 2018 12:14 AM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] removing "make parallelize" meta patch Austin, Yes please, I created a storyboard for the removal of those patches here: https://storyboard.openstack.org/#!/story/2003803 Thanks Sau! On 10/08/2018 07:49 AM, Rowsell, Brent wrote: > Austin, > > No objections on removing these. > > Brent > > -----Original Message----- > From: Sun, Austin [mailto:austin.sun at intel.com] > Sent: Monday, October 8, 2018 4:54 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] removing "make parallelize" meta patch > > Hi All: > There are several meta patches to enable make in parallel , do we have any concern to remove it ? > > [0] > ./base/openssh/centos/meta_patches/0001-Further-parallelize-openssh-bu > ild.patch [1] > ./base/shadow-utils/centos/meta_patches/0001-Further-parallelize-shado > w-utils-build.patch [2] > ./base/sudo/centos/meta_patches/0003-Further-parallelize-sudo-build.pa > tch > > Thanks. > Austin Sun. > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Tue Oct 9 15:01:06 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 9 Oct 2018 15:01:06 +0000 Subject: [Starlingx-discuss] Controller recovery question Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E4637@fmsmsx101.amr.corp.intel.com> Hi I'm running a test scenario where a critical process is killed and I have to verify that Controller is recovered, with below steps I'm seeing that swat happens correctly but controller with the process killed keeps rebooting and process killed is not regenerated, my question is if the scenario is valid and the sm script should be regenerated. ==================================================================== Prep: Perform the following steps on the standy controller: a) Locate a critical process that is pmon managed: -> cd /etc/pmon.d -> grep -r "severity = critical" b) E.g. cat /etc/pmon.d/sm.conf c) Locate the name of the daemon pid file or startup script file that is configured in the conf file: e.g. server.pid-file = "/var/run/lighttpd.pid" or script = /etc/init.d/sm c) Rename the startup script file so that the critical process is not respawned after it is killed: e.g. mv /etc/init.d/sm /etc/init.d/sm~ Execution: 1) Kill the critical process on the standby controller sudo pkill sm **** EXPECTED ********* 1) Verify that the controller goes for a reset and recovers. ========================================================================= Regards, José From ada.cabrales at intel.com Tue Oct 9 15:27:58 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Tue, 9 Oct 2018 15:27:58 +0000 Subject: [Starlingx-discuss] New mailing list for help? Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E716FDEC2@FMSMSX108.amr.corp.intel.com> Hello, We've exchanging and asking information on this list about how StarlingX works on the administration side. Should we create a different mailing list (like starlingx-help) for all these questions? Regards Ada From juan.carlos.alonso at intel.com Tue Oct 9 15:38:51 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Tue, 9 Oct 2018 15:38:51 +0000 Subject: [Starlingx-discuss] snmp-api.log Message-ID: <8557B550001AFB46A43A0CCC314BF8514424627E@FMSMSX108.amr.corp.intel.com> There is a test case to check audit logging configuration and if its logs are written in /var/log/snmp-api.log file. I could not find snmp-api.log in such path (nor any), is there another path where the file could be generated? Steps: SNMP system details was created (site-contact, site-location) Community was created. Community details could display. There is something else do to in order to generate snmp-api.log file? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtroyer at gmail.com Tue Oct 9 15:40:05 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Tue, 9 Oct 2018 10:40:05 -0500 Subject: [Starlingx-discuss] New mailing list for help? In-Reply-To: <4F6AACE4B0F173488D033B02A8BB5B7E716FDEC2@FMSMSX108.amr.corp.intel.com> References: <4F6AACE4B0F173488D033B02A8BB5B7E716FDEC2@FMSMSX108.amr.corp.intel.com> Message-ID: On Tue, Oct 9, 2018 at 10:28 AM Cabrales, Ada wrote: > We've exchanging and asking information on this list about how StarlingX works on the administration side. Should we create a different mailing list (like starlingx-help) for all these questions? I'd suggest no... OpenStack is getting ready to consolidate most of thier lists back into a single list (-dev, -operators, -sigs?) leaving only -announce separate. A big part of the motivation to do that is to keep visibility better for the various groups involved, ie segregating operators and developers turned out to cause a lot of cross-posting to both lists to cover both audiences. Getting people into the habit of tagging messages with topics goes a long way to being able to pick out messages you either specifically want to see or want to ignore. Plus, the volume of our -discuss list is still very low. dt -- Dean Troyer dtroyer at gmail.com From ada.cabrales at intel.com Tue Oct 9 15:48:32 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Tue, 9 Oct 2018 15:48:32 +0000 Subject: [Starlingx-discuss] New mailing list for help? In-Reply-To: References: <4F6AACE4B0F173488D033B02A8BB5B7E716FDEC2@FMSMSX108.amr.corp.intel.com> Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E716FDF04@FMSMSX108.amr.corp.intel.com> I didn't know about the consolidation. That's fine. We'll keep posting questions here. And reinforce the use of tags. A. > -----Original Message----- > From: Dean Troyer [mailto:dtroyer at gmail.com] > Sent: Tuesday, October 9, 2018 10:40 AM > To: Cabrales, Ada > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] New mailing list for help? > > On Tue, Oct 9, 2018 at 10:28 AM Cabrales, Ada > wrote: > > We've exchanging and asking information on this list about how StarlingX > works on the administration side. Should we create a different mailing list > (like starlingx-help) for all these questions? > > I'd suggest no... OpenStack is getting ready to consolidate most of thier lists > back into a single list (-dev, -operators, -sigs?) leaving only -announce > separate. A big part of the motivation to do that is to keep visibility better for > the various groups involved, ie segregating operators and developers turned > out to cause a lot of cross-posting to both lists to cover both audiences. > > Getting people into the habit of tagging messages with topics goes a long way > to being able to pick out messages you either specifically want to see or want > to ignore. > > Plus, the volume of our -discuss list is still very low. > > dt > > -- > > Dean Troyer > dtroyer at gmail.com From jose.perez.carranza at intel.com Tue Oct 9 16:11:50 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 9 Oct 2018 16:11:50 +0000 Subject: [Starlingx-discuss] StarlingX alarms fields displayed Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E46B3@fmsmsx101.amr.corp.intel.com> Hi All I have a question about the alarms on StarlingX, in one of the scenarios that I'm executing says: - "Verify that entity-instance-field and suppression set to True not shown (system alarm-list, system alarm-show )" but actually the "Supression | Ture " is displayed, is this a correct behavior and I should update the scenario, or a Launchpad should be created to address this? Regards, José From scott.little at windriver.com Tue Oct 9 20:19:31 2018 From: scott.little at windriver.com (Scott Little) Date: Tue, 9 Oct 2018 16:19:31 -0400 Subject: [Starlingx-discuss] dnsmasq patch for DHCP6 config In-Reply-To: <5f266bb8-cdd6-9680-a6ac-3a40c7d16175@linux.intel.com> References: <5f266bb8-cdd6-9680-a6ac-3a40c7d16175@linux.intel.com> Message-ID: <23161b5d-d934-e8e1-344f-2873984ad0ee@windriver.com> nsmasq: allow updates to hostname in ipv6 leases     When a new lease is acquired by a new node this triggers an IP address     allocation but no hostname is associated with the lease.  A new lease update is     sent to sysinv and then a some later time sysinv will persist that record in to     the dnsmasq.host file.  When the dnsmasq.host file is updated dnsmasq is sent     the HUP signal to get it to re-read its config file.  At that time dnsmasq     attempts to look at the dnsmasq.conf file to find host records that may have     additional information that could be used to update existing leases.  That is,     if it finds a lease that is missing a hostname it will update it based on the     dnsmasq.hosts record.     This mechanism works fine for IPv4 but it is not supported for IPv6. This     update adds the necessary support to dnsmasq to allow the aforementioned     mechanism to work with IPv6.     This change in dnsmasq has also inadvertantly fixed an issue with an IPv6 DHCP     host record hostname being used to update an IPv4 lease.  This was causing     hostname lookups to return both an IPOv4 (pxe) and IPv6 (mgmt) address for the     same node. On 18-10-05 07:22 PM, Saul Wold wrote: > > Scott, Brent: > > You have a patch against dnsmasq that adds some code for handling IPv6 > addrs in lease_update_from_configs(). > > Can you provide any more background, was this reported upstream? What > exactly is this fixing? > > stx-integ/base/dnsmasq/centos/patches/dnsmasq-update-ipv6-leases-from-config.patch > > > Thanks >    Sau! From scott.little at windriver.com Tue Oct 9 20:20:07 2018 From: scott.little at windriver.com (Scott Little) Date: Tue, 9 Oct 2018 16:20:07 -0400 Subject: [Starlingx-discuss] dnsmasq patch for DHCP6 config In-Reply-To: <5f266bb8-cdd6-9680-a6ac-3a40c7d16175@linux.intel.com> References: <5f266bb8-cdd6-9680-a6ac-3a40c7d16175@linux.intel.com> Message-ID: dnsmasq: allow updates to hostname in ipv6 leases     When a new lease is acquired by a new node this triggers an IP address     allocation but no hostname is associated with the lease.  A new lease update is     sent to sysinv and then a some later time sysinv will persist that record in to     the dnsmasq.host file.  When the dnsmasq.host file is updated dnsmasq is sent     the HUP signal to get it to re-read its config file.  At that time dnsmasq     attempts to look at the dnsmasq.conf file to find host records that may have     additional information that could be used to update existing leases.  That is,     if it finds a lease that is missing a hostname it will update it based on the     dnsmasq.hosts record.     This mechanism works fine for IPv4 but it is not supported for IPv6. This     update adds the necessary support to dnsmasq to allow the aforementioned     mechanism to work with IPv6.     This change in dnsmasq has also inadvertantly fixed an issue with an IPv6 DHCP     host record hostname being used to update an IPv4 lease.  This was causing     hostname lookups to return both an IPOv4 (pxe) and IPv6 (mgmt) address for the     same node. On 18-10-05 07:22 PM, Saul Wold wrote: > > Scott, Brent: > > You have a patch against dnsmasq that adds some code for handling IPv6 > addrs in lease_update_from_configs(). > > Can you provide any more background, was this reported upstream? What > exactly is this fixing? > > stx-integ/base/dnsmasq/centos/patches/dnsmasq-update-ipv6-leases-from-config.patch > > > Thanks >    Sau! From Numan.Waheed at windriver.com Tue Oct 9 21:11:30 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Tue, 9 Oct 2018 21:11:30 +0000 Subject: [Starlingx-discuss] STX.2018.10 Test Results - 20181009 Message-ID: <3CAA827B7A79BA46B15B280EC82088FE4820BF10@ALA-MBD.corp.ad.wrs.com> Please find below the test results for stx.2018.10 Sanity: Sanity Results Config All-in-One Simplex Build ID 2018-10-08_01-52-01 Total 11 Pass 11 Fail 0 Bugs None Config All-in-One Duplex Build ID 2018-10-08_01-52-01 Total 15 Pass 15 Fail 0 Bugs None Config Multi Node Build ID 2018-10-08_01-52-01 Total 23 Pass 23 Fail 0 Bugs None Config Multi Node with CEPH Build ID 2018-10-08_01-52-01 Total 24 Pass 24 Fail 0 Bugs None Manual Regression: Domains Total Test Cases Passed Failed Total Executed Launchpad Backup and Restore 4 0 1 1 https://bugs.launchpad.net/starlingx/+bug/1796404 Distributed Cloud 30 12 0 12 Installation and Config 5 0 0 0 MTC 19 19 0 19 Networking 29 4 3 7 https://bugs.launchpad.net/starlingx/+bug/1796198 https://bugs.launchpad.net/starlingx/+bug/1790968 Nova 35 9 3 12 https://bugs.launchpad.net/starlingx/+bug/1796393 Security 17 3 0 3 Storage 11 1 2 3 https://bugs.launchpad.net/starlingx/+bug/1795962 System Inventory 8 6 0 6 System Test 11 4 0 4 Total 169 57 10 67 Regards, [WR logo signiture] Numan Waheed, Manager, Titanium Cloud Testing, Wind River Direct: 613.270.5762 * Mobile: 613.793.5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.jpg Type: image/jpeg Size: 1416 bytes Desc: image003.jpg URL: From liuchang77 at chinaunicom.cn Wed Oct 10 02:33:58 2018 From: liuchang77 at chinaunicom.cn (liuchang77 at chinaunicom.cn) Date: Wed, 10 Oct 2018 10:33:58 +0800 Subject: [Starlingx-discuss] Doubt about creating starling iso Message-ID: <201810101033581160498@chinaunicom.cn> Hi, all During my time studying starlingx. I have encountered various problems when trying to create a starlingx iso. So I thought of a problem, why do I have to build iso myself? What if there is a generic iso? I hope someone can help me solve this doubt. Thx! Chang Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: From liang.a.fang at intel.com Wed Oct 10 08:35:06 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Wed, 10 Oct 2018 08:35:06 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? Message-ID: Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang From liang.a.fang at intel.com Wed Oct 10 08:49:06 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Wed, 10 Oct 2018 08:49:06 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? In-Reply-To: References: Message-ID: Related review URL: https://review.openstack.org/#/c/605014/ https://review.openstack.org/#/c/605006/ Compare with upstream glance, starlingx added "Store" column as below +-----+--------+-------+-----------------+-----+------------+--------+---------+ | ID | Name | Store | Disk_format | ... | Size | Status |Owner| +-----+--------+-------+-----------------+-----+------------+--------+---------+ | xxx | img1 | file | raw | ... | 10737418 | active | xxx | | xxx | img2 | file | raw | ... | 5086345 | active | xxx | +-----+--------+-------+-----------------+-----+------------+--------+---------+ Regards Liang -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:35 PM To: jack.ding at windriver.com; robert.church at windriver.com; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the use case of these 2 patches? Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Wed Oct 10 12:50:12 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Wed, 10 Oct 2018 12:50:12 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? In-Reply-To: References: Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A94B0@ALA-MBD.corp.ad.wrs.com> Liang, This was a small usability improvement to help the user distinguish the glance backend type in a multi backend environment. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:49 AM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Related review URL: https://review.openstack.org/#/c/605014/ https://review.openstack.org/#/c/605006/ Compare with upstream glance, starlingx added "Store" column as below +-----+--------+-------+-----------------+-----+------------+--------+---------+ | ID | Name | Store | Disk_format | ... | Size | Status |Owner| +-----+--------+-------+-----------------+-----+------------+--------+---------+ | xxx | img1 | file | raw | ... | 10737418 | active | xxx | | xxx | img2 | file | raw | ... | 5086345 | active | xxx | +-----+--------+-------+-----------------+-----+------------+--------+---------+ Regards Liang -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:35 PM To: jack.ding at windriver.com; robert.church at windriver.com; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the use case of these 2 patches? Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Wed Oct 10 12:54:05 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Wed, 10 Oct 2018 12:54:05 +0000 Subject: [Starlingx-discuss] Doubt about creating starling iso In-Reply-To: <201810101033581160498@chinaunicom.cn> References: <201810101033581160498@chinaunicom.cn> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A94EB@ALA-MBD.corp.ad.wrs.com> Liu, You raise a valid point. We are working on publishing pre-built iso’s but that is not available yet. This will be announced on the mailing list when available. In the meantime, you would need to build one. For the issues you are experiencing please share on the mailing list so we can attempt to help. Brent From: liuchang77 at chinaunicom.cn [mailto:liuchang77 at chinaunicom.cn] Sent: Tuesday, October 9, 2018 10:34 PM To: starlingx-discuss Subject: [Starlingx-discuss] Doubt about creating starling iso Hi, all During my time studying starlingx. I have encountered various problems when trying to create a starlingx iso. So I thought of a problem, why do I have to build iso myself? What if there is a generic iso? I hope someone can help me solve this doubt. Thx! Chang Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: From cindy.xie at intel.com Wed Oct 10 13:36:42 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Wed, 10 Oct 2018 13:36:42 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D69087@SHSMSX104.ccr.corp.intel.com> Agenda & Notes for 10/10 meeting: 1. Bugs for 2018.10 review (Ghada) no pending storyboard & bugs. Cheers! 2. non-OpenStack patch analysis status (Saul/Zhipeng) init related patches, current build related ones merged by Austin; Zhipeng is looking into solutions for the init related patches to remove from individual RPMs. deep dive to some upstream source level patches and analysis WIP. Brent submitted several new stories and please mark it as "stx-non-distro" tag. 3. Fuzz patch guideline and follow-up actions (Brent/Shuicheng) Going forward we should put guideline for strictly no fuzz patch. AR: Cindy to put a developer guideline to: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines https://storyboard.openstack.org/#!/story/2003917 include all fuzz patches, most of them are merged. Shuicheng will add more tasks (few left) within this week. There is not much we can do w/ the upstreamed patches from sRPM for now. story for upgrade rpm to v14 is not required for now. rpm v14 requires OpenSSH v1.1 and we cannot upgrade from a lower version. For now, the system limitation is that we cannot install the same package with different version simultaneously. 4. Kernel driver upgrade status (Shuicheng) https://storyboard.openstack.org/#!/story/2003597, all patches are ready and pending for testing. Intel GDC do not have appropriate test systems available. AR: Cindy to coordinate Ada about the HW required to run the testing. Mellanox driver version: how to define the version# to upgrade? choose the latest version. usually the version we pick shall be aligned w/ the version in the DPDK supported. Need to look for the version from the Mellanox page. AR: Shuicheng to check the version to make sure it compatible w/ DPDK version in our current mainline. 5. Ceph upgrade status (Vivian/Dehao) - no update due to holiday 6. Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other From cesar.lara at intel.com Wed Oct 10 14:23:09 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 10 Oct 2018 14:23:09 +0000 Subject: [Starlingx-discuss] Doubt about creating starling iso In-Reply-To: <201810101033581160498@chinaunicom.cn> References: <201810101033581160498@chinaunicom.cn> Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052D921@fmsmsx104.amr.corp.intel.com> Unfortunately, today we don’t have a way to distribute binaries or iso images, we are working in a solution for this, hopefully, we will have something in the next few weeks. In the meantime, please share with this list the errors and blockers that you have so we can help you, I have a team dedicated to test the iso creation and the build so they can totally help you with your deployment. Regards Cesar Lara From: liuchang77 at chinaunicom.cn [mailto:liuchang77 at chinaunicom.cn] Sent: Tuesday, October 9, 2018 9:34 PM To: starlingx-discuss Subject: [Starlingx-discuss] Doubt about creating starling iso Hi, all During my time studying starlingx. I have encountered various problems when trying to create a starlingx iso. So I thought of a problem, why do I have to build iso myself? What if there is a generic iso? I hope someone can help me solve this doubt. Thx! Chang Liu -------------- next part -------------- An HTML attachment was scrubbed... URL: From cindy.xie at intel.com Wed Oct 10 14:37:30 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Wed, 10 Oct 2018 14:37:30 +0000 Subject: [Starlingx-discuss] OpenStack upgrade to Rocky Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D691FB@SHSMSX104.ccr.corp.intel.com> Do we have this planned for 2019.03 release? -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Wed Oct 10 14:37:04 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Wed, 10 Oct 2018 14:37:04 +0000 Subject: [Starlingx-discuss] [ Test ] stx.2018.10 test plan In-Reply-To: References: <4F6AACE4B0F173488D033B02A8BB5B7E716E81DD@fmsmsx104.amr.corp.intel.com> Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E716FE6D4@FMSMSX108.amr.corp.intel.com> Going back to this topic, the steps for the tests are published here [0] Just consider some steps (or commands) might not be 100% accurate. We are still working on adjusting them. However, you can have a good idea about what the test is about. Let me know if you have questions. [0] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Ada > -----Original Message----- > From: kai [mailto:li.kai at 99cloud.net] > Sent: Thursday, September 27, 2018 9:43 PM > To: Cabrales, Ada ; starlingx- > discuss at lists.starlingx.io > Cc: Hu, Wei W ; Sun, Xianghui > ; wuyx36 at chinaunicom.cn; Xie, Cindy > ; Wang, Shane ; Ding, Jian- > feng ; Zeng, Hongli ; Jin, > Yuntong ; 'lvhz7' ; Zhao, > Forrest ; 'yanky5' ; '陈 > 丹' ; '王友祥' ; > huang.shuquan at 99cloud.net; Liang, Maggie ; > 'jeffrey.zhang' ; '龚永生' > ; 'jing.liuqing' ; > '99cloud' > Subject: 答复: [Starlingx-discuss] [ Test ] stx.2018.10 test plan > > Hi, Ada, where can we find the detail test steps for each test case? > > -----邮件原件----- > 发件人: Cabrales, Ada > 发送时间: 2018年9月22日 7:54 > 收件人: starlingx-discuss at lists.starlingx.io > 主题: [Starlingx-discuss] [ Test ] stx.2018.10 test plan > > Hello, > > I'm posting the test plan proposed for the October Release in the wiki [0]. > We have 238 tests so far, mostly focus on the stx services. As we mentioned > before, the tests will be run manually by our team for this release. The > expectation is to increase the number of automated tests along time. > > We are considering to cover the four configurations (Simplex, Duplex, > Multinode local storage, Multinode external storage) both on bare metal and > virtual. However, we are experiencing setup problems with multinode bare > metal and simplex. We are digging into it. We'll have more information next > week. > > Please take a look at it and let us know what else must be included. > > Thanks > Ada > > [0] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From david.a.cobbley at intel.com Wed Oct 10 14:44:33 2018 From: david.a.cobbley at intel.com (Cobbley, David A) Date: Wed, 10 Oct 2018 14:44:33 +0000 Subject: [Starlingx-discuss] OpenStack upgrade to Rocky Message-ID: <53720C37-FFC5-414F-A542-3304B1BF9FF1@intel.com> I believe the community discussion, including at the PTG, was to skip Rocky and move directly to Stein in Q1. -- David C From: "Xie, Cindy" Date: Wednesday, October 10, 2018 at 7:40 AM To: "starlingx-discuss at lists.starlingx.io" Subject: [Starlingx-discuss] OpenStack upgrade to Rocky Do we have this planned for 2019.03 release? -------------- next part -------------- An HTML attachment was scrubbed... URL: From cindy.xie at intel.com Wed Oct 10 14:47:21 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Wed, 10 Oct 2018 14:47:21 +0000 Subject: [Starlingx-discuss] OpenStack upgrade to Rocky In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2A9C70@ALA-MBD.corp.ad.wrs.com> References: <2FD5DDB5A04D264C80D42CA35194914F35D691FB@SHSMSX104.ccr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2A9C70@ALA-MBD.corp.ad.wrs.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D69250@SHSMSX104.ccr.corp.intel.com> OK, good to know that - then 2019.03 release will stick w/ Pike. Thx. - cindy From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Wednesday, October 10, 2018 10:42 PM To: Xie, Cindy ; starlingx-discuss at lists.starlingx.io Subject: RE: OpenStack upgrade to Rocky Cindy, We discussed this topic at the PTG. The conclusion reached there was that we would not be upgrading to Rocky. The candidate release would be Stein which would be after the March release. Brent From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 10:38 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] OpenStack upgrade to Rocky Do we have this planned for 2019.03 release? -------------- next part -------------- An HTML attachment was scrubbed... URL: From Brent.Rowsell at windriver.com Wed Oct 10 14:42:14 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Wed, 10 Oct 2018 14:42:14 +0000 Subject: [Starlingx-discuss] OpenStack upgrade to Rocky In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F35D691FB@SHSMSX104.ccr.corp.intel.com> References: <2FD5DDB5A04D264C80D42CA35194914F35D691FB@SHSMSX104.ccr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2A9C70@ALA-MBD.corp.ad.wrs.com> Cindy, We discussed this topic at the PTG. The conclusion reached there was that we would not be upgrading to Rocky. The candidate release would be Stein which would be after the March release. Brent From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 10:38 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] OpenStack upgrade to Rocky Do we have this planned for 2019.03 release? -------------- next part -------------- An HTML attachment was scrubbed... URL: From liang.a.fang at intel.com Wed Oct 10 15:44:19 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Wed, 10 Oct 2018 15:44:19 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2A94B0@ALA-MBD.corp.ad.wrs.com> References: <2588653EBDFFA34B982FAF00F1B4844EBB2A94B0@ALA-MBD.corp.ad.wrs.com> Message-ID: OK, thank you Brent, let me try my best to convince community to accept these two patches. Regards Liang -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Wednesday, October 10, 2018 8:50 PM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Liang, This was a small usability improvement to help the user distinguish the glance backend type in a multi backend environment. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:49 AM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Related review URL: https://review.openstack.org/#/c/605014/ https://review.openstack.org/#/c/605006/ Compare with upstream glance, starlingx added "Store" column as below +-----+--------+-------+-----------------+-----+------------+--------+---------+ | ID | Name | Store | Disk_format | ... | Size | Status |Owner| +-----+--------+-------+-----------------+-----+------------+--------+---------+ | xxx | img1 | file | raw | ... | 10737418 | active | xxx | | xxx | img2 | file | raw | ... | 5086345 | active | xxx | +-----+--------+-------+-----------------+-----+------------+--------+---------+ Regards Liang -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:35 PM To: jack.ding at windriver.com; robert.church at windriver.com; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the use case of these 2 patches? Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From erich.cordoba.malibran at intel.com Wed Oct 10 20:12:59 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Wed, 10 Oct 2018 20:12:59 +0000 Subject: [Starlingx-discuss] Patch to support different address witdhs in DPDK Message-ID: <7c0998d84c9afa74347a0de01c96f38571d1396e.camel@intel.com> Hi all, This bug[1] was failing (simplex in the baremetal case) because the /etc/init.d/compute_config was unable to configure the compute causing a "degraded" state and nova-compute to be disabled. While running the compute_config script a lot of kernel messages like this appeared: DMAR: intel_iommu_map: iommu width (39) is not sufficient for the mapped address (7fbdc0000000) It turns out that this is a known issue in DPDK[2] and a patch is available[3]. I applied this patch into the openvswitch package and then we were able to solve the problem. My question is, what would be the best way to handle this patch? - Should we patch our openvswitch package? : we get openvswitch from centos and that version doesn't include this patch. We can apply this patch here and wait for an upgrade in upstream to remove it. - Should be look for a newer openvswitch version? We could look for a newer version but this one won't be a centos standard version. What do you think? Thanks in advance -Erich [1] https://bugs.launchpad.net/starlingx/+bug/1796420 [2] http://mails.dpdk.org/archives/stable/2018-January/004109.html [3] https://github.com/DPDK/dpdk/commit/54a328f552ff2e0098c3f96f9e32302 675f2bcf4 From cesar.lara at intel.com Wed Oct 10 20:32:58 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 10 Oct 2018 20:32:58 +0000 Subject: [Starlingx-discuss] [build][meetings] Agenda for 10/11/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052E063@fmsmsx104.amr.corp.intel.com> Agenda - cengn hosting status - test cycle - stxb tool spec - opens Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From juan.carlos.alonso at intel.com Wed Oct 10 20:41:56 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Wed, 10 Oct 2018 20:41:56 +0000 Subject: [Starlingx-discuss] snmp-api.log In-Reply-To: <8557B550001AFB46A43A0CCC314BF8514424627E@FMSMSX108.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF8514424627E@FMSMSX108.amr.corp.intel.com> Message-ID: <8557B550001AFB46A43A0CCC314BF8514EEFF95D@fmsmsx101.amr.corp.intel.com> Hi, Any update or light about this? Regards. Juan Carlos Alonso From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Tuesday, October 9, 2018 10:39 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] snmp-api.log There is a test case to check audit logging configuration and if its logs are written in /var/log/snmp-api.log file. I could not find snmp-api.log in such path (nor any), is there another path where the file could be generated? Steps: SNMP system details was created (site-contact, site-location) Community was created. Community details could display. There is something else do to in order to generate snmp-api.log file? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Wed Oct 10 20:52:25 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 10 Oct 2018 20:52:25 +0000 Subject: [Starlingx-discuss] [build] use case discussion for tomorrow's meeting Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052E0C4@fmsmsx104.amr.corp.intel.com> Hi build team I put the use cases in a presentation so we can all have something to discuss to, these use cases were sent to the mailing list a couple weeks ago and I want to make sure that are still valid since we are going to use them to propose a refactor of stxb tool based on this information. Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: stx-use case discussion.pdf Type: application/pdf Size: 25289 bytes Desc: stx-use case discussion.pdf URL: From corey at corebits.ca Wed Oct 10 21:55:41 2018 From: corey at corebits.ca (Corey Erickson) Date: Wed, 10 Oct 2018 15:55:41 -0600 Subject: [Starlingx-discuss] 2018.10 build success Message-ID: Hi All, Just wanted to send a quick note to say that I have now been able to build the r/2018.10 branch successfully. I hope to fully test it over the next few weeks as time permits. Best Regards, Corey Erickson -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Wed Oct 10 22:11:51 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 10 Oct 2018 22:11:51 +0000 Subject: [Starlingx-discuss] 2018.10 build success In-Reply-To: References: Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052E2DA@fmsmsx104.amr.corp.intel.com> Awesome, thank you for sharing, to me this is great news (and is almost 5pm here, so cheers). Regards Cesar Lara From: Corey Erickson [mailto:corey at corebits.ca] Sent: Wednesday, October 10, 2018 4:56 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] 2018.10 build success Hi All, Just wanted to send a quick note to say that I have now been able to build the r/2018.10 branch successfully. I hope to fully test it over the next few weeks as time permits. Best Regards, Corey Erickson -------------- next part -------------- An HTML attachment was scrubbed... URL: From erich.cordoba.malibran at intel.com Wed Oct 10 22:12:58 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Wed, 10 Oct 2018 22:12:58 +0000 Subject: [Starlingx-discuss] 2018.10 build success In-Reply-To: References: Message-ID: Hi Corey, that's really good news :) We are in a testing cycle so expect to have updates in the branch. Any change will be notified through the mailing list. -Erich On Wed, 2018-10-10 at 15:55 -0600, Corey Erickson wrote: > Hi All, > > Just wanted to send a quick note to say that I have now been able to > build the r/2018.10 branch successfully. > > I hope to fully test it over the next few weeks as time permits. > > Best Regards, > Corey Erickson > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From erich.cordoba.malibran at intel.com Wed Oct 10 22:22:03 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Wed, 10 Oct 2018 22:22:03 +0000 Subject: [Starlingx-discuss] Doubt about creating starling iso In-Reply-To: <201810101033581160498@chinaunicom.cn> References: <201810101033581160498@chinaunicom.cn> Message-ID: <717d0b177af1c2117a0ea3c12bc694b921074035.camel@intel.com> Hi Chang, welcome! As other peopled said, we are in the process to provide builds, I would like to say that one of the use cases of StarlingX is that you can use it as framework to develop or adapt your edge computing solution. In this use case it will be necessary to build your own ISO. So, as a starting point it will be nice to have precompiled ISOs to download (work in progress) but once you get more experienced with the environment you may want to build your own ISO. -Erich On Wed, 2018-10-10 at 10:33 +0800, liuchang77 at chinaunicom.cn wrote: > Hi, all > > During my time studying starlingx. I have encountered various > problems when trying to create a starlingx iso. So I thought of a > problem, why do I have to build iso myself? What if there is a > generic iso? I hope someone can help me solve this doubt. > Thx! > > Chang Liu > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Numan.Waheed at windriver.com Wed Oct 10 20:39:12 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Wed, 10 Oct 2018 20:39:12 +0000 Subject: [Starlingx-discuss] Testing Status - 20181010 Message-ID: <3CAA827B7A79BA46B15B280EC82088FE4820EC28@ALA-MBD.corp.ad.wrs.com> Please find below testing status for 20181010. Sanity: Sanity Results Config All-in-One Simplex Build ID 2018-10-09_01-52-01 Total 11 Pass 11 Fail 0 Bugs None Config All-in-One Duplex Build ID 2018-10-08_01-52-01 Total 15 Pass 15 Fail 0 Bugs None Config Multi Node Build ID 2018-10-08_01-52-01 Total 23 Pass 23 Fail 0 Bugs None Config Multi Node with CEPH Build ID 2018-10-09_01-52-01 Total 24 Pass 24 Fail 0 Bugs None Manual Regression: Domains Total Test Cases Passed Failed Total Executed Launchpad Backup and Restore 4 3 1 4 https://bugs.launchpad.net/starlingx/+bug/1796404 Distributed Cloud 30 12 0 12 Installation and Config 5 0 0 0 MTC 19 19 0 19 Networking 29 5 3 8 https://bugs.launchpad.net/starlingx/+bug/1796198 https://bugs.launchpad.net/starlingx/+bug/1790968 Nova 35 9 3 12 https://bugs.launchpad.net/starlingx/+bug/1796393 Security 17 3 0 3 Storage 11 3 3 6 https://bugs.launchpad.net/starlingx/+bug/1795962 https://bugs.launchpad.net/starlingx/+bug/1796913 System Inventory 8 7 1 8 System Test 11 5 0 5 Total 169 66 11 77 Automated Regression: Build Used: 2018-10-05_01-52-00 Domains Passed Failed Total Executed Comments Heat 30 0 30 Horizon 57 0 57 MTC General 87 4 91 MTC Process Monitoring 66 0 66 Networking 80 24 104 https://bugs.launchpad.net/starlingx/+bug/1790968 https://bugs.launchpad.net/starlingx/+bug/1796952 Nova 447 11 458 https://bugs.launchpad.net/starlingx/+bug/1791685 https://bugs.launchpad.net/starlingx/+bug/1792945 https://bugs.launchpad.net/starlingx/+bug/1792999 REST API 229 1 230 https://bugs.launchpad.net/starlingx/+bug/1789979 Security 49 38 87 https://bugs.launchpad.net/starlingx/+bug/1792418 Storage 74 1 75 https://bugs.launchpad.net/starlingx/+bug/1796912 Sysinv 22 1 23 https://bugs.launchpad.net/starlingx/+bug/1792418 Total 1141 80 1221 Pass Rate = 93.4%, Adjusted Pass Rate = 96.9% Regards, [WR logo signiture] Numan Waheed, Manager, Titanium Cloud Testing, Wind River Direct: 613.270.5762 * Mobile: 613.793.5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.jpg Type: image/jpeg Size: 1416 bytes Desc: image003.jpg URL: From dtroyer at gmail.com Wed Oct 10 23:40:07 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Wed, 10 Oct 2018 18:40:07 -0500 Subject: [Starlingx-discuss] Patch to support different address witdhs in DPDK In-Reply-To: <7c0998d84c9afa74347a0de01c96f38571d1396e.camel@intel.com> References: <7c0998d84c9afa74347a0de01c96f38571d1396e.camel@intel.com> Message-ID: On Wed, Oct 10, 2018 at 3:18 PM Cordoba Malibran, Erich wrote: > - Should be look for a newer openvswitch version? We could look for a > newer version but this one won't be a centos standard version. We already keep more current versions of libvirt and qemu than are supplied with CentOS because there are certain advantages, including performance. openvswitch may fall in to that category too, particularly with the reliance we will have on DPDK. I'm not sure what the criteria should be for making that kind of decision, although I do expect we should have a high bar to meet to do that.. I would like to hear the thoughts from those who know these bits well. The last time I build an openvswitch package was at the Bexar summit... dt -- Dean Troyer dtroyer at gmail.com From Brent.Rowsell at windriver.com Thu Oct 11 00:49:02 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Thu, 11 Oct 2018 00:49:02 +0000 Subject: [Starlingx-discuss] Patch to support different address witdhs in DPDK In-Reply-To: References: <7c0998d84c9afa74347a0de01c96f38571d1396e.camel@intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2ABB35@ALA-MBD.corp.ad.wrs.com> The networking team should weigh in here with the recommended path forward. Brent -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Wednesday, October 10, 2018 7:40 PM To: Cordoba Malibran, Erich Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Patch to support different address witdhs in DPDK On Wed, Oct 10, 2018 at 3:18 PM Cordoba Malibran, Erich wrote: > - Should be look for a newer openvswitch version? We could look for a > newer version but this one won't be a centos standard version. We already keep more current versions of libvirt and qemu than are supplied with CentOS because there are certain advantages, including performance. openvswitch may fall in to that category too, particularly with the reliance we will have on DPDK. I'm not sure what the criteria should be for making that kind of decision, although I do expect we should have a high bar to meet to do that.. I would like to hear the thoughts from those who know these bits well. The last time I build an openvswitch package was at the Bexar summit... dt -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ada.cabrales at intel.com Thu Oct 11 01:22:43 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Thu, 11 Oct 2018 01:22:43 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-10-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C34E253@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-10-19-r-2018.10.iso - 31 TCs PASS, 0 TCs FAIL - Simplex Bare metal working after a DPDK patch applied. Launchpad [0] updated. - Multinode local and external storage bare metal configs stable after changes on the switch configuration. Executing test cases now. - Having problems on VM/instances creation on Multinode external storage with ceph, virtual environment [1]. Checking if there's replicated on Bare metal config. - namespace is not reasgined on swact [2] - seen on Duplex bare metal config. - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 0/2/78/80 Simplex virtual - 0/10/61/71 Duplex bare metal - 2/31/67/100 Duplex virtual - 1/41/63/105 Multinode local storage bare metal - 1/34/82/117 Multinode local storage virtual - 4/57/57/118 Multinode external storage bare metal - 0/3/114/117 Multinode external storage virtual - 3/16/91/110 Full report in the wiki [3] [0] https://bugs.launchpad.net/starlingx/+bug/1796420 [1] https://bugs.launchpad.net/starlingx/+bug/1797234 [2] https://bugs.launchpad.net/starlingx/+bug/1797217 [3] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary From Greg.Waines at windriver.com Thu Oct 11 10:40:34 2018 From: Greg.Waines at windriver.com (Waines, Greg) Date: Thu, 11 Oct 2018 10:40:34 +0000 Subject: [Starlingx-discuss] snmp-api.log Message-ID: Run some SNMP GET commands against the stx-fault MIB. Greg. From: "Alonso, Juan Carlos" Date: Wednesday, October 10, 2018 at 4:41 PM To: "starlingx-discuss at lists.starlingx.io" Subject: Re: [Starlingx-discuss] snmp-api.log Hi, Any update or light about this? Regards. Juan Carlos Alonso From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Tuesday, October 9, 2018 10:39 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] snmp-api.log There is a test case to check audit logging configuration and if its logs are written in /var/log/snmp-api.log file. I could not find snmp-api.log in such path (nor any), is there another path where the file could be generated? Steps: SNMP system details was created (site-contact, site-location) Community was created. Community details could display. There is something else do to in order to generate snmp-api.log file? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matt.Peters at windriver.com Thu Oct 11 15:01:43 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Thu, 11 Oct 2018 15:01:43 +0000 Subject: [Starlingx-discuss] Patch to support different address witdhs in DPDK In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2ABB35@ALA-MBD.corp.ad.wrs.com> References: <7c0998d84c9afa74347a0de01c96f38571d1396e.camel@intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2ABB35@ALA-MBD.corp.ad.wrs.com> Message-ID: The patch is not for openvswitch itself and is only for the bundled DPDK distribution that is included in the CentOS specific SRPM (currently based on 17.11.0 - old snapshot). Currently CentOS is not a distributing a newer version of the openvswitch package that includes a newer version of DPDK. I don't think we want to up version the DPDK distribution in this package within StarlingX since this would deviate from a supported lineup of Centos. However, that being said, the recommended version from OVS is DPDK 17.11.3. I think the short term solution is we should include this patch in StarlingX to address the immediate issue. Longer term, we would want to push to get the CentOS openvswitch SRPM package updated with a newer version of DPDK 17.11 (currently at 17.11.4) to have all available bug fixes in the dpdk-stable branch. Regards, Matt > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Wednesday, October 10, 2018 8:49 PM > To: Dean Troyer; Cordoba Malibran, Erich > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] Patch to support different address witdhs in > DPDK > > The networking team should weigh in here with the recommended path > forward. > > Brent > > -----Original Message----- > From: Dean Troyer [mailto:dtroyer at gmail.com] > Sent: Wednesday, October 10, 2018 7:40 PM > To: Cordoba Malibran, Erich > Cc: starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] Patch to support different address witdhs in > DPDK > > On Wed, Oct 10, 2018 at 3:18 PM Cordoba Malibran, Erich > wrote: > > - Should be look for a newer openvswitch version? We could look for a > > newer version but this one won't be a centos standard version. > > We already keep more current versions of libvirt and qemu than are supplied > with CentOS because there are certain advantages, including performance. > openvswitch may fall in to that category too, particularly with the reliance we > will have on DPDK. I'm not sure what the criteria should be for making that > kind of decision, although I do expect we should have a high bar to meet to > do that.. > > I would like to hear the thoughts from those who know these bits well. > The last time I build an openvswitch package was at the Bexar summit... > > dt > > -- > > Dean Troyer > dtroyer at gmail.com > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cesar.lara at intel.com Thu Oct 11 16:50:39 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Thu, 11 Oct 2018 16:50:39 +0000 Subject: [Starlingx-discuss] [build][meetings] Meeting Minutes 10/11/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471052EE1D@fmsmsx104.amr.corp.intel.com> 10/11/2018 Attendees Saul, Scott, Jason, Ken, Erich, Memo, Chuy, Mario, Marcela, Hayde, Abraham, Cesar - Cengn hosting status - ken Exploration phase to make sure what we are going to offer initially on Cengn, we are prioritizing the availability for the mirror, and getting prepared to start utilizing that mirror instead of local based mirrors. The team is focusing over the planning of the new build structure that was proposed in which we split the OS from the flock components. Cengn is potentially having a bare metal server for hosting ready by the end of this week and coordinate with StarlingX build team to start populating this server with a mirror, Scott acting as the interface between StarlingX and Cengn. We might need to propose a separate spec for build infrastructure and get feedback from the community. AR - We need to analyze the signing of the modified RPMs being distributed in the Cengn mirror, for unmodified packages we will use CentOS signatures. - test cycle for next release Do we have open issues on the build system right now? No - stxb tool spec Should we move forward with stxb tool? Yes, we should provide the community with an easy to use tool to build isos for developers and contributors assuming the end users will download finished isos from Cengn mirror We need to build a spec around this tool and the use cases that it will address, based on this feedback we will architect a robust solution for this tool. The team took a few ARs from this discussion AR - take a look on the doc sent to ML and review test cases there to make sure they all still make sense AR - Send potential use cases to ML for discussion to be included in the architecture, sent those by Monday EOD Use cases will be written in an statement fashion and not in a question form. Case 3A and 3B from the document should be down to - detect if there any changes in the artifacts or packages in the mirror and act accordingly - opens Intel team is working on enabling StarlingX on Clear Linux and as part of this effort we are using Koji as a build system, we found some commonalities that led us to believe we could use is as part of our build system. Once we get past the discovery phase we will put together a POC to evaluate this. May need to consider this as part of the multi OS effort to generate RPMs and Deb packages for consumption. Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From abraham.arce.moreno at intel.com Thu Oct 11 17:51:54 2018 From: abraham.arce.moreno at intel.com (Arce Moreno, Abraham) Date: Thu, 11 Oct 2018 17:51:54 +0000 Subject: [Starlingx-discuss] [Doc] restapi-doc wadl based package removal Message-ID: We have already transitioned our StarlingX API documentation [0] based on OpenStack API documentation guidelines [1]. It is time to remove the previous restapi-doc package based on wadl format using maven to generate it. We would appreciate your support to review these changes and point out any missing dependency, the change set is found here [2] [0] https://wiki.openstack.org/wiki/StarlingX/Developer_Guide/API_Documentation#Tracking [1] https://docs.openstack.org/doc-contrib-guide/api-guides.html [2] https://review.openstack.org/#/q/topic:restapi-doc-wadl+(status:open+OR+status:merged) From shuicheng.lin at intel.com Fri Oct 12 06:46:31 2018 From: shuicheng.lin at intel.com (Lin, Shuicheng) Date: Fri, 12 Oct 2018 06:46:31 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F35D69087@SHSMSX104.ccr.corp.intel.com> References: <2FD5DDB5A04D264C80D42CA35194914F35D69087@SHSMSX104.ccr.corp.intel.com> Message-ID: <9700A18779F35F49AF027300A49E7C765FE0EE0B@SHSMSX101.ccr.corp.intel.com> Hi Ghada/Brent, I try to go through the doc in mellanox website, but cannot find the matrix between mlx driver and DPDK. Could you do me a favor where could I get it? Thanks. Here I get the latest mlx driver: http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers From the dpdk package (dpdk-17.05.2.tar.gz) in our code, I get below info, which seems obsolete. " Currently supported by DPDK: - Mellanox OFED version: **4.0-2.0.0.0** " Best Regards Shuicheng -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 9:37 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Agenda & Notes for 10/10 meeting: 1. Bugs for 2018.10 review (Ghada) no pending storyboard & bugs. Cheers! 2. non-OpenStack patch analysis status (Saul/Zhipeng) init related patches, current build related ones merged by Austin; Zhipeng is looking into solutions for the init related patches to remove from individual RPMs. deep dive to some upstream source level patches and analysis WIP. Brent submitted several new stories and please mark it as "stx-non-distro" tag. 3. Fuzz patch guideline and follow-up actions (Brent/Shuicheng) Going forward we should put guideline for strictly no fuzz patch. AR: Cindy to put a developer guideline to: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines https://storyboard.openstack.org/#!/story/2003917 include all fuzz patches, most of them are merged. Shuicheng will add more tasks (few left) within this week. There is not much we can do w/ the upstreamed patches from sRPM for now. story for upgrade rpm to v14 is not required for now. rpm v14 requires OpenSSH v1.1 and we cannot upgrade from a lower version. For now, the system limitation is that we cannot install the same package with different version simultaneously. 4. Kernel driver upgrade status (Shuicheng) https://storyboard.openstack.org/#!/story/2003597, all patches are ready and pending for testing. Intel GDC do not have appropriate test systems available. AR: Cindy to coordinate Ada about the HW required to run the testing. Mellanox driver version: how to define the version# to upgrade? choose the latest version. usually the version we pick shall be aligned w/ the version in the DPDK supported. Need to look for the version from the Mellanox page. AR: Shuicheng to check the version to make sure it compatible w/ DPDK version in our current mainline. 5. Ceph upgrade status (Vivian/Dehao) - no update due to holiday 6. Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From kailun.qin at intel.com Fri Oct 12 10:33:16 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Fri, 12 Oct 2018 10:33:16 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for e5712fe Message-ID: Hi Matt, We're looking into the patch e5712fe, which indicates a possible inconsistency/mismatch of tag values between dnsmasq option files and dnsmasq command line. >From the commit message: "This seems to be happening because of how dhcp_rpc.py::_group_by_network_id would behave if the list of subnets coming from the database was ordered different or contained different values." After checking, we found that the latest Neutron seems to already have subnets sorted: https://github.com/openstack/neutron/blob/master/neutron/api/rpc/handlers/dhcp_rpc.py#L150L154. Would you please kindly help double check on this? Let me know if any question. Thank you! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matt.Peters at windriver.com Fri Oct 12 13:08:08 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Fri, 12 Oct 2018 13:08:08 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for e5712fe In-Reply-To: References: Message-ID: <9299A6BD-9EC4-48EE-A71E-67051508A7EC@windriver.com> Hello Kailun, I agree that the change looks like it should address the same issue. The change from e5712fe may have been carried forward between rebases without realizing that it was addressed on the server side. Have you retested the scenario without the change for e5712fe? If things are working as expected, than it can be dropped. Regards, Matt From: "Qin, Kailun" Date: Friday, October 12, 2018 at 6:33 AM To: "Peters, Matt" Cc: "Le, Huifeng" , "starlingx-discuss at lists.starlingx.io" Subject: Question about patch upstreaming for e5712fe Hi Matt, We’re looking into the patch e5712fe, which indicates a possible inconsistency/mismatch of tag values between dnsmasq option files and dnsmasq command line. From the commit message: “This seems to be happening because of how dhcp_rpc.py::_group_by_network_id would behave if the list of subnets coming from the database was ordered different or contained different values.” After checking, we found that the latest Neutron seems to already have subnets sorted: https://github.com/openstack/neutron/blob/master/neutron/api/rpc/handlers/dhcp_rpc.py#L150L154. Would you please kindly help double check on this? Let me know if any question. Thank you! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From Allain.Legacy at windriver.com Fri Oct 12 13:10:33 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Fri, 12 Oct 2018 13:10:33 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for e5712fe In-Reply-To: References: Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC0D57B@ALA-MBD.corp.ad.wrs.com> Yes, your analysis is correct. Our patch (e5712fe) can be dropped because the issue has now been addressed by upstream commit (e45add7b07e5c72c43047d18e10af1c4ae307b0f). Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Friday, October 12, 2018 6:33 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Question about patch upstreaming for e5712fe Hi Matt, We're looking into the patch e5712fe, which indicates a possible inconsistency/mismatch of tag values between dnsmasq option files and dnsmasq command line. >From the commit message: "This seems to be happening because of how dhcp_rpc.py::_group_by_network_id would behave if the list of subnets coming from the database was ordered different or contained different values." After checking, we found that the latest Neutron seems to already have subnets sorted: https://github.com/openstack/neutron/blob/master/neutron/api/rpc/handlers/dhcp_rpc.py#L150L154. Would you please kindly help double check on this? Let me know if any question. Thank you! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From kailun.qin at intel.com Fri Oct 12 10:23:11 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Fri, 12 Oct 2018 10:23:11 +0000 Subject: [Starlingx-discuss] Network Segment Range Management - Spec Rework Review Message-ID: Hi Matt, Please kindly find the attached spec rework for StarlingX provider network management feature upstreaming [1][2]. Any further comments or suggestions would be greatly appreciated. BR, Kailun [1] https://review.openstack.org/599980 [2] https://review.openstack.org/579411 From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Thursday, October 11, 2018 7:38 PM To: Qin, Kailun Cc: Jolliffe, Ian ; Rowsell, Brent ; Khalil, Ghada ; Guo, Ruijing ; Le, Huifeng ; Xu, Chenjie ; Zhao, Forrest Subject: Re: Provider Segment Range Management - Spec Rework Review Hi Kailun, Attached are my latest review comments. Regards, Matt From: "Qin, Kailun" > Date: Tuesday, October 9, 2018 at 11:27 AM To: "Peters, Matt" > Cc: "Jolliffe, Ian" >, Brent Rowsell >, Ghada Khalil >, "Guo, Ruijing" >, "Le, Huifeng" >, "Xu, Chenjie" >, "Zhao, Forrest" > Subject: RE: Provider Segment Range Management - Spec Rework Review Matt, Thank you for all the comments. Please kindly see the updated spec in the attached. Let me know if any further suggestion. BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 9, 2018 9:17 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: RE: Provider Segment Range Management - Spec Rework Review See inline. From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Monday, October 08, 2018 5:40 AM To: Peters, Matt Cc: Jolliffe, Ian; Rowsell, Brent; Khalil, Ghada; Guo, Ruijing; Le, Huifeng; Xu, Chenjie; Zhao, Forrest Subject: RE: Provider Segment Range Management - Spec Rework Review Hi Matt, I am working on the update of spec. I have two main questions in response to your comments: 1. You recommended to use the generalized “network_segment_ranges” rather than including provider in the DB/API name and description. Do you also expect the whole spec to avoid using the “provider” terminology and we pick “network segment range management” as the global subject? [MP>] I think we should try to avoid linking this specifically to provider networks to avoid the same confusion and contention we received with the original spec. 2. I proposed an extended to attribute to the resource “networks” which you think is not required, as the network entity already has the “provider” extension that can be used to assign the required provider attributes upon creation. I agree with you. But the initiative of introducing this attribute, is to provide users with admin privilege the ability to specify the underlying provider segment *range* when creating networks (NOT just a specific segment ID -- what the “provider” extension does). This aims to cover the scenarios described in UseCase III of the spec, which are also proposed in this bp: https://blueprints.launchpad.net/neutron/+spec/extend-api-for-hpb. Our proposal is related w/ segment range and has a larger scope, which is able to cover the cases described in the cited bp if with that attribute. Do you suggest that we cover UseCase III in our bp? [MP>] I don’t feel we need to support a range at the network level. I feel the tenant level scope is sufficient and solves the use-cases we have encountered. If a customer really wants to select a segment ID for a network, they can use the provider extension rather than having it specifically selected from a range. I think we should focus on the main use case of having pools of managed segment ranges that can either be shared or assigned to a specific tenant. Please let me know what your thoughts are. Great thanks! BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 1, 2018 10:07 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: RE: Provider Segment Range Management - Spec Rework Review Attached are my review comments and suggested changes. NOTE: I had to change the format in order to embed my review comments, but it can continue to be an RST file format. Regards, Matt From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Thursday, September 27, 2018 10:36 AM To: Peters, Matt Cc: Jolliffe, Ian; Rowsell, Brent; Khalil, Ghada; Guo, Ruijing; Le, Huifeng; Qin, Kailun; Xu, Chenjie; Zhao, Forrest Subject: Provider Segment Range Management - Spec Rework Review Hi Matt, Please kindly find in the attached the drafted rework of provider segment range management spec. I made the updates directly on the upstream spec patch so that figures/tables/format can be kept as is. Excuse me for the inconvenience to make comments in a rst file. Please kindly leave a tag where you would like to comment so that I can filter them out. Great thanks! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: network-segment-range-management_v3.docx Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 21897 bytes Desc: network-segment-range-management_v3.docx URL: From Matt.Peters at windriver.com Fri Oct 12 13:23:53 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Fri, 12 Oct 2018 13:23:53 +0000 Subject: [Starlingx-discuss] Network Segment Range Management - Spec Rework Review In-Reply-To: References: Message-ID: <370612D7-ADDA-4F2C-9662-8BA53071FBB1@windriver.com> Hi Kailun, I have no further review comments. Once you receive the final feedback from others, I think it is ready to push it for review in the neutron-specs. Regards, Matt From: "Qin, Kailun" Date: Friday, October 12, 2018 at 6:24 AM To: "Peters, Matt" Cc: "starlingx-discuss at lists.starlingx.io" Subject: Network Segment Range Management - Spec Rework Review Hi Matt, Please kindly find the attached spec rework for StarlingX provider network management feature upstreaming [1][2]. Any further comments or suggestions would be greatly appreciated. BR, Kailun [1] https://review.openstack.org/599980 [2] https://review.openstack.org/579411 From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Thursday, October 11, 2018 7:38 PM To: Qin, Kailun Cc: Jolliffe, Ian ; Rowsell, Brent ; Khalil, Ghada ; Guo, Ruijing ; Le, Huifeng ; Xu, Chenjie ; Zhao, Forrest Subject: Re: Provider Segment Range Management - Spec Rework Review Hi Kailun, Attached are my latest review comments. Regards, Matt From: "Qin, Kailun" > Date: Tuesday, October 9, 2018 at 11:27 AM To: "Peters, Matt" > Cc: "Jolliffe, Ian" >, Brent Rowsell >, Ghada Khalil >, "Guo, Ruijing" >, "Le, Huifeng" >, "Xu, Chenjie" >, "Zhao, Forrest" > Subject: RE: Provider Segment Range Management - Spec Rework Review Matt, Thank you for all the comments. Please kindly see the updated spec in the attached. Let me know if any further suggestion. BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 9, 2018 9:17 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: RE: Provider Segment Range Management - Spec Rework Review See inline. From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Monday, October 08, 2018 5:40 AM To: Peters, Matt Cc: Jolliffe, Ian; Rowsell, Brent; Khalil, Ghada; Guo, Ruijing; Le, Huifeng; Xu, Chenjie; Zhao, Forrest Subject: RE: Provider Segment Range Management - Spec Rework Review Hi Matt, I am working on the update of spec. I have two main questions in response to your comments: 1. You recommended to use the generalized “network_segment_ranges” rather than including provider in the DB/API name and description. Do you also expect the whole spec to avoid using the “provider” terminology and we pick “network segment range management” as the global subject? [MP>] I think we should try to avoid linking this specifically to provider networks to avoid the same confusion and contention we received with the original spec. 2. I proposed an extended to attribute to the resource “networks” which you think is not required, as the network entity already has the “provider” extension that can be used to assign the required provider attributes upon creation. I agree with you. But the initiative of introducing this attribute, is to provide users with admin privilege the ability to specify the underlying provider segment *range* when creating networks (NOT just a specific segment ID -- what the “provider” extension does). This aims to cover the scenarios described in UseCase III of the spec, which are also proposed in this bp: https://blueprints.launchpad.net/neutron/+spec/extend-api-for-hpb. Our proposal is related w/ segment range and has a larger scope, which is able to cover the cases described in the cited bp if with that attribute. Do you suggest that we cover UseCase III in our bp? [MP>] I don’t feel we need to support a range at the network level. I feel the tenant level scope is sufficient and solves the use-cases we have encountered. If a customer really wants to select a segment ID for a network, they can use the provider extension rather than having it specifically selected from a range. I think we should focus on the main use case of having pools of managed segment ranges that can either be shared or assigned to a specific tenant. Please let me know what your thoughts are. Great thanks! BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 1, 2018 10:07 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: RE: Provider Segment Range Management - Spec Rework Review Attached are my review comments and suggested changes. NOTE: I had to change the format in order to embed my review comments, but it can continue to be an RST file format. Regards, Matt From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Thursday, September 27, 2018 10:36 AM To: Peters, Matt Cc: Jolliffe, Ian; Rowsell, Brent; Khalil, Ghada; Guo, Ruijing; Le, Huifeng; Qin, Kailun; Xu, Chenjie; Zhao, Forrest Subject: Provider Segment Range Management - Spec Rework Review Hi Matt, Please kindly find in the attached the drafted rework of provider segment range management spec. I made the updates directly on the upstream spec patch so that figures/tables/format can be kept as is. Excuse me for the inconvenience to make comments in a rst file. Please kindly leave a tag where you would like to comment so that I can filter them out. Great thanks! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From Frank.Miller at windriver.com Fri Oct 12 14:06:28 2018 From: Frank.Miller at windriver.com (Miller, Frank) Date: Fri, 12 Oct 2018 14:06:28 +0000 Subject: [Starlingx-discuss] Containerizing StarlingX Infrastructure Specs Message-ID: StarlingX Community: Last week Brent shared the Containerization project approach [1]. Specs for a subset of features on this project are starting to be posted for reviews [2]. Feedback from interested community members would be welcome and appreciated. [1]: https://wiki.openstack.org/wiki/Containerizing_StarlingX_Infrastructure [2]: https://review.openstack.org/#/q/projects:openstack/stx-specs Frank Miller Project Lead, StarlingX Containers -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Fri Oct 12 15:30:53 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Fri, 12 Oct 2018 15:30:53 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-11-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C34FBD1@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-11-20-r-2018.10.iso - 31 TCs PASS, 0 TCs FAIL - Simplex virtual getting affected by " openvswitch fails to allocate memory pool in virtual environment " [0] - unable to have this configuration enabled. - "Admin credentials are setup on both controllers after swact" [1] - seen on duplex bare metal and multinode bare metal. - Multinode external storage, both bare metal and virtual blocked by issue [2] (not able to create instances) - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 0/15/65/80 Simplex virtual - 0/10/61/71 Duplex bare metal - 1/41/58/100 Duplex virtual - 1/50/54/105 Multinode local storage bare metal - 1/51/66/118 Multinode local storage virtual - 4/57/57/118 Multinode external storage bare metal - 0/20/97/117 Multinode external storage virtual - 2/32/76/121 -- 11 blocked Full report in the wiki [3] [0] https://bugs.launchpad.net/starlingx/+bug/1796380 [1] https://bugs.launchpad.net/starlingx/+bug/1797422 [2] https://bugs.launchpad.net/starlingx/+bug/1797234 [3] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary From ada.cabrales at intel.com Fri Oct 12 16:49:36 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Fri, 12 Oct 2018 16:49:36 +0000 Subject: [Starlingx-discuss] Controller recovery question In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E4637@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E4637@fmsmsx101.amr.corp.intel.com> Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C34FC3B@fmsmsx104.amr.corp.intel.com> Asking again: does anyone know ... If the /etc/init.d/sm file in a controller is lost, is the system able to recover it by itself? Regards A. > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Tuesday, October 9, 2018 10:01 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Controller recovery question > > Hi > > I'm running a test scenario where a critical process is killed and I have to > verify that Controller is recovered, with below steps I'm seeing that swat > happens correctly but controller with the process killed keeps rebooting and > process killed is not regenerated, my question is if the scenario is valid and > the sm script should be regenerated. > > ============================================================== > ====== > Prep: > > Perform the following steps on the standy controller: > > a) Locate a critical process that is pmon managed: > > -> cd /etc/pmon.d > -> grep -r "severity = critical" > > b) E.g. cat /etc/pmon.d/sm.conf > c) Locate the name of the daemon pid file or startup script file that is > configured in the conf file: > e.g. server.pid-file = "/var/run/lighttpd.pid" or script = /etc/init.d/sm > c) Rename the startup script file so that the critical process is not respawned > after it is killed: > e.g. mv /etc/init.d/sm /etc/init.d/sm~ > > Execution: > 1) Kill the critical process on the standby controller sudo pkill sm > > > **** EXPECTED ********* > 1) Verify that the controller goes for a reset and recovers. > ============================================================== > =========== > > Regards, > José > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From dtroyer at gmail.com Fri Oct 12 18:23:45 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Fri, 12 Oct 2018 13:23:45 -0500 Subject: [Starlingx-discuss] [tsc][governance] Governance repo updates Message-ID: For everyone interested in setting up the StarlingX Governance: I have begun some updates to the stx-governance repo: [0] makes some changes to the document structure by adding a 'tsc' directory under /reference to leave space for other things to be co-located here (like a StarlingX-specific WG or something similar) [0] also adds some text to the landing page that parallels the OpenStack Governance a bit and starts the list of document links on the TSC landing page with the Four Opens and our Charter. I am not using :toctree: here because we will have a mixture of external links and internal documents. [0] also adds a Creative Commons Attribution 3.0 Unported (CC-BY-3.0) LICENSE file to the top of the repo similar to how we have LICENSE files containing the Apache license in our code repos. This raises a question of if we should also include CC-BY-3.0 in code repos that we publish documentation for to cover the docs? [1] adds some code taken from the TC and UC governance repos that produces a table of TSC members from a data file. This adds Python to the repo so all of the usual support bits are also included here. I anticipate doing something similar in the future for the project list but first we have to create the project list... [2] adds the "house rules" doc we talked about in the 11Oct2018 TSC meeting. I have generalized this a bit from the OpenStack TC usage to include the things we decide pertaining to operation of TSC functions that are not included in the Charter. I think details do not belong in charters but still need to be recorded. I do not expect to merge this review as-is, please add suggestions or thoughts here... I plan to continue fleshing out this repo over the next couple of days, if you have suggestions for priorities please include them here...this is what is on my list today: * StarlingX Project list (projects that we are the upstream for, excluding all forks - they should be in a separate list) * Release naming * Project Testing (we have thus far adopted the OpenStack PTI wholesale, point there [3]) * Licensing requirements (point to OpenStack[4]?) * Election process (not the top priority (yet)) dt [0] https://review.openstack.org/#/c/609814/ [1] https://review.openstack.org/#/c/609817/ [2] https://review.openstack.org/#/c/609842/ [3] https://governance.openstack.org/tc/reference/project-testing-interface.html [4] https://governance.openstack.org/tc/reference/licensing.html -- Dean Troyer dtroyer at gmail.com From Allain.Legacy at windriver.com Fri Oct 12 18:59:37 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Fri, 12 Oct 2018 18:59:37 +0000 Subject: [Starlingx-discuss] Controller recovery question In-Reply-To: <4F6AACE4B0F173488D033B02A8BB5B7E7C34FC3B@fmsmsx104.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E4637@fmsmsx101.amr.corp.intel.com> <4F6AACE4B0F173488D033B02A8BB5B7E7C34FC3B@fmsmsx104.amr.corp.intel.com> Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC0DA07@ALA-MBD.corp.ad.wrs.com> /etc/init.d/sm is a packaged file; it is not created dynamically. Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279  fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5   > -----Original Message----- > From: Cabrales, Ada [mailto:ada.cabrales at intel.com] > Sent: Friday, October 12, 2018 12:50 PM > To: Perez Carranza, Jose; starlingx-discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] Controller recovery question > > Asking again: does anyone know ... > > If the /etc/init.d/sm file in a controller is lost, is the system able to recover it > by itself? > > Regards > A. > > > > -----Original Message----- > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > Sent: Tuesday, October 9, 2018 10:01 AM > > To: starlingx-discuss at lists.starlingx.io > > Subject: [Starlingx-discuss] Controller recovery question > > > > Hi > > > > I'm running a test scenario where a critical process is killed and I > > have to verify that Controller is recovered, with below steps I'm > > seeing that swat happens correctly but controller with the process > > killed keeps rebooting and process killed is not regenerated, my > > question is if the scenario is valid and the sm script should be regenerated. > > > > > ========================================================== > ==== > > ====== > > Prep: > > > > Perform the following steps on the standy controller: > > > > a) Locate a critical process that is pmon managed: > > > > -> cd /etc/pmon.d > > -> grep -r "severity = critical" > > > > b) E.g. cat /etc/pmon.d/sm.conf > > c) Locate the name of the daemon pid file or startup script file that > > is configured in the conf file: > > e.g. server.pid-file = "/var/run/lighttpd.pid" or script = > > /etc/init.d/sm > > c) Rename the startup script file so that the critical process is not > > respawned after it is killed: > > e.g. mv /etc/init.d/sm /etc/init.d/sm~ > > > > Execution: > > 1) Kill the critical process on the standby controller sudo pkill sm > > > > > > **** EXPECTED ********* > > 1) Verify that the controller goes for a reset and recovers. > > > ========================================================== > ==== > > =========== > > > > Regards, > > José > > > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From scott.little at windriver.com Fri Oct 12 19:17:34 2018 From: scott.little at windriver.com (Scott Little) Date: Fri, 12 Oct 2018 15:17:34 -0400 Subject: [Starlingx-discuss] [Build] Build Avoidance In-Reply-To: <06fb400c-4e93-c765-6385-9e579e2b3ae7@windriver.com> References: <06fb400c-4e93-c765-6385-9e579e2b3ae7@windriver.com> Message-ID: I've posted the Build Avoidance wiki https://wiki.openstack.org/wiki/StarlingX/Developer_Guide#Build_Avoidance Reviews welcome. We can relocate it if there is a better home folks can suggest. Scott On 18-08-27 04:17 PM, Scott Little wrote: > > Build Avoidance, a build tool improvement.* > * > > *Purpose:* > >    Greatly reduce build times after a repo sync for designers working > within a regional office.  For a new workspace, build-pkgs typically > requires 3+ hours, build avoidance typically reduces this step to ~20min. > > *Limitations:* > >    Little or no benefit for designers who refresh a pre-existing > workspace at least daily. (download_mirror.sh, repo sync, > generate-cgcs-centos-repo.sh, build-pkgs, build-iso). > >    Not likely to be useful to solo designers, or teleworkers that wish > to compile on there home computers.  WAN speeds are generally to slow. > > *Method (in brief):* > > 1) Reference builds > >    - A server performs a regular (daily?), automated builds using > existing methods.  Call these the reference builds. >    - The builds are timestamped, and preserved for some time. (weeks?) >    - A build CONTEXT is captured, consisting of the SHA of each and > every git that contributed to the build. >    - For each package built, a file shall capture he md5sums of all > the source code inputs to the build of that package. >    - All these build products are accessible locally (e.g. a regional > office) via rsync (other protocols can be added later) > > 2) Designers >    - build-pkgs --build-avoidance   ... will request a build avoidance > build. >    - Additional arguments, and/or environment variables, and/or a > config file unique to the regional office, are used to specify a URL > to the reference builds. >    - build-pkgs will: >      = From newest to oldest, scan the CONTEXTs of the various > reference builds.  Select the first (most recent) context which > satisfies:  For every git, the SHA specified in the CONTEXT is present. >      = The selected context might be slightly out of date, but not by > more than a day (assuming daily reference builds). >      = If the context has not been previously downloaded, then > download it now.  Meaning download select portions of the reference > build workspace into the designer's workspace.  This includes all the > SRPMS, RPMS, MD5SUMS, and misc supporting files. (~10 min over office LAN) >      = The designer may have additional commits not present in the > reference build, or uncommitted changes.   Affected packages will > identified by the differing md5sum's, and the package is re-built. (5+ > min, depending on what packages have changed) > > > *Requirement:* > >    - The regional office implements an automated build that pulls the > latest StarlingX software and builds it on a regular basis. e.g. a > daily.  Perhaps implemented by Jenkins, cron, or similar tools. >    - Each build is saved to a unique directory, and preserved for a > time that is reflective of how long a designer might be expected to > work on a private branch without syncronizing with the master > branch.   e.g. 2 weeks. >    - The MY_WORKSPACE directory for the build shall have a common root > directory, and a leaf directory that is a time stamp of format > YYYY-MM-DD_hh-mm-ss.  e.g. > MY_WORKSPACE=/localdisk/loadbuild/jenkins/StarlingX/2018-07-19_11-30-21 >    - Designers can access all build products over the internal network > of the regional office.  The current prototype employs rsync. Other > protocols that can efficiently share/copy/transfer large directories > of content can be added as needed. > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Fri Oct 12 20:11:46 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Fri, 12 Oct 2018 20:11:46 +0000 Subject: [Starlingx-discuss] Controller recovery question In-Reply-To: <70A7408C6E1BFB41B192A929744D8523BAC0DA07@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E4637@fmsmsx101.amr.corp.intel.com> <4F6AACE4B0F173488D033B02A8BB5B7E7C34FC3B@fmsmsx104.amr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0DA07@ALA-MBD.corp.ad.wrs.com> Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C34FCEF@fmsmsx104.amr.corp.intel.com> Thanks :) > -----Original Message----- > From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] > Sent: Friday, October 12, 2018 2:00 PM > To: Cabrales, Ada ; Perez Carranza, Jose > ; starlingx-discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Controller recovery question > > /etc/init.d/sm is a packaged file; it is not created dynamically. > > Regards, > Allain > > Allain Legacy, Software Developer, Wind River direct 613.270.2279  fax > 613.492.7870 skype allain.legacy > 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 > > > > > > -----Original Message----- > > From: Cabrales, Ada [mailto:ada.cabrales at intel.com] > > Sent: Friday, October 12, 2018 12:50 PM > > To: Perez Carranza, Jose; starlingx-discuss at lists.starlingx.io > > Subject: Re: [Starlingx-discuss] Controller recovery question > > > > Asking again: does anyone know ... > > > > If the /etc/init.d/sm file in a controller is lost, is the system able > > to recover it by itself? > > > > Regards > > A. > > > > > > > -----Original Message----- > > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > > Sent: Tuesday, October 9, 2018 10:01 AM > > > To: starlingx-discuss at lists.starlingx.io > > > Subject: [Starlingx-discuss] Controller recovery question > > > > > > Hi > > > > > > I'm running a test scenario where a critical process is killed and I > > > have to verify that Controller is recovered, with below steps I'm > > > seeing that swat happens correctly but controller with the process > > > killed keeps rebooting and process killed is not regenerated, my > > > question is if the scenario is valid and the sm script should be > regenerated. > > > > > > > > ========================================================== > > ==== > > > ====== > > > Prep: > > > > > > Perform the following steps on the standy controller: > > > > > > a) Locate a critical process that is pmon managed: > > > > > > -> cd /etc/pmon.d > > > -> grep -r "severity = critical" > > > > > > b) E.g. cat /etc/pmon.d/sm.conf > > > c) Locate the name of the daemon pid file or startup script file > > > that is configured in the conf file: > > > e.g. server.pid-file = "/var/run/lighttpd.pid" or script = > > > /etc/init.d/sm > > > c) Rename the startup script file so that the critical process is > > > not respawned after it is killed: > > > e.g. mv /etc/init.d/sm /etc/init.d/sm~ > > > > > > Execution: > > > 1) Kill the critical process on the standby controller sudo pkill sm > > > > > > > > > **** EXPECTED ********* > > > 1) Verify that the controller goes for a reset and recovers. > > > > > ========================================================== > > ==== > > > =========== > > > > > > Regards, > > > José > > > > > > > > > _______________________________________________ > > > Starlingx-discuss mailing list > > > Starlingx-discuss at lists.starlingx.io > > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Numan.Waheed at windriver.com Fri Oct 12 20:30:44 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Fri, 12 Oct 2018 20:30:44 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 20181012 Message-ID: <3CAA827B7A79BA46B15B280EC82088FE4820FA3E@ALA-MBD.corp.ad.wrs.com> Below is the testing status for 20181012. Sanity: Sanity Results Config All-in-One Simplex Build ID 2018-10-10_01-52-00 Total 11 Pass 11 Fail 0 Bugs None Config All-in-One Duplex Build ID 2018-10-12_01-52-00 Total 15 Pass 15 Fail 0 Bugs None Config Multi Node Build ID 2018-10-08_01-52-01 Total 23 Pass 23 Fail 0 Bugs None Config Multi Node with CEPH Build ID 2018-10-10_01-52-00 Total 24 Pass 24 Fail 0 Bugs None Manual Regression: Domains Total Test Cases Passed Failed Total Executed Launchpad Backup and Restore 4 3 1 4 https://bugs.launchpad.net/starlingx/+bug/1796404 Distributed Cloud 30 15 0 15 Installation and Config 5 3 2 5 https://bugs.launchpad.net/starlingx/+bug/1797187 https://bugs.launchpad.net/starlingx/+bug/1797608 MTC 19 19 0 19 Networking 22 10 4 14 https://bugs.launchpad.net/starlingx/+bug/1796198 https://bugs.launchpad.net/starlingx/+bug/1790968 Nova 35 15 6 21 https://bugs.launchpad.net/starlingx/+bug/1796393 Security 17 8 3 11 https://bugs.launchpad.net/starlingx/+bug/1797204 https://bugs.launchpad.net/starlingx/+bug/1797219 https://bugs.launchpad.net/starlingx/+bug/1797569 Storage 11 4 5 9 https://bugs.launchpad.net/starlingx/+bug/1795962 https://bugs.launchpad.net/starlingx/+bug/1796913 https://bugs.launchpad.net/starlingx/+bug/1797151 https://bugs.launchpad.net/starlingx/+bug/1797391 System Inventory 8 7 1 8 https://bugs.launchpad.net/starlingx/+bug/1797187 System Test 10 8 0 8 Total 161 92 22 114 Thanks, [WR logo signiture] Numan Waheed, Manager, Titanium Cloud Testing, Wind River Direct: 613.270.5762 * Mobile: 613.793.5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.jpg Type: image/jpeg Size: 1416 bytes Desc: image003.jpg URL: From ada.cabrales at intel.com Fri Oct 12 23:08:29 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Fri, 12 Oct 2018 23:08:29 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-12-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C350F75@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-12-21-r-2018.10.iso - 31 TCs PASS, 0 TCs FAIL - Multinode external storage, both bare metal and virtual unblocked. Launchpad [0] updated with steps followed. - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 0/31/47/78 Simplex virtual - 0/10/61/71 Duplex bare metal - 1/41/58/100 Duplex virtual - 1/56/47/104 Multinode local storage bare metal - 1/51/65/117 Multinode local storage virtual - 4/57/57/118 Multinode external storage bare metal - 0/48/69/117 Multinode external storage virtual - 2/40/78/120 Full report in the wiki [1] [0] https://bugs.launchpad.net/starlingx/+bug/1797234 [1] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary Happy weekend! A. From kailun.qin at intel.com Mon Oct 15 01:58:32 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Mon, 15 Oct 2018 01:58:32 +0000 Subject: [Starlingx-discuss] Network Segment Range Management - Spec Rework Review In-Reply-To: <370612D7-ADDA-4F2C-9662-8BA53071FBB1@windriver.com> References: <370612D7-ADDA-4F2C-9662-8BA53071FBB1@windriver.com> Message-ID: Hi Matt, Thanks a lot for all the feedbacks. I’ve pushed the updated spec for review and it is now available @ https://review.openstack.org/599980. BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Friday, October 12, 2018 9:24 PM To: Qin, Kailun Cc: starlingx-discuss at lists.starlingx.io Subject: Re: Network Segment Range Management - Spec Rework Review Hi Kailun, I have no further review comments. Once you receive the final feedback from others, I think it is ready to push it for review in the neutron-specs. Regards, Matt From: "Qin, Kailun" > Date: Friday, October 12, 2018 at 6:24 AM To: "Peters, Matt" > Cc: "starlingx-discuss at lists.starlingx.io" > Subject: Network Segment Range Management - Spec Rework Review Hi Matt, Please kindly find the attached spec rework for StarlingX provider network management feature upstreaming [1][2]. Any further comments or suggestions would be greatly appreciated. BR, Kailun [1] https://review.openstack.org/599980 [2] https://review.openstack.org/579411 From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Thursday, October 11, 2018 7:38 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: Re: Provider Segment Range Management - Spec Rework Review Hi Kailun, Attached are my latest review comments. Regards, Matt From: "Qin, Kailun" > Date: Tuesday, October 9, 2018 at 11:27 AM To: "Peters, Matt" > Cc: "Jolliffe, Ian" >, Brent Rowsell >, Ghada Khalil >, "Guo, Ruijing" >, "Le, Huifeng" >, "Xu, Chenjie" >, "Zhao, Forrest" > Subject: RE: Provider Segment Range Management - Spec Rework Review Matt, Thank you for all the comments. Please kindly see the updated spec in the attached. Let me know if any further suggestion. BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 9, 2018 9:17 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: RE: Provider Segment Range Management - Spec Rework Review See inline. From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Monday, October 08, 2018 5:40 AM To: Peters, Matt Cc: Jolliffe, Ian; Rowsell, Brent; Khalil, Ghada; Guo, Ruijing; Le, Huifeng; Xu, Chenjie; Zhao, Forrest Subject: RE: Provider Segment Range Management - Spec Rework Review Hi Matt, I am working on the update of spec. I have two main questions in response to your comments: 1. You recommended to use the generalized “network_segment_ranges” rather than including provider in the DB/API name and description. Do you also expect the whole spec to avoid using the “provider” terminology and we pick “network segment range management” as the global subject? [MP>] I think we should try to avoid linking this specifically to provider networks to avoid the same confusion and contention we received with the original spec. 2. I proposed an extended to attribute to the resource “networks” which you think is not required, as the network entity already has the “provider” extension that can be used to assign the required provider attributes upon creation. I agree with you. But the initiative of introducing this attribute, is to provide users with admin privilege the ability to specify the underlying provider segment *range* when creating networks (NOT just a specific segment ID -- what the “provider” extension does). This aims to cover the scenarios described in UseCase III of the spec, which are also proposed in this bp: https://blueprints.launchpad.net/neutron/+spec/extend-api-for-hpb. Our proposal is related w/ segment range and has a larger scope, which is able to cover the cases described in the cited bp if with that attribute. Do you suggest that we cover UseCase III in our bp? [MP>] I don’t feel we need to support a range at the network level. I feel the tenant level scope is sufficient and solves the use-cases we have encountered. If a customer really wants to select a segment ID for a network, they can use the provider extension rather than having it specifically selected from a range. I think we should focus on the main use case of having pools of managed segment ranges that can either be shared or assigned to a specific tenant. Please let me know what your thoughts are. Great thanks! BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 1, 2018 10:07 PM To: Qin, Kailun > Cc: Jolliffe, Ian >; Rowsell, Brent >; Khalil, Ghada >; Guo, Ruijing >; Le, Huifeng >; Xu, Chenjie >; Zhao, Forrest > Subject: RE: Provider Segment Range Management - Spec Rework Review Attached are my review comments and suggested changes. NOTE: I had to change the format in order to embed my review comments, but it can continue to be an RST file format. Regards, Matt From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Thursday, September 27, 2018 10:36 AM To: Peters, Matt Cc: Jolliffe, Ian; Rowsell, Brent; Khalil, Ghada; Guo, Ruijing; Le, Huifeng; Qin, Kailun; Xu, Chenjie; Zhao, Forrest Subject: Provider Segment Range Management - Spec Rework Review Hi Matt, Please kindly find in the attached the drafted rework of provider segment range management spec. I made the updates directly on the upstream spec patch so that figures/tables/format can be kept as is. Excuse me for the inconvenience to make comments in a rst file. Please kindly leave a tag where you would like to comment so that I can filter them out. Great thanks! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From kailun.qin at intel.com Mon Oct 15 02:05:07 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Mon, 15 Oct 2018 02:05:07 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for e5712fe In-Reply-To: <70A7408C6E1BFB41B192A929744D8523BAC0D57B@ALA-MBD.corp.ad.wrs.com> References: <70A7408C6E1BFB41B192A929744D8523BAC0D57B@ALA-MBD.corp.ad.wrs.com> Message-ID: Hi Allain and Matt, Thanks for confirming. We've retested the scenario w/o the change e5712fe and cannot reproduce the issue. I think it's OK to drop the original patch e5712fe. BR, Kailun From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Friday, October 12, 2018 9:11 PM To: Qin, Kailun ; Peters, Matt Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for e5712fe Yes, your analysis is correct. Our patch (e5712fe) can be dropped because the issue has now been addressed by upstream commit (e45add7b07e5c72c43047d18e10af1c4ae307b0f). Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Friday, October 12, 2018 6:33 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Question about patch upstreaming for e5712fe Hi Matt, We're looking into the patch e5712fe, which indicates a possible inconsistency/mismatch of tag values between dnsmasq option files and dnsmasq command line. >From the commit message: "This seems to be happening because of how dhcp_rpc.py::_group_by_network_id would behave if the list of subnets coming from the database was ordered different or contained different values." After checking, we found that the latest Neutron seems to already have subnets sorted: https://github.com/openstack/neutron/blob/master/neutron/api/rpc/handlers/dhcp_rpc.py#L150L154. Would you please kindly help double check on this? Let me know if any question. Thank you! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From huifeng.le at intel.com Mon Oct 15 02:23:54 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Mon, 15 Oct 2018 02:23:54 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> Hi Matt, We're looking into the patch 1e9a089, which target to solve the issue of "CGTS-2408: bypass policy check when setting attribute to None." After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: "type: uuid_or_none" (2) the "enforce_policy" should be set to "True" and the policy needs to be defined in etc/policy.json (3) the check policy uses "tenant_id" for owner check The only found attribute which will be impacted by this issue in STX is "wrs-tm:qos" which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What's your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From kailun.qin at intel.com Mon Oct 15 09:00:12 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Mon, 15 Oct 2018 09:00:12 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for a38f899 Message-ID: Hi Matt, Allain, I am analyzing the patch a38f899 for upstreaming. It seems to me that parts of the patch have already been addressed by upstream, while some of the rest may need your double confirm/clarification. Please kindly check the attached analysis report for further details. Let me know if any question. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Analysis Report for patch a38f899.docx Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 17208 bytes Desc: Analysis Report for patch a38f899.docx URL: From zhipengs.liu at intel.com Mon Oct 15 09:26:38 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Mon, 15 Oct 2018 09:26:38 +0000 Subject: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. Message-ID: <93814834B4855241994F290E959305C752FC8ABD@SHSMSX104.ccr.corp.intel.com> Hi all, Recently I have analyzed all init patches of non-openstack patches that you can filter by "init" in "Analysis" column of below sheet https://docs.google.com/spreadsheets/d/1nKnkweuxcqvVOoRcpnTYMVUUv1RoAugOWXMjB7VIrfc/edit#gid=1953568313 I have added my comments in column of "Comments from zhipeng" 1) The comments colored with blue means we can add service to /etc/systemd/system/ so that it can be started during power on. 2) The comments colored with orange means we can ADD config files or scripts to specific system folder. 3) The comments colored with green means we need to use our customized config files or scripts instead of default ones. For both case 1 and 2 mentioned above, we just need to consider initial installation scenario. (22 patches) I worked out a method to remove related patches like below steps. a) Use platform-util RPM to copy all related config/service files to /tmp folder. Add config/service files to platform-util/centos/files/ Modify platform-util.spec to copy all related config/service files to /tmp folder. b) Add a post-installation script to kickstart config file. (This script will be called by anaconda after software installation finished.) This post-installation script will copy config/service files to specific system folder. c) Then we can remove related patches. For case 3, we need to consider both initial installation and SW update scenarios. (44 patches) a) Initial installation We can do the same as we do for case 1 and 2 b) SW update Add a post-update script that will be called by sw-patch-agent after sw update finished. Use post-update script to copy customized config files or scripts to specific system folders, overwrite default ones. So far, I have started working on case 1 and 2, finished nfs-utils (Story: 2003768/Task: 26461) and verified deployment. You can review my patch below. https://review.openstack.org/#/c/610456/ https://review.openstack.org/#/c/610459/ Any concern? Welcome to add your comment and proposal! Thanks! Zhipeng From cindy.xie at intel.com Mon Oct 15 10:38:36 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Mon, 15 Oct 2018 10:38:36 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F35D69087@SHSMSX104.ccr.corp.intel.com> References: <2FD5DDB5A04D264C80D42CA35194914F35D69087@SHSMSX104.ccr.corp.intel.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D6F3DC@SHSMSX104.ccr.corp.intel.com> Fuzz patch guideline has been added: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines#StarlingX_Code_Submission_Guidelines thx. - cindy -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 9:37 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Agenda & Notes for 10/10 meeting: 1. Bugs for 2018.10 review (Ghada) no pending storyboard & bugs. Cheers! 2. non-OpenStack patch analysis status (Saul/Zhipeng) init related patches, current build related ones merged by Austin; Zhipeng is looking into solutions for the init related patches to remove from individual RPMs. deep dive to some upstream source level patches and analysis WIP. Brent submitted several new stories and please mark it as "stx-non-distro" tag. 3. Fuzz patch guideline and follow-up actions (Brent/Shuicheng) Going forward we should put guideline for strictly no fuzz patch. AR: Cindy to put a developer guideline to: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines https://storyboard.openstack.org/#!/story/2003917 include all fuzz patches, most of them are merged. Shuicheng will add more tasks (few left) within this week. There is not much we can do w/ the upstreamed patches from sRPM for now. story for upgrade rpm to v14 is not required for now. rpm v14 requires OpenSSH v1.1 and we cannot upgrade from a lower version. For now, the system limitation is that we cannot install the same package with different version simultaneously. 4. Kernel driver upgrade status (Shuicheng) https://storyboard.openstack.org/#!/story/2003597, all patches are ready and pending for testing. Intel GDC do not have appropriate test systems available. AR: Cindy to coordinate Ada about the HW required to run the testing. Mellanox driver version: how to define the version# to upgrade? choose the latest version. usually the version we pick shall be aligned w/ the version in the DPDK supported. Need to look for the version from the Mellanox page. AR: Shuicheng to check the version to make sure it compatible w/ DPDK version in our current mainline. 5. Ceph upgrade status (Vivian/Dehao) - no update due to holiday 6. Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From chris.friesen at windriver.com Mon Oct 15 16:25:59 2018 From: chris.friesen at windriver.com (Chris Friesen) Date: Mon, 15 Oct 2018 10:25:59 -0600 Subject: [Starlingx-discuss] stx-specs git repo directory structure Message-ID: Hi, In the "stx-specs" git repo [1] the directory structure currently looks like "specs/approved". That's fine for this release, but once we get to next release we're going to have "old" specs still in there making it hard to keep track of which spec is for which release. I'm wondering whether we want to add a release-specific subdirectory like other OpenStack projects have. So specs being submitted now would go into "specs/2019.3/approved". Thanks, Chris [1] https://github.com/openstack/stx-specs From Matt.Peters at windriver.com Mon Oct 15 16:55:23 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Mon, 15 Oct 2018 16:55:23 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> Message-ID: Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don't even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We're looking into the patch 1e9a089, which target to solve the issue of "CGTS-2408: bypass policy check when setting attribute to None." After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: "type: uuid_or_none" (2) the "enforce_policy" should be set to "True" and the policy needs to be defined in etc/policy.json (3) the check policy uses "tenant_id" for owner check The only found attribute which will be impacted by this issue in STX is "wrs-tm:qos" which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What's your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matt.Peters at windriver.com Mon Oct 15 17:05:29 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Mon, 15 Oct 2018 17:05:29 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for a38f899 In-Reply-To: References: Message-ID: Based on the original commit message (see below), it would fall into scenario #2. However, as the commit message indicates, this is based on race conditions when running tempest, therefore it may not be easily reproducible. CGTS-3885: dhcp: handle concurrent port creation errors Running tempest nose tests exposes a race condition between the server deleting a subnet and an agent creating a DHCP port. This problem manifests itself in different ways depending on the whether the error is returned to the subnet delete operation or the DHCP port creation. This Jira (3885) addresses how the error is handled by the agent when the subnet is concurrently deleted at the server. The changes to address the error that is returned to the API client are in CGTS-3432. Because of how the server and agents are designed there is a lack of determinism in how this particular issue manifests itself. The variants manifest themselves based on when the subnet is deleted in relation to when the port is created. 1) if the port is returned with no fixed IP address then this results in the agent failing to setup the default route in the namespace because there is no local IP address to use to contact the gateway IP address. 2) if the port is updated after creation to no longer have an IP address then 'reload_allocations' action is invoked instead of the 'disabled' action. This leads to a failure to reenable because the subnet no longer exits. 3) if doing a 'disable' action it is possible that previous 'restart', 'enable', or 'reload' actions did not get far enough to create the namespace. For this reason we should not throw an error if the namespace does not exist. 4) if attempting a 'reload_allocations' we should fail the request if the namespace does not exist. 5) if attempting a 'reload_allocations' with retain_port=True we should not delete the configuration files because the vswitch port name is stored in the 'interface' file. If an error occurs during the enable action we will not be able to 'unplug' the port from the vswitch because we will not know which name to use during the 'unplug'. This will leak ports in the vswitch. From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Monday, October 15, 2018 5:00 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Question about patch upstreaming for a38f899 Hi Matt, Allain, I am analyzing the patch a38f899 for upstreaming. It seems to me that parts of the patch have already been addressed by upstream, while some of the rest may need your double confirm/clarification. Please kindly check the attached analysis report for further details. Let me know if any question. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From corey at corebits.ca Mon Oct 15 17:20:54 2018 From: corey at corebits.ca (Corey Erickson) Date: Mon, 15 Oct 2018 11:20:54 -0600 Subject: [Starlingx-discuss] STX + Tungsten Fabric Message-ID: Hi All, Curious to know if anyone is active in working on SDN integrations such as Tungsten Fabric ? Thanks, Corey Erickson -------------- next part -------------- An HTML attachment was scrubbed... URL: From claire at openstack.org Mon Oct 15 17:42:49 2018 From: claire at openstack.org (Claire Massey) Date: Mon, 15 Oct 2018 12:42:49 -0500 Subject: [Starlingx-discuss] Invitation, October 24 StarlingX First Release & Community On-Boarding Webinar Message-ID: <323CACC5-032C-4728-9626-B8E6AF6DF536@openstack.org> Hi everyone, On October 24 at 8:00am PDT we will hold a Community On-Boarding webinar to coincide with the first release for the StarlingX project. Ian Jolliffe and Bruce Jones will lead the on-boarding presentation. Please mark your calendars and plan to join us! Meeting link: https://zoom.us/j/112003649 . The webinar will be recorded if you’re unable to make it. More information and news around the project’s first release will be shared on October 24. Stay tuned! Thanks, Claire -------------- next part -------------- An HTML attachment was scrubbed... URL: From claire at openstack.org Mon Oct 15 17:42:44 2018 From: claire at openstack.org (claire at openstack.org) Date: Mon, 15 Oct 2018 17:42:44 +0000 Subject: [Starlingx-discuss] Invitation: StarlingX First Release, Community Webinar @ Wed Oct 24, 2018 10am - 11am (CDT) (starlingx-discuss@lists.starlingx.io) Message-ID: <000000000000f0653a057847f626@google.com> You have been invited to the following event. Title: StarlingX First Release, Community Webinar https://etherpad.openstack.org/p/openstack-community-meeting When: Wed Oct 24, 2018 10am – 11am Central Time - Chicago Where: https://zoom.us/j/112003649 Calendar: starlingx-discuss at lists.starlingx.io Who: * claire at openstack.org - organizer * Ildiko Vancsa * Allison Price * Ian Jolliffe * scott.w.doenecke at intel.com * Bruce E Jones * Chris Hoge * Jeff * glenn.seiler at windriver.com * Jennifer Fowler * Lauren Sell * Robert Cathey * starlingx-discuss at lists.starlingx.io Event details: https://www.google.com/calendar/event?action=VIEW&eid=XzhoMTNnZ3BvOGwxM2NiOWw4Z3I0MmI5azhoMzNpYmExOGtvazZiYTY3MHEzZ2NxNDg1MGthYzFnNmsgc3Rhcmxpbmd4LWRpc2N1c3NAbGlzdHMuc3Rhcmxpbmd4Lmlv&tok=MjAjY2xhaXJlQG9wZW5zdGFjay5vcmc0YjdkMzYwMzNkY2NjNjAxNGFmYjQ4Y2MzMGY4NGU3NGRkNmI0OTU3&ctz=America%2FChicago&hl=en&es=0 Invitation from Google Calendar: https://www.google.com/calendar/ You are receiving this courtesy email at the account starlingx-discuss at lists.starlingx.io because you are an attendee of this event. To stop receiving future updates for this event, decline this event. Alternatively you can sign up for a Google account at https://www.google.com/calendar/ and control your notification settings for your entire calendar. Forwarding this invitation could allow any recipient to modify your RSVP response. Learn more at https://support.google.com/calendar/answer/37135#forwarding -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 3160 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: invite.ics Type: application/ics Size: 3220 bytes Desc: not available URL: From dtroyer at gmail.com Mon Oct 15 17:45:33 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 15 Oct 2018 12:45:33 -0500 Subject: [Starlingx-discuss] stx-specs git repo directory structure In-Reply-To: References: Message-ID: On Mon, Oct 15, 2018 at 11:27 AM Chris Friesen wrote: > I'm wondering whether we want to add a release-specific subdirectory > like other OpenStack projects have. So specs being submitted now would > go into "specs/2019.3/approved". Yes we should do that, probably sooner than later if we want to be able to deep-link to them after they are approved. It looks like there are 5 in the queue right now, anyone care either way if we wait for those to be approved or cause them all to need a rebase? dt -- Dean Troyer dtroyer at gmail.com From Frank.Miller at windriver.com Mon Oct 15 17:50:15 2018 From: Frank.Miller at windriver.com (Miller, Frank) Date: Mon, 15 Oct 2018 17:50:15 +0000 Subject: [Starlingx-discuss] stx-specs git repo directory structure In-Reply-To: References: Message-ID: Looks like 1 is merged already and 6 are in the queue. I think we can let the commit to create the 2019.03 directory merge first and ask the 6 people with outstanding specs to rebase. Chris - as part of your commit to create the 2019.03 directory can you also move the one spec that already merged into the new 2019.03 directory? Frank -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Monday, October 15, 2018 1:46 PM To: Friesen, Chris Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] stx-specs git repo directory structure On Mon, Oct 15, 2018 at 11:27 AM Chris Friesen wrote: > I'm wondering whether we want to add a release-specific subdirectory > like other OpenStack projects have. So specs being submitted now > would go into "specs/2019.3/approved". Yes we should do that, probably sooner than later if we want to be able to deep-link to them after they are approved. It looks like there are 5 in the queue right now, anyone care either way if we wait for those to be approved or cause them all to need a rebase? dt -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From scott.little at windriver.com Mon Oct 15 18:56:43 2018 From: scott.little at windriver.com (Scott Little) Date: Mon, 15 Oct 2018 14:56:43 -0400 Subject: [Starlingx-discuss] repo sync failures: 503 error in ima-evm-utils.git Message-ID: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> We are currently seeing the following error from 'repo sync' up here in Ottawa.    fatal: unable to access 'https://git.code.sf.net/p/linux-ima/ima-evm-utils.git/': The requested URL returned error: 503 Timestamp is roughly 2018-10-15 18:45 UTC Can other sites see this ? *Workaround:* If you already have an established repo, including ima, you can use the following to refresh your repo ...    repo sync  --optimized-fetch This works because ima in our manifest is anchored to a specific tag/sha, which you already have downloaded, so no need to refresh that specific git. -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtroyer at gmail.com Mon Oct 15 19:26:23 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 15 Oct 2018 14:26:23 -0500 Subject: [Starlingx-discuss] repo sync failures: 503 error in ima-evm-utils.git In-Reply-To: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> References: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> Message-ID: On Mon, Oct 15, 2018 at 1:59 PM Scott Little wrote: > We are currently seeing the following error from 'repo sync' up here in Ottawa. > fatal: unable to access 'https://git.code.sf.net/p/linux-ima/ima-evm-utils.git/': The requested URL returned error: 503 I am able to git clone that URL from home just now... (Google Fiber if it matters) dt -- Dean Troyer dtroyer at gmail.com From scott.little at windriver.com Mon Oct 15 19:24:42 2018 From: scott.little at windriver.com (Scott Little) Date: Mon, 15 Oct 2018 15:24:42 -0400 Subject: [Starlingx-discuss] repo sync failures: 503 error in ima-evm-utils.git In-Reply-To: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> References: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> Message-ID: It looks like ima is no longer compiled by StarlingX. grep -r ima $MY_REPO/stx/stx*/centos_pkg_dirs*    << nothing found >> We are now using externally built binaries. grep ima-evm-utils $MY_REPO/../stx-tools/centos-mirror-tools/*.lst .../stx-tools/centos-mirror-tools/rpms_centos3rdparties.lst:ima-evm-utils-1.1-2.el7.x86_64.rpm .../stx-tools/centos-mirror-tools/rpms_centos3rdparties.lst:ima-evm-utils-devel-1.1-2.el7.x86_64.rpm I believe ima can safely be removed from the manifest. Scott On 18-10-15 02:56 PM, Scott Little wrote: > > We are currently seeing the following error from 'repo sync' up here > in Ottawa. > >    fatal: unable to access > 'https://git.code.sf.net/p/linux-ima/ima-evm-utils.git/': The > requested URL returned error: 503 > > Timestamp is roughly 2018-10-15 18:45 UTC > > Can other sites see this ? > > > *Workaround:* > > If you already have an established repo, including ima, you can use > the following to refresh your repo ... > >    repo sync  --optimized-fetch > > This works because ima in our manifest is anchored to a specific > tag/sha, which you already have downloaded, so no need to refresh that > specific git. > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From chris.friesen at windriver.com Mon Oct 15 19:30:20 2018 From: chris.friesen at windriver.com (Chris Friesen) Date: Mon, 15 Oct 2018 13:30:20 -0600 Subject: [Starlingx-discuss] stx-specs git repo directory structure In-Reply-To: References: Message-ID: <633234e1-f4c7-8d36-7207-5b2d0519d018@windriver.com> On 10/15/2018 11:50 AM, Miller, Frank wrote: > Looks like 1 is merged already and 6 are in the queue. I think we can let the commit to create the 2019.03 directory merge first and ask the 6 people with outstanding specs to rebase. > > Chris - as part of your commit to create the 2019.03 directory can you also move the one spec that already merged into the new 2019.03 directory? Done. In case anyone is interested, review is at: https://review.openstack.org/#/c/610669 Chris From dtroyer at gmail.com Mon Oct 15 19:35:46 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 15 Oct 2018 14:35:46 -0500 Subject: [Starlingx-discuss] repo sync failures: 503 error in ima-evm-utils.git In-Reply-To: References: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> Message-ID: On Mon, Oct 15, 2018 at 2:27 PM Scott Little wrote: > It looks like ima is no longer compiled by StarlingX. ... > I believe ima can safely be removed from the manifest. Even better! https://review.openstack.org/610710 Should this be backported to r/2018.10 too? dt -- Dean Troyer dtroyer at gmail.com From scott.little at windriver.com Mon Oct 15 19:58:51 2018 From: scott.little at windriver.com (Scott Little) Date: Mon, 15 Oct 2018 15:58:51 -0400 Subject: [Starlingx-discuss] repo sync failures: 503 error in ima-evm-utils.git In-Reply-To: References: <539f9fc7-99c1-24cc-68a0-165a2e298018@windriver.com> Message-ID: Yes, please backport. Scott On 18-10-15 03:35 PM, Dean Troyer wrote: > On Mon, Oct 15, 2018 at 2:27 PM Scott Little wrote: >> It looks like ima is no longer compiled by StarlingX. > ... > >> I believe ima can safely be removed from the manifest. > Even better! > > https://review.openstack.org/610710 > > Should this be backported to r/2018.10 too? > > dt > From Numan.Waheed at windriver.com Mon Oct 15 21:21:58 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Mon, 15 Oct 2018 21:21:58 +0000 Subject: [Starlingx-discuss] [ TEST ] October release testing status - 20181015 Message-ID: <3CAA827B7A79BA46B15B280EC82088FE4821089A@ALA-MBD.corp.ad.wrs.com> Please find below testing status for 20181015. Sanity Results ============== Config: All-in-One Simplex Build ID: 2018-10-12_01-52-00 Total: 11 Pass: 11 Fail: 0 Bugs: None ========================================== Config: All-in-One Duplex Build ID: 2018-10-14_01-52-00 Total: 15 Pass: 15 Fail: 0 Bugs: None ========================================== Config: Multi Node Build ID: 2018-10-08_01-52-01 Total: 23 Pass: 23 Fail: 0 Bugs: None ========================================== Config: Multi Node with CEPH Build ID: 2018-10-10_01-52-00 Total: 24 Pass: 24 Fail: 0 Bugs: None ========================================== Manual Regression: Domains Total Passed Failed Executed Launchpad ========================================================================= Backup and Restore 4 3 1 4 https://bugs.launchpad.net/starlingx/+bug/1796404 Distributed Cloud 30 19 4 23 https://bugs.launchpad.net/starlingx/+bug/1797960 Installation and Config 5 3 2 5 https://bugs.launchpad.net/starlingx/+bug/1797187 https://bugs.launchpad.net/starlingx/+bug/1797608 MTC 19 19 0 19 Networking 22 10 5 15 https://bugs.launchpad.net/starlingx/+bug/1796198 https://bugs.launchpad.net/starlingx/+bug/1790968 Nova 35 15 6 21 https://bugs.launchpad.net/starlingx/+bug/1796393 Security 17 9 2 11 https://bugs.launchpad.net/starlingx/+bug/1797204 https://bugs.launchpad.net/starlingx/+bug/1797219 https://bugs.launchpad.net/starlingx/+bug/1797569 Storage 11 4 5 9 https://bugs.launchpad.net/starlingx/+bug/1795962 https://bugs.launchpad.net/starlingx/+bug/1796913 https://bugs.launchpad.net/starlingx/+bug/1797151 https://bugs.launchpad.net/starlingx/+bug/1797391 System Inventory 8 7 1 8 https://bugs.launchpad.net/starlingx/+bug/1797187 System Test 10 8 1 9 ========================================================================= Total 161 97 27 124 Thanks, Numan Waheed, Manager, Titanium Cloud Testing Wind River direct 613 270 5762, mobile 613 793 5476 -------------- next part -------------- An HTML attachment was scrubbed... URL: From sgw at linux.intel.com Mon Oct 15 23:54:28 2018 From: sgw at linux.intel.com (Saul Wold) Date: Mon, 15 Oct 2018 16:54:28 -0700 Subject: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. In-Reply-To: <93814834B4855241994F290E959305C752FC8ABD@SHSMSX104.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FC8ABD@SHSMSX104.ccr.corp.intel.com> Message-ID: <2051f79c-b681-8bd1-83b7-8c2bf4b47d58@linux.intel.com> On 10/15/2018 02:26 AM, Liu, ZhipengS wrote: > Hi all, > > Recently I have analyzed all init patches of non-openstack patches that you can filter by "init" in "Analysis" column of below sheet > https://docs.google.com/spreadsheets/d/1nKnkweuxcqvVOoRcpnTYMVUUv1RoAugOWXMjB7VIrfc/edit#gid=1953568313 > I have added my comments in column of "Comments from zhipeng" > > 1) The comments colored with blue means we can add service to /etc/systemd/system/ so that it can be started during power on. > 2) The comments colored with orange means we can ADD config files or scripts to specific system folder. > 3) The comments colored with green means we need to use our customized config files or scripts instead of default ones. > > For both case 1 and 2 mentioned above, we just need to consider initial installation scenario. (22 patches) > I worked out a method to remove related patches like below steps. > a) Use platform-util RPM to copy all related config/service files to /tmp folder. > Add config/service files to platform-util/centos/files/ > Modify platform-util.spec to copy all related config/service files to /tmp folder. I don't think we should put everything into one RPM, I would consider creating a new RPM per package, if a package as both a case 1 & case 2, they can go in the same RPM. Ultimately, I think we need to break platform-utils down a little. For example the memcached.service should be moved to a stx-memcached RPM. Looking back I probably should have nacked that patch. > b) Add a post-installation script to kickstart config file. (This script will be called by anaconda after software installation finished.) > This post-installation script will copy config/service files to specific system folder. These should be individual scripts that are called by a standard mechanism from the kickstart post-install, this way they can be re-used by other software installers for Multi-OS support. > c) Then we can remove related patches. > > For case 3, we need to consider both initial installation and SW update scenarios. (44 patches) > a) Initial installation > We can do the same as we do for case 1 and 2 > b) SW update > Add a post-update script that will be called by sw-patch-agent after sw update finished. > Use post-update script to copy customized config files or scripts to specific system folders, overwrite default ones. > > So far, I have started working on case 1 and 2, finished nfs-utils (Story: 2003768/Task: 26461) and verified deployment. You can review my patch below. > https://review.openstack.org/#/c/610456/ > https://review.openstack.org/#/c/610459/ > I added additional review comments to these as well. Sau! > Any concern? Welcome to add your comment and proposal! > > Thanks! > Zhipeng > > > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From ada.cabrales at intel.com Tue Oct 16 00:01:18 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Tue, 16 Oct 2018 00:01:18 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-15-2018 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C352DBF@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-15-24-r-2018.10.iso - 31 TCs PASS, 0 TCs FAIL - Multinode external storage on virtual environment having the problem with creating instances again. Reviewing comments from Frank Miller and debugging a bit more - Launchpad is [0] - Looks like our duplex virtual also hit the problem with openvswitch [1] - checking on this. - For " Admin credentials are setup on both controllers after swact" [2] - even though this is an expected behavior, we think this might be a security issue. We'll let the TSC to decide on this. - New Launchpad: "MemPageSIze can be empty on a flavor" on duplex bare metal [3] - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 0/40/37/77 Simplex virtual - 0/10/61/71 Duplex bare metal - 1/57/42/100 Duplex virtual - 1/56/48/105 Multinode local storage bare metal - 0/67/50/117 Multinode local storage virtual - 6/59/57/122 Multinode external storage bare metal - 0/70/47/117 Multinode external storage virtual - 2/46/72/120 Full report in the wiki [4] [0] https://bugs.launchpad.net/starlingx/+bug/1797234 [1] https://bugs.launchpad.net/starlingx/+bug/1796380 [2] https://bugs.launchpad.net/starlingx/+bug/1797961 [3] https://bugs.launchpad.net/starlingx/+bug/1797961 [4] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary A. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From huifeng.le at intel.com Tue Oct 16 01:45:56 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Tue, 16 Oct 2018 01:45:56 +0000 Subject: [Starlingx-discuss] Analysis report about Network Trunk feature for StartlingX upstreaming In-Reply-To: <45BCE356-0DAC-4C66-9244-4EED06478182@windriver.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D4EB222@SHSMSX104.ccr.corp.intel.com> <9A85D2917C58154C960D95352B22818BAB577F1B@fmsmsx115.amr.corp.intel.com> <76647BD697F40748B1FA4F56DA02AA0B4D4EB495@SHSMSX104.ccr.corp.intel.com> <304C67CB-BAE4-4E42-A613-1C848535F166@windriver.com> <76647BD697F40748B1FA4F56DA02AA0B4D4EDBFE@SHSMSX104.ccr.corp.intel.com> <45BCE356-0DAC-4C66-9244-4EED06478182@windriver.com> Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D520224@SHSMSX104.ccr.corp.intel.com> Matt, Just FYI. Below is the community feedback (https://bugs.launchpad.net/neutron/+bug/1797368) the different behavior is aware by community and it is preferred to keep the current design. Thanks much! “ Hi, thanks for reporting this bug. In reality this is working as intended. the Trunk admin status is simply a locking mechanism for management operations on the trunk as it's been articulated on the documentation:https://docs.openstack.org/neutron/pike/admin/config-trunking.html In particular: When admin_state is set to DOWN, the user is blocked from performing operations on the trunk. admin_state is set by the user and should not be used to monitor the health of the trunk. Having said that, I can see the confusion and a different expectation some users may have. Neutron resources admin states are not necessarily used for blocking the datapath (I think another example of that might neutron router floating IP ports, but I am no longer sure). The reason why this was designed as was mainly simplicity and robustness. Ensuring that the operation of turning up/down the admin status for trunk worked reliably and atomically across all sub-ports in a trunk is not straightforward, therefore in the absence of a strong use case the existing semantic was chosen. “ Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Wednesday, September 19, 2018 9:15 PM To: Le, Huifeng Cc: Zhao, Forrest ; starlingx-discuss at lists.starlingx.io Subject: Re: Analysis report about Network Trunk feature for StartlingX upstreaming Hello Huifeng, I wanted to follow-up on item #1 below (ba9d9f60a7a2665194cacb92a05e0acd2dc3de41: Add rpc notification for trunk updates). The concern is that there is a difference in behavior between the port and trunks, that may impact the user experience and not just impact the agent/server behavior. The problem is that it is not symmetric with setting admin_setup_up on a Port. If you set that attribute to False on a Port then that port is disabled in the vswitch; packets are no longer sent/received. Based on that expectation, then it would make sense that setting admin_state_up=False on a trunk would disable that trunk on the vswitch (i.e., stop processing VLAN packets arriving from that VM instance), but that is not what happens; it continues to be operational. It is our opinion that this is an incorrect behavior that is worth correcting. Based on the above behavioral difference, I think it makes sense to pursue this with the neutron team since they may want to align on this behavior as well. If the neutron team rejects this change, then we can align on the current upstream behavior. Regards, Matt From: "Peters, Matt" > Date: Friday, September 7, 2018 at 11:00 AM To: "Le, Huifeng" >, "Jolliffe, Ian" >, "Jones, Bruce E" >, Brent Rowsell > Cc: "Zhao, Forrest" >, "Troyer, Dean" >, "starlingx-discuss at lists.starlingx.io" > Subject: RE: Analysis report about Network Trunk feature for StartlingX upstreaming See inline for specific responses to the review information. Responses marked with [MP>] From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, August 19, 2018 10:32 PM To: Jolliffe, Ian; Jones, Bruce E; Rowsell, Brent; Peters, Matt Cc: Zhao, Forrest; Troyer, Dean; starlingx-discuss at lists.starlingx.io Subject: RE: Analysis report about Network Trunk feature for StartlingX upstreaming Ian, Thanks very much for the comments. some comments below for you reference, and please help to review, thanks much! Best Regards, Le, Huifeng From: Jolliffe, Ian [mailto:Ian.Jolliffe at windriver.com] Sent: Saturday, August 18, 2018 4:16 AM To: Le, Huifeng >; Jones, Bruce E >; Rowsell, Brent >; Peters, Matt > Cc: Zhao, Forrest >; Troyer, Dean >; starlingx-discuss at lists.starlingx.io Subject: Re: Analysis report about Network Trunk feature for StartlingX upstreaming Hi Huifeng; Thanks for the updates/analysis, comments below. Ian Ian/Brent/Matt, We did analysis about the Network trunk related patches for StartingX upstream, below are the suggestions for upstreaming, could you please help to review and comment? Thanks much! 1. ba9d9f60a7a2665194cacb92a05e0acd2dc3de41: Add rpc notification for trunk updates Function: sent notification to the agent when a trunk is updated Analysis: (1)Trunk’s AFTER_UPDATE event is generated for API call: PUT /v2.0/trunks/{trunk-id} The update request is only for changing fields like name, description or admin_state_up. Setting the admin_state_up to False locks the trunk in that it prevents operations such as adding/removing subports. In Neutron upstream, admin_state_up is used in server side, e.g. add_subports, remove subports, delete_trunk and not used in agent side (2)OVS trunk agent driver uses OVSDB event to handle trunk event, no need to manually trigger trunk update event (3)Linux trunk agent driver will handle trunk update event triggered by server, while it will need apply the patch only in case admin_state_up update need to be handled Suggestion: Not a bug for Neutron upstream, suggest not to upstream If this is not upstreamed, are the dependencies or changes required in the StarlingX code base? What are the implications of not upstreaming? [hle2] for STX, trunk_updated event will force the trunk’s parent-port to refresh (e.g. handle_trunks->mark_port_for_refresh(trunk['port_id']) etc.) to get the new “admin_state_up” value from server and this value will be used in handle_updated_port() to determine whether it is allowed to update port/device status in server side. “admin_state_up” is mainly used to control operation at neutron server side like add_subports, remove subports, delete_trunk etc. and all these 3 operations will force port to refresh (handle_trunks/handle_subports->mark_port_for_refresh), so suppose, the general flow will not be impacted whether to handle trunk_updated event or not. But in some wired cases, add “admin_state_up” check in agent side may cause issues (please help to review whether it is make sense), e.g for below calling flow (suppose trunk’s ‘admin_state_up’ is ‘up’): (1) add_subports (2) set “admin_state_up” to ‘down’, step(1) may fail to set device’s state at agent side in case aws agent’s handle_updated_port() (in daemon loop) executed after step (2) So to my understanding: (1) if using OVS agent in STX, no impact for not upstream (2) if using AVS agent + STX, suggest removing “admin_state_up” check in AVS agent (in function handle_updated_port () of avs/agent.py) like below. if trunk_details and trunk_details['admin_state_up']: … [MP>] Thanks for the detailed analysis. The suggested change will need to be tested to see if it full resolves the original issue. However, since the driver could be used by other agents (beyond just the AVS agent), I think it would not hurt to have the additional notification sent to complete the driver definition for the set of RPC notifications. 2. 6955351c5eca6e37061fb0140d11ea53693fe0e1: Add support to delete bound network Function: enable delete trunk if it is can_be_trunked (not bounded or driver’s can_trunk_bound_port=true) Analysis: Applied for LinuxBridge Driver and AVS bridge Driver (can_trunk_bound_port=True), no impact for OVSTrunkDriver (can_trunk_bound_port=False). workaround also available for linux bridge (e.g. unbind the port first then delete the trunk) Suggestion: it is a low priority bug for Neutron upstream (only applied for linux bridge and workround available), suggest not to upstream I think you need to propose a fix. Or this will need to be carried long term. [hle2] yes, let’s try to propose a fix for upstream. [MP>] Agree. 3. 43a684946e781a25d21a4f50b8dc67d61be42809: Enable trunk service by default Function: add “trunk” in DEFAULT_SERVICE_PLUGINS Analysis: It is a deploy configuration for downstream product Suggestion: Not a bug for Neutron upstream, suggest not to upstream Agree [MP>] Agree. 4. c54d804792f10b7f505de6794274c4df4768f6f0: Include trunk presence in port details Function: add trunk_port (bool) flag in port_details to identify whether this port is a parent port for a trunk Analysis: It is a performance improvement for AVS agent by reducing RPC call from agent to server. OVS agent has different implementation with no improvement by introducing this field Suggestion: Not a bug for Neutron upstream, suggest not to upstream Agree [MP>] Agree. 5. 3eed837ebd236e6b1959ea88d9ab5322c9eef6b9: Ignore trunk subports on same vlan as vlan-subnet ports Function: Ignore trunk subports on same vlan as vlan-subnet ports Analysis: It is a bug fix for AVS agent Suggestion: Not a bug for Neutron upstream, suggest not to upstream Agree [MP>] Agree. Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhipengs.liu at intel.com Tue Oct 16 02:23:26 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Tue, 16 Oct 2018 02:23:26 +0000 Subject: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. In-Reply-To: <2051f79c-b681-8bd1-83b7-8c2bf4b47d58@linux.intel.com> References: <93814834B4855241994F290E959305C752FC8ABD@SHSMSX104.ccr.corp.intel.com> <2051f79c-b681-8bd1-83b7-8c2bf4b47d58@linux.intel.com> Message-ID: <93814834B4855241994F290E959305C752FC90C4@SHSMSX104.ccr.corp.intel.com> Hi Saul and Brent, Really appreciate for your comment! Please see my inline comment. For case 3, sw update scenario. Furtherly, I'd like to do as below. Define a post-patch script RPM in below folder. cgcs-root/stx/stx-update/patch-scripts/ I see that we already have post-patch script running mechanism. This script will be copied to %_runtime_patch_scripts=/run/patching/patch-scripts After SW-UPDATE, it will be called by sw-patch-agent Thx! -Zhipeng -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: 2018年10月16日 7:54 To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. On 10/15/2018 02:26 AM, Liu, ZhipengS wrote: > Hi all, > > Recently I have analyzed all init patches of non-openstack patches > that you can filter by "init" in "Analysis" column of below sheet > https://docs.google.com/spreadsheets/d/1nKnkweuxcqvVOoRcpnTYMVUUv1RoAu > gOWXMjB7VIrfc/edit#gid=1953568313 I have added my comments in column > of "Comments from zhipeng" > > 1) The comments colored with blue means we can add service to /etc/systemd/system/ so that it can be started during power on. > 2) The comments colored with orange means we can ADD config files or scripts to specific system folder. > 3) The comments colored with green means we need to use our customized config files or scripts instead of default ones. > > For both case 1 and 2 mentioned above, we just need to consider > initial installation scenario. (22 patches) I worked out a method to remove related patches like below steps. > a) Use platform-util RPM to copy all related config/service files to /tmp folder. > Add config/service files to platform-util/centos/files/ > Modify platform-util.spec to copy all related config/service files to /tmp folder. [Saul] I don't think we should put everything into one RPM, I would consider creating a new RPM per package, if a package as both a case 1 & case 2, they can go in the same RPM. Ultimately, I think we need to break platform-utils down a little. For example the memcached.service should be moved to a stx-memcached RPM. Looking back I probably should have nacked that patch. [Zhipeng] Now, we already put memcached.service related copy script in platform-utils spec file. That's why I did the same for nfs-utils. As talked to you before, we can define %package for nfs-utils and other packages respectively, then build out different RPMs However, here we just use this platform-util spec to copy service to destination folder, is it really necessary to let it build out RPM for every third-party package. What's the purpose? @Brent also mentioned we'd better use tis-extensions. From the comment of tis-extensions below, it seems to be designed especially for this purpose. # # The tis-extensions group of packages is intended to allow us to # add files to "extend" thirdparty packages, such as by packaging # custom systemd files into /etc/systemd to override the originals # without modifying or rebuilding the thirdparty package. # So, for this point, could you help to get it aligned with Brent? Then we can start work on other patches soon. > b) Add a post-installation script to kickstart config file. (This script will be called by anaconda after software installation finished.) > This post-installation script will copy config/service files to specific system folder. [Saul ]These should be individual scripts that are called by a standard mechanism from the kickstart post-install, this way they can be re-used by other software installers for Multi-OS support. [Zhipeng] Yes! Currently, I just added the scriptlet in bsp-files/kickstarts/post_common.cfg. I'd like to add a new ks.cfg file like "bsp-files/kickstarts/post_3rdparty_package_install.cfg" to own this post-install script. ( KS_SCRIPT_POST) BTW, this is the standard mechanism that kickstart will parse this cfg file and get this post-script called after SW installation finished. > c) Then we can remove related patches. > > For case 3, we need to consider both initial installation and SW > update scenarios. (44 patches) > a) Initial installation > We can do the same as we do for case 1 and 2 > b) SW update > Add a post-update script that will be called by sw-patch-agent after sw update finished. > Use post-update script to copy customized config files or scripts to specific system folders, overwrite default ones. > > So far, I have started working on case 1 and 2, finished nfs-utils (Story: 2003768/Task: 26461) and verified deployment. You can review my patch below. > https://review.openstack.org/#/c/610456/ > https://review.openstack.org/#/c/610459/ > I added additional review comments to these as well. Sau! > Any concern? Welcome to add your comment and proposal! > > Thanks! > Zhipeng > > > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Tue Oct 16 06:13:00 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Tue, 16 Oct 2018 06:13:00 +0000 Subject: [Starlingx-discuss] Call for agenda: Weekly StarlingX non-OpenStack Distro meeting Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D706C8@SHSMSX104.ccr.corp.intel.com> Bruce, Ghada & Saul Other than the regular agenda we have below, I think we at least have 30min slot open, please propose the topics you'd like to discuss in this meeting, I can add into tomorrow's meeting. - Bugs for 2018.10 review - Non-openstack patch analysis & refactoring update - Ceph upgrade update. Thx. - cindy -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 17, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other From michel.thebeau at windriver.com Tue Oct 16 11:58:09 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Tue, 16 Oct 2018 07:58:09 -0400 Subject: [Starlingx-discuss] StarlingX alarms fields displayed In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E46B3@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E46B3@fmsmsx101.amr.corp.intel.com> Message-ID: <4c8d45c6-1bcd-aca7-291d-e952b9fe30f7@windriver.com> Hi Jose, I found text of your question in this web document: https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Test #85 is presented as the first in a series intended to test alarm suppression, and that this first test is intended the verify the alarm presentation without suppression. I'm not sure that the test instruction is complete or that the "suppression" field is the right one to look at for this test case.  I've CC'd Tao for her experience on the subject of alarm, events and event suppression. M On 2018-10-09 12:11 p.m., Perez Carranza, Jose wrote: > Hi All > > I have a question about the alarms on StarlingX, in one of the scenarios that I'm executing says: - "Verify that entity-instance-field and suppression set to True not shown (system alarm-list, system alarm-show )" but actually the "Supression | Ture " is displayed, is this a correct behavior and I should update the scenario, or a Launchpad should be created to address this? > > Regards, > José > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Allain.Legacy at windriver.com Tue Oct 16 12:23:09 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Tue, 16 Oct 2018 12:23:09 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> I am not sure. The qos_policy_id attribute does not have "enforce_policy" set in its attribute definition so I am going to guess that there is no checking at all for it. I don't have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don't even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We're looking into the patch 1e9a089, which target to solve the issue of "CGTS-2408: bypass policy check when setting attribute to None." After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: "type: uuid_or_none" (2) the "enforce_policy" should be set to "True" and the policy needs to be defined in etc/policy.json (3) the check policy uses "tenant_id" for owner check The only found attribute which will be impacted by this issue in STX is "wrs-tm:qos" which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What's your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From Brent.Rowsell at windriver.com Tue Oct 16 12:49:23 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Tue, 16 Oct 2018 12:49:23 +0000 Subject: [Starlingx-discuss] Call for agenda: Weekly StarlingX non-OpenStack Distro meeting In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F35D706C8@SHSMSX104.ccr.corp.intel.com> References: <2FD5DDB5A04D264C80D42CA35194914F35D706C8@SHSMSX104.ccr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2B5824@ALA-MBD.corp.ad.wrs.com> Add - Status of driver upversion Brent -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Tuesday, October 16, 2018 2:13 AM To: Khalil, Ghada ; Jones, Bruce E ; Wold, Saul ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Call for agenda: Weekly StarlingX non-OpenStack Distro meeting Bruce, Ghada & Saul Other than the regular agenda we have below, I think we at least have 30min slot open, please propose the topics you'd like to discuss in this meeting, I can add into tomorrow's meeting. - Bugs for 2018.10 review - Non-openstack patch analysis & refactoring update - Ceph upgrade update. Thx. - cindy -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 17, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Tue Oct 16 13:22:16 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Tue, 16 Oct 2018 13:22:16 +0000 Subject: [Starlingx-discuss] Call for agenda: Weekly StarlingX non-OpenStack Distro meeting In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2B5824@ALA-MBD.corp.ad.wrs.com> References: <2FD5DDB5A04D264C80D42CA35194914F35D706C8@SHSMSX104.ccr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B5824@ALA-MBD.corp.ad.wrs.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D70E84@SHSMSX104.ccr.corp.intel.com> OK, let me put it onto our agenda. But we are still waiting for the testing resource to be available to do testing before we merge them. @shuicheng may have more details. Thx. - cindy -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Tuesday, October 16, 2018 8:49 PM To: Xie, Cindy ; Khalil, Ghada ; Jones, Bruce E ; Wold, Saul ; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Call for agenda: Weekly StarlingX non-OpenStack Distro meeting Add - Status of driver upversion Brent -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Tuesday, October 16, 2018 2:13 AM To: Khalil, Ghada ; Jones, Bruce E ; Wold, Saul ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Call for agenda: Weekly StarlingX non-OpenStack Distro meeting Bruce, Ghada & Saul Other than the regular agenda we have below, I think we at least have 30min slot open, please propose the topics you'd like to discuss in this meeting, I can add into tomorrow's meeting. - Bugs for 2018.10 review - Non-openstack patch analysis & refactoring update - Ceph upgrade update. Thx. - cindy -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 17, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Don.Penney at windriver.com Tue Oct 16 13:58:30 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Tue, 16 Oct 2018 13:58:30 +0000 Subject: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. In-Reply-To: <93814834B4855241994F290E959305C752FC90C4@SHSMSX104.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FC8ABD@SHSMSX104.ccr.corp.intel.com> <2051f79c-b681-8bd1-83b7-8c2bf4b47d58@linux.intel.com> <93814834B4855241994F290E959305C752FC90C4@SHSMSX104.ccr.corp.intel.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3C94B8@ALA-MBD.corp.ad.wrs.com> Hi Zhipeng, Sorry, I was on vacation last week and have been playing catch-up. This patch script mechanism you're referring to is not for such purpose. It is for in-service patching only (no reboot patching), and is only for restarting services after a host has been patched and only if that host will not be rebooted as part of that patch. In fact, the StarlingX patching mechanism should not be part of your consideration here at all. Why are you planning to copy files to a /tmp dir at all here, or using the kickstarts at all? I would see this effort as being one of the following: Service files: a) If we just want to append something to a packaged service file, create an override. For example, if we wanted to add an After=sw-patch.service to a system xyz.service file, we can package a file named /etc/systemd/system/xyz.service.d/add-sw-patch-dependency.conf that contains: [Unit] After=sw-patch.service b) If we want to completely replace a packaged system service file, like xyz.service, we can package our version to /etc/systemd/system/xyz.service c) If we're patching a system package to make changes to a configuration file, we need to ask whether this change is necessary immediately upon installation, or whether the change can be made by puppet when the manifest is applied. Ideally, most cases should be the latter. If we believe it is the former, we'll need to consider other options, and ensure we can handle it in a way that will be patchable at runtime. -----Original Message----- From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: Monday, October 15, 2018 10:23 PM To: Saul Wold; Rowsell, Brent Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. Hi Saul and Brent, Really appreciate for your comment! Please see my inline comment. For case 3, sw update scenario. Furtherly, I'd like to do as below. Define a post-patch script RPM in below folder. cgcs-root/stx/stx-update/patch-scripts/ I see that we already have post-patch script running mechanism. This script will be copied to %_runtime_patch_scripts=/run/patching/patch-scripts After SW-UPDATE, it will be called by sw-patch-agent Thx! -Zhipeng -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: 2018年10月16日 7:54 To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. On 10/15/2018 02:26 AM, Liu, ZhipengS wrote: > Hi all, > > Recently I have analyzed all init patches of non-openstack patches > that you can filter by "init" in "Analysis" column of below sheet > https://docs.google.com/spreadsheets/d/1nKnkweuxcqvVOoRcpnTYMVUUv1RoAu > gOWXMjB7VIrfc/edit#gid=1953568313 I have added my comments in column > of "Comments from zhipeng" > > 1) The comments colored with blue means we can add service to /etc/systemd/system/ so that it can be started during power on. > 2) The comments colored with orange means we can ADD config files or scripts to specific system folder. > 3) The comments colored with green means we need to use our customized config files or scripts instead of default ones. > > For both case 1 and 2 mentioned above, we just need to consider > initial installation scenario. (22 patches) I worked out a method to remove related patches like below steps. > a) Use platform-util RPM to copy all related config/service files to /tmp folder. > Add config/service files to platform-util/centos/files/ > Modify platform-util.spec to copy all related config/service files to /tmp folder. [Saul] I don't think we should put everything into one RPM, I would consider creating a new RPM per package, if a package as both a case 1 & case 2, they can go in the same RPM. Ultimately, I think we need to break platform-utils down a little. For example the memcached.service should be moved to a stx-memcached RPM. Looking back I probably should have nacked that patch. [Zhipeng] Now, we already put memcached.service related copy script in platform-utils spec file. That's why I did the same for nfs-utils. As talked to you before, we can define %package for nfs-utils and other packages respectively, then build out different RPMs However, here we just use this platform-util spec to copy service to destination folder, is it really necessary to let it build out RPM for every third-party package. What's the purpose? @Brent also mentioned we'd better use tis-extensions. From the comment of tis-extensions below, it seems to be designed especially for this purpose. # # The tis-extensions group of packages is intended to allow us to # add files to "extend" thirdparty packages, such as by packaging # custom systemd files into /etc/systemd to override the originals # without modifying or rebuilding the thirdparty package. # So, for this point, could you help to get it aligned with Brent? Then we can start work on other patches soon. > b) Add a post-installation script to kickstart config file. (This script will be called by anaconda after software installation finished.) > This post-installation script will copy config/service files to specific system folder. [Saul ]These should be individual scripts that are called by a standard mechanism from the kickstart post-install, this way they can be re-used by other software installers for Multi-OS support. [Zhipeng] Yes! Currently, I just added the scriptlet in bsp-files/kickstarts/post_common.cfg. I'd like to add a new ks.cfg file like "bsp-files/kickstarts/post_3rdparty_package_install.cfg" to own this post-install script. ( KS_SCRIPT_POST) BTW, this is the standard mechanism that kickstart will parse this cfg file and get this post-script called after SW installation finished. > c) Then we can remove related patches. > > For case 3, we need to consider both initial installation and SW > update scenarios. (44 patches) > a) Initial installation > We can do the same as we do for case 1 and 2 > b) SW update > Add a post-update script that will be called by sw-patch-agent after sw update finished. > Use post-update script to copy customized config files or scripts to specific system folders, overwrite default ones. > > So far, I have started working on case 1 and 2, finished nfs-utils (Story: 2003768/Task: 26461) and verified deployment. You can review my patch below. > https://review.openstack.org/#/c/610456/ > https://review.openstack.org/#/c/610459/ > I added additional review comments to these as well. Sau! > Any concern? Welcome to add your comment and proposal! > > Thanks! > Zhipeng > > > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From huifeng.le at intel.com Tue Oct 16 14:08:32 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Tue, 16 Oct 2018 14:08:32 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): "QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects." >From the default policy.json, e.g. "create/update/delete" operation is for admin only and "get" operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt ; Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have "enforce_policy" set in its attribute definition so I am going to guess that there is no checking at all for it. I don't have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don't even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We're looking into the patch 1e9a089, which target to solve the issue of "CGTS-2408: bypass policy check when setting attribute to None." After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: "type: uuid_or_none" (2) the "enforce_policy" should be set to "True" and the policy needs to be defined in etc/policy.json (3) the check policy uses "tenant_id" for owner check The only found attribute which will be impacted by this issue in STX is "wrs-tm:qos" which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What's your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From Matt.Peters at windriver.com Tue Oct 16 14:13:18 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Tue, 16 Oct 2018 14:13:18 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> Message-ID: <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> I was specifically referring to the network attribute that refers to the QoS policy ID, rather than the RBAC rules related to the policies themselves. I don’t see anything in the policy.json that restricts the setting of the network attribute, so I am assuming it is falling back to the network entity policy itself rather than an individual property rule. From: "Le, Huifeng" Date: Tuesday, October 16, 2018 at 10:08 AM To: Allain Legacy , "Peters, Matt" Cc: "starlingx-discuss at lists.starlingx.io" Subject: RE: Question about patch upstreaming for 1e9a089 Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): “QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects.” From the default policy.json, e.g. “create/update/delete” operation is for admin only and “get” operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt ; Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have “enforce_policy” set in its attribute definition so I am going to guess that there is no checking at all for it. I don’t have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don’t even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We’re looking into the patch 1e9a089, which target to solve the issue of “CGTS-2408: bypass policy check when setting attribute to None.” After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: “type: uuid_or_none” (2) the “enforce_policy” should be set to “True” and the policy needs to be defined in etc/policy.json (3) the check policy uses “tenant_id” for owner check The only found attribute which will be impacted by this issue in STX is “wrs-tm:qos” which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What’s your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1808 bytes Desc: image001.png URL: From zhipengs.liu at intel.com Tue Oct 16 15:15:21 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Tue, 16 Oct 2018 15:15:21 +0000 Subject: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3C94B8@ALA-MBD.corp.ad.wrs.com> References: <93814834B4855241994F290E959305C752FC8ABD@SHSMSX104.ccr.corp.intel.com> <2051f79c-b681-8bd1-83b7-8c2bf4b47d58@linux.intel.com> <93814834B4855241994F290E959305C752FC90C4@SHSMSX104.ccr.corp.intel.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3C94B8@ALA-MBD.corp.ad.wrs.com> Message-ID: <93814834B4855241994F290E959305C752FC958C@SHSMSX104.ccr.corp.intel.com> Hi Penney, Thanks for your kind reply! See my inline comment. Thx! -Zhipeng -----Original Message----- From: Penney, Don [mailto:Don.Penney at windriver.com] Sent: 2018年10月16日 21:59 To: Liu, ZhipengS ; Saul Wold ; Rowsell, Brent Cc: starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. Hi Zhipeng, Sorry, I was on vacation last week and have been playing catch-up. This patch script mechanism you're referring to is not for such purpose. It is for in-service patching only (no reboot patching), and is only for restarting services after a host has been patched and only if that host will not be rebooted as part of that patch. In fact, the StarlingX patching mechanism should not be part of your consideration here at all. [Zhipeng] Got clear! Why are you planning to copy files to a /tmp dir at all here, or using the kickstarts at all? I would see this effort as being one of the following: [zhipeng] As some patch also refers to changing the default config files and need to keep platform-util installed after nfs-utils(let custom files overwrite default ones) Instead, we can also use %require in platform-utils to keep this order. Service files: a) If we just want to append something to a packaged service file, create an override. For example, if we wanted to add an After=sw-patch.service to a system xyz.service file, we can package a file named /etc/systemd/system/xyz.service.d/add-sw-patch-dependency.conf that contains: [Unit] After=sw-patch.service b) If we want to completely replace a packaged system service file, like xyz.service, we can package our version to /etc/systemd/system/xyz.service [Zhipeng] How to package it, use platform-utils, or utilities/tis-extensions, or separate RPM to package config file to specific system folder. What's your preferred method or proposal? c) If we're patching a system package to make changes to a configuration file, we need to ask whether this change is necessary immediately upon installation, or whether the change can be made by puppet when the manifest is applied. Ideally, most cases should be the latter. If we believe it is the former, we'll need to consider other options, and ensure we can handle it in a way that will be patchable at runtime. [Zhipeng] Using puppet to change config file should be the same as script did, right? (Both copy our version configuration file to specific folder to overwrite default ones that are installed earlier) It still confuse me for sw update scenario, I'd like to get answer from you. For example if we change default configuration file of nfs-utils through script or puppet. After we upgrade nfs-utils to new version through sw-patch service. How to apply customized configuration file to it again? -----Original Message----- From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: Monday, October 15, 2018 10:23 PM To: Saul Wold; Rowsell, Brent Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. Hi Saul and Brent, Really appreciate for your comment! Please see my inline comment. For case 3, sw update scenario. Furtherly, I'd like to do as below. Define a post-patch script RPM in below folder. cgcs-root/stx/stx-update/patch-scripts/ I see that we already have post-patch script running mechanism. This script will be copied to %_runtime_patch_scripts=/run/patching/patch-scripts After SW-UPDATE, it will be called by sw-patch-agent Thx! -Zhipeng -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: 2018年10月16日 7:54 To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Work on reducing non-openstack patches that related to config/service/init files. On 10/15/2018 02:26 AM, Liu, ZhipengS wrote: > Hi all, > > Recently I have analyzed all init patches of non-openstack patches > that you can filter by "init" in "Analysis" column of below sheet > https://docs.google.com/spreadsheets/d/1nKnkweuxcqvVOoRcpnTYMVUUv1RoAu > gOWXMjB7VIrfc/edit#gid=1953568313 I have added my comments in column > of "Comments from zhipeng" > > 1) The comments colored with blue means we can add service to /etc/systemd/system/ so that it can be started during power on. > 2) The comments colored with orange means we can ADD config files or scripts to specific system folder. > 3) The comments colored with green means we need to use our customized config files or scripts instead of default ones. > > For both case 1 and 2 mentioned above, we just need to consider > initial installation scenario. (22 patches) I worked out a method to remove related patches like below steps. > a) Use platform-util RPM to copy all related config/service files to /tmp folder. > Add config/service files to platform-util/centos/files/ > Modify platform-util.spec to copy all related config/service files to /tmp folder. [Saul] I don't think we should put everything into one RPM, I would consider creating a new RPM per package, if a package as both a case 1 & case 2, they can go in the same RPM. Ultimately, I think we need to break platform-utils down a little. For example the memcached.service should be moved to a stx-memcached RPM. Looking back I probably should have nacked that patch. [Zhipeng] Now, we already put memcached.service related copy script in platform-utils spec file. That's why I did the same for nfs-utils. As talked to you before, we can define %package for nfs-utils and other packages respectively, then build out different RPMs However, here we just use this platform-util spec to copy service to destination folder, is it really necessary to let it build out RPM for every third-party package. What's the purpose? @Brent also mentioned we'd better use tis-extensions. From the comment of tis-extensions below, it seems to be designed especially for this purpose. # # The tis-extensions group of packages is intended to allow us to # add files to "extend" thirdparty packages, such as by packaging # custom systemd files into /etc/systemd to override the originals # without modifying or rebuilding the thirdparty package. # So, for this point, could you help to get it aligned with Brent? Then we can start work on other patches soon. > b) Add a post-installation script to kickstart config file. (This script will be called by anaconda after software installation finished.) > This post-installation script will copy config/service files to specific system folder. [Saul ]These should be individual scripts that are called by a standard mechanism from the kickstart post-install, this way they can be re-used by other software installers for Multi-OS support. [Zhipeng] Yes! Currently, I just added the scriptlet in bsp-files/kickstarts/post_common.cfg. I'd like to add a new ks.cfg file like "bsp-files/kickstarts/post_3rdparty_package_install.cfg" to own this post-install script. ( KS_SCRIPT_POST) BTW, this is the standard mechanism that kickstart will parse this cfg file and get this post-script called after SW installation finished. > c) Then we can remove related patches. > > For case 3, we need to consider both initial installation and SW > update scenarios. (44 patches) > a) Initial installation > We can do the same as we do for case 1 and 2 > b) SW update > Add a post-update script that will be called by sw-patch-agent after sw update finished. > Use post-update script to copy customized config files or scripts to specific system folders, overwrite default ones. > > So far, I have started working on case 1 and 2, finished nfs-utils (Story: 2003768/Task: 26461) and verified deployment. You can review my patch below. > https://review.openstack.org/#/c/610456/ > https://review.openstack.org/#/c/610459/ > I added additional review comments to these as well. Sau! > Any concern? Welcome to add your comment and proposal! > > Thanks! > Zhipeng > > > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Tao.Liu at windriver.com Tue Oct 16 15:27:12 2018 From: Tao.Liu at windriver.com (Liu, Tao) Date: Tue, 16 Oct 2018 15:27:12 +0000 Subject: [Starlingx-discuss] StarlingX alarms fields displayed In-Reply-To: <4c8d45c6-1bcd-aca7-291d-e952b9fe30f7@windriver.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E46B3@fmsmsx101.amr.corp.intel.com> <4c8d45c6-1bcd-aca7-291d-e952b9fe30f7@windriver.com> Message-ID: <7242A3DC72E453498E3D783BBB134C3E9DD81B88@ALA-MBD.corp.ad.wrs.com> Hi Jose, I think you should update the scenarios with the some details: I believe the scenario that you refer is intended to test the system suppresses the alarms that associated with a host when the host is locked. The alarm entity-instance-id field shows whether the alarm is associated with the host and the suppression field shows whether it allows to be suppressed by the system. This is different than the suppression_status field shown in the alarm detail. The suppression_status is driven by the user action ( I.e. via CLi command fm event-suppress). The test procedure would be: Prior to lock compute-0, check if there are any alarms that located on compute-0 ( for example entity-instance-id: host=compute-0.filesystem=) fm alarm-list --uuid => display the alarm list with uuid field fm alarm show => it should the alarm detail ( include suppression field) for a given alarm After compute-0 is locked, check the alarm list and ensure those alarms are not shown in the list. Regards, Tao -----Original Message----- From: Thebeau, Michel Sent: Tuesday, October 16, 2018 7:58 AM To: Perez Carranza, Jose; starlingx-discuss at lists.starlingx.io; Liu, Tao Subject: Re: [Starlingx-discuss] StarlingX alarms fields displayed Hi Jose, I found text of your question in this web document: https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Test #85 is presented as the first in a series intended to test alarm suppression, and that this first test is intended the verify the alarm presentation without suppression. I'm not sure that the test instruction is complete or that the "suppression" field is the right one to look at for this test case.  I've CC'd Tao for her experience on the subject of alarm, events and event suppression. M On 2018-10-09 12:11 p.m., Perez Carranza, Jose wrote: > Hi All > > I have a question about the alarms on StarlingX, in one of the scenarios that I'm executing says: - "Verify that field and suppression set to True not shown (system alarm-list, system alarm-show )" but actually the "Supression | Ture " is displayed, is this a correct behavior and I should update the scenario, or a Launchpad should be created to address this? > > Regards, > José > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From michel.thebeau at windriver.com Tue Oct 16 16:36:00 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Tue, 16 Oct 2018 12:36:00 -0400 Subject: [Starlingx-discuss] StarlingX alarms fields displayed In-Reply-To: <7242A3DC72E453498E3D783BBB134C3E9DD81B88@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E46B3@fmsmsx101.amr.corp.intel.com> <4c8d45c6-1bcd-aca7-291d-e952b9fe30f7@windriver.com> <7242A3DC72E453498E3D783BBB134C3E9DD81B88@ALA-MBD.corp.ad.wrs.com> Message-ID: Thanks Tao, That's interesting.  I did not know about the system suppressing alarms for a locked host. M On 2018-10-16 11:27 a.m., Liu, Tao wrote: > Hi Jose, > > I think you should update the scenarios with the some details: > > I believe the scenario that you refer is intended to test the system suppresses the alarms that associated with a host when the host is locked. The alarm entity-instance-id field shows whether the alarm is associated with the host and the suppression field shows whether it allows to be suppressed by the system. This is different than the suppression_status field shown in the alarm detail. The suppression_status is driven by the user action ( I.e. via CLi command fm event-suppress). > > The test procedure would be: > Prior to lock compute-0, check if there are any alarms that located on compute-0 ( for example entity-instance-id: host=compute-0.filesystem=) > > fm alarm-list --uuid => display the alarm list with uuid field > > fm alarm show => it should the alarm detail ( include suppression field) for a given alarm > > After compute-0 is locked, check the alarm list and ensure those alarms are not shown in the list. > > > Regards, > Tao > > -----Original Message----- > From: Thebeau, Michel > Sent: Tuesday, October 16, 2018 7:58 AM > To: Perez Carranza, Jose; starlingx-discuss at lists.starlingx.io; Liu, Tao > Subject: Re: [Starlingx-discuss] StarlingX alarms fields displayed > > Hi Jose, > > I found text of your question in this web document: > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > Test #85 is presented as the first in a series intended to test alarm > suppression, and that this first test is intended the verify the alarm > presentation without suppression. > > I'm not sure that the test instruction is complete or that the > "suppression" field is the right one to look at for this test case. > I've CC'd Tao for her experience on the subject of alarm, events and > event suppression. > > M > > > On 2018-10-09 12:11 p.m., Perez Carranza, Jose wrote: >> Hi All >> >> I have a question about the alarms on StarlingX, in one of the scenarios that I'm executing says: - "Verify that field and suppression set to True not shown (system alarm-list, system alarm-show )" but actually the "Supression | Ture " is displayed, is this a correct behavior and I should update the scenario, or a Launchpad should be created to address this? >> >> Regards, >> José >> >> >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From jose.perez.carranza at intel.com Tue Oct 16 16:40:47 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 16 Oct 2018 16:40:47 +0000 Subject: [Starlingx-discuss] StarlingX alarms fields displayed In-Reply-To: <7242A3DC72E453498E3D783BBB134C3E9DD81B88@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E46B3@fmsmsx101.amr.corp.intel.com> <4c8d45c6-1bcd-aca7-291d-e952b9fe30f7@windriver.com> <7242A3DC72E453498E3D783BBB134C3E9DD81B88@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E59D5@fmsmsx101.amr.corp.intel.com> Thanks Michel, Tao I will update the test scenario with your feedback Regards, José > -----Original Message----- > From: Liu, Tao [mailto:Tao.Liu at windriver.com] > Sent: Tuesday, October 16, 2018 10:27 AM > To: Thebeau, Michel ; Perez Carranza, Jose > ; starlingx-discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] StarlingX alarms fields displayed > > Hi Jose, > > I think you should update the scenarios with the some details: > > I believe the scenario that you refer is intended to test the system suppresses > the alarms that associated with a host when the host is locked. The alarm > entity-instance-id field shows whether the alarm is associated with the host > and the suppression field shows whether it allows to be suppressed by the > system. This is different than the suppression_status field shown in the alarm > detail. The suppression_status is driven by the user action ( I.e. via CLi > command fm event-suppress). > > The test procedure would be: > Prior to lock compute-0, check if there are any alarms that located on > compute-0 ( for example entity-instance-id: host=compute- > 0.filesystem=) > > fm alarm-list --uuid => display the alarm list with uuid field > > fm alarm show => it should the alarm detail ( include suppression > field) for a given alarm > > After compute-0 is locked, check the alarm list and ensure those alarms are not > shown in the list. > > > Regards, > Tao > > -----Original Message----- > From: Thebeau, Michel > Sent: Tuesday, October 16, 2018 7:58 AM > To: Perez Carranza, Jose; starlingx-discuss at lists.starlingx.io; Liu, Tao > Subject: Re: [Starlingx-discuss] StarlingX alarms fields displayed > > Hi Jose, > > I found text of your question in this web document: > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > Test #85 is presented as the first in a series intended to test alarm suppression, > and that this first test is intended the verify the alarm presentation without > suppression. > > I'm not sure that the test instruction is complete or that the "suppression" field > is the right one to look at for this test case. I've CC'd Tao for her experience on > the subject of alarm, events and event suppression. > > M > > > On 2018-10-09 12:11 p.m., Perez Carranza, Jose wrote: > > Hi All > > > > I have a question about the alarms on StarlingX, in one of the scenarios that > I'm executing says: - "Verify that field and suppression set to True not shown > (system alarm-list, system alarm-show )" but actually the "Supression | > Ture " is displayed, is this a correct behavior and I should update the scenario, > or a Launchpad should be created to address this? > > > > Regards, > > José > > > > > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From bruce.e.jones at intel.com Tue Oct 16 17:50:37 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Tue, 16 Oct 2018 17:50:37 +0000 Subject: [Starlingx-discuss] Agenda open for the October 17th Community Call Message-ID: <9A85D2917C58154C960D95352B22818BB1E7DB79@ORSMSX158.amr.corp.intel.com> The agenda is open for our call this week: https://etherpad.openstack.org/p/stx-status. Please feel free to add any items that need to be discussed, any help needed, etc.. Brucej PS. Maui is awesome. -------------- next part -------------- An HTML attachment was scrubbed... URL: From claire at openstack.org Tue Oct 16 18:29:15 2018 From: claire at openstack.org (claire at openstack.org) Date: Tue, 16 Oct 2018 18:29:15 +0000 Subject: [Starlingx-discuss] Updated invitation: Bi-Weekly StarlingX Marketing/Community Content Work @ Every 2 weeks from 10am to 11am on Wednesday from Wed Oct 17 to Wed Nov 7 (CDT) (starlingx-discuss@lists.starlingx.io) Message-ID: <000000000000294f3205785cbb8f@google.com> This event has been changed. Title: Bi-Weekly StarlingX Marketing/Community Content Work Agenda & Notes: https://etherpad.openstack.org/p/stx-2018-marketing-content Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/470924647 Or iPhone one-tap : US: +16699006833,,470924647# or +16468769923,,470924647# Or Telephone: Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 Meeting ID: 470 924 647 International numbers available: https://zoom.us/u/RIBXdQZH When: Every 2 weeks from 10am to 11am on Wednesday from Wed Oct 17 to Wed Nov 7 Central Time - Chicago (changed) Where: https://zoom.us/j/470924647 Calendar: starlingx-discuss at lists.starlingx.io Who: * claire at openstack.org - organizer * Allison Price * glenn.seiler at windriver.com * Ildiko Vancsa * Jeff * scott.w.doenecke at intel.com * starlingx-discuss at lists.starlingx.io * Travis V Broughton * Lauren Sell * Bruce E Jones Event details: https://www.google.com/calendar/event?action=VIEW&eid=XzZzcDM4ZHBvNjRzNDJiYTU2a3JrY2I5azZvc2owYjlwNzRxamliOXA4NHMzZWRobDZjcWo0Y3E1NzBsbGtqcWY5bDZrYWhhazk1NzRlamlsOWwxNGFraHE2Z3JqMGU5aTZncjM4ZG8gc3Rhcmxpbmd4LWRpc2N1c3NAbGlzdHMuc3Rhcmxpbmd4Lmlv&tok=MjAjY2xhaXJlQG9wZW5zdGFjay5vcmcyOTNkMDQ5ZjYxYzVjZjM4MjU3MDRjNDQ0ZDI3MzRkYjE0MWY2ZTA1&ctz=America%2FChicago&hl=en&es=0 Invitation from Google Calendar: https://www.google.com/calendar/ You are receiving this courtesy email at the account starlingx-discuss at lists.starlingx.io because you are an attendee of this event. To stop receiving future updates for this event, decline this event. Alternatively you can sign up for a Google account at https://www.google.com/calendar/ and control your notification settings for your entire calendar. Forwarding this invitation could allow any recipient to modify your RSVP response. Learn more at https://support.google.com/calendar/answer/37135#forwarding -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/calendar Size: 4028 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: invite.ics Type: application/ics Size: 4119 bytes Desc: not available URL: From jose.perez.carranza at intel.com Tue Oct 16 19:27:27 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 16 Oct 2018 19:27:27 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> Hi I'm trying to execute scenario 206 [1] that indicates that I need to launch instance with a flavor with "hw:wrs:shared_vcpu", I created a flavor with "hw:wrs:shared_vcpu=1" but at the moment of create the instance below error is displayed.. Is there any extra config or parameter that I need to apply before creating the instance ? Error (Duplex Bare-Metal): ====== No valid host was found. There are not enough hosts available. controller-0: (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required by instance cell 0, Shared vCPU not enabled on host cell 0, required by instance cell 0, controller-1: (NUM ===== 1- https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Regards, José From Brent.Rowsell at windriver.com Tue Oct 16 19:28:56 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Tue, 16 Oct 2018 19:28:56 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> Did you provision your host with a shared cpu ? Brent -----Original Message----- From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] Sent: Tuesday, October 16, 2018 3:27 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Help on shared vCPU usage Hi I'm trying to execute scenario 206 [1] that indicates that I need to launch instance with a flavor with "hw:wrs:shared_vcpu", I created a flavor with "hw:wrs:shared_vcpu=1" but at the moment of create the instance below error is displayed.. Is there any extra config or parameter that I need to apply before creating the instance ? Error (Duplex Bare-Metal): ====== No valid host was found. There are not enough hosts available. controller-0: (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required by instance cell 0, Shared vCPU not enabled on host cell 0, required by instance cell 0, controller-1: (NUM ===== 1- https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Regards, José _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Tue Oct 16 19:38:53 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 16 Oct 2018 19:38:53 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B9A@fmsmsx101.amr.corp.intel.com> I'm not sure about it, I just follow the procedure documented on StarlingX [1] wiki, I don’t see there any label for shared vcpu, here is what my controller and the system has configured [2], What I need to do that kind of provisioning? 1- https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Duplex 2- http://paste.openstack.org/show/732291/ Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Tuesday, October 16, 2018 2:29 PM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > Did you provision your host with a shared cpu ? > > Brent > > > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Tuesday, October 16, 2018 3:27 PM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Help on shared vCPU usage > > Hi > > I'm trying to execute scenario 206 [1] that indicates that I need to launch > instance with a flavor with "hw:wrs:shared_vcpu", I created a flavor with > "hw:wrs:shared_vcpu=1" but at the moment of create the instance below > error is displayed.. Is there any extra config or parameter that I need to apply > before creating the instance ? > > Error (Duplex Bare-Metal): > ====== > No valid host was found. There are not enough hosts available. controller-0: > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required by > instance cell 0, Shared vCPU not enabled on host cell 0, required by instance > cell 0, controller-1: (NUM ===== > > 1- https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > Regards, > José > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Tue Oct 16 20:02:25 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Tue, 16 Oct 2018 20:02:25 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B9A@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B9A@fmsmsx101.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2B7390@ALA-MBD.corp.ad.wrs.com> system host-cpu-list -----Original Message----- From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] Sent: Tuesday, October 16, 2018 3:39 PM To: Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Help on shared vCPU usage I'm not sure about it, I just follow the procedure documented on StarlingX [1] wiki, I don’t see there any label for shared vcpu, here is what my controller and the system has configured [2], What I need to do that kind of provisioning? 1- https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Duplex 2- http://paste.openstack.org/show/732291/ Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Tuesday, October 16, 2018 2:29 PM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > Did you provision your host with a shared cpu ? > > Brent > > > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Tuesday, October 16, 2018 3:27 PM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Help on shared vCPU usage > > Hi > > I'm trying to execute scenario 206 [1] that indicates that I need to > launch instance with a flavor with "hw:wrs:shared_vcpu", I created a > flavor with "hw:wrs:shared_vcpu=1" but at the moment of create the > instance below error is displayed.. Is there any extra config or > parameter that I need to apply before creating the instance ? > > Error (Duplex Bare-Metal): > ====== > No valid host was found. There are not enough hosts available. controller-0: > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required > by instance cell 0, Shared vCPU not enabled on host cell 0, required > by instance cell 0, controller-1: (NUM ===== > > 1- > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instruc > tions > > Regards, > José > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Tue Oct 16 20:09:44 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 16 Oct 2018 20:09:44 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2B7390@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B9A@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7390@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5BCB@fmsmsx101.amr.corp.intel.com> Here is the list [1] So I need to use system host-cpu-modify and configure -p0 with the number of cores and assig the function as shared? 1- http://paste.openstack.org/show/732294/ Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Tuesday, October 16, 2018 3:02 PM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > system host-cpu-list > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Tuesday, October 16, 2018 3:39 PM > To: Rowsell, Brent ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > I'm not sure about it, I just follow the procedure documented on StarlingX [1] > wiki, I don’t see there any label for shared vcpu, here is what my controller and > the system has configured [2], What I need to do that kind of provisioning? > > 1- > https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environ > ment/Duplex > 2- http://paste.openstack.org/show/732291/ > > Regards, > José > > > > -----Original Message----- > > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > > Sent: Tuesday, October 16, 2018 2:29 PM > > To: Perez Carranza, Jose ; starlingx- > > discuss at lists.starlingx.io > > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > > > Did you provision your host with a shared cpu ? > > > > Brent > > > > > > > > -----Original Message----- > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > Sent: Tuesday, October 16, 2018 3:27 PM > > To: starlingx-discuss at lists.starlingx.io > > Subject: [Starlingx-discuss] Help on shared vCPU usage > > > > Hi > > > > I'm trying to execute scenario 206 [1] that indicates that I need to > > launch instance with a flavor with "hw:wrs:shared_vcpu", I created a > > flavor with "hw:wrs:shared_vcpu=1" but at the moment of create the > > instance below error is displayed.. Is there any extra config or > > parameter that I need to apply before creating the instance ? > > > > Error (Duplex Bare-Metal): > > ====== > > No valid host was found. There are not enough hosts available. controller-0: > > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required > > by instance cell 0, Shared vCPU not enabled on host cell 0, required > > by instance cell 0, controller-1: (NUM ===== > > > > 1- > > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instruc > > tions > > > > Regards, > > José > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From juan.carlos.alonso at intel.com Tue Oct 16 20:14:08 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Tue, 16 Oct 2018 20:14:08 +0000 Subject: [Starlingx-discuss] Delete snapshot and volume Message-ID: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> Hi, The test 211[1] indicates that an Image should be created, then a Volume should be created from the image, then a Snapshot should be created from the Volume and finally an Instance should be launch from the Snapshot. The test indicates that we try to delete the Snapshot and Volume while the Instance is running, Snapshot nor Volume cannot be deleted. But when tried found that the Snapshot can be deleted, but the Volume cannot. Is this behavior correct? Those are the commands I used: openstack image create --disk-format qcow2 --file /home/wrsroot/cirros-0.4.0-x86_64-disk.img --public cirros openstack volume create --image cirros --availability-zone nova --size 1 cirros-vol openstack volume snapshot create --name cirros-snap cirros-vol openstack server create --volume cirros-vol --flavor m1.small --availibility-zone nova vm-cirros 1 https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From Brent.Rowsell at windriver.com Tue Oct 16 20:16:22 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Tue, 16 Oct 2018 20:16:22 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5BCB@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B9A@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7390@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5BCB@fmsmsx101.amr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2B748B@ALA-MBD.corp.ad.wrs.com> The output show you have no shared provisioned. For this scenario, one is required. With the node locked. system host-cpu-modify -f shared -p0 1 Brent -----Original Message----- From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] Sent: Tuesday, October 16, 2018 4:10 PM To: Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Help on shared vCPU usage Here is the list [1] So I need to use system host-cpu-modify and configure -p0 with the number of cores and assig the function as shared? 1- http://paste.openstack.org/show/732294/ Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Tuesday, October 16, 2018 3:02 PM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > system host-cpu-list > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Tuesday, October 16, 2018 3:39 PM > To: Rowsell, Brent ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > I'm not sure about it, I just follow the procedure documented on > StarlingX [1] wiki, I don’t see there any label for shared vcpu, here > is what my controller and the system has configured [2], What I need to do that kind of provisioning? > > 1- > https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_E > nviron > ment/Duplex > 2- http://paste.openstack.org/show/732291/ > > Regards, > José > > > > -----Original Message----- > > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > > Sent: Tuesday, October 16, 2018 2:29 PM > > To: Perez Carranza, Jose ; starlingx- > > discuss at lists.starlingx.io > > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > > > Did you provision your host with a shared cpu ? > > > > Brent > > > > > > > > -----Original Message----- > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > Sent: Tuesday, October 16, 2018 3:27 PM > > To: starlingx-discuss at lists.starlingx.io > > Subject: [Starlingx-discuss] Help on shared vCPU usage > > > > Hi > > > > I'm trying to execute scenario 206 [1] that indicates that I need to > > launch instance with a flavor with "hw:wrs:shared_vcpu", I created a > > flavor with "hw:wrs:shared_vcpu=1" but at the moment of create the > > instance below error is displayed.. Is there any extra config or > > parameter that I need to apply before creating the instance ? > > > > Error (Duplex Bare-Metal): > > ====== > > No valid host was found. There are not enough hosts available. controller-0: > > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, > > required by instance cell 0, Shared vCPU not enabled on host cell 0, > > required by instance cell 0, controller-1: (NUM ===== > > > > 1- > > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instr > > uc > > tions > > > > Regards, > > José > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Tue Oct 16 20:24:18 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Tue, 16 Oct 2018 20:24:18 +0000 Subject: [Starlingx-discuss] Help on shared vCPU usage In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2B748B@ALA-MBD.corp.ad.wrs.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B62@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7159@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5B9A@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B7390@ALA-MBD.corp.ad.wrs.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5BCB@fmsmsx101.amr.corp.intel.com> <2588653EBDFFA34B982FAF00F1B4844EBB2B748B@ALA-MBD.corp.ad.wrs.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5C01@fmsmsx101.amr.corp.intel.com> Thanks a lot... I will update my test procedure !! Regards, José > -----Original Message----- > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > Sent: Tuesday, October 16, 2018 3:16 PM > To: Perez Carranza, Jose ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > The output show you have no shared provisioned. > For this scenario, one is required. > > With the node locked. > > system host-cpu-modify -f shared -p0 1 > > Brent > > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Tuesday, October 16, 2018 4:10 PM > To: Rowsell, Brent ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > Here is the list [1] > > So I need to use system host-cpu-modify and configure -p0 with the number of > cores and assig the function as shared? > > 1- http://paste.openstack.org/show/732294/ > > Regards, > José > > > > -----Original Message----- > > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > > Sent: Tuesday, October 16, 2018 3:02 PM > > To: Perez Carranza, Jose ; starlingx- > > discuss at lists.starlingx.io > > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > > > system host-cpu-list > > > > -----Original Message----- > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > Sent: Tuesday, October 16, 2018 3:39 PM > > To: Rowsell, Brent ; starlingx- > > discuss at lists.starlingx.io > > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > > > I'm not sure about it, I just follow the procedure documented on > > StarlingX [1] wiki, I don’t see there any label for shared vcpu, here > > is what my controller and the system has configured [2], What I need to do > that kind of provisioning? > > > > 1- > > https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_E > > nviron > > ment/Duplex > > 2- http://paste.openstack.org/show/732291/ > > > > Regards, > > José > > > > > > > -----Original Message----- > > > From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] > > > Sent: Tuesday, October 16, 2018 2:29 PM > > > To: Perez Carranza, Jose ; starlingx- > > > discuss at lists.starlingx.io > > > Subject: RE: [Starlingx-discuss] Help on shared vCPU usage > > > > > > Did you provision your host with a shared cpu ? > > > > > > Brent > > > > > > > > > > > > -----Original Message----- > > > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > > > Sent: Tuesday, October 16, 2018 3:27 PM > > > To: starlingx-discuss at lists.starlingx.io > > > Subject: [Starlingx-discuss] Help on shared vCPU usage > > > > > > Hi > > > > > > I'm trying to execute scenario 206 [1] that indicates that I need to > > > launch instance with a flavor with "hw:wrs:shared_vcpu", I created a > > > flavor with "hw:wrs:shared_vcpu=1" but at the moment of create the > > > instance below error is displayed.. Is there any extra config or > > > parameter that I need to apply before creating the instance ? > > > > > > Error (Duplex Bare-Metal): > > > ====== > > > No valid host was found. There are not enough hosts available. controller- > 0: > > > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, > > > required by instance cell 0, Shared vCPU not enabled on host cell 0, > > > required by instance cell 0, controller-1: (NUM ===== > > > > > > 1- > > > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instr > > > uc > > > tions > > > > > > Regards, > > > José > > > > > > _______________________________________________ > > > Starlingx-discuss mailing list > > > Starlingx-discuss at lists.starlingx.io > > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Numan.Waheed at windriver.com Tue Oct 16 21:23:41 2018 From: Numan.Waheed at windriver.com (Waheed, Numan) Date: Tue, 16 Oct 2018 21:23:41 +0000 Subject: [Starlingx-discuss] [ TEST ] October release testing status - 20181016 Message-ID: <3CAA827B7A79BA46B15B280EC82088FE48211084@ALA-MBD.corp.ad.wrs.com> Please find below testing status for 20181016. Sanity Results =========================================== Config: All-in-One Simplex Build ID: 2018-10-15_01-52-00 Total: 11 Pass: 11 Fail: 0 Bugs: None =========================================== Config: All-in-One Duplex Build ID: 2018-10-14_01-52-00 Total: 15 Pass: 15 Fail: 0 Bugs: None =========================================== Config: Multi Node Build ID: 2018-10-08_01-52-01 Total: 23 Pass: 23 Fail: 0 Bugs: None =========================================== Config: Multi Node with CEPH Build ID: 2018-10-15_01-52-00 Total: 24 Pass: 24 Fail: 0 Bugs: None Manual Regression: Domains Total Passed Failed Executed Launchpad ================================================================================================ Backup and Restore 4 3 1 4 https://bugs.launchpad.net/starlingx/+bug/1796404 Distributed Cloud 30 19 4 23 https://bugs.launchpad.net/starlingx/+bug/1797960 Installation and Config 5 3 2 5 https://bugs.launchpad.net/starlingx/+bug/1797187 https://bugs.launchpad.net/starlingx/+bug/1797608 MTC 19 19 0 19 Networking 22 18 3 21 https://bugs.launchpad.net/starlingx/+bug/1796198 https://bugs.launchpad.net/starlingx/+bug/1790968 Nova 35 21 7 28 https://bugs.launchpad.net/starlingx/+bug/1796393 Security 17 9 2 11 https://bugs.launchpad.net/starlingx/+bug/1797204 https://bugs.launchpad.net/starlingx/+bug/1797219 https://bugs.launchpad.net/starlingx/+bug/1797569 Storage 11 5 5 10 https://bugs.launchpad.net/starlingx/+bug/1795962 https://bugs.launchpad.net/starlingx/+bug/1796913 https://bugs.launchpad.net/starlingx/+bug/1797151 https://bugs.launchpad.net/starlingx/+bug/1797391 System Inventory 8 7 1 8 https://bugs.launchpad.net/starlingx/+bug/1797187 System Test 10 8 1 9 ================================================================================================= Total 161 112 26 138 Thanks, Numan. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ghada.Khalil at windriver.com Tue Oct 16 22:38:07 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Tue, 16 Oct 2018 22:38:07 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting >> mellanox info Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46AA87@ALA-MBD.corp.ad.wrs.com> Hi Shuicheng, I found the recommended Mellanox versions in the DPDK documentation. You are correct in that this information is not explicitly listed on the Mellanox website. For DPDK 17.11, it lists the Mellanox OFED version as 4.2 https://doc.dpdk.org/guides-17.11/nics/mlx5.html See section 21.5.1.2 For DPDK 18.02 & 18.05, the versions listed are: 4.2 & 4.3 https://doc.dpdk.org/guides-18.02/nics/mlx5.html https://doc.dpdk.org/guides-18.05/nics/mlx5.html For DPDK 18.08, the versions listed are: 4.3 & 4.4 https://doc.dpdk.org/guides-18.08/nics/mlx5.html In StarlingX, we are currently using dpdk 17.11 (bundled w/ ovs-dpdk). The mellanox driver version we are using is 4.3.1.0.1.1 Regards, Ghada -----Original Message----- From: Lin, Shuicheng [mailto:shuicheng.lin at intel.com] Sent: Friday, October 12, 2018 2:47 AM To: Xie, Cindy; starlingx-discuss at lists.starlingx.io; Khalil, Ghada; Rowsell, Brent Subject: RE: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Hi Ghada/Brent, I try to go through the doc in mellanox website, but cannot find the matrix between mlx driver and DPDK. Could you do me a favor where could I get it? Thanks. Here I get the latest mlx driver: http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers From the dpdk package (dpdk-17.05.2.tar.gz) in our code, I get below info, which seems obsolete. " Currently supported by DPDK: - Mellanox OFED version: **4.0-2.0.0.0** " Best Regards Shuicheng -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 9:37 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Agenda & Notes for 10/10 meeting: 1. Bugs for 2018.10 review (Ghada) no pending storyboard & bugs. Cheers! 2. non-OpenStack patch analysis status (Saul/Zhipeng) init related patches, current build related ones merged by Austin; Zhipeng is looking into solutions for the init related patches to remove from individual RPMs. deep dive to some upstream source level patches and analysis WIP. Brent submitted several new stories and please mark it as "stx-non-distro" tag. 3. Fuzz patch guideline and follow-up actions (Brent/Shuicheng) Going forward we should put guideline for strictly no fuzz patch. AR: Cindy to put a developer guideline to: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines https://storyboard.openstack.org/#!/story/2003917 include all fuzz patches, most of them are merged. Shuicheng will add more tasks (few left) within this week. There is not much we can do w/ the upstreamed patches from sRPM for now. story for upgrade rpm to v14 is not required for now. rpm v14 requires OpenSSH v1.1 and we cannot upgrade from a lower version. For now, the system limitation is that we cannot install the same package with different version simultaneously. 4. Kernel driver upgrade status (Shuicheng) https://storyboard.openstack.org/#!/story/2003597, all patches are ready and pending for testing. Intel GDC do not have appropriate test systems available. AR: Cindy to coordinate Ada about the HW required to run the testing. Mellanox driver version: how to define the version# to upgrade? choose the latest version. usually the version we pick shall be aligned w/ the version in the DPDK supported. Need to look for the version from the Mellanox page. AR: Shuicheng to check the version to make sure it compatible w/ DPDK version in our current mainline. 5. Ceph upgrade status (Vivian/Dehao) - no update due to holiday 6. Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ghada.Khalil at windriver.com Tue Oct 16 22:48:04 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Tue, 16 Oct 2018 22:48:04 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting >> mellanox info Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46AAA0@ALA-MBD.corp.ad.wrs.com> My apologies ...I included the wrong version for starlingx below. The correct version is: 4.3.3.0.2.1 srcrpm name: mlnx-ofa_kernel-4.3-OFED.4.3.3.0.2.1.gcf60532.tis.1.src.rpm -----Original Message----- From: Khalil, Ghada Sent: Tuesday, October 16, 2018 6:38 PM To: 'Lin, Shuicheng'; Xie, Cindy; starlingx-discuss at lists.starlingx.io; Rowsell, Brent Subject: RE: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting >> mellanox info Hi Shuicheng, I found the recommended Mellanox versions in the DPDK documentation. You are correct in that this information is not explicitly listed on the Mellanox website. For DPDK 17.11, it lists the Mellanox OFED version as 4.2 https://doc.dpdk.org/guides-17.11/nics/mlx5.html See section 21.5.1.2 For DPDK 18.02 & 18.05, the versions listed are: 4.2 & 4.3 https://doc.dpdk.org/guides-18.02/nics/mlx5.html https://doc.dpdk.org/guides-18.05/nics/mlx5.html For DPDK 18.08, the versions listed are: 4.3 & 4.4 https://doc.dpdk.org/guides-18.08/nics/mlx5.html In StarlingX, we are currently using dpdk 17.11 (bundled w/ ovs-dpdk). The mellanox driver version we are using is 4.3.1.0.1.1 Regards, Ghada -----Original Message----- From: Lin, Shuicheng [mailto:shuicheng.lin at intel.com] Sent: Friday, October 12, 2018 2:47 AM To: Xie, Cindy; starlingx-discuss at lists.starlingx.io; Khalil, Ghada; Rowsell, Brent Subject: RE: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Hi Ghada/Brent, I try to go through the doc in mellanox website, but cannot find the matrix between mlx driver and DPDK. Could you do me a favor where could I get it? Thanks. Here I get the latest mlx driver: http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers From the dpdk package (dpdk-17.05.2.tar.gz) in our code, I get below info, which seems obsolete. " Currently supported by DPDK: - Mellanox OFED version: **4.0-2.0.0.0** " Best Regards Shuicheng -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 9:37 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Agenda & Notes for 10/10 meeting: 1. Bugs for 2018.10 review (Ghada) no pending storyboard & bugs. Cheers! 2. non-OpenStack patch analysis status (Saul/Zhipeng) init related patches, current build related ones merged by Austin; Zhipeng is looking into solutions for the init related patches to remove from individual RPMs. deep dive to some upstream source level patches and analysis WIP. Brent submitted several new stories and please mark it as "stx-non-distro" tag. 3. Fuzz patch guideline and follow-up actions (Brent/Shuicheng) Going forward we should put guideline for strictly no fuzz patch. AR: Cindy to put a developer guideline to: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines https://storyboard.openstack.org/#!/story/2003917 include all fuzz patches, most of them are merged. Shuicheng will add more tasks (few left) within this week. There is not much we can do w/ the upstreamed patches from sRPM for now. story for upgrade rpm to v14 is not required for now. rpm v14 requires OpenSSH v1.1 and we cannot upgrade from a lower version. For now, the system limitation is that we cannot install the same package with different version simultaneously. 4. Kernel driver upgrade status (Shuicheng) https://storyboard.openstack.org/#!/story/2003597, all patches are ready and pending for testing. Intel GDC do not have appropriate test systems available. AR: Cindy to coordinate Ada about the HW required to run the testing. Mellanox driver version: how to define the version# to upgrade? choose the latest version. usually the version we pick shall be aligned w/ the version in the DPDK supported. Need to look for the version from the Mellanox page. AR: Shuicheng to check the version to make sure it compatible w/ DPDK version in our current mainline. 5. Ceph upgrade status (Vivian/Dehao) - no update due to holiday 6. Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ada.cabrales at intel.com Tue Oct 16 23:28:30 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Tue, 16 Oct 2018 23:28:30 +0000 Subject: [Starlingx-discuss] [ Test ] October release testing status - 10-16-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C35399B@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-16-25-r-2018.10.iso - 31 TCs PASS, 0 TCs FAIL - Simplex and Duplex virtual environment blocked by the openvswitch problem [0]. - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 2/55/21/78 Simplex virtual - 0/10/61/71 Duplex bare metal - 1/70/29/100 Duplex virtual - 2/60/43/105 Multinode local storage bare metal - 3/80/35/118 Multinode local storage virtual - 7/73/45/125 Multinode external storage bare metal - 0/76/41/117 Multinode external storage virtual - 2/52/66/120 Full report in the wiki [1] [0] https://bugs.launchpad.net/starlingx/+bug/1796380 [1] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary A. From shuicheng.lin at intel.com Wed Oct 17 02:05:12 2018 From: shuicheng.lin at intel.com (Lin, Shuicheng) Date: Wed, 17 Oct 2018 02:05:12 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting >> mellanox info In-Reply-To: <151EE31B9FCCA54397A757BC674650F0BA46AAA0@ALA-MBD.corp.ad.wrs.com> References: <151EE31B9FCCA54397A757BC674650F0BA46AAA0@ALA-MBD.corp.ad.wrs.com> Message-ID: <9700A18779F35F49AF027300A49E7C765FE129A9@SHSMSX101.ccr.corp.intel.com> Thanks Ghada for the info. So I will keep current mellanox driver, since it is already higher than the version supported by dpdk. Best Regards Shuicheng -----Original Message----- From: Khalil, Ghada [mailto:Ghada.Khalil at windriver.com] Sent: Wednesday, October 17, 2018 6:48 AM To: Lin, Shuicheng ; Xie, Cindy ; starlingx-discuss at lists.starlingx.io; Rowsell, Brent Subject: RE: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting >> mellanox info My apologies ...I included the wrong version for starlingx below. The correct version is: 4.3.3.0.2.1 srcrpm name: mlnx-ofa_kernel-4.3-OFED.4.3.3.0.2.1.gcf60532.tis.1.src.rpm -----Original Message----- From: Khalil, Ghada Sent: Tuesday, October 16, 2018 6:38 PM To: 'Lin, Shuicheng'; Xie, Cindy; starlingx-discuss at lists.starlingx.io; Rowsell, Brent Subject: RE: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting >> mellanox info Hi Shuicheng, I found the recommended Mellanox versions in the DPDK documentation. You are correct in that this information is not explicitly listed on the Mellanox website. For DPDK 17.11, it lists the Mellanox OFED version as 4.2 https://doc.dpdk.org/guides-17.11/nics/mlx5.html See section 21.5.1.2 For DPDK 18.02 & 18.05, the versions listed are: 4.2 & 4.3 https://doc.dpdk.org/guides-18.02/nics/mlx5.html https://doc.dpdk.org/guides-18.05/nics/mlx5.html For DPDK 18.08, the versions listed are: 4.3 & 4.4 https://doc.dpdk.org/guides-18.08/nics/mlx5.html In StarlingX, we are currently using dpdk 17.11 (bundled w/ ovs-dpdk). The mellanox driver version we are using is 4.3.1.0.1.1 Regards, Ghada -----Original Message----- From: Lin, Shuicheng [mailto:shuicheng.lin at intel.com] Sent: Friday, October 12, 2018 2:47 AM To: Xie, Cindy; starlingx-discuss at lists.starlingx.io; Khalil, Ghada; Rowsell, Brent Subject: RE: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Hi Ghada/Brent, I try to go through the doc in mellanox website, but cannot find the matrix between mlx driver and DPDK. Could you do me a favor where could I get it? Thanks. Here I get the latest mlx driver: http://www.mellanox.com/page/products_dyn?product_family=26&mtag=linux_sw_drivers From the dpdk package (dpdk-17.05.2.tar.gz) in our code, I get below info, which seems obsolete. " Currently supported by DPDK: - Mellanox OFED version: **4.0-2.0.0.0** " Best Regards Shuicheng -----Original Message----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Wednesday, October 10, 2018 9:37 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting Agenda & Notes for 10/10 meeting: 1. Bugs for 2018.10 review (Ghada) no pending storyboard & bugs. Cheers! 2. non-OpenStack patch analysis status (Saul/Zhipeng) init related patches, current build related ones merged by Austin; Zhipeng is looking into solutions for the init related patches to remove from individual RPMs. deep dive to some upstream source level patches and analysis WIP. Brent submitted several new stories and please mark it as "stx-non-distro" tag. 3. Fuzz patch guideline and follow-up actions (Brent/Shuicheng) Going forward we should put guideline for strictly no fuzz patch. AR: Cindy to put a developer guideline to: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines https://storyboard.openstack.org/#!/story/2003917 include all fuzz patches, most of them are merged. Shuicheng will add more tasks (few left) within this week. There is not much we can do w/ the upstreamed patches from sRPM for now. story for upgrade rpm to v14 is not required for now. rpm v14 requires OpenSSH v1.1 and we cannot upgrade from a lower version. For now, the system limitation is that we cannot install the same package with different version simultaneously. 4. Kernel driver upgrade status (Shuicheng) https://storyboard.openstack.org/#!/story/2003597, all patches are ready and pending for testing. Intel GDC do not have appropriate test systems available. AR: Cindy to coordinate Ada about the HW required to run the testing. Mellanox driver version: how to define the version# to upgrade? choose the latest version. usually the version we pick shall be aligned w/ the version in the DPDK supported. Need to look for the version from the Mellanox page. AR: Shuicheng to check the version to make sure it compatible w/ DPDK version in our current mainline. 5. Ceph upgrade status (Vivian/Dehao) - no update due to holiday 6. Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; 'Ildiko Vancsa'; starlingx-discuss at lists.starlingx.io Cc: Perez Rodriguez, Humberto I; Hu, Yong; Zhu, Vivian; 'Chen, Jacky'; 'Leo Xu'; 'Waines, Greg'; 'Eslimi, Dariush'; 'Komiyama, Takeo'; Martinez Monroy, Elio; Jones, Bruce E; Hernandez Gonzalez, Fernando; Hu, Wei W; Qi, Mingyuan; 'Young, Ken'; Arce Moreno, Abraham; 'Seiler, Glenn'; Gomez, Juan P; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 10, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From huifeng.le at intel.com Wed Oct 17 06:39:18 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Wed, 17 Oct 2018 06:39:18 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 0b7c63a Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D520557@SHSMSX104.ccr.corp.intel.com> Hi Matt, We're looking into the patch 0b7c63a, which target to solve the issue of "Plugin: force a refresh of port db object" but failed to reproduce the issue, attachment is the POC steps and result, could you please help to check what's the gap in our POC and share your test steps to reproduce the issue? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: patch-0b7c63a POC.DOCX Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 16703 bytes Desc: patch-0b7c63a POC.DOCX URL: From huifeng.le at intel.com Wed Oct 17 08:52:16 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Wed, 17 Oct 2018 08:52:16 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D52059D@SHSMSX104.ccr.corp.intel.com> Matt, Yes, your thought is correct - the check is not using individual property rule. When creating network with a qos_policy_id, neutron will try to get the qos policy object within current context (e.g. by QosCoreResourceExtension. _get_policy_obj) then raises QosPolicyNotFound exception if qos_policy_id is not owned by the project. During my test, I can see below error raised by neutron: {"NeutronError": { "message": "QoS policy bda58768-7785-4ed3-b9b1-768d7979cc70 could not be found.", "type": "QosPolicyNotFound", "detail": "" }} Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 16, 2018 10:13 PM To: Le, Huifeng ; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Re: Question about patch upstreaming for 1e9a089 I was specifically referring to the network attribute that refers to the QoS policy ID, rather than the RBAC rules related to the policies themselves. I don’t see anything in the policy.json that restricts the setting of the network attribute, so I am assuming it is falling back to the network entity policy itself rather than an individual property rule. From: "Le, Huifeng" > Date: Tuesday, October 16, 2018 at 10:08 AM To: Allain Legacy >, "Peters, Matt" > Cc: "starlingx-discuss at lists.starlingx.io" > Subject: RE: Question about patch upstreaming for 1e9a089 Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): “QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects.” From the default policy.json, e.g. “create/update/delete” operation is for admin only and “get” operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt >; Le, Huifeng > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have “enforce_policy” set in its attribute definition so I am going to guess that there is no checking at all for it. I don’t have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don’t even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We’re looking into the patch 1e9a089, which target to solve the issue of “CGTS-2408: bypass policy check when setting attribute to None.” After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: “type: uuid_or_none” (2) the “enforce_policy” should be set to “True” and the policy needs to be defined in etc/policy.json (3) the check policy uses “tenant_id” for owner check The only found attribute which will be impacted by this issue in STX is “wrs-tm:qos” which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What’s your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1808 bytes Desc: image001.png URL: From zhipengs.liu at intel.com Wed Oct 17 09:11:09 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Wed, 17 Oct 2018 09:11:09 +0000 Subject: [Starlingx-discuss] Work on reducing config patch for ntp Message-ID: <93814834B4855241994F290E959305C752FCA9C6@SHSMSX104.ccr.corp.intel.com> Hi Penny, I'd like to discuss with you about this 2 patches. ntpd-started-no-g-option.patch //change /etc/sysconfig/ntpd ntp-spec-add-TiS-custom-config-files.patch // add ntp.conf file to /etc/ Can we remove this 2 patches? In stx/stx-config/puppet-manifests/src/modules/platform/manifests/ntp.pp, it already generates ntp.conf file by puppet file { 'ntp_config': ensure => file, path => '/etc/ntp.conf', mode => '0640', content => template('platform/ntp.conf.server.erb'), } For /etc/sysconfig/ntpd, I'd like to add this custom file in ntp.pp, then we can remove related patch. Any comment from you? Thanks! BTW, I saw some other packages also have pp file in stx/stx-config/puppet-manifests/src/modules/platform/manifests/ So I can do some modification on existed pp to change/add some config file, right? For SW patching scenario, it is OK for non-in-service update. After reboot, puppet will apply config again. But for in-service update, without reboot, how to apply this custom file again after patching. That's why I thought of using patching-script. Do you have any other proposal? Thanks! zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From cindy.xie at intel.com Wed Oct 17 13:40:09 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Wed, 17 Oct 2018 13:40:09 +0000 Subject: [Starlingx-discuss] Minutes: Weekly StarlingX non-OpenStack Distro meeting, 10/17 Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D728BF@SHSMSX104.ccr.corp.intel.com> Agenda & notes for 10/17 meeting: - Bugs for 2018.10 review (Ghada) Lshell component will be deprecated: https://bugs.launchpad.net/starlingx/+bug/1795451 Shuicheng to work offline to address the patch reviews. - non-OpenStack patch analysis status (Saul/Zhipeng) not much update/changes from Saul's side. 1st patch submitted by Zhipeng for nfs-util: https://review.openstack.org/#/c/610459/. Following Penney's review comments, will use puppet for the configuration. Still working on after-patching scripts. For in-service sw-update may not require reboot, the puppet scripts will not be run. Brent's comments: may depends on type of changes. Will discuss in the mailing list for the design details. needs to change the existing puppet config files, no need to create folders for puppet config files. Need to discuss case by case. Zhipeng want to use platform-util to config the destinations scripts. tis-extension is designed for 3rd-party. Saul wants to have each changes for each RPM files. not to use platform-util. We may have to document the design of how to handle it. Shall be based on scenario. NTP can be a special case. Zhipeng to work on a design doc for the general cases, and also identify the special cases how to handle it. - Kernel driver upgrade update (Shuicheng) mellanox driver upgrade has been dropped. only 3 drivers will be upgraded: QAT, DRBD, e1000e drivers are the ones being upgraded. Still waiting for the test resource to be available before we can merge them. - Ceph upgrade status (Vivian/Dehao) Not much updated. Dehao lost the access to github. Checking patching list and rebase it next. Will follow the requirements to cherry pick those. Check if there is any puppet changes required to config the newer version of Ceph. Except for Python-ceph, it's not depends on Puppet. - Opens (All): None -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; Chen, Yan; Wold, Saul; Sun, Austin; Wang, Yi C; starlingx-discuss at lists.starlingx.io; Lin, Shuicheng; Khalil, Ghada; Rowsell, Brent; Somerville, Jim; 'Ildiko Vancsa' Cc: 'Leo Xu'; Zhu, Vivian; 'Waines, Greg'; Qi, Mingyuan; 'Chen, Jacky'; 'Young, Ken'; Hernandez Gonzalez, Fernando; 'Komiyama, Takeo'; Perez Rodriguez, Humberto I; 'Seiler, Glenn'; Arce Moreno, Abraham; Hu, Yong; Gomez, Juan P; Jones, Bruce E; Hu, Wei W; 'Eslimi, Dariush'; Martinez Monroy, Elio; Shang, Dehao; Cobbley, David A Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 17, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other From Matt.Peters at windriver.com Wed Oct 17 13:43:12 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Wed, 17 Oct 2018 13:43:12 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D52059D@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> <76647BD697F40748B1FA4F56DA02AA0B4D52059D@SHSMSX104.ccr.corp.intel.com> Message-ID: Hi Huifeng, What is the behavior if you modify the policy to None (using openstack network set --no-qos-policy )? -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 4:52 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Matt, Yes, your thought is correct - the check is not using individual property rule. When creating network with a qos_policy_id, neutron will try to get the qos policy object within current context (e.g. by QosCoreResourceExtension. _get_policy_obj) then raises QosPolicyNotFound exception if qos_policy_id is not owned by the project. During my test, I can see below error raised by neutron: {"NeutronError": { "message": "QoS policy bda58768-7785-4ed3-b9b1-768d7979cc70 could not be found.", "type": "QosPolicyNotFound", "detail": "" }} Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 16, 2018 10:13 PM To: Le, Huifeng ; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Re: Question about patch upstreaming for 1e9a089 I was specifically referring to the network attribute that refers to the QoS policy ID, rather than the RBAC rules related to the policies themselves. I don’t see anything in the policy.json that restricts the setting of the network attribute, so I am assuming it is falling back to the network entity policy itself rather than an individual property rule. From: "Le, Huifeng" > Date: Tuesday, October 16, 2018 at 10:08 AM To: Allain Legacy >, "Peters, Matt" > Cc: "starlingx-discuss at lists.starlingx.io" > Subject: RE: Question about patch upstreaming for 1e9a089 Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): “QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects.” From the default policy.json, e.g. “create/update/delete” operation is for admin only and “get” operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt >; Le, Huifeng > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have “enforce_policy” set in its attribute definition so I am going to guess that there is no checking at all for it. I don’t have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don’t even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We’re looking into the patch 1e9a089, which target to solve the issue of “CGTS-2408: bypass policy check when setting attribute to None.” After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: “type: uuid_or_none” (2) the “enforce_policy” should be set to “True” and the policy needs to be defined in etc/policy.json (3) the check policy uses “tenant_id” for owner check The only found attribute which will be impacted by this issue in STX is “wrs-tm:qos” which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What’s your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1808 bytes Desc: image001.png URL: From huifeng.le at intel.com Wed Oct 17 13:54:18 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Wed, 17 Oct 2018 13:54:18 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> <76647BD697F40748B1FA4F56DA02AA0B4D52059D@SHSMSX104.ccr.corp.intel.com> Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D52060C@SHSMSX104.ccr.corp.intel.com> Matt, the command can be executed success and the qos_policy_id will be set to “None” after the operation. thanks! Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Wednesday, October 17, 2018 9:43 PM To: Le, Huifeng ; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Hi Huifeng, What is the behavior if you modify the policy to None (using openstack network set --no-qos-policy )? -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 4:52 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Matt, Yes, your thought is correct - the check is not using individual property rule. When creating network with a qos_policy_id, neutron will try to get the qos policy object within current context (e.g. by QosCoreResourceExtension. _get_policy_obj) then raises QosPolicyNotFound exception if qos_policy_id is not owned by the project. During my test, I can see below error raised by neutron: {"NeutronError": { "message": "QoS policy bda58768-7785-4ed3-b9b1-768d7979cc70 could not be found.", "type": "QosPolicyNotFound", "detail": "" }} Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 16, 2018 10:13 PM To: Le, Huifeng >; Legacy, Allain > Cc: starlingx-discuss at lists.starlingx.io Subject: Re: Question about patch upstreaming for 1e9a089 I was specifically referring to the network attribute that refers to the QoS policy ID, rather than the RBAC rules related to the policies themselves. I don’t see anything in the policy.json that restricts the setting of the network attribute, so I am assuming it is falling back to the network entity policy itself rather than an individual property rule. From: "Le, Huifeng" > Date: Tuesday, October 16, 2018 at 10:08 AM To: Allain Legacy >, "Peters, Matt" > Cc: "starlingx-discuss at lists.starlingx.io" > Subject: RE: Question about patch upstreaming for 1e9a089 Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): “QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects.” From the default policy.json, e.g. “create/update/delete” operation is for admin only and “get” operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt >; Le, Huifeng > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have “enforce_policy” set in its attribute definition so I am going to guess that there is no checking at all for it. I don’t have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don’t even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We’re looking into the patch 1e9a089, which target to solve the issue of “CGTS-2408: bypass policy check when setting attribute to None.” After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: “type: uuid_or_none” (2) the “enforce_policy” should be set to “True” and the policy needs to be defined in etc/policy.json (3) the check policy uses “tenant_id” for owner check The only found attribute which will be impacted by this issue in STX is “wrs-tm:qos” which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What’s your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1808 bytes Desc: image001.png URL: From Matt.Peters at windriver.com Wed Oct 17 14:12:41 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Wed, 17 Oct 2018 14:12:41 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 0b7c63a In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D520557@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D520557@SHSMSX104.ccr.corp.intel.com> Message-ID: This patch is only applicable for "guest vlans", which has been deprecated and replaced with trunk ports. Therefore this patch can be abandoned. From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 2:39 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io Subject: Question about patch upstreaming for 0b7c63a Hi Matt, We're looking into the patch 0b7c63a, which target to solve the issue of "Plugin: force a refresh of port db object" but failed to reproduce the issue, attachment is the POC steps and result, could you please help to check what's the gap in our POC and share your test steps to reproduce the issue? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Wed Oct 17 14:35:42 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Wed, 17 Oct 2018 14:35:42 +0000 Subject: [Starlingx-discuss] StarlingX community call notes Oct 17 2018 Message-ID: <9A85D2917C58154C960D95352B22818BB1E7E771@ORSMSX158.amr.corp.intel.com> Agenda and Notes for the Oct 17th call * Release build and testing update o https://bugs.launchpad.net/starlingx/+bugs?field.tag=stx.2018.10] o All currently open bugs are in progress o Testing expected to complete this week except for virtual environment which is blocked by https://bugs.launchpad.net/starlingx/+bug/1796380 * Release notes status o Visible in the main documents page: https://docs.starlingx.io/ o Work is in progress to add a StarlingX theme to the pages * Multi-OS meeting - Please join if you are interested * Cengn update o Progressing well - Ken updated that they've met with Cengn, working with them to get access to a bare metal server, focusing first on the mirror available in time for the Nov Summit. * StarlingX webinar next week (10/24) - please attend! * Dean is posting changes to stx-governance - please review. -------------- next part -------------- An HTML attachment was scrubbed... URL: From juan.carlos.alonso at intel.com Wed Oct 17 15:47:47 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Wed, 17 Oct 2018 15:47:47 +0000 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> Message-ID: <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> Hello, Does anybody has faced the same behavior? I am wondering if it is possible to delete a snapshot, or not, when an instance is up and running. In my site, I could, but need to know if this is correct or not. Regards. Juan Carlos Alonso From: Alonso, Juan Carlos Sent: Tuesday, October 16, 2018 3:14 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Delete snapshot and volume Hi, The test 211[1] indicates that an Image should be created, then a Volume should be created from the image, then a Snapshot should be created from the Volume and finally an Instance should be launch from the Snapshot. The test indicates that we try to delete the Snapshot and Volume while the Instance is running, Snapshot nor Volume cannot be deleted. But when tried found that the Snapshot can be deleted, but the Volume cannot. Is this behavior correct? Those are the commands I used: openstack image create --disk-format qcow2 --file /home/wrsroot/cirros-0.4.0-x86_64-disk.img --public cirros openstack volume create --image cirros --availability-zone nova --size 1 cirros-vol openstack volume snapshot create --name cirros-snap cirros-vol openstack server create --volume cirros-vol --flavor m1.small --availibility-zone nova vm-cirros 1 https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From michel.thebeau at windriver.com Wed Oct 17 16:22:29 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Wed, 17 Oct 2018 12:22:29 -0400 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> Message-ID: <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> Hi Juan Carlos, I had inquired internally at WR. My current understanding is that one does not "launch an instance from a volume snapshot", except in the form presented by Horizon interface.  In that case, the behaviour of Horizon form is to create a volume from volume snapshot and launch the instance from the new volume. In your email you listed CLI commands and the command you listed showed that you launched the instance from volume, not from volume snapshot.  So there's no reason to believe the volume snapshot should be locked from deletion. Moreover, it doesn't make sense for an instance to run from a volume snapshot - that's not what a  snapshot is.  The snapshot itself should be a static thing.  You could probably have used the "openstack volume create --snapshot ..." command option to create a volume from volume snapshot, and then launch an instance from the new volume.  The volume snapshot in this case would not be locked from deletion, neither the original volume from which the snapshot was based, but only the new volume you created. I hope this information is accurate and helpful  :)  I'll forward the response to the guys I inquired with, and they can correct me here if anything I wrote is amiss. M On 2018-10-17 11:47 a.m., Alonso, Juan Carlos wrote: > Hello, > > Does anybody has faced the same behavior? > I am wondering if it is possible to delete a snapshot, or not, when an instance is up and running. > In my site, I could, but need to know if this is correct or not. > > Regards. > Juan Carlos Alonso > > From: Alonso, Juan Carlos > Sent: Tuesday, October 16, 2018 3:14 PM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Delete snapshot and volume > > Hi, > > The test 211[1] indicates that an Image should be created, then a Volume should be created from the image, then a Snapshot should be created from the Volume and finally an Instance should be launch from the Snapshot. > The test indicates that we try to delete the Snapshot and Volume while the Instance is running, Snapshot nor Volume cannot be deleted. > But when tried found that the Snapshot can be deleted, but the Volume cannot. > > Is this behavior correct? > > Those are the commands I used: > > openstack image create --disk-format qcow2 --file /home/wrsroot/cirros-0.4.0-x86_64-disk.img --public cirros > openstack volume create --image cirros --availability-zone nova --size 1 cirros-vol > openstack volume snapshot create --name cirros-snap cirros-vol > openstack server create --volume cirros-vol --flavor m1.small --availibility-zone nova vm-cirros > > > 1 https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > Regards. > Juan Carlos Alonso > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From jose.perez.carranza at intel.com Wed Oct 17 16:48:52 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Wed, 17 Oct 2018 16:48:52 +0000 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5DCD@fmsmsx101.amr.corp.intel.com> I was able to delete the snapshot even when 2 instances were launched from there following below steps: === $ openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- 0.4.0-x86_64-disk.img --public cirros $ openstack volume create --image cirros --availability-zone nova --size 1 cirros-vol $ openstack volume snapshot create --name cirros-snap cirros-vol $ nova boot --flavor --snapshot --nic net-id= $ openstack snapshot delete cirros-snap === So seems like the snapshot is not directly linked to the instance created. Regards, José > -----Original Message----- > From: Michel Thebeau [mailto:michel.thebeau at windriver.com] > Sent: Wednesday, October 17, 2018 11:22 AM > To: Alonso, Juan Carlos ; 'starlingx- > discuss at lists.starlingx.io' > Subject: Re: [Starlingx-discuss] Delete snapshot and volume > > Hi Juan Carlos, > > I had inquired internally at WR. > > My current understanding is that one does not "launch an instance from a > volume snapshot", except in the form presented by Horizon interface.  In that > case, the behaviour of Horizon form is to create a volume from volume > snapshot and launch the instance from the new volume. > > In your email you listed CLI commands and the command you listed showed > that you launched the instance from volume, not from volume snapshot. So > there's no reason to believe the volume snapshot should be locked from > deletion. > > Moreover, it doesn't make sense for an instance to run from a volume > snapshot - that's not what a  snapshot is.  The snapshot itself should be a static > thing.  You could probably have used the "openstack volume create --snapshot > ..." command option to create a volume from volume snapshot, and then > launch an instance from the new volume.  The volume snapshot in this case > would not be locked from deletion, neither the original volume from which > the snapshot was based, but only the new volume you created. > > I hope this information is accurate and helpful  :)  I'll forward the response to > the guys I inquired with, and they can correct me here if anything I wrote is > amiss. > > M > > > > On 2018-10-17 11:47 a.m., Alonso, Juan Carlos wrote: > > Hello, > > > > Does anybody has faced the same behavior? > > I am wondering if it is possible to delete a snapshot, or not, when an instance > is up and running. > > In my site, I could, but need to know if this is correct or not. > > > > Regards. > > Juan Carlos Alonso > > > > From: Alonso, Juan Carlos > > Sent: Tuesday, October 16, 2018 3:14 PM > > To: starlingx-discuss at lists.starlingx.io > > Subject: [Starlingx-discuss] Delete snapshot and volume > > > > Hi, > > > > The test 211[1] indicates that an Image should be created, then a Volume > should be created from the image, then a Snapshot should be created from > the Volume and finally an Instance should be launch from the Snapshot. > > The test indicates that we try to delete the Snapshot and Volume while the > Instance is running, Snapshot nor Volume cannot be deleted. > > But when tried found that the Snapshot can be deleted, but the Volume > cannot. > > > > Is this behavior correct? > > > > Those are the commands I used: > > > > openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- > 0.4.0-x86_64-disk.img --public cirros > > openstack volume create --image cirros --availability-zone nova --size 1 > cirros-vol > > openstack volume snapshot create --name cirros-snap cirros-vol > > openstack server create --volume cirros-vol --flavor m1.small --availibility- > zone nova vm-cirros > > > > > > 1 > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > > > Regards. > > Juan Carlos Alonso > > > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From michel.thebeau at windriver.com Wed Oct 17 17:15:11 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Wed, 17 Oct 2018 13:15:11 -0400 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E5DCD@fmsmsx101.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5DCD@fmsmsx101.amr.corp.intel.com> Message-ID: <7dcc9990-c05d-07c2-7db6-4a88bb2f741c@windriver.com> Hi José, Thanks for the nova boot command.  I see that "nova boot --snapshot" also creates a new volume and launches the instance from the new volume.  I agree that the snapshot is not in use by the instance.  The "openstack server show" and "nova show" commands list the new volume created. M On 2018-10-17 12:48 p.m., Perez Carranza, Jose wrote: > I was able to delete the snapshot even when 2 instances were launched from there following below steps: > > === > $ openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- > 0.4.0-x86_64-disk.img --public cirros > $ openstack volume create --image cirros --availability-zone nova --size 1 > cirros-vol > $ openstack volume snapshot create --name cirros-snap cirros-vol > $ nova boot --flavor --snapshot --nic net-id= > $ openstack snapshot delete cirros-snap > === > > So seems like the snapshot is not directly linked to the instance created. > > Regards, > José > > >> -----Original Message----- >> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >> Sent: Wednesday, October 17, 2018 11:22 AM >> To: Alonso, Juan Carlos ; 'starlingx- >> discuss at lists.starlingx.io' >> Subject: Re: [Starlingx-discuss] Delete snapshot and volume >> >> Hi Juan Carlos, >> >> I had inquired internally at WR. >> >> My current understanding is that one does not "launch an instance from a >> volume snapshot", except in the form presented by Horizon interface.  In that >> case, the behaviour of Horizon form is to create a volume from volume >> snapshot and launch the instance from the new volume. >> >> In your email you listed CLI commands and the command you listed showed >> that you launched the instance from volume, not from volume snapshot. So >> there's no reason to believe the volume snapshot should be locked from >> deletion. >> >> Moreover, it doesn't make sense for an instance to run from a volume >> snapshot - that's not what a  snapshot is.  The snapshot itself should be a static >> thing.  You could probably have used the "openstack volume create --snapshot >> ..." command option to create a volume from volume snapshot, and then >> launch an instance from the new volume.  The volume snapshot in this case >> would not be locked from deletion, neither the original volume from which >> the snapshot was based, but only the new volume you created. >> >> I hope this information is accurate and helpful  :)  I'll forward the response to >> the guys I inquired with, and they can correct me here if anything I wrote is >> amiss. >> >> M >> >> >> >> On 2018-10-17 11:47 a.m., Alonso, Juan Carlos wrote: >>> Hello, >>> >>> Does anybody has faced the same behavior? >>> I am wondering if it is possible to delete a snapshot, or not, when an instance >> is up and running. >>> In my site, I could, but need to know if this is correct or not. >>> >>> Regards. >>> Juan Carlos Alonso >>> >>> From: Alonso, Juan Carlos >>> Sent: Tuesday, October 16, 2018 3:14 PM >>> To: starlingx-discuss at lists.starlingx.io >>> Subject: [Starlingx-discuss] Delete snapshot and volume >>> >>> Hi, >>> >>> The test 211[1] indicates that an Image should be created, then a Volume >> should be created from the image, then a Snapshot should be created from >> the Volume and finally an Instance should be launch from the Snapshot. >>> The test indicates that we try to delete the Snapshot and Volume while the >> Instance is running, Snapshot nor Volume cannot be deleted. >>> But when tried found that the Snapshot can be deleted, but the Volume >> cannot. >>> Is this behavior correct? >>> >>> Those are the commands I used: >>> >>> openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- >> 0.4.0-x86_64-disk.img --public cirros >>> openstack volume create --image cirros --availability-zone nova --size 1 >> cirros-vol >>> openstack volume snapshot create --name cirros-snap cirros-vol >>> openstack server create --volume cirros-vol --flavor m1.small --availibility- >> zone nova vm-cirros >>> >>> 1 >> https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions >>> Regards. >>> Juan Carlos Alonso >>> >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From juan.carlos.alonso at intel.com Wed Oct 17 17:30:12 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Wed, 17 Oct 2018 17:30:12 +0000 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <7dcc9990-c05d-07c2-7db6-4a88bb2f741c@windriver.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5DCD@fmsmsx101.amr.corp.intel.com> <7dcc9990-c05d-07c2-7db6-4a88bb2f741c@windriver.com> Message-ID: <8557B550001AFB46A43A0CCC314BF8514EF0F8E4@FMSMSX108.amr.corp.intel.com> Hi, I tested again based on your feedback and I could check that the snapshot created from the volume, create another volume for each instance launched. Such new volume is used for the instance, so the original volume and snapshot can be deleted without no problems, as Jose said they are not linked to the instance. Thank you for your help. Regards. Juan Carlos Alonso -----Original Message----- From: Michel Thebeau [mailto:michel.thebeau at windriver.com] Sent: Wednesday, October 17, 2018 12:15 PM To: Perez Carranza, Jose ; Alonso, Juan Carlos ; 'starlingx-discuss at lists.starlingx.io' Subject: Re: [Starlingx-discuss] Delete snapshot and volume Hi José, Thanks for the nova boot command.  I see that "nova boot --snapshot" also creates a new volume and launches the instance from the new volume.  I agree that the snapshot is not in use by the instance.  The "openstack server show" and "nova show" commands list the new volume created. M On 2018-10-17 12:48 p.m., Perez Carranza, Jose wrote: > I was able to delete the snapshot even when 2 instances were launched from there following below steps: > > === > $ openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- > 0.4.0-x86_64-disk.img --public cirros $ openstack volume create > --image cirros --availability-zone nova --size 1 > cirros-vol > $ openstack volume snapshot create --name cirros-snap cirros-vol $ > nova boot --flavor --snapshot --nic > net-id= $ openstack snapshot delete > cirros-snap === > > So seems like the snapshot is not directly linked to the instance created. > > Regards, > José > > >> -----Original Message----- >> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >> Sent: Wednesday, October 17, 2018 11:22 AM >> To: Alonso, Juan Carlos ; 'starlingx- >> discuss at lists.starlingx.io' >> Subject: Re: [Starlingx-discuss] Delete snapshot and volume >> >> Hi Juan Carlos, >> >> I had inquired internally at WR. >> >> My current understanding is that one does not "launch an instance >> from a volume snapshot", except in the form presented by Horizon >> interface.  In that case, the behaviour of Horizon form is to create >> a volume from volume snapshot and launch the instance from the new volume. >> >> In your email you listed CLI commands and the command you listed >> showed that you launched the instance from volume, not from volume >> snapshot. So there's no reason to believe the volume snapshot should >> be locked from deletion. >> >> Moreover, it doesn't make sense for an instance to run from a volume >> snapshot - that's not what a  snapshot is.  The snapshot itself >> should be a static thing.  You could probably have used the >> "openstack volume create --snapshot ..." command option to create a >> volume from volume snapshot, and then launch an instance from the new >> volume.  The volume snapshot in this case would not be locked from >> deletion, neither the original volume from which the snapshot was based, but only the new volume you created. >> >> I hope this information is accurate and helpful  :)  I'll forward the >> response to the guys I inquired with, and they can correct me here if >> anything I wrote is amiss. >> >> M >> >> >> >> On 2018-10-17 11:47 a.m., Alonso, Juan Carlos wrote: >>> Hello, >>> >>> Does anybody has faced the same behavior? >>> I am wondering if it is possible to delete a snapshot, or not, when >>> an instance >> is up and running. >>> In my site, I could, but need to know if this is correct or not. >>> >>> Regards. >>> Juan Carlos Alonso >>> >>> From: Alonso, Juan Carlos >>> Sent: Tuesday, October 16, 2018 3:14 PM >>> To: starlingx-discuss at lists.starlingx.io >>> Subject: [Starlingx-discuss] Delete snapshot and volume >>> >>> Hi, >>> >>> The test 211[1] indicates that an Image should be created, then a >>> Volume >> should be created from the image, then a Snapshot should be created >> from the Volume and finally an Instance should be launch from the Snapshot. >>> The test indicates that we try to delete the Snapshot and Volume >>> while the >> Instance is running, Snapshot nor Volume cannot be deleted. >>> But when tried found that the Snapshot can be deleted, but the >>> Volume >> cannot. >>> Is this behavior correct? >>> >>> Those are the commands I used: >>> >>> openstack image create --disk-format qcow2 --file >>> /home/wrsroot/cirros- >> 0.4.0-x86_64-disk.img --public cirros >>> openstack volume create --image cirros --availability-zone nova >>> --size 1 >> cirros-vol >>> openstack volume snapshot create --name cirros-snap cirros-vol >>> openstack server create --volume cirros-vol --flavor m1.small >>> --availibility- >> zone nova vm-cirros >>> >>> 1 >> https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instru >> ctions >>> Regards. >>> Juan Carlos Alonso >>> >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Daniel.Badea at windriver.com Wed Oct 17 18:21:22 2018 From: Daniel.Badea at windriver.com (Badea, Daniel) Date: Wed, 17 Oct 2018 18:21:22 +0000 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <8557B550001AFB46A43A0CCC314BF8514EF0F8E4@FMSMSX108.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5DCD@fmsmsx101.amr.corp.intel.com> <7dcc9990-c05d-07c2-7db6-4a88bb2f741c@windriver.com>, <8557B550001AFB46A43A0CCC314BF8514EF0F8E4@FMSMSX108.amr.corp.intel.com> Message-ID: <9174DAE490321844AE273F6AD001E3EA9D7F1D37@ALA-MBD.corp.ad.wrs.com> Hi Juan, openstack volume snapshot create does not have a --name parameter so your example will not run. It does have a --volume parameter but then the order of names is reversed: openstack volume snapshot create --volume cirros-vol cirros-snap Regards, Daniel Badea ________________________________________ From: Alonso, Juan Carlos [juan.carlos.alonso at intel.com] Sent: Wednesday, October 17, 2018 20:30 To: Thebeau, Michel; Perez Carranza, Jose; 'starlingx-discuss at lists.starlingx.io' Subject: Re: [Starlingx-discuss] Delete snapshot and volume Hi, I tested again based on your feedback and I could check that the snapshot created from the volume, create another volume for each instance launched. Such new volume is used for the instance, so the original volume and snapshot can be deleted without no problems, as Jose said they are not linked to the instance. Thank you for your help. Regards. Juan Carlos Alonso -----Original Message----- From: Michel Thebeau [mailto:michel.thebeau at windriver.com] Sent: Wednesday, October 17, 2018 12:15 PM To: Perez Carranza, Jose ; Alonso, Juan Carlos ; 'starlingx-discuss at lists.starlingx.io' Subject: Re: [Starlingx-discuss] Delete snapshot and volume Hi José, Thanks for the nova boot command. I see that "nova boot --snapshot" also creates a new volume and launches the instance from the new volume. I agree that the snapshot is not in use by the instance. The "openstack server show" and "nova show" commands list the new volume created. M On 2018-10-17 12:48 p.m., Perez Carranza, Jose wrote: > I was able to delete the snapshot even when 2 instances were launched from there following below steps: > > === > $ openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- > 0.4.0-x86_64-disk.img --public cirros $ openstack volume create > --image cirros --availability-zone nova --size 1 > cirros-vol > $ openstack volume snapshot create --name cirros-snap cirros-vol $ > nova boot --flavor --snapshot --nic > net-id= $ openstack snapshot delete > cirros-snap === > > So seems like the snapshot is not directly linked to the instance created. > > Regards, > José > > >> -----Original Message----- >> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >> Sent: Wednesday, October 17, 2018 11:22 AM >> To: Alonso, Juan Carlos ; 'starlingx- >> discuss at lists.starlingx.io' >> Subject: Re: [Starlingx-discuss] Delete snapshot and volume >> >> Hi Juan Carlos, >> >> I had inquired internally at WR. >> >> My current understanding is that one does not "launch an instance >> from a volume snapshot", except in the form presented by Horizon >> interface. In that case, the behaviour of Horizon form is to create >> a volume from volume snapshot and launch the instance from the new volume. >> >> In your email you listed CLI commands and the command you listed >> showed that you launched the instance from volume, not from volume >> snapshot. So there's no reason to believe the volume snapshot should >> be locked from deletion. >> >> Moreover, it doesn't make sense for an instance to run from a volume >> snapshot - that's not what a snapshot is. The snapshot itself >> should be a static thing. You could probably have used the >> "openstack volume create --snapshot ..." command option to create a >> volume from volume snapshot, and then launch an instance from the new >> volume. The volume snapshot in this case would not be locked from >> deletion, neither the original volume from which the snapshot was based, but only the new volume you created. >> >> I hope this information is accurate and helpful :) I'll forward the >> response to the guys I inquired with, and they can correct me here if >> anything I wrote is amiss. >> >> M >> >> >> >> On 2018-10-17 11:47 a.m., Alonso, Juan Carlos wrote: >>> Hello, >>> >>> Does anybody has faced the same behavior? >>> I am wondering if it is possible to delete a snapshot, or not, when >>> an instance >> is up and running. >>> In my site, I could, but need to know if this is correct or not. >>> >>> Regards. >>> Juan Carlos Alonso >>> >>> From: Alonso, Juan Carlos >>> Sent: Tuesday, October 16, 2018 3:14 PM >>> To: starlingx-discuss at lists.starlingx.io >>> Subject: [Starlingx-discuss] Delete snapshot and volume >>> >>> Hi, >>> >>> The test 211[1] indicates that an Image should be created, then a >>> Volume >> should be created from the image, then a Snapshot should be created >> from the Volume and finally an Instance should be launch from the Snapshot. >>> The test indicates that we try to delete the Snapshot and Volume >>> while the >> Instance is running, Snapshot nor Volume cannot be deleted. >>> But when tried found that the Snapshot can be deleted, but the >>> Volume >> cannot. >>> Is this behavior correct? >>> >>> Those are the commands I used: >>> >>> openstack image create --disk-format qcow2 --file >>> /home/wrsroot/cirros- >> 0.4.0-x86_64-disk.img --public cirros >>> openstack volume create --image cirros --availability-zone nova >>> --size 1 >> cirros-vol >>> openstack volume snapshot create --name cirros-snap cirros-vol >>> openstack server create --volume cirros-vol --flavor m1.small >>> --availibility- >> zone nova vm-cirros >>> >>> 1 >> https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instru >> ctions >>> Regards. >>> Juan Carlos Alonso >>> >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From juan.carlos.alonso at intel.com Wed Oct 17 18:45:46 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Wed, 17 Oct 2018 18:45:46 +0000 Subject: [Starlingx-discuss] Delete snapshot and volume In-Reply-To: <9174DAE490321844AE273F6AD001E3EA9D7F1D37@ALA-MBD.corp.ad.wrs.com> References: <8557B550001AFB46A43A0CCC314BF8514EF0E5AB@FMSMSX108.amr.corp.intel.com> <8557B550001AFB46A43A0CCC314BF8514EF0F845@FMSMSX108.amr.corp.intel.com> <775afc0c-3840-05d0-2961-fead802e8aed@windriver.com> <0A5D9A624DF90343892F8F3FE7DE525A2A8E5DCD@fmsmsx101.amr.corp.intel.com> <7dcc9990-c05d-07c2-7db6-4a88bb2f741c@windriver.com>, <8557B550001AFB46A43A0CCC314BF8514EF0F8E4@FMSMSX108.amr.corp.intel.com> <9174DAE490321844AE273F6AD001E3EA9D7F1D37@ALA-MBD.corp.ad.wrs.com> Message-ID: <8557B550001AFB46A43A0CCC314BF8514EF0F914@FMSMSX108.amr.corp.intel.com> Hi Daniel, Yes, you right, I made a mistake when write the commands on this email. The command I used was: "openstack snapshot create --name ", but seems that such command has been deprecated or will be. Command you mention should be used now. Regards. Juan Carlos Alonso -----Original Message----- From: Badea, Daniel [mailto:Daniel.Badea at windriver.com] Sent: Wednesday, October 17, 2018 1:21 PM To: Alonso, Juan Carlos ; Thebeau, Michel ; Perez Carranza, Jose ; 'starlingx-discuss at lists.starlingx.io' Subject: RE: [Starlingx-discuss] Delete snapshot and volume Hi Juan, openstack volume snapshot create does not have a --name parameter so your example will not run. It does have a --volume parameter but then the order of names is reversed: openstack volume snapshot create --volume cirros-vol cirros-snap Regards, Daniel Badea ________________________________________ From: Alonso, Juan Carlos [juan.carlos.alonso at intel.com] Sent: Wednesday, October 17, 2018 20:30 To: Thebeau, Michel; Perez Carranza, Jose; 'starlingx-discuss at lists.starlingx.io' Subject: Re: [Starlingx-discuss] Delete snapshot and volume Hi, I tested again based on your feedback and I could check that the snapshot created from the volume, create another volume for each instance launched. Such new volume is used for the instance, so the original volume and snapshot can be deleted without no problems, as Jose said they are not linked to the instance. Thank you for your help. Regards. Juan Carlos Alonso -----Original Message----- From: Michel Thebeau [mailto:michel.thebeau at windriver.com] Sent: Wednesday, October 17, 2018 12:15 PM To: Perez Carranza, Jose ; Alonso, Juan Carlos ; 'starlingx-discuss at lists.starlingx.io' Subject: Re: [Starlingx-discuss] Delete snapshot and volume Hi José, Thanks for the nova boot command. I see that "nova boot --snapshot" also creates a new volume and launches the instance from the new volume. I agree that the snapshot is not in use by the instance. The "openstack server show" and "nova show" commands list the new volume created. M On 2018-10-17 12:48 p.m., Perez Carranza, Jose wrote: > I was able to delete the snapshot even when 2 instances were launched from there following below steps: > > === > $ openstack image create --disk-format qcow2 --file /home/wrsroot/cirros- > 0.4.0-x86_64-disk.img --public cirros $ openstack volume create > --image cirros --availability-zone nova --size 1 > cirros-vol > $ openstack volume snapshot create --name cirros-snap cirros-vol $ > nova boot --flavor --snapshot --nic > net-id= $ openstack snapshot delete > cirros-snap === > > So seems like the snapshot is not directly linked to the instance created. > > Regards, > José > > >> -----Original Message----- >> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >> Sent: Wednesday, October 17, 2018 11:22 AM >> To: Alonso, Juan Carlos ; 'starlingx- >> discuss at lists.starlingx.io' >> Subject: Re: [Starlingx-discuss] Delete snapshot and volume >> >> Hi Juan Carlos, >> >> I had inquired internally at WR. >> >> My current understanding is that one does not "launch an instance >> from a volume snapshot", except in the form presented by Horizon >> interface. In that case, the behaviour of Horizon form is to create >> a volume from volume snapshot and launch the instance from the new volume. >> >> In your email you listed CLI commands and the command you listed >> showed that you launched the instance from volume, not from volume >> snapshot. So there's no reason to believe the volume snapshot should >> be locked from deletion. >> >> Moreover, it doesn't make sense for an instance to run from a volume >> snapshot - that's not what a snapshot is. The snapshot itself >> should be a static thing. You could probably have used the >> "openstack volume create --snapshot ..." command option to create a >> volume from volume snapshot, and then launch an instance from the new >> volume. The volume snapshot in this case would not be locked from >> deletion, neither the original volume from which the snapshot was based, but only the new volume you created. >> >> I hope this information is accurate and helpful :) I'll forward the >> response to the guys I inquired with, and they can correct me here if >> anything I wrote is amiss. >> >> M >> >> >> >> On 2018-10-17 11:47 a.m., Alonso, Juan Carlos wrote: >>> Hello, >>> >>> Does anybody has faced the same behavior? >>> I am wondering if it is possible to delete a snapshot, or not, when >>> an instance >> is up and running. >>> In my site, I could, but need to know if this is correct or not. >>> >>> Regards. >>> Juan Carlos Alonso >>> >>> From: Alonso, Juan Carlos >>> Sent: Tuesday, October 16, 2018 3:14 PM >>> To: starlingx-discuss at lists.starlingx.io >>> Subject: [Starlingx-discuss] Delete snapshot and volume >>> >>> Hi, >>> >>> The test 211[1] indicates that an Image should be created, then a >>> Volume >> should be created from the image, then a Snapshot should be created >> from the Volume and finally an Instance should be launch from the Snapshot. >>> The test indicates that we try to delete the Snapshot and Volume >>> while the >> Instance is running, Snapshot nor Volume cannot be deleted. >>> But when tried found that the Snapshot can be deleted, but the >>> Volume >> cannot. >>> Is this behavior correct? >>> >>> Those are the commands I used: >>> >>> openstack image create --disk-format qcow2 --file >>> /home/wrsroot/cirros- >> 0.4.0-x86_64-disk.img --public cirros >>> openstack volume create --image cirros --availability-zone nova >>> --size 1 >> cirros-vol >>> openstack volume snapshot create --name cirros-snap cirros-vol >>> openstack server create --volume cirros-vol --flavor m1.small >>> --availibility- >> zone nova vm-cirros >>> >>> 1 >> https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instru >> ctions >>> Regards. >>> Juan Carlos Alonso >>> >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss >> >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cesar.lara at intel.com Wed Oct 17 23:53:13 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 17 Oct 2018 23:53:13 +0000 Subject: [Starlingx-discuss] [build][meetings] Agenda for 10/18/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB4710534D52@fmsmsx104.amr.corp.intel.com> Build team meeting Agenda - Cengn status for external mirror - Use cases review/discussion - Opens I believe this will be a fairly short meeting, hope to see you there. Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From shuicheng.lin at intel.com Thu Oct 18 00:31:05 2018 From: shuicheng.lin at intel.com (Lin, Shuicheng) Date: Thu, 18 Oct 2018 00:31:05 +0000 Subject: [Starlingx-discuss] sanlock switch to rpm instead of src rpm Message-ID: <9700A18779F35F49AF027300A49E7C765FE12DC8@SHSMSX101.ccr.corp.intel.com> Hi Scott, I created a story to replace sanlock src rpm with rpm. https://storyboard.openstack.org/#!/story/2004109 The reason is that I find sanlock is used just in build stage, but not included in ISO image. I find there are some patches in sanlock created by you. Could you share me the background for these patches and is it ok to switch to rpm package? Thanks. Best Regards Shuicheng -------------- next part -------------- An HTML attachment was scrubbed... URL: From huifeng.le at intel.com Thu Oct 18 01:36:26 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Thu, 18 Oct 2018 01:36:26 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D52060C@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> <76647BD697F40748B1FA4F56DA02AA0B4D52059D@SHSMSX104.ccr.corp.intel.com> <76647BD697F40748B1FA4F56DA02AA0B4D52060C@SHSMSX104.ccr.corp.intel.com> Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D520767@SHSMSX104.ccr.corp.intel.com> Matt, Do you think the current upstream Qos behavior meet STX requirements? or any more improvement required to be checked with community? Thanks much! Best Regards, Le, Huifeng From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 9:54 PM To: Peters, Matt ; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Matt, the command can be executed success and the qos_policy_id will be set to “None” after the operation. thanks! Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Wednesday, October 17, 2018 9:43 PM To: Le, Huifeng >; Legacy, Allain > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Hi Huifeng, What is the behavior if you modify the policy to None (using openstack network set --no-qos-policy )? -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 4:52 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Matt, Yes, your thought is correct - the check is not using individual property rule. When creating network with a qos_policy_id, neutron will try to get the qos policy object within current context (e.g. by QosCoreResourceExtension. _get_policy_obj) then raises QosPolicyNotFound exception if qos_policy_id is not owned by the project. During my test, I can see below error raised by neutron: {"NeutronError": { "message": "QoS policy bda58768-7785-4ed3-b9b1-768d7979cc70 could not be found.", "type": "QosPolicyNotFound", "detail": "" }} Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 16, 2018 10:13 PM To: Le, Huifeng >; Legacy, Allain > Cc: starlingx-discuss at lists.starlingx.io Subject: Re: Question about patch upstreaming for 1e9a089 I was specifically referring to the network attribute that refers to the QoS policy ID, rather than the RBAC rules related to the policies themselves. I don’t see anything in the policy.json that restricts the setting of the network attribute, so I am assuming it is falling back to the network entity policy itself rather than an individual property rule. From: "Le, Huifeng" > Date: Tuesday, October 16, 2018 at 10:08 AM To: Allain Legacy >, "Peters, Matt" > Cc: "starlingx-discuss at lists.starlingx.io" > Subject: RE: Question about patch upstreaming for 1e9a089 Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): “QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects.” From the default policy.json, e.g. “create/update/delete” operation is for admin only and “get” operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt >; Le, Huifeng > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have “enforce_policy” set in its attribute definition so I am going to guess that there is no checking at all for it. I don’t have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don’t even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We’re looking into the patch 1e9a089, which target to solve the issue of “CGTS-2408: bypass policy check when setting attribute to None.” After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: “type: uuid_or_none” (2) the “enforce_policy” should be set to “True” and the policy needs to be defined in etc/policy.json (3) the check policy uses “tenant_id” for owner check The only found attribute which will be impacted by this issue in STX is “wrs-tm:qos” which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What’s your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1808 bytes Desc: image001.png URL: From ada.cabrales at intel.com Thu Oct 18 01:46:53 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Thu, 18 Oct 2018 01:46:53 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-17-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C354458@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-17-26-r-2018.10.iso- 31 TCs PASS, 0 TCs FAIL - Simplex and Duplex virtual environment still waiting for the the openvswitch problem [0] to be fixed. A fix has been proposed [1] - Found a problem changing the MTU size on the data network, on bare metal configs [2] - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 2/61/14/77 Simplex virtual - 0/10/61/71 Duplex bare metal - 1/77/22/100 Duplex virtual - 3/73/29/105 Multinode local storage bare metal - 3/84/31/118 Multinode local storage virtual - 6/76/43/125 Multinode external storage bare metal - 2/83/32/117 Multinode external storage virtual - 2/63/55/120 Full report in the wiki [3] [0] https://bugs.launchpad.net/starlingx/+bug/1796380 [1] https://review.openstack.org/#/c/611391/ [2] https://bugs.launchpad.net/starlingx/+bug/1798442 [3] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary A. From Brent.Rowsell at windriver.com Thu Oct 18 02:21:18 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Thu, 18 Oct 2018 02:21:18 +0000 Subject: [Starlingx-discuss] sanlock switch to rpm instead of src rpm In-Reply-To: <9700A18779F35F49AF027300A49E7C765FE12DC8@SHSMSX101.ccr.corp.intel.com> References: <9700A18779F35F49AF027300A49E7C765FE12DC8@SHSMSX101.ccr.corp.intel.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2BB3E2@ALA-MBD.corp.ad.wrs.com> This looks like some hold over code from an older OS. You can remove. Brent From: Lin, Shuicheng [mailto:shuicheng.lin at intel.com] Sent: Wednesday, October 17, 2018 8:31 PM To: Little, Scott ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] sanlock switch to rpm instead of src rpm Hi Scott, I created a story to replace sanlock src rpm with rpm. https://storyboard.openstack.org/#!/story/2004109 The reason is that I find sanlock is used just in build stage, but not included in ISO image. I find there are some patches in sanlock created by you. Could you share me the background for these patches and is it ok to switch to rpm package? Thanks. Best Regards Shuicheng -------------- next part -------------- An HTML attachment was scrubbed... URL: From shuicheng.lin at intel.com Thu Oct 18 07:05:00 2018 From: shuicheng.lin at intel.com (Lin, Shuicheng) Date: Thu, 18 Oct 2018 07:05:00 +0000 Subject: [Starlingx-discuss] question about system upgrade Message-ID: <9700A18779F35F49AF027300A49E7C765FE12EE8@SHSMSX101.ccr.corp.intel.com> Hi all, I try to execute "system load-import" in controller node, and it shows "Metadata file not found". And from code, it is due to there is no "/upgrades/metadata.xml" in the ISO. Anybody know how to generate ISO with this file? Thanks. Best Regards Shuicheng -------------- next part -------------- An HTML attachment was scrubbed... URL: From kailun.qin at intel.com Thu Oct 18 09:29:10 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Thu, 18 Oct 2018 09:29:10 +0000 Subject: [Starlingx-discuss] Questions about StarlingX OVS-DPDK firewall driver enhancement In-Reply-To: References: Message-ID: Hi Matt, To follow up with the performance concern of (kernel-space) conntrack used in option #3. I’ve checked and evaluated the user-space conntrack functionality in OVS and in Neutron native OVS firewall. Please kindly see the attached report for details. We can have another sync in the stx-networking weekly meeting later. Thanks! BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, September 17, 2018 6:41 PM To: Qin, Kailun ; starlingx-discuss at lists.starlingx.io Subject: Re: Questions about StarlingX OVS-DPDK firewall driver enhancement Hello Kailun, I’m fine with exploring option #3 as an alternative to #2. I wasn’t aware that the native driver fully supported the netdev datapath, but if it can be supported, then I agree it would be a better choice. Regards, Matt From: "Qin, Kailun" > Date: Thursday, September 6, 2018 at 10:09 AM To: "Peters, Matt" >, "starlingx-discuss at lists.starlingx.io" > Subject: Questions about StarlingX OVS-DPDK firewall driver enhancement Hi Matt, We are doing some ground work for https://storyboard.openstack.org/#!/story/2002944 regarding the StarlingX OVS-DPDK firewall driver enhancement. To better understand the requirement, we may need your help on the following question. There are 3 security group implementations in neutron OVS-DPDK agent. 1. Iptables based security group The OVS agent and Compute service use a Linux bridge between each instance (VM) and the OVS integration bridge br-int to implement security groups. This implementation is stateful but cause scalability and performance problems. 2. Openflow based security group This implementation is stateless and based in OpenFlow 'learn action', which uses a firewall driver in networking-ovs-dpdk project. https://github.com/openstack/networking-ovs-dpdk/blob/master/networking_ovs_dpdk/agent/ovs_dpdk_firewall.py 3. Openflow + conntrack based security group This implementation is stateful, which uses a native firewall driver. https://docs.openstack.org/newton/networking-guide/config-ovsfwdriver.html http://docs.openvswitch.org/en/latest/tutorials/ovs-conntrack/ The requirement from the enhancement description is that we're going to adopt implementation option 2 for StarlingX OVS-DPDK security group. Compared with option 3 which is stateful and uses a native firewall driver, what is our consideration on taking option 2 as our target? Please kindly see the attached doc for further details. Just let me know if anything unclear. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Userspace Conntrack Investigation Report.docx Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 10772 bytes Desc: Userspace Conntrack Investigation Report.docx URL: From zhipengs.liu at intel.com Thu Oct 18 13:32:14 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Thu, 18 Oct 2018 13:32:14 +0000 Subject: [Starlingx-discuss] Work on reducing config patch for ntp In-Reply-To: <93814834B4855241994F290E959305C752FCA9C6@SHSMSX104.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FCA9C6@SHSMSX104.ccr.corp.intel.com> Message-ID: <93814834B4855241994F290E959305C752FCB35C@SHSMSX104.ccr.corp.intel.com> Hi Penny, Any comment on ntp? For nfs-utils, I have updated patch below. Could you pls help to review it? https://review.openstack.org/610459 I just used a new separate package (nfs-utils-config) instead of platform-utils to package Config files. Since the config file in nfs-utils is marked as %config(noreplace) , patching in future will Not overwrite it. So there is no issue for patching scenario. No need to use kickstart script and patching script for this case. Thanks for your proposal! Zhipeng From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: 2018年10月17日 17:11 To: Penney, Don Cc: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Work on reducing config patch for ntp Hi Penny, I’d like to discuss with you about this 2 patches. ntpd-started-no-g-option.patch //change /etc/sysconfig/ntpd ntp-spec-add-TiS-custom-config-files.patch // add ntp.conf file to /etc/ Can we remove this 2 patches? In stx/stx-config/puppet-manifests/src/modules/platform/manifests/ntp.pp, it already generates ntp.conf file by puppet file { 'ntp_config': ensure => file, path => '/etc/ntp.conf', mode => '0640', content => template('platform/ntp.conf.server.erb'), } For /etc/sysconfig/ntpd, I’d like to add this custom file in ntp.pp, then we can remove related patch. Any comment from you? Thanks! BTW, I saw some other packages also have pp file in stx/stx-config/puppet-manifests/src/modules/platform/manifests/ So I can do some modification on existed pp to change/add some config file, right? For SW patching scenario, it is OK for non-in-service update. After reboot, puppet will apply config again. But for in-service update, without reboot, how to apply this custom file again after patching. That’s why I thought of using patching-script. Do you have any other proposal? Thanks! zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matt.Peters at windriver.com Thu Oct 18 14:08:22 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Thu, 18 Oct 2018 14:08:22 +0000 Subject: [Starlingx-discuss] Questions about StarlingX OVS-DPDK firewall driver enhancement In-Reply-To: References: Message-ID: Hi Kailun, The code change required to use the ‘openvswitch’ driver would be in stx-config. Currently the default driver is ‘noop’, so once the performance testing is complete and verified, this can be updated to be the default. https://git.openstack.org/cgit/openstack/stx-config/tree/puppet-manifests/src/hieradata/compute.yaml#n21 Regards, Matt From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Thursday, October 18, 2018 5:29 AM To: Peters, Matt; starlingx-discuss at lists.starlingx.io Subject: RE: Questions about StarlingX OVS-DPDK firewall driver enhancement Hi Matt, To follow up with the performance concern of (kernel-space) conntrack used in option #3. I’ve checked and evaluated the user-space conntrack functionality in OVS and in Neutron native OVS firewall. Please kindly see the attached report for details. We can have another sync in the stx-networking weekly meeting later. Thanks! BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, September 17, 2018 6:41 PM To: Qin, Kailun ; starlingx-discuss at lists.starlingx.io Subject: Re: Questions about StarlingX OVS-DPDK firewall driver enhancement Hello Kailun, I’m fine with exploring option #3 as an alternative to #2. I wasn’t aware that the native driver fully supported the netdev datapath, but if it can be supported, then I agree it would be a better choice. Regards, Matt From: "Qin, Kailun" > Date: Thursday, September 6, 2018 at 10:09 AM To: "Peters, Matt" >, "starlingx-discuss at lists.starlingx.io" > Subject: Questions about StarlingX OVS-DPDK firewall driver enhancement Hi Matt, We are doing some ground work for https://storyboard.openstack.org/#!/story/2002944 regarding the StarlingX OVS-DPDK firewall driver enhancement. To better understand the requirement, we may need your help on the following question. There are 3 security group implementations in neutron OVS-DPDK agent. 1. Iptables based security group The OVS agent and Compute service use a Linux bridge between each instance (VM) and the OVS integration bridge br-int to implement security groups. This implementation is stateful but cause scalability and performance problems. 2. Openflow based security group This implementation is stateless and based in OpenFlow 'learn action', which uses a firewall driver in networking-ovs-dpdk project. https://github.com/openstack/networking-ovs-dpdk/blob/master/networking_ovs_dpdk/agent/ovs_dpdk_firewall.py 3. Openflow + conntrack based security group This implementation is stateful, which uses a native firewall driver. https://docs.openstack.org/newton/networking-guide/config-ovsfwdriver.html http://docs.openvswitch.org/en/latest/tutorials/ovs-conntrack/ The requirement from the enhancement description is that we're going to adopt implementation option 2 for StarlingX OVS-DPDK security group. Compared with option 3 which is stateful and uses a native firewall driver, what is our consideration on taking option 2 as our target? Please kindly see the attached doc for further details. Just let me know if anything unclear. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matt.Peters at windriver.com Thu Oct 18 14:15:24 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Thu, 18 Oct 2018 14:15:24 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D520767@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D51FF95@SHSMSX104.ccr.corp.intel.com> <70A7408C6E1BFB41B192A929744D8523BAC0F610@ALA-MBD.corp.ad.wrs.com> <76647BD697F40748B1FA4F56DA02AA0B4D520356@SHSMSX104.ccr.corp.intel.com> <6F368FDA-A3AD-472C-93B9-279D494D88CA@windriver.com> <76647BD697F40748B1FA4F56DA02AA0B4D52059D@SHSMSX104.ccr.corp.intel.com> <76647BD697F40748B1FA4F56DA02AA0B4D52060C@SHSMSX104.ccr.corp.intel.com> <76647BD697F40748B1FA4F56DA02AA0B4D520767@SHSMSX104.ccr.corp.intel.com> Message-ID: Hi Huifeng, It looks like the required enforcement is occurring without the policy. With the deprecation of the QoS extension (https://storyboard.openstack.org/#!/story/2003948) , we can also align to the current upstream behavior to do a similar enforcement without this policy entry and also revert this change. Thanks, Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 9:36 PM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Matt, Do you think the current upstream Qos behavior meet STX requirements? or any more improvement required to be checked with community? Thanks much! Best Regards, Le, Huifeng From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 9:54 PM To: Peters, Matt ; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Matt, the command can be executed success and the qos_policy_id will be set to “None” after the operation. thanks! Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Wednesday, October 17, 2018 9:43 PM To: Le, Huifeng >; Legacy, Allain > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Hi Huifeng, What is the behavior if you modify the policy to None (using openstack network set --no-qos-policy )? -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Wednesday, October 17, 2018 4:52 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 Matt, Yes, your thought is correct - the check is not using individual property rule. When creating network with a qos_policy_id, neutron will try to get the qos policy object within current context (e.g. by QosCoreResourceExtension. _get_policy_obj) then raises QosPolicyNotFound exception if qos_policy_id is not owned by the project. During my test, I can see below error raised by neutron: {"NeutronError": { "message": "QoS policy bda58768-7785-4ed3-b9b1-768d7979cc70 could not be found.", "type": "QosPolicyNotFound", "detail": "" }} Best Regards, Le, Huifeng From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Tuesday, October 16, 2018 10:13 PM To: Le, Huifeng >; Legacy, Allain > Cc: starlingx-discuss at lists.starlingx.io Subject: Re: Question about patch upstreaming for 1e9a089 I was specifically referring to the network attribute that refers to the QoS policy ID, rather than the RBAC rules related to the policies themselves. I don’t see anything in the policy.json that restricts the setting of the network attribute, so I am assuming it is falling back to the network entity policy itself rather than an individual property rule. From: "Le, Huifeng" > Date: Tuesday, October 16, 2018 at 10:08 AM To: Allain Legacy >, "Peters, Matt" > Cc: "starlingx-discuss at lists.starlingx.io" > Subject: RE: Question about patch upstreaming for 1e9a089 Matt & Allain, I think it is the current upstream design (https://docs.openstack.org/neutron/latest/admin/config-qos.html): “QoS policies are only created by admins with the default policy.json. Therefore, you should have the cloud operator set them up on behalf of the cloud projects.” From the default policy.json, e.g. “create/update/delete” operation is for admin only and “get” operation is for any user. Do you see any gap with this design? "create_policy": "rule:admin_only", "get_policy": "rule:regular_user", "update_policy": "rule:admin_only", "delete_policy": "rule:admin_only", Best Regards, Le, Huifeng From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Tuesday, October 16, 2018 8:23 PM To: Peters, Matt >; Le, Huifeng > Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for 1e9a089 I am not sure. The qos_policy_id attribute does not have “enforce_policy” set in its attribute definition so I am going to guess that there is no checking at all for it. I don’t have a system setup to test this at the moment, but I am going to guess that you could end up attaching a network to a qos_policy_id that is owned by a different tenant. If so, then that is an issue that should be addressed. Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 12:55 PM To: Le, Huifeng Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for 1e9a089 Hi Allain, Do you know why the standard qos_policy_id attribute of the network would not hit this problem? I don’t even see them reference this field in the policy.json, so does it just permit anything to be changed by the tenant unless it is explicitly defined? http://crucible.wrs.com/changelog/CGCS_github_neutron?cs=1e9a089 -Matt From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Sunday, October 14, 2018 10:24 PM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io; Qin, Kailun Subject: Question about patch upstreaming for 1e9a089 Hi Matt, We’re looking into the patch 1e9a089, which target to solve the issue of “CGTS-2408: bypass policy check when setting attribute to None.” After investigating this issue, it is found that this issue will only happen for attribute which need to meet below conditions: (1) the validate rule is: “type: uuid_or_none” (2) the “enforce_policy” should be set to “True” and the policy needs to be defined in etc/policy.json (3) the check policy uses “tenant_id” for owner check The only found attribute which will be impacted by this issue in STX is “wrs-tm:qos” which will be removed later after STX applying upstreaming QOS API. Since no other impacted attributes can be found in upstream (or could you please share with us if you see any other attributes which may be impacted by this issue?), What’s your proposal for upstreaming or Do you think holding on this fix is make sense for now? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1808 bytes Desc: image001.png URL: From kailun.qin at intel.com Thu Oct 18 14:31:24 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Thu, 18 Oct 2018 14:31:24 +0000 Subject: [Starlingx-discuss] Questions about StarlingX OVS-DPDK firewall driver enhancement In-Reply-To: References: Message-ID: Matt, Thanks a lot for the information. I’ll keep you updated for the performance testing. BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Thursday, October 18, 2018 10:08 PM To: Qin, Kailun ; starlingx-discuss at lists.starlingx.io Subject: RE: Questions about StarlingX OVS-DPDK firewall driver enhancement Hi Kailun, The code change required to use the ‘openvswitch’ driver would be in stx-config. Currently the default driver is ‘noop’, so once the performance testing is complete and verified, this can be updated to be the default. https://git.openstack.org/cgit/openstack/stx-config/tree/puppet-manifests/src/hieradata/compute.yaml#n21 Regards, Matt From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Thursday, October 18, 2018 5:29 AM To: Peters, Matt; starlingx-discuss at lists.starlingx.io Subject: RE: Questions about StarlingX OVS-DPDK firewall driver enhancement Hi Matt, To follow up with the performance concern of (kernel-space) conntrack used in option #3. I’ve checked and evaluated the user-space conntrack functionality in OVS and in Neutron native OVS firewall. Please kindly see the attached report for details. We can have another sync in the stx-networking weekly meeting later. Thanks! BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, September 17, 2018 6:41 PM To: Qin, Kailun >; starlingx-discuss at lists.starlingx.io Subject: Re: Questions about StarlingX OVS-DPDK firewall driver enhancement Hello Kailun, I’m fine with exploring option #3 as an alternative to #2. I wasn’t aware that the native driver fully supported the netdev datapath, but if it can be supported, then I agree it would be a better choice. Regards, Matt From: "Qin, Kailun" > Date: Thursday, September 6, 2018 at 10:09 AM To: "Peters, Matt" >, "starlingx-discuss at lists.starlingx.io" > Subject: Questions about StarlingX OVS-DPDK firewall driver enhancement Hi Matt, We are doing some ground work for https://storyboard.openstack.org/#!/story/2002944 regarding the StarlingX OVS-DPDK firewall driver enhancement. To better understand the requirement, we may need your help on the following question. There are 3 security group implementations in neutron OVS-DPDK agent. 1. Iptables based security group The OVS agent and Compute service use a Linux bridge between each instance (VM) and the OVS integration bridge br-int to implement security groups. This implementation is stateful but cause scalability and performance problems. 2. Openflow based security group This implementation is stateless and based in OpenFlow 'learn action', which uses a firewall driver in networking-ovs-dpdk project. https://github.com/openstack/networking-ovs-dpdk/blob/master/networking_ovs_dpdk/agent/ovs_dpdk_firewall.py 3. Openflow + conntrack based security group This implementation is stateful, which uses a native firewall driver. https://docs.openstack.org/newton/networking-guide/config-ovsfwdriver.html http://docs.openvswitch.org/en/latest/tutorials/ovs-conntrack/ The requirement from the enhancement description is that we're going to adopt implementation option 2 for StarlingX OVS-DPDK security group. Compared with option 3 which is stateful and uses a native firewall driver, what is our consideration on taking option 2 as our target? Please kindly see the attached doc for further details. Just let me know if anything unclear. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Thu Oct 18 15:57:00 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Thu, 18 Oct 2018 15:57:00 +0000 Subject: [Starlingx-discuss] Oct 18 2018 StarlingX TSC meeting notes Message-ID: <9A85D2917C58154C960D95352B22818BB1E7F56B@ORSMSX158.amr.corp.intel.com> 10/18/18 * Bill Zvonar has joined the WR team as a Program Manager for StarlingX * Project priority review and next steps - Bruce o Bruce to send out his updated list as an Etherpad, asking folks to tag changes made, team to use as initial planning doc for 2019.03. * Governance feedback & changes - Bruce o Active reviews seem to address all current feedback. o Rendered document is visible at https://docs.starlingx.io/governance/reference/tsc/stx_charter.html o Link to reviews for stx-governance: https://review.openstack.org/#/q/project:openstack/stx-governance,n,z * Feature content for the 2019.03 release - Brent * Follow-up on spec process - Brent o Good discussion last week. As we build up spec volume we need to make sure the process is working. ? AR Bruce - align feature dev process with house rules. ? Spec approval should have both TSC approval and key expert (Tech Lead(s)), with no negative feedback. ? Consider multiple levels of specs - higher level specs for direction - approved by TSC - lower level specs for implementation - approved by PL(s). ? We are very early in this process. Let's run the process and adjust as needed. * Containerization specs and steps to obtain final approval - Frank o There are 6 specs for this project in progress, approved by Brent +2 and by the project team. Ready for TSC review. ? List of all pending spec reviews: https://review.openstack.org/#/q/status:open+AND+project:%255Eopenstack/stx-specs ? AR Team to review these specs -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ovidiu.Poncea at windriver.com Thu Oct 18 16:29:32 2018 From: Ovidiu.Poncea at windriver.com (Poncea, Ovidiu) Date: Thu, 18 Oct 2018 16:29:32 +0000 Subject: [Starlingx-discuss] Request to merge a small change in stx-ceph Message-ID: <4C60D9C5C8176C47874FFF36647AA19E9D5DC2F0@ALA-MBD.corp.ad.wrs.com> Hi Dean, Please take a look over https://github.com/starlingx-staging/stx-ceph/pull/4 We need this as part of ongoing activity in: https://review.openstack.org/#/c/610047/3/specs/approved/containerization-2002844-CEPH-persistent-storage-backend-for-Kubernetes.rst The change modifies StarlingX specific files in the "wrs" folder of stx-ceph. We plan on moving these to another location but for now we need this change in ceph in order for ceph on a 1 node configuration to work. Thank you, Ovidiu PS: Change was approved by arch and reviewed by Eric MacDonalds, the mtc responsible as it touches one of its scripts. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Thu Oct 18 16:41:02 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Thu, 18 Oct 2018 16:41:02 +0000 Subject: [Starlingx-discuss] Oct 18 2018 StarlingX TSC meeting notes In-Reply-To: <9A85D2917C58154C960D95352B22818BB1E7F56B@ORSMSX158.amr.corp.intel.com> References: <9A85D2917C58154C960D95352B22818BB1E7F56B@ORSMSX158.amr.corp.intel.com> Message-ID: <9A85D2917C58154C960D95352B22818BB1E7F62E@ORSMSX158.amr.corp.intel.com> The Project Priority working document (Etherpad) is now available online [0]. It is also linked to from the wiki Project Priority page [1]. The StarlingX Feature Development Process [2] has been updated to align with the TSC House Rules. Brucej [0] https://etherpad.openstack.org/p/stx-2019.03-priorities [1] https://wiki.openstack.org/wiki/StarlingX/Project_Priorities [2] https://wiki.openstack.org/wiki/StarlingX/Feature_Development_Process#Specification_Process From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Thursday, October 18, 2018 8:57 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Oct 18 2018 StarlingX TSC meeting notes 10/18/18 * Bill Zvonar has joined the WR team as a Program Manager for StarlingX * Project priority review and next steps - Bruce o Bruce to send out his updated list as an Etherpad, asking folks to tag changes made, team to use as initial planning doc for 2019.03. * Governance feedback & changes - Bruce o Active reviews seem to address all current feedback. o Rendered document is visible at https://docs.starlingx.io/governance/reference/tsc/stx_charter.html o Link to reviews for stx-governance: https://review.openstack.org/#/q/project:openstack/stx-governance,n,z * Feature content for the 2019.03 release - Brent * Follow-up on spec process - Brent o Good discussion last week. As we build up spec volume we need to make sure the process is working. ? AR Bruce - align feature dev process with house rules. ? Spec approval should have both TSC approval and key expert (Tech Lead(s)), with no negative feedback. ? Consider multiple levels of specs - higher level specs for direction - approved by TSC - lower level specs for implementation - approved by PL(s). ? We are very early in this process. Let's run the process and adjust as needed. * Containerization specs and steps to obtain final approval - Frank o There are 6 specs for this project in progress, approved by Brent +2 and by the project team. Ready for TSC review. ? List of all pending spec reviews: https://review.openstack.org/#/q/status:open+AND+project:%255Eopenstack/stx-specs ? AR Team to review these specs -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ghada.Khalil at windriver.com Thu Oct 18 17:03:12 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Thu, 18 Oct 2018 17:03:12 +0000 Subject: [Starlingx-discuss] Minutes: StarlingX Networking Weekly Meeting -- 10/18 Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46D3B0@ALA-MBD.corp.ad.wrs.com> Meeting agenda/minutes are captured at: https://etherpad.openstack.org/p/stx-networking Team Meeting Agenda/Notes - Oct 17/2018 * Upstreaming Status Updates * Vlan trunk (ba9d9f60): admin_state_up behavior in attached email (Huifeng) ? Recommendation is to abandon the bug and align with upstream behavior. * 1e9a089 in attached email (Huifeng) ? Recommendation is to abandon the bug since it is not required for network qos_policy_id enforcement * a38f899 in attached email (Kailun) and 4 patch (2 RFE and 2 bugs) upstreaming status ? Core reviewers required to move reviews forward * OVS-DPDK firewall. (Kailun) ? Functionality available in OVS to handle CT states for 'openvswitch' firewall driver ? Performance testing required ? Matt to send code reference to stx-config firewall driver configuration * 4ae5a58: latest status @https://bugs.launchpad.net/neutron/+bug/1793653 (Chenjie); 9f926a5: RFE draft under internal review and depends on 4ae5a58 (Chenjie). ? Matt to review bug report comments and RFE * e547a7e, ebd41fd: patch reworked according to comments from core reviewer (Chenjie) * Update the master spreadsheet with the missing neutron repo's * Derek is working with Dean to get the list of commits for the repo's in order to add them to the spreadsheet * 4 repo's are currently missing: ? https://github.com/starlingx-staging/stx-networking-bgpvpn ? https://github.com/starlingx-staging/stx-networking-sfc ? https://github.com/starlingx-staging/stx-networking-odl ? https://github.com/starlingx-staging/stx-ryu * Adding Co-Authored-By on reviews -- email sent to Forrest * Reviewed the process with team and agreed to start adding WR co-authors for the neutron * Recommended Process ? The Intel prime looks up the original author in the WR patch * Ghada can help identify authors when needed ? The Intel prime adds the original author as a co-author in the gerrit review * Co-Authored-By: Name ? The Intel prime also adds the original author and Matt Peters as a review of the gerrit review * Containerized Openstack * Specs are in progress: https://review.openstack.org/#/q/projects:openstack/stx-specs * Storyboard Container Stories: https://storyboard.openstack.org/#!/search?tags=stx.containers * Joseph Richard (WR) is currently working on neutron helm chart overrides ? Chart Override Spec: https://review.openstack.org/#/c/609849/ ? Once the neutron code is ready, we will add Ruijing to the gerrit review * There is currently no plan for ovs-dpdk to be running in a container ? Rationale for deferring: * Kubernetes / Docker does not currently handle several host level properties / resources that are required for DPDK based applications: o Hugepages - actively being worked in the K8S community o CPU Isolation and Pinning - isolcpus is a kernel level option that isn't exposed to individual containers o Device Plugins - NIC resource assignments to Pods - currently depends on SRIOV PF CNI o NUMA topology - NIC and CPU awareness for Pod scheduling o Kernel Modules - Generally kernel module loading, but also specifically VFIO and IOMMU support * No other immediate networking work items are needed. Ghada to keep the team in the loop on what is happening with respect to networking. Ghada Khalil, Manager, Titanium Cloud, Wind River direct 613.270.2273 skype ghada.khalil.ottawa 350 Terry Fox Drive, Suite 200, Kanata, ON K2K 2W5 -------------- next part -------------- An HTML attachment was scrubbed... URL: From Matt.Peters at windriver.com Thu Oct 18 19:16:56 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Thu, 18 Oct 2018 19:16:56 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for a38f899 In-Reply-To: References: Message-ID: Hi Kailun, Looking at the remaining items that you had in the attachment, I agree with your assessment of the changes. I don't have any further information on how to reproduce the first item, other than to run the tempest tests in a multi-core environment repeatedly to try and trigger the race condition. However, since the tests may have changed from the time of the original commit, it may be that this scenario can no longer be triggered. -Matt From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 1:05 PM To: Qin, Kailun; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for a38f899 Based on the original commit message (see below), it would fall into scenario #2. However, as the commit message indicates, this is based on race conditions when running tempest, therefore it may not be easily reproducible. CGTS-3885: dhcp: handle concurrent port creation errors Running tempest nose tests exposes a race condition between the server deleting a subnet and an agent creating a DHCP port. This problem manifests itself in different ways depending on the whether the error is returned to the subnet delete operation or the DHCP port creation. This Jira (3885) addresses how the error is handled by the agent when the subnet is concurrently deleted at the server. The changes to address the error that is returned to the API client are in CGTS-3432. Because of how the server and agents are designed there is a lack of determinism in how this particular issue manifests itself. The variants manifest themselves based on when the subnet is deleted in relation to when the port is created. 1) if the port is returned with no fixed IP address then this results in the agent failing to setup the default route in the namespace because there is no local IP address to use to contact the gateway IP address. 2) if the port is updated after creation to no longer have an IP address then 'reload_allocations' action is invoked instead of the 'disabled' action. This leads to a failure to reenable because the subnet no longer exits. 3) if doing a 'disable' action it is possible that previous 'restart', 'enable', or 'reload' actions did not get far enough to create the namespace. For this reason we should not throw an error if the namespace does not exist. 4) if attempting a 'reload_allocations' we should fail the request if the namespace does not exist. 5) if attempting a 'reload_allocations' with retain_port=True we should not delete the configuration files because the vswitch port name is stored in the 'interface' file. If an error occurs during the enable action we will not be able to 'unplug' the port from the vswitch because we will not know which name to use during the 'unplug'. This will leak ports in the vswitch. From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Monday, October 15, 2018 5:00 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Question about patch upstreaming for a38f899 Hi Matt, Allain, I am analyzing the patch a38f899 for upstreaming. It seems to me that parts of the patch have already been addressed by upstream, while some of the rest may need your double confirm/clarification. Please kindly check the attached analysis report for further details. Let me know if any question. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From jose.perez.carranza at intel.com Thu Oct 18 20:39:16 2018 From: jose.perez.carranza at intel.com (Perez Carranza, Jose) Date: Thu, 18 Oct 2018 20:39:16 +0000 Subject: [Starlingx-discuss] Question about neutron connectivity-test error Message-ID: <0A5D9A624DF90343892F8F3FE7DE525A2A8E62B9@fmsmsx101.amr.corp.intel.com> Hi When I run command "neutron providernet-connectivity-test-list" I received an error --> "Provider network testing is not supported in this configuration.". Is there any specific configuration that I should apply to my controller or provider network? http://paste.openstack.org/show/732446/ Regards, José > -----Original Message----- > From: Perez Carranza, Jose > Sent: Tuesday, October 16, 2018 2:27 PM > To: starlingx-discuss at lists.starlingx.io > Subject: Help on shared vCPU usage > > Hi > > I'm trying to execute scenario 206 [1] that indicates that I need to launch > instance with a flavor with "hw:wrs:shared_vcpu", I created a flavor with > "hw:wrs:shared_vcpu=1" but at the moment of create the instance below > error is displayed.. Is there any extra config or parameter that I need to apply > before creating the instance ? > > Error (Duplex Bare-Metal): > ====== > No valid host was found. There are not enough hosts available. controller-0: > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required by > instance cell 0, Shared vCPU not enabled on host cell 0, required by instance > cell 0, controller-1: (NUM ===== > > 1- https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > Regards, > José From Don.Penney at windriver.com Thu Oct 18 22:12:16 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Thu, 18 Oct 2018 22:12:16 +0000 Subject: [Starlingx-discuss] Work on reducing config patch for ntp In-Reply-To: <93814834B4855241994F290E959305C752FCB35C@SHSMSX104.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FCA9C6@SHSMSX104.ccr.corp.intel.com> <93814834B4855241994F290E959305C752FCB35C@SHSMSX104.ccr.corp.intel.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3CBE6E@ALA-MBD.corp.ad.wrs.com> I think we should be ok with dropping patches to NTP config files from the packaging, leaving it to puppet instead. We’ll just want to make sure ntp is disabled upon initial install, which could be done in a kickstart, and enabled by puppet when the config file has been updated. We’ve added hooks in the installer/kickstart to call ntpdate during installation, so that the system time should be synced with the controller during initial install. We don’t generally update packaged config files in a patch. If we want to make customizations, we do it via puppet. All config files should be packaged as %config(noreplace), so any changes wouldn’t make it to a modified file anyway. Our goal for all thirdparty packages should be to have the packaged config file untouched, with necessary changes made via puppet. There should be few exceptions, like NFS, where we’d need changes to the config file before puppet runs. As for the puppet manifests, if there’s an existing manifest related to the config file you want to add, then it should be ok to modify an existing file. If not, you may need to add a new .pp file. And use of the ERB templates is preferred, as it generates the whole file in one shot, meaning it’s much cleaner if we need to apply or remove patches (no risk of stale data). Let me know if you have any questions. Or if anyone has a different opinion, feel free to comment. Cheers, Don. From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: Thursday, October 18, 2018 9:32 AM To: Liu, ZhipengS; Penney, Don Cc: starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Work on reducing config patch for ntp Hi Penny, Any comment on ntp? For nfs-utils, I have updated patch below. Could you pls help to review it? https://review.openstack.org/610459 I just used a new separate package (nfs-utils-config) instead of platform-utils to package Config files. Since the config file in nfs-utils is marked as %config(noreplace) , patching in future will Not overwrite it. So there is no issue for patching scenario. No need to use kickstart script and patching script for this case. Thanks for your proposal! Zhipeng From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: 2018年10月17日 17:11 To: Penney, Don Cc: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Work on reducing config patch for ntp Hi Penny, I’d like to discuss with you about this 2 patches. ntpd-started-no-g-option.patch //change /etc/sysconfig/ntpd ntp-spec-add-TiS-custom-config-files.patch // add ntp.conf file to /etc/ Can we remove this 2 patches? In stx/stx-config/puppet-manifests/src/modules/platform/manifests/ntp.pp, it already generates ntp.conf file by puppet file { 'ntp_config': ensure => file, path => '/etc/ntp.conf', mode => '0640', content => template('platform/ntp.conf.server.erb'), } For /etc/sysconfig/ntpd, I’d like to add this custom file in ntp.pp, then we can remove related patch. Any comment from you? Thanks! BTW, I saw some other packages also have pp file in stx/stx-config/puppet-manifests/src/modules/platform/manifests/ So I can do some modification on existed pp to change/add some config file, right? For SW patching scenario, it is OK for non-in-service update. After reboot, puppet will apply config again. But for in-service update, without reboot, how to apply this custom file again after patching. That’s why I thought of using patching-script. Do you have any other proposal? Thanks! zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Thu Oct 18 23:11:47 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Thu, 18 Oct 2018 23:11:47 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-18-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C354CE1@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-18-27-r-2018.10.iso 29 TCs PASS, 2 TCs FAIL. Controller-0 shows `config-out-of-date` - debug in progress. - Fix for the openvswitch problem on virtual environment has been released. Working on testing it. - "Killing glance.api.pid process is not creating any kind of fm alarm-list" created [0] - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 2/68/7/77 Simplex virtual - 0/10/61/71 Duplex bare metal - 1/85/13/99 Duplex virtual - 3/74/27/104 Multinode local storage bare metal - 4/92/23/119 Multinode local storage virtual - 6/76/43/125 Multinode external storage bare metal - 3/88/26/117 Multinode external storage virtual - 3/81/36/120 Full report in the wiki [1] [0] https://bugs.launchpad.net/starlingx/+bug/1798631 [1] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary A. From Ghada.Khalil at windriver.com Thu Oct 18 23:13:59 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Thu, 18 Oct 2018 23:13:59 +0000 Subject: [Starlingx-discuss] Fix for ovs memory allocation failure is merged in master Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46D6CF@ALA-MBD.corp.ad.wrs.com> Hello all, I know that several people have been waiting for a fix for Launchpad: https://bugs.launchpad.net/starlingx/+bug/1796380 This is to let you know that the fix has been merged in master this afternoon. We will merge the code in the r/2018.10 release branch tomorrow. Regards, Ghada -------------- next part -------------- An HTML attachment was scrubbed... URL: From kailun.qin at intel.com Fri Oct 19 00:05:30 2018 From: kailun.qin at intel.com (Qin, Kailun) Date: Fri, 19 Oct 2018 00:05:30 +0000 Subject: [Starlingx-discuss] Question about patch upstreaming for a38f899 In-Reply-To: References: Message-ID: Hi Matt, Thank you for the comments and advice. I'll give it a try on the reproduction and keep you synced with the status. BR, Kailun From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Friday, October 19, 2018 3:17 AM To: Qin, Kailun ; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: RE: Question about patch upstreaming for a38f899 Hi Kailun, Looking at the remaining items that you had in the attachment, I agree with your assessment of the changes. I don't have any further information on how to reproduce the first item, other than to run the tempest tests in a multi-core environment repeatedly to try and trigger the race condition. However, since the tests may have changed from the time of the original commit, it may be that this scenario can no longer be triggered. -Matt From: Peters, Matt [mailto:Matt.Peters at windriver.com] Sent: Monday, October 15, 2018 1:05 PM To: Qin, Kailun; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Question about patch upstreaming for a38f899 Based on the original commit message (see below), it would fall into scenario #2. However, as the commit message indicates, this is based on race conditions when running tempest, therefore it may not be easily reproducible. CGTS-3885: dhcp: handle concurrent port creation errors Running tempest nose tests exposes a race condition between the server deleting a subnet and an agent creating a DHCP port. This problem manifests itself in different ways depending on the whether the error is returned to the subnet delete operation or the DHCP port creation. This Jira (3885) addresses how the error is handled by the agent when the subnet is concurrently deleted at the server. The changes to address the error that is returned to the API client are in CGTS-3432. Because of how the server and agents are designed there is a lack of determinism in how this particular issue manifests itself. The variants manifest themselves based on when the subnet is deleted in relation to when the port is created. 1) if the port is returned with no fixed IP address then this results in the agent failing to setup the default route in the namespace because there is no local IP address to use to contact the gateway IP address. 2) if the port is updated after creation to no longer have an IP address then 'reload_allocations' action is invoked instead of the 'disabled' action. This leads to a failure to reenable because the subnet no longer exits. 3) if doing a 'disable' action it is possible that previous 'restart', 'enable', or 'reload' actions did not get far enough to create the namespace. For this reason we should not throw an error if the namespace does not exist. 4) if attempting a 'reload_allocations' we should fail the request if the namespace does not exist. 5) if attempting a 'reload_allocations' with retain_port=True we should not delete the configuration files because the vswitch port name is stored in the 'interface' file. If an error occurs during the enable action we will not be able to 'unplug' the port from the vswitch because we will not know which name to use during the 'unplug'. This will leak ports in the vswitch. From: Qin, Kailun [mailto:kailun.qin at intel.com] Sent: Monday, October 15, 2018 5:00 AM To: Peters, Matt; Legacy, Allain Cc: starlingx-discuss at lists.starlingx.io Subject: Question about patch upstreaming for a38f899 Hi Matt, Allain, I am analyzing the patch a38f899 for upstreaming. It seems to me that parts of the patch have already been addressed by upstream, while some of the rest may need your double confirm/clarification. Please kindly check the attached analysis report for further details. Let me know if any question. Thanks a lot! BR, Kailun -------------- next part -------------- An HTML attachment was scrubbed... URL: From chenjie.xu at intel.com Fri Oct 19 02:58:19 2018 From: chenjie.xu at intel.com (Xu, Chenjie) Date: Fri, 19 Oct 2018 02:58:19 +0000 Subject: [Starlingx-discuss] Community Feedbacks for RFE(patch 4ae5a58) for StartlingX upstreaming Message-ID: Hi Matt, For patch 4ae5a58, the RFE “Enable other subprojects to extend l2pop fdb information” has been draft and is under reviewing for now. Community provide some feedbacks and could you please help review and comment! 1. The link for the RFE is below: https://bugs.launchpad.net/neutron/+bug/1793653 2. The link for the code is below: https://review.openstack.org/#/c/599319/ 3. The link for the BP mentioned by Slawek is below: https://blueprints.launchpad.net/neutron/+spec/native-l2pop 4. The link for the RFE related to the BP is below: https://bugs.launchpad.net/neutron/+bug/1630981 For patch 9f926a5, the RFE “Add l2pop support for floating ip resources” has been draft and is under internal reviewing for now. This RFE depends on the RFE “Enable other subprojects to extend l2pop fdb information” Best Regards, Xu, Chenjie -------------- next part -------------- An HTML attachment was scrubbed... URL: From liang.a.fang at intel.com Fri Oct 19 03:16:44 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Fri, 19 Oct 2018 03:16:44 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? In-Reply-To: References: <2588653EBDFFA34B982FAF00F1B4844EBB2A94B0@ALA-MBD.corp.ad.wrs.com> Message-ID: Hi Brent and all I brought these two patches to yesterday's glance meeting, some update here: - Regarding the commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Community suggest to add option "-vv " to "glance image-list", something like "glance -vv image-list". They don't want to show store info in "-v" because in most situation store info is the same, so there will be same values showed in the column in whole page, something like: | ... | img1 | rbd | ... | ... | img2 | rbd | ... | ... | img3 | rbd | ... | ... | img4 | rbd | ... | ... | img5 | rbd | ... ... ... I think it make sense to show this info in -vv, not -v Do you think is this achieved your original purpose of the patch? If so, I will go ahead to update the patch for "-vv". - Regarding the commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Cores mentioned it has been implemented in rocky version which introduced a new feature "multi-store". I tried to setup the multi-store environment and verified this, the store info indeed saved, and the info can be found in database table: image_locations, something like {"backend": "ceph"}, see below: ===================================================================================== mysql> select * from image_locations where status = 'active'; +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ | id | image_id | value | created_at | updated_at | deleted_at | deleted | meta_data | status | +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ | 1 | 617f2880-f5b9-491f-b569-d2bd088c0bd5 | file:///opt/stack/data/glance/images/617f2880-f5b9-491f-b569-d2bd088c0bd5 | 2018-10-17 03:15:46 | 2018-10-17 03:15:46 | NULL | 0 | {} | active | | 9 | 0de67910-661e-4a2c-b5de-bcdc4d4a62db | rbd://b7b109e5-6c2d-4060-b81f-bd5c6fe57e6e/pool1/0de67910-661e-4a2c-b5de-bcdc4d4a62db/snap | 2018-10-18 06:45:24 | 2018-10-18 06:45:24 | NULL | 0 | {} | active | | 13 | e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | file:///opt/stack/data/glance/images/e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | 2018-10-18 08:40:18 | 2018-10-18 08:40:18 | NULL | 0 | {"backend": "locallvm"} | active | | 14 | 4804a6b7-6392-4232-9a9e-897d1eeef589 | rbd://b7b109e5-6c2d-4060-b81f-bd5c6fe57e6e/pool1/4804a6b7-6392-4232-9a9e-897d1eeef589/snap | 2018-10-18 08:41:57 | 2018-10-18 08:41:57 | NULL | 0 | {"backend": "ceph"} | active | +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ 4 rows in set (0.00 sec) ===================================================================================== Also Cmd "glance image-show" can show the store id like: "stores | ceph", see below: stack at ceph01liang:~$ glance image-show 4804a6b7-6392-4232-9a9e-897d1eeef589 +------------------+----------------------------------------------------------------------------------+ | Property | Value | +------------------+----------------------------------------------------------------------------------+ | checksum | f8ab98ff5e73ebab884d80c9dc9c7290 | | container_format | bare | | created_at | 2018-10-18T08:41:55Z | | disk_format | raw | | id | 4804a6b7-6392-4232-9a9e-897d1eeef589 | | min_disk | 0 | | min_ram | 0 | | name | img2-ceph | | os_hash_algo | sha512 | | os_hash_value | f0fd1b50420dce4ca382ccfbb528eef3a38bbeff00b54e95e3876b9bafe7ed2d6f919ca35d9046d4 | | | 37c6d2d8698b1174a335fbd66035bb3edc525d2cdb187232 | | os_hidden | False | | owner | 60a12b1788ad44468afd983f89a5f8dc | | protected | False | | size | 13267968 | | status | active | | stores | ceph | | tags | [] | | updated_at | 2018-10-18T08:41:57Z | | virtual_size | Not available | | visibility | public | +------------------+----------------------------------------------------------------------------------+ So, the store info already saved in latest code, do you think is this achieved your original purpose of the patch? If so, I guess the patch can be marked as merged. Thanks Liang -----Original Message----- From: Fang, Liang A Sent: Wednesday, October 10, 2018 11:44 PM To: Rowsell, Brent ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? OK, thank you Brent, let me try my best to convince community to accept these two patches. Regards Liang -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Wednesday, October 10, 2018 8:50 PM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Liang, This was a small usability improvement to help the user distinguish the glance backend type in a multi backend environment. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:49 AM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Related review URL: https://review.openstack.org/#/c/605014/ https://review.openstack.org/#/c/605006/ Compare with upstream glance, starlingx added "Store" column as below +-----+--------+-------+-----------------+-----+------------+--------+---------+ | ID | Name | Store | Disk_format | ... | Size | Status |Owner| +-----+--------+-------+-----------------+-----+------------+--------+---------+ | xxx | img1 | file | raw | ... | 10737418 | active | xxx | | xxx | img2 | file | raw | ... | 5086345 | active | xxx | +-----+--------+-------+-----------------+-----+------------+--------+---------+ Regards Liang -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:35 PM To: jack.ding at windriver.com; robert.church at windriver.com; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the use case of these 2 patches? Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From longqian.zhao at intel.com Fri Oct 19 10:26:21 2018 From: longqian.zhao at intel.com (Zhao, Longqian) Date: Fri, 19 Oct 2018 10:26:21 +0000 Subject: [Starlingx-discuss] Why am I don't download hundreds of package from the Internet? Message-ID: <6325970F810676468E0EC5D8ED76D2190B7B97AC@SHSMSX101.ccr.corp.intel.com> Hi, my problem is following phenomenon: 1.Firstly, I follow the steps as required: a. sudo apt-get update b. Installation Requirements and Dependencies c. Install Public SSH Key d. Install stx-tools project e. Create a Workspace Directory f. Setup Repository Docker Container 2. When I exec the step: "Download Packages" g. docker exec -it /bin/bash h. bash download_mirror.sh 3. Through below two steps, Finally, I have hundreds of package in the L1_failmoved_url_centos.log. I have put this file in the attachment. 4. I hope you can help me slove this problem I have spend in 20 days. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: L1_failmoved_url_centos.log Type: application/octet-stream Size: 443463 bytes Desc: L1_failmoved_url_centos.log URL: From longqian.zhao at intel.com Fri Oct 19 10:34:25 2018 From: longqian.zhao at intel.com (Zhao, Longqian) Date: Fri, 19 Oct 2018 10:34:25 +0000 Subject: [Starlingx-discuss] Why am I don't download hundreds of package in the step: "bash download_mirror.sh" Message-ID: <6325970F810676468E0EC5D8ED76D2190B7BB7DA@SHSMSX101.ccr.corp.intel.com> Hi, when I install environment about StarlingX. My problem is following phenomenon: 1.Firstly, I follow the steps as required: a. sudo apt-get update b. Installation Requirements and Dependencies c. Install Public SSH Key d. Install stx-tools project e. Create a Workspace Directory f. Setup Repository Docker Container 2. When I exec the step: "Download Packages" g. docker exec -it /bin/bash h. bash download_mirror.sh 3. Through below two steps, Finally, I have hundreds of package in the L1_failmoved_url_centos.log. I have put this file in the attachment. 4. I hope you can help me slove this problem I have spend in 20 days. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: L1_failmoved_url_centos.rar Type: application/octet-stream Size: 34983 bytes Desc: L1_failmoved_url_centos.rar URL: From Matt.Peters at windriver.com Fri Oct 19 16:33:44 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Fri, 19 Oct 2018 16:33:44 +0000 Subject: [Starlingx-discuss] Question about neutron connectivity-test error In-Reply-To: <0A5D9A624DF90343892F8F3FE7DE525A2A8E62B9@fmsmsx101.amr.corp.intel.com> References: <0A5D9A624DF90343892F8F3FE7DE525A2A8E62B9@fmsmsx101.amr.corp.intel.com> Message-ID: Hello José, The provider network audits are disabled by default in StarlingX since there is no corresponding agent driver for OVS-DPDK. Therefore the audits cannot be supported until an associated driver for OVS-DPDK is developed. Regards, Matt > -----Original Message----- > From: Perez Carranza, Jose [mailto:jose.perez.carranza at intel.com] > Sent: Thursday, October 18, 2018 4:39 PM > To: 'starlingx-discuss at lists.starlingx.io' > Subject: [Starlingx-discuss] Question about neutron connectivity-test error > > Hi > > When I run command "neutron providernet-connectivity-test-list" I received > an error --> "Provider network testing is not supported in this > configuration.". Is there any specific configuration that I should apply to my > controller or provider network? > > http://paste.openstack.org/show/732446/ > > Regards, > José > > > > -----Original Message----- > > From: Perez Carranza, Jose > > Sent: Tuesday, October 16, 2018 2:27 PM > > To: starlingx-discuss at lists.starlingx.io > > Subject: Help on shared vCPU usage > > > > Hi > > > > I'm trying to execute scenario 206 [1] that indicates that I need to launch > > instance with a flavor with "hw:wrs:shared_vcpu", I created a flavor with > > "hw:wrs:shared_vcpu=1" but at the moment of create the instance below > > error is displayed.. Is there any extra config or parameter that I need to > apply > > before creating the instance ? > > > > Error (Duplex Bare-Metal): > > ====== > > No valid host was found. There are not enough hosts available. controller- > 0: > > (NUMATopologyFilter) Shared vCPU not enabled on host cell 1, required by > > instance cell 0, Shared vCPU not enabled on host cell 0, required by > instance > > cell 0, controller-1: (NUM ===== > > > > 1- > https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > > > > Regards, > > José > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ricardo.o.perez at intel.com Fri Oct 19 17:45:17 2018 From: ricardo.o.perez at intel.com (Perez, Ricardo O) Date: Fri, 19 Oct 2018 17:45:17 +0000 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages Message-ID: Hello, I'm trying to execute a test, which his goal is to made use of huge pages (2M / 1G) launching a VM. I have already verified the configuration of the controllers GRUB options and we have defined the variable hugepagesz=1G, however, inside the same parameters list, there is a "default_hugepagesz=2M". The default hugepagesz, has been modified to 1G, and the compute has been restarted with no issues. The flavor has been modified, with the following option: " hw:mem_page_size=104875" (1G) page. However when I launch a VM using the modified flavor it creates the VM, but if I check the field values of the VM using " openstack server show ". I'm getting the following value for pgsize: wrs-res:topology | node:0, 512MB, pgsize:2M, vcpus:0, pol:sha | So, seems like it's still taking the 2M pagesize instead of the 1G. What I have already tried is to: * Change also the image metadata: hw:mem_page_size=large and also 104875. Result, VM launched successfully, but pgsize value still 2M. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=104875, launch the VM with ERROR status. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=2048, launch the VM with success, , pgsize value is 2M, which is expected, ok. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=small launch the VM with success, , pgsize value is 4k, which is expected, ok. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=any launch the VM with success, , pgsize value is 2M, which is expected, ok. Could be this an issue trying to use 1G mem pages or do I'm missing some other configuration parameter? Thanks in advance for your help -------------- next part -------------- An HTML attachment was scrubbed... URL: From Ghada.Khalil at windriver.com Fri Oct 19 18:05:08 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Fri, 19 Oct 2018 18:05:08 +0000 Subject: [Starlingx-discuss] Fix for ovs memory allocation failure is merged in master In-Reply-To: <151EE31B9FCCA54397A757BC674650F0BA46D6CF@ALA-MBD.corp.ad.wrs.com> References: <151EE31B9FCCA54397A757BC674650F0BA46D6CF@ALA-MBD.corp.ad.wrs.com> Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46DB98@ALA-MBD.corp.ad.wrs.com> FYI. The fix has been merged in r/2018.10 release branch: https://review.openstack.org/#/c/611873/ From: Khalil, Ghada [mailto:Ghada.Khalil at windriver.com] Sent: Thursday, October 18, 2018 7:14 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Fix for ovs memory allocation failure is merged in master Hello all, I know that several people have been waiting for a fix for Launchpad: https://bugs.launchpad.net/starlingx/+bug/1796380 This is to let you know that the fix has been merged in master this afternoon. We will merge the code in the r/2018.10 release branch tomorrow. Regards, Ghada -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Fri Oct 19 18:07:59 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Fri, 19 Oct 2018 18:07:59 +0000 Subject: [Starlingx-discuss] Fix for ovs memory allocation failure is merged in master In-Reply-To: <151EE31B9FCCA54397A757BC674650F0BA46DB98@ALA-MBD.corp.ad.wrs.com> References: <151EE31B9FCCA54397A757BC674650F0BA46D6CF@ALA-MBD.corp.ad.wrs.com> <151EE31B9FCCA54397A757BC674650F0BA46DB98@ALA-MBD.corp.ad.wrs.com> Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C354F25@fmsmsx104.amr.corp.intel.com> Thanks, Ghada! We will work on getting an ISO. A. From: Khalil, Ghada [mailto:Ghada.Khalil at windriver.com] Sent: Friday, October 19, 2018 1:05 PM To: Khalil, Ghada ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Fix for ovs memory allocation failure is merged in master FYI. The fix has been merged in r/2018.10 release branch: https://review.openstack.org/#/c/611873/ From: Khalil, Ghada [mailto:Ghada.Khalil at windriver.com] Sent: Thursday, October 18, 2018 7:14 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Fix for ovs memory allocation failure is merged in master Hello all, I know that several people have been waiting for a fix for Launchpad: https://bugs.launchpad.net/starlingx/+bug/1796380 This is to let you know that the fix has been merged in master this afternoon. We will merge the code in the r/2018.10 release branch tomorrow. Regards, Ghada -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Fri Oct 19 18:18:33 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Fri, 19 Oct 2018 18:18:33 +0000 Subject: [Starlingx-discuss] Why am I don't download hundreds of package from the Internet? In-Reply-To: <6325970F810676468E0EC5D8ED76D2190B7B97AC@SHSMSX101.ccr.corp.intel.com> References: <6325970F810676468E0EC5D8ED76D2190B7B97AC@SHSMSX101.ccr.corp.intel.com> Message-ID: <9A85D2917C58154C960D95352B22818BB1E80489@ORSMSX158.amr.corp.intel.com> Hello Longqian. What you are seeing is normal and expected behavior. StarlingX is built from many open source components and to build it, you have to be able to satisfy all of the build dependencies for all of them. That's what the download_mirror.sh script does. Unfortunately the script downloads a _lot_ of packages, from an open source eco-system that is constantly changing. We monitor the script and update it daily but you may still find issues. Please contact Cindy Xie who leads the Intel Shanghai team for StarlingX. Someone on her team can work with you to get you past the issues you are hitting. Long term we are working on build system improvements that will address this issue. brucej From: Zhao, Longqian [mailto:longqian.zhao at intel.com] Sent: Friday, October 19, 2018 3:26 AM To: starlingx-discuss at lists.starlingx.io Cc: Zhao, Longqian Subject: [Starlingx-discuss] Why am I don't download hundreds of package from the Internet? Hi, my problem is following phenomenon: 1.Firstly, I follow the steps as required: a. sudo apt-get update b. Installation Requirements and Dependencies c. Install Public SSH Key d. Install stx-tools project e. Create a Workspace Directory f. Setup Repository Docker Container 2. When I exec the step: "Download Packages" g. docker exec -it /bin/bash h. bash download_mirror.sh 3. Through below two steps, Finally, I have hundreds of package in the L1_failmoved_url_centos.log. I have put this file in the attachment. 4. I hope you can help me slove this problem I have spend in 20 days. -------------- next part -------------- An HTML attachment was scrubbed... URL: From michel.thebeau at windriver.com Fri Oct 19 18:39:27 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Fri, 19 Oct 2018 14:39:27 -0400 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: References: Message-ID: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: > Hello, > > I'm trying to execute a test, which his goal is to made use of huge pages (2M / 1G) launching a VM. > > I have already verified the configuration of the controllers GRUB options and we have defined the variable hugepagesz=1G, however, inside the same parameters list, there is a "default_hugepagesz=2M". The default hugepagesz, has been modified to 1G, and the compute has been restarted with no issues. Hi Ricardo, You wrote, "controllers GRUB options" then "the compute has been restarted with no issues".  Is there a GRUB option one can configure on the controller that is for the computes?  I don't remember seeing one.  If you're actually changing /etc/default/grub, I doubt that's supported. I did not find a procedure for changing default_hugepagesz  on controllers or computes grub configuration. It's supposed to be an option, for example through in horizon interface or 'system host-memory-modify'... but you didn't quote either of those. > > The flavor has been modified, with the following option: " hw:mem_page_size=104875" (1G) page. Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but not quite. M > > However when I launch a VM using the modified flavor it creates the VM, but if I check the field values of the VM using " openstack server show ". I'm getting the following value for pgsize: > wrs-res:topology | node:0, 512MB, pgsize:2M, vcpus:0, pol:sha | > > So, seems like it's still taking the 2M pagesize instead of the 1G. > > What I have already tried is to: > > * Change also the image metadata: hw:mem_page_size=large and also 104875. Result, VM launched successfully, but pgsize value still 2M. There's the interesting "104875" again. > * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=104875, launch the VM with ERROR status. 104875 > * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=2048, launch the VM with success, , pgsize value is 2M, which is expected, ok. > > * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=small launch the VM with success, , pgsize value is 4k, which is expected, ok. > > * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=any launch the VM with success, , pgsize value is 2M, which is expected, ok. > > Could be this an issue trying to use 1G mem pages or do I'm missing some other configuration parameter? > > Thanks in advance for your help > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Brent.Rowsell at windriver.com Fri Oct 19 18:50:36 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Fri, 19 Oct 2018 18:50:36 +0000 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: References: Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2BEEC8@ALA-MBD.corp.ad.wrs.com> If you want to use 1G huge pages, you need to provision 1G huge pages in sysinv for the compute and then specify the page size extra spec when launching the VM. system host-memory-modify usage: system host-memory-modify [-m ] [-2M <2M hugepages number>] [-1G <1G hugepages number>] You will have to reduce the number of 2M huge pages to accommodate the 1G requirements. You should not be manually modifying grub args. Brent From: Perez, Ricardo O [mailto:ricardo.o.perez at intel.com] Sent: Friday, October 19, 2018 1:45 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Question - Using 1Gb mem pages Hello, I'm trying to execute a test, which his goal is to made use of huge pages (2M / 1G) launching a VM. I have already verified the configuration of the controllers GRUB options and we have defined the variable hugepagesz=1G, however, inside the same parameters list, there is a "default_hugepagesz=2M". The default hugepagesz, has been modified to 1G, and the compute has been restarted with no issues. The flavor has been modified, with the following option: " hw:mem_page_size=104875" (1G) page. However when I launch a VM using the modified flavor it creates the VM, but if I check the field values of the VM using " openstack server show ". I'm getting the following value for pgsize: wrs-res:topology | node:0, 512MB, pgsize:2M, vcpus:0, pol:sha | So, seems like it's still taking the 2M pagesize instead of the 1G. What I have already tried is to: * Change also the image metadata: hw:mem_page_size=large and also 104875. Result, VM launched successfully, but pgsize value still 2M. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=104875, launch the VM with ERROR status. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=2048, launch the VM with success, , pgsize value is 2M, which is expected, ok. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=small launch the VM with success, , pgsize value is 4k, which is expected, ok. * Leave the GRUB default values, and use modified flavor to hw:mem_page_size=any launch the VM with success, , pgsize value is 2M, which is expected, ok. Could be this an issue trying to use 1G mem pages or do I'm missing some other configuration parameter? Thanks in advance for your help -------------- next part -------------- An HTML attachment was scrubbed... URL: From Brent.Rowsell at windriver.com Fri Oct 19 19:42:23 2018 From: Brent.Rowsell at windriver.com (Rowsell, Brent) Date: Fri, 19 Oct 2018 19:42:23 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? In-Reply-To: References: <2588653EBDFFA34B982FAF00F1B4844EBB2A94B0@ALA-MBD.corp.ad.wrs.com> Message-ID: <2588653EBDFFA34B982FAF00F1B4844EBB2BF09A@ALA-MBD.corp.ad.wrs.com> Hi Lisa, See inline. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Thursday, October 18, 2018 11:17 PM To: Rowsell, Brent ; Ding, Jack ; Church, Robert ; 'starlingx-discuss at lists.starlingx.io' Cc: Li, Xiaoyan Subject: RE: What's the use case of these 2 patches? Hi Brent and all I brought these two patches to yesterday's glance meeting, some update here: - Regarding the commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Community suggest to add option "-vv " to "glance image-list", something like "glance -vv image-list". They don't want to show store info in "-v" because in most situation store info is the same, so there will be same values showed in the column in whole page, something like: | ... | img1 | rbd | ... | ... | img2 | rbd | ... | ... | img3 | rbd | ... | ... | img4 | rbd | ... | ... | img5 | rbd | ... ... ... I think it make sense to show this info in -vv, not -v Do you think is this achieved your original purpose of the patch? If so, I will go ahead to update the patch for "-vv". [BR] I think that is reasonable. - Regarding the commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Cores mentioned it has been implemented in rocky version which introduced a new feature "multi-store". I tried to setup the multi-store environment and verified this, the store info indeed saved, and the info can be found in database table: image_locations, something like {"backend": "ceph"}, see below: ===================================================================================== mysql> select * from image_locations where status = 'active'; +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ | id | image_id | value | created_at | updated_at | deleted_at | deleted | meta_data | status | +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ | 1 | 617f2880-f5b9-491f-b569-d2bd088c0bd5 | file:///opt/stack/data/glance/images/617f2880-f5b9-491f-b569-d2bd088c0bd5 | 2018-10-17 03:15:46 | 2018-10-17 03:15:46 | NULL | 0 | {} | active | | 9 | 0de67910-661e-4a2c-b5de-bcdc4d4a62db | rbd://b7b109e5-6c2d-4060-b81f-bd5c6fe57e6e/pool1/0de67910-661e-4a2c-b5de-bcdc4d4a62db/snap | 2018-10-18 06:45:24 | 2018-10-18 06:45:24 | NULL | 0 | {} | active | | 13 | e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | file:///opt/stack/data/glance/images/e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | 2018-10-18 08:40:18 | 2018-10-18 08:40:18 | NULL | 0 | {"backend": "locallvm"} | active | | 14 | 4804a6b7-6392-4232-9a9e-897d1eeef589 | rbd://b7b109e5-6c2d-4060-b81f-bd5c6fe57e6e/pool1/4804a6b7-6392-4232-9a9e-897d1eeef589/snap | 2018-10-18 08:41:57 | 2018-10-18 08:41:57 | NULL | 0 | {"backend": "ceph"} | active | +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ 4 rows in set (0.00 sec) ===================================================================================== Also Cmd "glance image-show" can show the store id like: "stores | ceph", see below: stack at ceph01liang:~$ glance image-show 4804a6b7-6392-4232-9a9e-897d1eeef589 +------------------+----------------------------------------------------------------------------------+ | Property | Value | +------------------+----------------------------------------------------------------------------------+ | checksum | f8ab98ff5e73ebab884d80c9dc9c7290 | | container_format | bare | | created_at | 2018-10-18T08:41:55Z | | disk_format | raw | | id | 4804a6b7-6392-4232-9a9e-897d1eeef589 | | min_disk | 0 | | min_ram | 0 | | name | img2-ceph | | os_hash_algo | sha512 | | os_hash_value | f0fd1b50420dce4ca382ccfbb528eef3a38bbeff00b54e95e3876b9bafe7ed2d6f919ca35d9046d4 | | | 37c6d2d8698b1174a335fbd66035bb3edc525d2cdb187232 | | os_hidden | False | | owner | 60a12b1788ad44468afd983f89a5f8dc | | protected | False | | size | 13267968 | | status | active | | stores | ceph | | tags | [] | | updated_at | 2018-10-18T08:41:57Z | | virtual_size | Not available | | visibility | public | +------------------+----------------------------------------------------------------------------------+ So, the store info already saved in latest code, do you think is this achieved your original purpose of the patch? If so, I guess the patch can be marked as merged. [BR] Yes, this addresses the intent of the original patch. You can update it to "Abandoned - Hold for Rebase" Thanks Liang -----Original Message----- From: Fang, Liang A Sent: Wednesday, October 10, 2018 11:44 PM To: Rowsell, Brent ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? OK, thank you Brent, let me try my best to convince community to accept these two patches. Regards Liang -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Wednesday, October 10, 2018 8:50 PM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Liang, This was a small usability improvement to help the user distinguish the glance backend type in a multi backend environment. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:49 AM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Related review URL: https://review.openstack.org/#/c/605014/ https://review.openstack.org/#/c/605006/ Compare with upstream glance, starlingx added "Store" column as below +-----+--------+-------+-----------------+-----+------------+--------+---------+ | ID | Name | Store | Disk_format | ... | Size | Status |Owner| +-----+--------+-------+-----------------+-----+------------+--------+---------+ | xxx | img1 | file | raw | ... | 10737418 | active | xxx | | xxx | img2 | file | raw | ... | 5086345 | active | xxx | +-----+--------+-------+-----------------+-----+------------+--------+---------+ Regards Liang -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:35 PM To: jack.ding at windriver.com; robert.church at windriver.com; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the use case of these 2 patches? Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ricardo.o.perez at intel.com Fri Oct 19 22:14:19 2018 From: ricardo.o.perez at intel.com (Perez, Ricardo O) Date: Fri, 19 Oct 2018 22:14:19 +0000 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> References: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> Message-ID: Hi Michael, Thanks for your answer, I just forgot to put the computes information about memory configuration, here is the info: Compute-0 VM Pages 0 Reserved for Platform: 8000 MiB Usable Total: 86387 MiB Available: 84787 MiB 4K Pages: Total: 230656 2M Hugepages: Total: 1271 Available: 983 1G Hugepages: Total: 80 Available: 80 1 Reserved for Platform: 2000 MiB Usable Total: 93698 MiB Available: 92674 MiB 4K Pages: Total: 29696 2M Hugepages: Total: 5319 Available: 5319 1G Hugepages: Total: 80 Available: 80 *********************************** Compute-1 Processor Memory VM Pages 0 Reserved for Platform: 8000 MiB Usable Total: 86504 MiB Available: 84968 MiB 4K Pages: Total: 0 2M Hugepages: Total: 1780 Available: 1524 1G Hugepages: Total: 80 Available: 80 1 Reserved for Platform: 2000 MiB Usable Total: 93552 MiB Available: 92528 MiB 4K Pages: Total: 0 2M Hugepages: Total: 5304 Available: 5304 1G Hugepages: Total: 80 Available: 80 Am I missing something else to made use of 1G memory pages ? Thanks -Richo > -----Original Message----- > From: Michel Thebeau [mailto:michel.thebeau at windriver.com] > Sent: Friday, October 19, 2018 1:39 PM > To: Perez, Ricardo O ; starlingx- > discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages > > On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: > > Hello, > > > > I'm trying to execute a test, which his goal is to made use of huge pages (2M / > 1G) launching a VM. > > > > I have already verified the configuration of the controllers GRUB options and > we have defined the variable hugepagesz=1G, however, inside the same > parameters list, there is a "default_hugepagesz=2M". The default hugepagesz, > has been modified to 1G, and the compute has been restarted with no issues. > > > Hi Ricardo, > > You wrote, "controllers GRUB options" then "the compute has been restarted > with no issues".  Is there a GRUB option one can configure on the controller that > is for the computes?  I don't remember seeing one. If you're actually changing > /etc/default/grub, I doubt that's supported. > I did not find a procedure for changing default_hugepagesz  on controllers or > computes grub configuration. > > It's supposed to be an option, for example through in horizon interface or > 'system host-memory-modify'... but you didn't quote either of those. > > > > > > The flavor has been modified, with the following option: " > hw:mem_page_size=104875" (1G) page. > > > Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but not > quite. > > > M > > > > > > However when I launch a VM using the modified flavor it creates the VM, but if > I check the field values of the VM using " openstack server show ". > I'm getting the following value for pgsize: > > wrs-res:topology | node:0, 512MB, pgsize:2M, vcpus:0, pol:sha | > > > > So, seems like it's still taking the 2M pagesize instead of the 1G. > > > > What I have already tried is to: > > > > * Change also the image metadata: hw:mem_page_size=large and also > 104875. Result, VM launched successfully, but pgsize value still 2M. > > There's the interesting "104875" again. > > > > * Leave the GRUB default values, and use modified flavor to > hw:mem_page_size=104875, launch the VM with ERROR status. > > 104875 > > > * Leave the GRUB default values, and use modified flavor to > hw:mem_page_size=2048, launch the VM with success, , pgsize value is 2M, > which is expected, ok. > > > > * Leave the GRUB default values, and use modified flavor to > hw:mem_page_size=small launch the VM with success, , pgsize value is 4k, > which is expected, ok. > > > > * Leave the GRUB default values, and use modified flavor to > hw:mem_page_size=any launch the VM with success, , pgsize value is 2M, which > is expected, ok. > > > > Could be this an issue trying to use 1G mem pages or do I'm missing some > other configuration parameter? > > > > Thanks in advance for your help > > > > > > _______________________________________________ > > Starlingx-discuss mailing list > > Starlingx-discuss at lists.starlingx.io > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From liang.a.fang at intel.com Sat Oct 20 00:11:06 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Sat, 20 Oct 2018 00:11:06 +0000 Subject: [Starlingx-discuss] What's the use case of these 2 patches? In-Reply-To: <2588653EBDFFA34B982FAF00F1B4844EBB2BF09A@ALA-MBD.corp.ad.wrs.com> References: <2588653EBDFFA34B982FAF00F1B4844EBB2A94B0@ALA-MBD.corp.ad.wrs.com> <2588653EBDFFA34B982FAF00F1B4844EBB2BF09A@ALA-MBD.corp.ad.wrs.com> Message-ID: Thanks Brent. Regards Liang -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Saturday, October 20, 2018 3:42 AM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; 'starlingx-discuss at lists.starlingx.io' Cc: Li, Xiaoyan Subject: RE: What's the use case of these 2 patches? Hi Lisa, See inline. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Thursday, October 18, 2018 11:17 PM To: Rowsell, Brent ; Ding, Jack ; Church, Robert ; 'starlingx-discuss at lists.starlingx.io' Cc: Li, Xiaoyan Subject: RE: What's the use case of these 2 patches? Hi Brent and all I brought these two patches to yesterday's glance meeting, some update here: - Regarding the commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Community suggest to add option "-vv " to "glance image-list", something like "glance -vv image-list". They don't want to show store info in "-v" because in most situation store info is the same, so there will be same values showed in the column in whole page, something like: | ... | img1 | rbd | ... | ... | img2 | rbd | ... | ... | img3 | rbd | ... | ... | img4 | rbd | ... | ... | img5 | rbd | ... ... ... I think it make sense to show this info in -vv, not -v Do you think is this achieved your original purpose of the patch? If so, I will go ahead to update the patch for "-vv". [BR] I think that is reasonable. - Regarding the commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Cores mentioned it has been implemented in rocky version which introduced a new feature "multi-store". I tried to setup the multi-store environment and verified this, the store info indeed saved, and the info can be found in database table: image_locations, something like {"backend": "ceph"}, see below: ===================================================================================== mysql> select * from image_locations where status = 'active'; +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ | id | image_id | value | created_at | updated_at | deleted_at | deleted | meta_data | status | +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ | 1 | 617f2880-f5b9-491f-b569-d2bd088c0bd5 | file:///opt/stack/data/glance/images/617f2880-f5b9-491f-b569-d2bd088c0bd5 | 2018-10-17 03:15:46 | 2018-10-17 03:15:46 | NULL | 0 | {} | active | | 9 | 0de67910-661e-4a2c-b5de-bcdc4d4a62db | rbd://b7b109e5-6c2d-4060-b81f-bd5c6fe57e6e/pool1/0de67910-661e-4a2c-b5de-bcdc4d4a62db/snap | 2018-10-18 06:45:24 | 2018-10-18 06:45:24 | NULL | 0 | {} | active | | 13 | e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | file:///opt/stack/data/glance/images/e9af91ba-4d8c-4a3c-b2d7-5fbb5421f065 | 2018-10-18 08:40:18 | 2018-10-18 08:40:18 | NULL | 0 | {"backend": "locallvm"} | active | | 14 | 4804a6b7-6392-4232-9a9e-897d1eeef589 | rbd://b7b109e5-6c2d-4060-b81f-bd5c6fe57e6e/pool1/4804a6b7-6392-4232-9a9e-897d1eeef589/snap | 2018-10-18 08:41:57 | 2018-10-18 08:41:57 | NULL | 0 | {"backend": "ceph"} | active | +----+--------------------------------------+--------------------------------------------------------------------------------------------+---------------------+---------------------+------------+---------+-------------------------+--------+ 4 rows in set (0.00 sec) ===================================================================================== Also Cmd "glance image-show" can show the store id like: "stores | ceph", see below: stack at ceph01liang:~$ glance image-show 4804a6b7-6392-4232-9a9e-897d1eeef589 +------------------+----------------------------------------------------------------------------------+ | Property | Value | +------------------+----------------------------------------------------------------------------------+ | checksum | f8ab98ff5e73ebab884d80c9dc9c7290 | | container_format | bare | | created_at | 2018-10-18T08:41:55Z | | disk_format | raw | | id | 4804a6b7-6392-4232-9a9e-897d1eeef589 | | min_disk | 0 | | min_ram | 0 | | name | img2-ceph | | os_hash_algo | sha512 | | os_hash_value | f0fd1b50420dce4ca382ccfbb528eef3a38bbeff00b54e95e3876b9bafe7ed2d6f919ca35d9046d4 | | | 37c6d2d8698b1174a335fbd66035bb3edc525d2cdb187232 | | os_hidden | False | | owner | 60a12b1788ad44468afd983f89a5f8dc | | protected | False | | size | 13267968 | | status | active | | stores | ceph | | tags | [] | | updated_at | 2018-10-18T08:41:57Z | | virtual_size | Not available | | visibility | public | +------------------+----------------------------------------------------------------------------------+ So, the store info already saved in latest code, do you think is this achieved your original purpose of the patch? If so, I guess the patch can be marked as merged. [BR] Yes, this addresses the intent of the original patch. You can update it to "Abandoned - Hold for Rebase" Thanks Liang -----Original Message----- From: Fang, Liang A Sent: Wednesday, October 10, 2018 11:44 PM To: Rowsell, Brent ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? OK, thank you Brent, let me try my best to convince community to accept these two patches. Regards Liang -----Original Message----- From: Rowsell, Brent [mailto:Brent.Rowsell at windriver.com] Sent: Wednesday, October 10, 2018 8:50 PM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Liang, This was a small usability improvement to help the user distinguish the glance backend type in a multi backend environment. Brent -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:49 AM To: Fang, Liang A ; Ding, Jack ; Church, Robert ; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: RE: What's the use case of these 2 patches? Related review URL: https://review.openstack.org/#/c/605014/ https://review.openstack.org/#/c/605006/ Compare with upstream glance, starlingx added "Store" column as below +-----+--------+-------+-----------------+-----+------------+--------+---------+ | ID | Name | Store | Disk_format | ... | Size | Status |Owner| +-----+--------+-------+-----------------+-----+------------+--------+---------+ | xxx | img1 | file | raw | ... | 10737418 | active | xxx | | xxx | img2 | file | raw | ... | 5086345 | active | xxx | +-----+--------+-------+-----------------+-----+------------+--------+---------+ Regards Liang -----Original Message----- From: Fang, Liang A [mailto:liang.a.fang at intel.com] Sent: Wednesday, October 10, 2018 4:35 PM To: jack.ding at windriver.com; robert.church at windriver.com; Rowsell, Brent ; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] What's the use case of these 2 patches? Hi I'm working on upstreaming below 2 patches and the code review was slightly blocked because community need to know the use case: Commit from glanceclient: 75967f4c9c9ca582277d036581effcac0c7ee5a9: Show glance_backend rbd or file. Commit from glance: f6ad3a44303a1bf9abec91e1d679cdd13fb9c4f6: Add store info in glance image-list and image-show Could anybody share the original intention of these two patches? What's the use case of saving/showing the store info? It seems end user cannot do anything with this information. Thanks! Regards Liang _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From aragorn at intel.com Sat Oct 20 09:34:30 2018 From: aragorn at intel.com (aragorn at intel.com) Date: 20 Oct 2018 02:34:30 -0700 Subject: [Starlingx-discuss] Automated Notification for: [mirror-downloader][#115] Missing packages - Results report Message-ID: <1649c0$2esqfc@orsmga007.jf.intel.com> [mirror-downloader][#115] Missing packages. Results report: - Missing: logs/L1_srpms_missing_centos3rdparties.log kernel-rt-3.10.0-862.11.6.rt56.819.el7.src.rpm python2-pysocks-1.6.7-1.el7.noarch.rpm - Missing GPG key: ./output/stx-r1/CentOS/pike/Source/libvirt-python-4.7.0-1.fc28.src.rpm: RSA sha1 ((MD5) PGP) md5 NOT OK (MISSING KEYS: (MD5) PGP#596bea5d) From ada.cabrales at intel.com Sat Oct 20 19:39:01 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Sat, 20 Oct 2018 19:39:01 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-19-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C355748@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-19-28-r-2018.10.iso 31 TCs PASS, 0 TCs FAIL. The 2 failures from stx-2018-10-18-27-r-2018.10.iso seems to be false positive. However, the root cause hasn't been found yet, still investigating. - Duplex virtual deployed and configured correctly after fix for openvswitch has been merged. - "Volume snapshot created from a cirros image can not be deleted with cli" [0] found in multinode with external storage on virtual configuration - Thanks to the team that is working over the weekend to increase our execution rate. - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 2/71/3/76 Simplex virtual - 0/10/61/71 Duplex bare metal - 2/89/7/98 Duplex virtual - 3/86/14/103 Multinode local storage bare metal - 4/106/8/118 Multinode local storage virtual - 7/94/25/126 Multinode external storage bare metal - 4/104/8/116 Multinode external storage virtual - 5/90/24/129 Full report in the wiki [1] [0] https://bugs.launchpad.net/starlingx/+bug/1798994 [1] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary A. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ian.Jolliffe at windriver.com Mon Oct 22 01:10:59 2018 From: Ian.Jolliffe at windriver.com (Jolliffe, Ian) Date: Mon, 22 Oct 2018 01:10:59 +0000 Subject: [Starlingx-discuss] STX + Tungsten Fabric In-Reply-To: References: Message-ID: <1B7CF6B0-0366-4598-82B4-27B346D339EA@windriver.com> Hi Corey; I am not currently aware of anyone working on this. Interesting idea for sure. Ian From: Corey Erickson Date: Monday, October 15, 2018 at 1:22 PM To: "starlingx-discuss at lists.starlingx.io" Subject: [Starlingx-discuss] STX + Tungsten Fabric Hi All, Curious to know if anyone is active in working on SDN integrations such as Tungsten Fabric ? Thanks, Corey Erickson -------------- next part -------------- An HTML attachment was scrubbed... URL: From liang.a.fang at intel.com Mon Oct 22 01:57:03 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Mon, 22 Oct 2018 01:57:03 +0000 Subject: [Starlingx-discuss] Anybody know how to create project manually in storyboard? Message-ID: Hi I want to do some backport work for stx-cinder, and want to trace the work in storyboard. But stx-cinder project is not in storyboard, do anybody know how to create this? I was told that stx-cinder is not in git.openstack.org, so it has not been created in storyboard automatically. Regards Liang -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhipengs.liu at intel.com Mon Oct 22 09:55:37 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Mon, 22 Oct 2018 09:55:37 +0000 Subject: [Starlingx-discuss] [Feature] Refactor init/config related non-openstack patches. Message-ID: <93814834B4855241994F290E959305C752FD9D75@SHSMSX104.ccr.corp.intel.com> Hi all, I'd like you to review the feature spec submitted by me for the topic of refactoring init/config related patches. https://review.openstack.org/#/c/612292/ For RPM patches related to adding or changing StarlingX specific systemd service files, configuration files or other types, 1) If they are not overwriting files already delivered by the upstream RPM, refactor and have an specific config rpm that delivers the new file instead of original patch 2) If they need to change files delivered by the upstream RPM, use puppet to change it instead of original patch. Related story is below link https://storyboard.openstack.org/#!/story/2003768 Thx! -zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhipengs.liu at intel.com Mon Oct 22 09:52:58 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Mon, 22 Oct 2018 09:52:58 +0000 Subject: [Starlingx-discuss] [Feature] Refactor init/config related non-openstack patches. Message-ID: <93814834B4855241994F290E959305C752FD9D37@SHSMSX104.ccr.corp.intel.com> Hi all, I'd like you to review the feature spec submitted by me for the topic of refactoring init/config related patches. https://review.openstack.org/#/c/612292/ For RPM patches related to adding or changing StarlingX specific systemd service files, configuration files or other types, 1) If they are not overwriting files already delivered by the upstream RPM, refactor and have an specific config rpm that delivers the new file instead of original patch 2) If they need to change files delivered by the upstream RPM, use puppet to change it instead of original patch. Related story is below link https://storyboard.openstack.org/#!/story/2003768 Also attached word version FYI. Thx! -zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Proposed solution for init patches.docx Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 436967 bytes Desc: Proposed solution for init patches.docx URL: From Matt.Peters at windriver.com Mon Oct 22 12:19:17 2018 From: Matt.Peters at windriver.com (Peters, Matt) Date: Mon, 22 Oct 2018 12:19:17 +0000 Subject: [Starlingx-discuss] Community Feedbacks for RFE(patch 4ae5a58) for StartlingX upstreaming In-Reply-To: References: Message-ID: Hello Chenjie, The referenced blueprint for native-l2pop I don’t think helps in this use-case since the extension framework being proposed for l2pop fdb entries is required for the BGP VPN use-case which is directly managing the FDB entries for l2pop. The native-l2pop wants to move the information to the port object and provide custom extensions at that level, but I think this use-case is best served with extensions to the FDB. -Matt From: "Xu, Chenjie" Date: Thursday, October 18, 2018 at 10:58 PM To: "Peters, Matt" Cc: "starlingx-discuss at lists.starlingx.io" Subject: [Starlingx-discuss] Community Feedbacks for RFE(patch 4ae5a58) for StartlingX upstreaming Hi Matt, For patch 4ae5a58, the RFE “Enable other subprojects to extend l2pop fdb information” has been draft and is under reviewing for now. Community provide some feedbacks and could you please help review and comment! 1. The link for the RFE is below: https://bugs.launchpad.net/neutron/+bug/1793653 2. The link for the code is below: https://review.openstack.org/#/c/599319/ 3. The link for the BP mentioned by Slawek is below: https://blueprints.launchpad.net/neutron/+spec/native-l2pop 4. The link for the RFE related to the BP is below: https://bugs.launchpad.net/neutron/+bug/1630981 For patch 9f926a5, the RFE “Add l2pop support for floating ip resources” has been draft and is under internal reviewing for now. This RFE depends on the RFE “Enable other subprojects to extend l2pop fdb information” Best Regards, Xu, Chenjie -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhipengs.liu at intel.com Mon Oct 22 13:07:34 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Mon, 22 Oct 2018 13:07:34 +0000 Subject: [Starlingx-discuss] [Feature] Refactor init/config related non-openstack patches. In-Reply-To: <93814834B4855241994F290E959305C752FD9D37@SHSMSX104.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FD9D37@SHSMSX104.ccr.corp.intel.com> Message-ID: <93814834B4855241994F290E959305C752FD9E10@SHSMSX104.ccr.corp.intel.com> Updated attachment! From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: 2018年10月22日 17:53 To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] [Feature] Refactor init/config related non-openstack patches. Hi all, I’d like you to review the feature spec submitted by me for the topic of refactoring init/config related patches. https://review.openstack.org/#/c/612292/ For RPM patches related to adding or changing StarlingX specific systemd service files, configuration files or other types, 1) If they are not overwriting files already delivered by the upstream RPM, refactor and have an specific config rpm that delivers the new file instead of original patch 2) If they need to change files delivered by the upstream RPM, use puppet to change it instead of original patch. Related story is below link https://storyboard.openstack.org/#!/story/2003768 Also attached word version FYI. Thx! -zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: Proposed solution for init patches.docx Type: application/vnd.openxmlformats-officedocument.wordprocessingml.document Size: 35531 bytes Desc: Proposed solution for init patches.docx URL: From ricardo.o.perez at intel.com Mon Oct 22 14:57:13 2018 From: ricardo.o.perez at intel.com (Perez, Ricardo O) Date: Mon, 22 Oct 2018 14:57:13 +0000 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: References: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> Message-ID: Hi team, any idea about what is missing here ? Thanks _Richo > -----Original Message----- > From: Perez, Ricardo O [mailto:ricardo.o.perez at intel.com] > Sent: Friday, October 19, 2018 5:14 PM > To: Michel Thebeau ; starlingx- > discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages > > Hi Michael, > > Thanks for your answer, I just forgot to put the computes information about > memory configuration, here is the info: > > Compute-0 > > VM Pages > 0 Reserved for Platform: 8000 MiB > Usable Total: 86387 MiB > Available: 84787 MiB > 4K Pages: > Total: 230656 > 2M Hugepages: > Total: 1271 > Available: 983 > 1G Hugepages: > Total: 80 > Available: 80 > > 1 Reserved for Platform: 2000 MiB > Usable Total: 93698 MiB > Available: 92674 MiB > 4K Pages: > Total: 29696 > 2M Hugepages: > Total: 5319 > Available: 5319 > 1G Hugepages: > Total: 80 > Available: 80 > > *********************************** > Compute-1 > Processor > Memory > VM Pages > 0 Reserved for Platform: 8000 MiB > Usable Total: 86504 MiB > Available: 84968 MiB > 4K Pages: > Total: 0 > 2M Hugepages: > Total: 1780 > Available: 1524 > 1G Hugepages: > Total: 80 > Available: 80 > > 1 Reserved for Platform: 2000 MiB > Usable Total: 93552 MiB > Available: 92528 MiB > 4K Pages: > Total: 0 > 2M Hugepages: > Total: 5304 > Available: 5304 > 1G Hugepages: > Total: 80 > Available: 80 > > Am I missing something else to made use of 1G memory pages ? > > Thanks > -Richo > > > -----Original Message----- > > From: Michel Thebeau [mailto:michel.thebeau at windriver.com] > > Sent: Friday, October 19, 2018 1:39 PM > > To: Perez, Ricardo O ; starlingx- > > discuss at lists.starlingx.io > > Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages > > > > On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: > > > Hello, > > > > > > I'm trying to execute a test, which his goal is to made use of huge > > > pages (2M / > > 1G) launching a VM. > > > > > > I have already verified the configuration of the controllers GRUB > > > options and > > we have defined the variable hugepagesz=1G, however, inside the same > > parameters list, there is a "default_hugepagesz=2M". The default > > hugepagesz, has been modified to 1G, and the compute has been restarted > with no issues. > > > > > > Hi Ricardo, > > > > You wrote, "controllers GRUB options" then "the compute has been > > restarted with no issues".  Is there a GRUB option one can configure > > on the controller that is for the computes?  I don't remember seeing > > one. If you're actually changing /etc/default/grub, I doubt that's supported. > > I did not find a procedure for changing default_hugepagesz  on > > controllers or computes grub configuration. > > > > It's supposed to be an option, for example through in horizon > > interface or 'system host-memory-modify'... but you didn't quote either of > those. > > > > > > > > > > The flavor has been modified, with the following option: " > > hw:mem_page_size=104875" (1G) page. > > > > > > Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but > > not quite. > > > > > > M > > > > > > > > > > However when I launch a VM using the modified flavor it creates the > > > VM, but if > > I check the field values of the VM using " openstack server show ". > > I'm getting the following value for pgsize: > > > wrs-res:topology | node:0, 512MB, pgsize:2M, vcpus:0, pol:sha | > > > > > > So, seems like it's still taking the 2M pagesize instead of the 1G. > > > > > > What I have already tried is to: > > > > > > * Change also the image metadata: hw:mem_page_size=large and also > > 104875. Result, VM launched successfully, but pgsize value still 2M. > > > > There's the interesting "104875" again. > > > > > > > * Leave the GRUB default values, and use modified flavor to > > hw:mem_page_size=104875, launch the VM with ERROR status. > > > > 104875 > > > > > * Leave the GRUB default values, and use modified flavor to > > hw:mem_page_size=2048, launch the VM with success, , pgsize value is > > 2M, which is expected, ok. > > > > > > * Leave the GRUB default values, and use modified flavor to > > hw:mem_page_size=small launch the VM with success, , pgsize value is > > 4k, which is expected, ok. > > > > > > * Leave the GRUB default values, and use modified flavor to > > hw:mem_page_size=any launch the VM with success, , pgsize value is 2M, > > which is expected, ok. > > > > > > Could be this an issue trying to use 1G mem pages or do I'm missing > > > some > > other configuration parameter? > > > > > > Thanks in advance for your help > > > > > > > > > _______________________________________________ > > > Starlingx-discuss mailing list > > > Starlingx-discuss at lists.starlingx.io > > > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From michel.thebeau at windriver.com Mon Oct 22 14:59:30 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Mon, 22 Oct 2018 10:59:30 -0400 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: References: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> Message-ID: <503670f0-6cfc-d277-acf1-3358bafd6519@windriver.com> Hi Ricardo, The documentation indicates to configure the the huge pages per numa node on the compute, and to configure the flavour extra specs for the VM.  I am not aware of anything else in particular. There's some tips and caveats in the document, but I don't see anything in your message to think of those. If there's any chance you are trying to do this under virtualization, I am not aware that that is currently supported. M On 2018-10-19 6:14 p.m., Perez, Ricardo O wrote: > Hi Michael, > > Thanks for your answer, I just forgot to put the computes information about memory configuration, here is the info: > > Compute-0 > > VM Pages > 0 Reserved for Platform: 8000 MiB > Usable Total: 86387 MiB > Available: 84787 MiB > 4K Pages: > Total: 230656 > 2M Hugepages: > Total: 1271 > Available: 983 > 1G Hugepages: > Total: 80 > Available: 80 > > 1 Reserved for Platform: 2000 MiB > Usable Total: 93698 MiB > Available: 92674 MiB > 4K Pages: > Total: 29696 > 2M Hugepages: > Total: 5319 > Available: 5319 > 1G Hugepages: > Total: 80 > Available: 80 > > *********************************** > Compute-1 > Processor > Memory > VM Pages > 0 Reserved for Platform: 8000 MiB > Usable Total: 86504 MiB > Available: 84968 MiB > 4K Pages: > Total: 0 > 2M Hugepages: > Total: 1780 > Available: 1524 > 1G Hugepages: > Total: 80 > Available: 80 > > 1 Reserved for Platform: 2000 MiB > Usable Total: 93552 MiB > Available: 92528 MiB > 4K Pages: > Total: 0 > 2M Hugepages: > Total: 5304 > Available: 5304 > 1G Hugepages: > Total: 80 > Available: 80 > > Am I missing something else to made use of 1G memory pages ? > > Thanks > -Richo > >> -----Original Message----- >> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >> Sent: Friday, October 19, 2018 1:39 PM >> To: Perez, Ricardo O ; starlingx- >> discuss at lists.starlingx.io >> Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages >> >> On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: >>> Hello, >>> >>> I'm trying to execute a test, which his goal is to made use of huge pages (2M / >> 1G) launching a VM. >>> I have already verified the configuration of the controllers GRUB options and >> we have defined the variable hugepagesz=1G, however, inside the same >> parameters list, there is a "default_hugepagesz=2M". The default hugepagesz, >> has been modified to 1G, and the compute has been restarted with no issues. >> >> >> Hi Ricardo, >> >> You wrote, "controllers GRUB options" then "the compute has been restarted >> with no issues".  Is there a GRUB option one can configure on the controller that >> is for the computes?  I don't remember seeing one. If you're actually changing >> /etc/default/grub, I doubt that's supported. >> I did not find a procedure for changing default_hugepagesz  on controllers or >> computes grub configuration. >> >> It's supposed to be an option, for example through in horizon interface or >> 'system host-memory-modify'... but you didn't quote either of those. >> >> >>> The flavor has been modified, with the following option: " >> hw:mem_page_size=104875" (1G) page. >> >> >> Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but not >> quite. >> >> >> M >> >> >>> However when I launch a VM using the modified flavor it creates the VM, but if >> I check the field values of the VM using " openstack server show ". >> I'm getting the following value for pgsize: >>> wrs-res:topology | node:0, 512MB, pgsize:2M, vcpus:0, pol:sha | >>> >>> So, seems like it's still taking the 2M pagesize instead of the 1G. >>> >>> What I have already tried is to: >>> >>> * Change also the image metadata: hw:mem_page_size=large and also >> 104875. Result, VM launched successfully, but pgsize value still 2M. >> >> There's the interesting "104875" again. >> >> >>> * Leave the GRUB default values, and use modified flavor to >> hw:mem_page_size=104875, launch the VM with ERROR status. >> >> 104875 >> >>> * Leave the GRUB default values, and use modified flavor to >> hw:mem_page_size=2048, launch the VM with success, , pgsize value is 2M, >> which is expected, ok. >>> * Leave the GRUB default values, and use modified flavor to >> hw:mem_page_size=small launch the VM with success, , pgsize value is 4k, >> which is expected, ok. >>> * Leave the GRUB default values, and use modified flavor to >> hw:mem_page_size=any launch the VM with success, , pgsize value is 2M, which >> is expected, ok. >>> Could be this an issue trying to use 1G mem pages or do I'm missing some >> other configuration parameter? >>> Thanks in advance for your help >>> >>> >>> _______________________________________________ >>> Starlingx-discuss mailing list >>> Starlingx-discuss at lists.starlingx.io >>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From michel.thebeau at windriver.com Mon Oct 22 16:18:02 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Mon, 22 Oct 2018 12:18:02 -0400 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: <503670f0-6cfc-d277-acf1-3358bafd6519@windriver.com> References: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> <503670f0-6cfc-d277-acf1-3358bafd6519@windriver.com> Message-ID: Hi Ricardo, I was able to select 4K and 2M pages in the following way (attached txt file).  I'll try to see if I can get a system up with 1G pages, and I'll follow up with you. M On 2018-10-22 10:59 a.m., Michel Thebeau wrote: > Hi Ricardo, > > The documentation indicates to configure the the huge pages per numa > node on the compute, and to configure the flavour extra specs for the > VM.  I am not aware of anything else in particular. There's some tips > and caveats in the document, but I don't see anything in your message > to think of those. > > If there's any chance you are trying to do this under virtualization, > I am not aware that that is currently supported. > > M > > > On 2018-10-19 6:14 p.m., Perez, Ricardo O wrote: >> Hi Michael, >> >> Thanks for your answer, I just forgot to put the computes information >> about memory configuration, here is the info: >> >> Compute-0 >> >>     VM Pages >> 0   Reserved for Platform: 8000 MiB >> Usable Total: 86387 MiB >> Available: 84787 MiB >>      4K Pages: >>      Total: 230656 >> 2M Hugepages: >>      Total: 1271 >>      Available: 983 >> 1G Hugepages: >>      Total: 80 >>      Available: 80 >> >> 1   Reserved for Platform: 2000 MiB >> Usable Total: 93698 MiB >> Available: 92674 MiB >>      4K Pages: >>      Total: 29696 >> 2M Hugepages: >>      Total: 5319 >>      Available: 5319 >> 1G Hugepages: >>      Total: 80 >>      Available: 80 >> >> *********************************** >> Compute-1 >> Processor >>      Memory >>      VM Pages >> 0   Reserved for Platform: 8000 MiB >> Usable Total: 86504 MiB >> Available: 84968 MiB >>      4K Pages: >>      Total: 0 >> 2M Hugepages: >>      Total: 1780 >>      Available: 1524 >> 1G Hugepages: >>      Total: 80 >>      Available: 80 >> >> 1   Reserved for Platform: 2000 MiB >> Usable Total: 93552 MiB >> Available: 92528 MiB >>      4K Pages: >>      Total: 0 >> 2M Hugepages: >>      Total: 5304 >>      Available: 5304 >> 1G Hugepages: >>      Total: 80 >>      Available: 80 >> >> Am I missing something else to made use of 1G memory pages ? >> >> Thanks >> -Richo >> >>> -----Original Message----- >>> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >>> Sent: Friday, October 19, 2018 1:39 PM >>> To: Perez, Ricardo O ; starlingx- >>> discuss at lists.starlingx.io >>> Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages >>> >>> On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: >>>> Hello, >>>> >>>> I'm trying to execute a test, which his goal is to made use of huge >>>> pages (2M / >>> 1G) launching a VM. >>>> I have already verified the configuration of the controllers GRUB >>>> options and >>> we have defined the variable hugepagesz=1G, however, inside the same >>> parameters list, there is a "default_hugepagesz=2M". The default >>> hugepagesz, >>> has been modified to 1G, and the compute has been restarted with no >>> issues. >>> >>> >>> Hi Ricardo, >>> >>> You wrote, "controllers GRUB options" then "the compute has been >>> restarted >>> with no issues".  Is there a GRUB option one can configure on the >>> controller that >>> is for the computes?  I don't remember seeing one. If you're >>> actually changing >>> /etc/default/grub, I doubt that's supported. >>> I did not find a procedure for changing default_hugepagesz  on >>> controllers or >>> computes grub configuration. >>> >>> It's supposed to be an option, for example through in horizon >>> interface or >>> 'system host-memory-modify'... but you didn't quote either of those. >>> >>> >>>> The flavor has been modified, with the following option: " >>> hw:mem_page_size=104875"  (1G) page. >>> >>> >>> Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but not >>> quite. >>> >>> >>> M >>> >>> >>>> However when I launch a VM using the modified flavor it creates the >>>> VM, but if >>> I check the field values of the VM using " openstack server show >>> ". >>> I'm getting the following value for pgsize: >>>> wrs-res:topology                    | node:0,   512MB, pgsize:2M, >>>> vcpus:0, pol:sha | >>>> >>>> So, seems like it's still taking the 2M pagesize instead of the 1G. >>>> >>>> What I have already tried is to: >>>> >>>> *        Change also the image metadata: hw:mem_page_size=large and >>>> also >>> 104875. Result, VM launched successfully, but pgsize value still 2M. >>> >>> There's the interesting "104875" again. >>> >>> >>>> *        Leave the GRUB default values, and use modified flavor to >>> hw:mem_page_size=104875, launch the VM with ERROR status. >>> >>> 104875 >>> >>>> *        Leave the GRUB default values, and use modified flavor to >>> hw:mem_page_size=2048, launch the VM with success, , pgsize value is >>> 2M, >>> which is expected, ok. >>>> *        Leave the GRUB default values, and use modified flavor to >>> hw:mem_page_size=small launch the VM with success, , pgsize value is >>> 4k, >>> which is  expected, ok. >>>> *        Leave the GRUB default values, and use modified flavor to >>> hw:mem_page_size=any launch the VM with success, , pgsize value is >>> 2M, which >>> is  expected, ok. >>>> Could be this an issue trying to use 1G mem pages or do I'm missing >>>> some >>> other configuration parameter? >>>> Thanks in advance for your help >>>> >>>> >>>> _______________________________________________ >>>> Starlingx-discuss mailing list >>>> Starlingx-discuss at lists.starlingx.io >>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > -------------- next part -------------- # nova flavor-key s.f1.256 set hw:mem_page_size=4 # nova flavor-key s.f1 set hw:mem_page_size=2048 # openstack flavor show s.f1.256 | grep extra | properties | aggregate_instance_extra_specs:storage='local_image', hw:mem_page_size='4' | # openstack flavor show s.f1 | grep extra | properties | aggregate_instance_extra_specs:storage='local_image', hw:mem_page_size='2048' | # launch 2 vms with either flavor # ... # nova show shouldbe4k | grep -i "pgsize\|hyper" | OS-EXT-SRV-ATTR:hypervisor_hostname | compute-0 | | wrs-res:topology | node:0, 256MB, pgsize:4K, vcpus:0, pol:sha # nova show shouldbe2M | grep -i "pgsize\|hyper" | OS-EXT-SRV-ATTR:hypervisor_hostname | compute-0 | | wrs-res:topology | node:0, 500MB, pgsize:2M, vcpus:0, pol:sha # system host-memory-show compute-0 0 +--------------------------------+--------------------------------------+ | Property | Value | +--------------------------------+--------------------------------------+ | Memory: Usable Total (MiB) | 14190 | | Platform (MiB) | 1600 | | Available (MiB) | 13178 | | Huge Pages Configured | True | | vSwitch Huge Pages: Size (MiB) | 2 | | Total | 256 | | Available | 0 | | VM Pages (4K): Total | 1824768 | | VM Huge Pages (2M): Total | 3275 | | Available | 3025 | | VM Huge Pages (1G): Total | 0 | | Available | None | From liang.a.fang at intel.com Mon Oct 22 16:51:03 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Mon, 22 Oct 2018 16:51:03 +0000 Subject: [Starlingx-discuss] Some PR will be created for python3 support of stx-cinder/stx-glance Message-ID: Hi Just let you know, in the following days, some patches will be backported from upstream repo to stx repo(stx-cinder/stx-glance) for python3 support. So you may see some PR of stx-cinder and stx-glance. And thanks for review these. Regards Liang -------------- next part -------------- An HTML attachment was scrubbed... URL: From ricardo.o.perez at intel.com Mon Oct 22 16:54:45 2018 From: ricardo.o.perez at intel.com (Perez, Ricardo O) Date: Mon, 22 Oct 2018 16:54:45 +0000 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: References: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> <503670f0-6cfc-d277-acf1-3358bafd6519@windriver.com> Message-ID: It's working now: wrsroot at compute-0 ~(keystone_admin)]$ openstack flavor set m1.small --property hw:mem_page_size=1GB [wrsroot at compute-0 ~(keystone_admin)]$ openstack server create --image cirros --flavor m1.small --nic net-id=net3 testvm [wrsroot at compute-0 ~(keystone_admin)]$ openstack server show testvm +-------------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | Field | Value | +-------------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | wrs-res:pci_devices | | | wrs-res:topology | node:0, 1024MB, pgsize:1G, vcpus:0-3, pol:sha | | wrs-res:vcpus | [4, 4, 4] | | wrs-sg:server_group | | +-------------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ Thanks ! -Richo > -----Original Message----- > From: Michel Thebeau [mailto:michel.thebeau at windriver.com] > Sent: Monday, October 22, 2018 11:18 AM > To: Perez, Ricardo O ; starlingx- > discuss at lists.starlingx.io > Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages > > Hi Ricardo, > > I was able to select 4K and 2M pages in the following way (attached txt file).  I'll > try to see if I can get a system up with 1G pages, and I'll follow up with you. > > > > M > > > On 2018-10-22 10:59 a.m., Michel Thebeau wrote: > > Hi Ricardo, > > > > The documentation indicates to configure the the huge pages per numa > > node on the compute, and to configure the flavour extra specs for the > > VM.  I am not aware of anything else in particular. There's some tips > > and caveats in the document, but I don't see anything in your message > > to think of those. > > > > If there's any chance you are trying to do this under virtualization, > > I am not aware that that is currently supported. > > > > M > > > > > > On 2018-10-19 6:14 p.m., Perez, Ricardo O wrote: > >> Hi Michael, > >> > >> Thanks for your answer, I just forgot to put the computes information > >> about memory configuration, here is the info: > >> > >> Compute-0 > >> > >>     VM Pages > >> 0   Reserved for Platform: 8000 MiB > >> Usable Total: 86387 MiB > >> Available: 84787 MiB > >>      4K Pages: > >>      Total: 230656 > >> 2M Hugepages: > >>      Total: 1271 > >>      Available: 983 > >> 1G Hugepages: > >>      Total: 80 > >>      Available: 80 > >> > >> 1   Reserved for Platform: 2000 MiB > >> Usable Total: 93698 MiB > >> Available: 92674 MiB > >>      4K Pages: > >>      Total: 29696 > >> 2M Hugepages: > >>      Total: 5319 > >>      Available: 5319 > >> 1G Hugepages: > >>      Total: 80 > >>      Available: 80 > >> > >> *********************************** > >> Compute-1 > >> Processor > >>      Memory > >>      VM Pages > >> 0   Reserved for Platform: 8000 MiB > >> Usable Total: 86504 MiB > >> Available: 84968 MiB > >>      4K Pages: > >>      Total: 0 > >> 2M Hugepages: > >>      Total: 1780 > >>      Available: 1524 > >> 1G Hugepages: > >>      Total: 80 > >>      Available: 80 > >> > >> 1   Reserved for Platform: 2000 MiB > >> Usable Total: 93552 MiB > >> Available: 92528 MiB > >>      4K Pages: > >>      Total: 0 > >> 2M Hugepages: > >>      Total: 5304 > >>      Available: 5304 > >> 1G Hugepages: > >>      Total: 80 > >>      Available: 80 > >> > >> Am I missing something else to made use of 1G memory pages ? > >> > >> Thanks > >> -Richo > >> > >>> -----Original Message----- > >>> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] > >>> Sent: Friday, October 19, 2018 1:39 PM > >>> To: Perez, Ricardo O ; starlingx- > >>> discuss at lists.starlingx.io > >>> Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages > >>> > >>> On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: > >>>> Hello, > >>>> > >>>> I'm trying to execute a test, which his goal is to made use of huge > >>>> pages (2M / > >>> 1G) launching a VM. > >>>> I have already verified the configuration of the controllers GRUB > >>>> options and > >>> we have defined the variable hugepagesz=1G, however, inside the same > >>> parameters list, there is a "default_hugepagesz=2M". The default > >>> hugepagesz, has been modified to 1G, and the compute has been > >>> restarted with no issues. > >>> > >>> > >>> Hi Ricardo, > >>> > >>> You wrote, "controllers GRUB options" then "the compute has been > >>> restarted with no issues".  Is there a GRUB option one can configure > >>> on the controller that is for the computes?  I don't remember seeing > >>> one. If you're actually changing /etc/default/grub, I doubt that's > >>> supported. > >>> I did not find a procedure for changing default_hugepagesz  on > >>> controllers or computes grub configuration. > >>> > >>> It's supposed to be an option, for example through in horizon > >>> interface or 'system host-memory-modify'... but you didn't quote > >>> either of those. > >>> > >>> > >>>> The flavor has been modified, with the following option: " > >>> hw:mem_page_size=104875"  (1G) page. > >>> > >>> > >>> Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but > >>> not quite. > >>> > >>> > >>> M > >>> > >>> > >>>> However when I launch a VM using the modified flavor it creates the > >>>> VM, but if > >>> I check the field values of the VM using " openstack server show > >>> ". > >>> I'm getting the following value for pgsize: > >>>> wrs-res:topology                    | node:0,   512MB, pgsize:2M, > >>>> vcpus:0, pol:sha | > >>>> > >>>> So, seems like it's still taking the 2M pagesize instead of the 1G. > >>>> > >>>> What I have already tried is to: > >>>> > >>>> *        Change also the image metadata: hw:mem_page_size=large and > >>>> also > >>> 104875. Result, VM launched successfully, but pgsize value still 2M. > >>> > >>> There's the interesting "104875" again. > >>> > >>> > >>>> *        Leave the GRUB default values, and use modified flavor to > >>> hw:mem_page_size=104875, launch the VM with ERROR status. > >>> > >>> 104875 > >>> > >>>> *        Leave the GRUB default values, and use modified flavor to > >>> hw:mem_page_size=2048, launch the VM with success, , pgsize value is > >>> 2M, which is expected, ok. > >>>> *        Leave the GRUB default values, and use modified flavor to > >>> hw:mem_page_size=small launch the VM with success, , pgsize value is > >>> 4k, which is  expected, ok. > >>>> *        Leave the GRUB default values, and use modified flavor to > >>> hw:mem_page_size=any launch the VM with success, , pgsize value is > >>> 2M, which is  expected, ok. > >>>> Could be this an issue trying to use 1G mem pages or do I'm missing > >>>> some > >>> other configuration parameter? > >>>> Thanks in advance for your help > >>>> > >>>> > >>>> _______________________________________________ > >>>> Starlingx-discuss mailing list > >>>> Starlingx-discuss at lists.starlingx.io > >>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discus > >>>> s > > From michel.thebeau at windriver.com Mon Oct 22 17:09:31 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Mon, 22 Oct 2018 13:09:31 -0400 Subject: [Starlingx-discuss] Question - Using 1Gb mem pages In-Reply-To: References: <230f1557-4579-2beb-817f-2fd077e486ea@windriver.com> <503670f0-6cfc-d277-acf1-3358bafd6519@windriver.com> Message-ID: <4df54fad-b64c-03cc-9d81-e46156d834ed@windriver.com> Hi Ricardo, That's cool. I had added cpu flag 'pdpe1gb' to compute under-virtualization, otherwise as persented by the virtualization wiki, to see if it would work:         Nehalem           +               And I was able to configure the compute and launch VMs with 1G pages (per the attached example).  ... I found the memory accounting or display to be a bit wonky though.  I'm not sure what to make of that.  I'll mention it to my bosses, but I'm sure my method is not a supported. ;) M On 2018-10-22 12:54 p.m., Perez, Ricardo O wrote: > It's working now: > > wrsroot at compute-0 ~(keystone_admin)]$ openstack flavor set m1.small --property hw:mem_page_size=1GB > [wrsroot at compute-0 ~(keystone_admin)]$ openstack server create --image cirros --flavor m1.small --nic net-id=net3 testvm > [wrsroot at compute-0 ~(keystone_admin)]$ openstack server show testvm > +-------------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ > | Field | Value | > +-------------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ > | wrs-res:pci_devices | | > | wrs-res:topology | node:0, 1024MB, pgsize:1G, vcpus:0-3, pol:sha | > | wrs-res:vcpus | [4, 4, 4] | > | wrs-sg:server_group | | > +-------------------------------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ > > > Thanks ! > -Richo > >> -----Original Message----- >> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >> Sent: Monday, October 22, 2018 11:18 AM >> To: Perez, Ricardo O ; starlingx- >> discuss at lists.starlingx.io >> Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages >> >> Hi Ricardo, >> >> I was able to select 4K and 2M pages in the following way (attached txt file).  I'll >> try to see if I can get a system up with 1G pages, and I'll follow up with you. >> >> >> >> M >> >> >> On 2018-10-22 10:59 a.m., Michel Thebeau wrote: >>> Hi Ricardo, >>> >>> The documentation indicates to configure the the huge pages per numa >>> node on the compute, and to configure the flavour extra specs for the >>> VM.  I am not aware of anything else in particular. There's some tips >>> and caveats in the document, but I don't see anything in your message >>> to think of those. >>> >>> If there's any chance you are trying to do this under virtualization, >>> I am not aware that that is currently supported. >>> >>> M >>> >>> >>> On 2018-10-19 6:14 p.m., Perez, Ricardo O wrote: >>>> Hi Michael, >>>> >>>> Thanks for your answer, I just forgot to put the computes information >>>> about memory configuration, here is the info: >>>> >>>> Compute-0 >>>> >>>>     VM Pages >>>> 0   Reserved for Platform: 8000 MiB >>>> Usable Total: 86387 MiB >>>> Available: 84787 MiB >>>>      4K Pages: >>>>      Total: 230656 >>>> 2M Hugepages: >>>>      Total: 1271 >>>>      Available: 983 >>>> 1G Hugepages: >>>>      Total: 80 >>>>      Available: 80 >>>> >>>> 1   Reserved for Platform: 2000 MiB >>>> Usable Total: 93698 MiB >>>> Available: 92674 MiB >>>>      4K Pages: >>>>      Total: 29696 >>>> 2M Hugepages: >>>>      Total: 5319 >>>>      Available: 5319 >>>> 1G Hugepages: >>>>      Total: 80 >>>>      Available: 80 >>>> >>>> *********************************** >>>> Compute-1 >>>> Processor >>>>      Memory >>>>      VM Pages >>>> 0   Reserved for Platform: 8000 MiB >>>> Usable Total: 86504 MiB >>>> Available: 84968 MiB >>>>      4K Pages: >>>>      Total: 0 >>>> 2M Hugepages: >>>>      Total: 1780 >>>>      Available: 1524 >>>> 1G Hugepages: >>>>      Total: 80 >>>>      Available: 80 >>>> >>>> 1   Reserved for Platform: 2000 MiB >>>> Usable Total: 93552 MiB >>>> Available: 92528 MiB >>>>      4K Pages: >>>>      Total: 0 >>>> 2M Hugepages: >>>>      Total: 5304 >>>>      Available: 5304 >>>> 1G Hugepages: >>>>      Total: 80 >>>>      Available: 80 >>>> >>>> Am I missing something else to made use of 1G memory pages ? >>>> >>>> Thanks >>>> -Richo >>>> >>>>> -----Original Message----- >>>>> From: Michel Thebeau [mailto:michel.thebeau at windriver.com] >>>>> Sent: Friday, October 19, 2018 1:39 PM >>>>> To: Perez, Ricardo O ; starlingx- >>>>> discuss at lists.starlingx.io >>>>> Subject: Re: [Starlingx-discuss] Question - Using 1Gb mem pages >>>>> >>>>> On 2018-10-19 1:45 p.m., Perez, Ricardo O wrote: >>>>>> Hello, >>>>>> >>>>>> I'm trying to execute a test, which his goal is to made use of huge >>>>>> pages (2M / >>>>> 1G) launching a VM. >>>>>> I have already verified the configuration of the controllers GRUB >>>>>> options and >>>>> we have defined the variable hugepagesz=1G, however, inside the same >>>>> parameters list, there is a "default_hugepagesz=2M". The default >>>>> hugepagesz, has been modified to 1G, and the compute has been >>>>> restarted with no issues. >>>>> >>>>> >>>>> Hi Ricardo, >>>>> >>>>> You wrote, "controllers GRUB options" then "the compute has been >>>>> restarted with no issues".  Is there a GRUB option one can configure >>>>> on the controller that is for the computes?  I don't remember seeing >>>>> one. If you're actually changing /etc/default/grub, I doubt that's >>>>> supported. >>>>> I did not find a procedure for changing default_hugepagesz  on >>>>> controllers or computes grub configuration. >>>>> >>>>> It's supposed to be an option, for example through in horizon >>>>> interface or 'system host-memory-modify'... but you didn't quote >>>>> either of those. >>>>> >>>>> >>>>>> The flavor has been modified, with the following option: " >>>>> hw:mem_page_size=104875"  (1G) page. >>>>> >>>>> >>>>> Is "104875" a typo?  It looks a lot like 1024 * 1024 = 1048576, but >>>>> not quite. >>>>> >>>>> >>>>> M >>>>> >>>>> >>>>>> However when I launch a VM using the modified flavor it creates the >>>>>> VM, but if >>>>> I check the field values of the VM using " openstack server show >>>>> ". >>>>> I'm getting the following value for pgsize: >>>>>> wrs-res:topology                    | node:0,   512MB, pgsize:2M, >>>>>> vcpus:0, pol:sha | >>>>>> >>>>>> So, seems like it's still taking the 2M pagesize instead of the 1G. >>>>>> >>>>>> What I have already tried is to: >>>>>> >>>>>> *        Change also the image metadata: hw:mem_page_size=large and >>>>>> also >>>>> 104875. Result, VM launched successfully, but pgsize value still 2M. >>>>> >>>>> There's the interesting "104875" again. >>>>> >>>>> >>>>>> *        Leave the GRUB default values, and use modified flavor to >>>>> hw:mem_page_size=104875, launch the VM with ERROR status. >>>>> >>>>> 104875 >>>>> >>>>>> *        Leave the GRUB default values, and use modified flavor to >>>>> hw:mem_page_size=2048, launch the VM with success, , pgsize value is >>>>> 2M, which is expected, ok. >>>>>> *        Leave the GRUB default values, and use modified flavor to >>>>> hw:mem_page_size=small launch the VM with success, , pgsize value is >>>>> 4k, which is  expected, ok. >>>>>> *        Leave the GRUB default values, and use modified flavor to >>>>> hw:mem_page_size=any launch the VM with success, , pgsize value is >>>>> 2M, which is  expected, ok. >>>>>> Could be this an issue trying to use 1G mem pages or do I'm missing >>>>>> some >>>>> other configuration parameter? >>>>>> Thanks in advance for your help >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> Starlingx-discuss mailing list >>>>>> Starlingx-discuss at lists.starlingx.io >>>>>> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discus >>>>>> s -------------- next part -------------- # nova flavor-key s.f1 set hw:mem_page_size=2048 # nova flavor-key s.f1.1G set hw:mem_page_size=1048576 # openstack flavor show s.f1 | grep extra | properties | aggregate_instance_extra_specs:storage='local_image', hw:mem_page_size='2048' | # nova flavor-show s.f1.1G | grep "extra" | extra_specs | {"hw:mem_page_size": "1048576", "aggregate_instance_extra_specs:storage": "local_image"} | # system host-memory-modify -m 2048 -2M 2048 -1G 4 compute-1 0 # system host-memory-show compute-1 -0 +--------------------------------+--------------------------------------+ | Property | Value | +--------------------------------+--------------------------------------+ | Memory: Usable Total (MiB) | 13747 | | Platform (MiB) | 2048 | | Available (MiB) | 12723 | | Huge Pages Configured | True | | vSwitch Huge Pages: Size (MiB) | 1024 | | Total | 1 | | Available | 0 | | VM Pages (4K): Total | 1158656 | | VM Huge Pages (2M): Total | 2048 | | Total Pending | 2048 | | Available | 1536 | | VM Huge Pages (1G): Total | 4 | | Total Pending | 4 | | Available | 5 | # fm alarm-list | grep compute-0 # launch 2 vms with either flavor # ... # nova show compute1-2M | grep -i "pgsize\|hyper" | OS-EXT-SRV-ATTR:hypervisor_hostname | compute-1 | | wrs-res:topology | node:0, 500MB, pgsize:2M, vcpus:0, unallocated, pol:sha # nova show shouldbe1G | grep -i "pgsize\|hyper" | OS-EXT-SRV-ATTR:hypervisor_hostname | compute-1 | | wrs-res:topology | node:0, 1024MB, pgsize:1G, vcpus:0, pol:sha # system host-memory-show compute-1 0 +--------------------------------+--------------------------------------+ | Property | Value | +--------------------------------+--------------------------------------+ | Memory: Usable Total (MiB) | 13742 | | Platform (MiB) | 2048 | | Available (MiB) | 11694 | | Huge Pages Configured | True | | vSwitch Huge Pages: Size (MiB) | 1024 | | Total | 1 | | Available | 0 | | VM Pages (4K): Total | 1158656 | | VM Huge Pages (2M): Total | 2048 | | Available | 1536 | | VM Huge Pages (1G): Total | 4 | | Available | 4 | #### # test memory accounting/display # # destroy the other VMs and start two more, both with flavor s.f1.1G # nova show asdf | grep -i "pgsize\|hyper" | OS-EXT-SRV-ATTR:hypervisor_hostname | compute-1 | | wrs-res:topology | node:0, 1024MB, pgsize:1G, vcpus:0, pol:sha # nova show qwer | grep -i "pgsize\|hyper" | OS-EXT-SRV-ATTR:hypervisor_hostname | compute-1 | | wrs-res:topology | node:0, 1024MB, pgsize:1G, vcpus:0, pol:sha # system host-memory-show compute-1 0 +--------------------------------+--------------------------------------+ | Property | Value | +--------------------------------+--------------------------------------+ | Memory: Usable Total (MiB) | 13742 | | Platform (MiB) | 2048 | | Available (MiB) | 10670 | | Huge Pages Configured | True | | vSwitch Huge Pages: Size (MiB) | 1024 | | Total | 1 | | Available | 0 | | VM Pages (4K): Total | 1158656 | | VM Huge Pages (2M): Total | 2048 | | Available | 1536 | | VM Huge Pages (1G): Total | 4 | | Available | 3 | From cesar.lara at intel.com Mon Oct 22 20:21:16 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Mon, 22 Oct 2018 20:21:16 +0000 Subject: [Starlingx-discuss] [build][meetings] build team meeting minutes 10/18/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471053ADA2@fmsmsx104.amr.corp.intel.com> Build team meeting 10/17/2018 Attendees Jason, Scott, Saul, Abraham , Memo, Hayde, Marcela, Mario, Chuy, Felipe, Erich, Cesar Agenda - Cengn status for external mirror - Use cases review/discussion - Opens Cengn Status - Waiting for access to bare metal server, once we have access we will start populating the mirror. Use case review/discussion - We will start with mail sent to ML by Jason[1], this will address most use cases coming from WR developers that have been working on StarlingX the most, Mexico team will meet to find a way to implement these use cases in code to feed the Stxb tool, this proposed mapping will be presented in a follow up session. The intention is to start coding necessary pieces in python, but since there is a concern about too many moving pieces in the build scripts , a full analysis is required to avoid losing functionality, that might lead to keep some of the bash scripting as part of the build tools. Opens - New steps required for build process to generate new kubernetes files and configuration, We will keep an eye on this requirement as is not crystal clear today, but we will need to add this in the near future. - Mexico team refactoring download script on python - probably the best one to start with, we might have a couple of efficiencies on this effort. -As part of multiOS effort we are setting up a Koji instance to work with Clear Linux - there are a complete set of extra instructions to make Koji work as a provider for rpms and ISO files, this will require further discussion, next week we will have Victor Rodriguez which is building up Koji for Clear Linux to give us a walk thru of the system and necessary steps to build the OS. [1] https://wiki.openstack.org/wiki/StarlingX/Build#Use_Cases_for_Stxb Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From cesar.lara at intel.com Mon Oct 22 21:19:31 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Mon, 22 Oct 2018 21:19:31 +0000 Subject: [Starlingx-discuss] [build]Build report ww42 Oct 14 - Oct 21 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471053AF43@fmsmsx104.amr.corp.intel.com> Build report Oct 14 - Oct 21 Intel build Release Branch 9 successful builds 0 Failed builds Master Branch 3 successful builds 1 Failed, expected error solved, config issues Changes - >From repo: stx-config.git ------------------------------------------------------------------------------------------------------------------------------------------------------- DATE COMMIT MESSAGE AUTHOR ------------------------------------------------------------------------------------------------------------------------------------------------------- 2018-10-15 13:58:36 -0400 a7cdb4c OVS: fix memory pool allocation for virtual environment steven.webster at windriver.com 2018-10-17 13:41:00 -0400 fb3f45d Set neutron dhcp host agent scheduler to host-based teresa.ho at windriver.com 2018-10-19 17:29:11 +0000 07d89cb Merge "OVS: fix memory pool allocation for virtual environment" into r/2018.10 zuul at review.openstack.org ------------------------------------------------------------------------------------------------------------------------------------------------------- Mirror Update [mirror-downloader][#115] Missing packages. Results report: - Missing: logs/L1_srpms_missing_centos3rdparties.log kernel-rt-3.10.0-862.11.6.rt56.819.el7.src.rpm python2-pysocks-1.6.7-1.el7.noarch.rpm - Missing GPG key ./output/stx-r1/CentOS/pike/Source/libvirt-python-4.7.0-1.fc28.src.rpm: RSA sha1 ((MD5) PGP) md5 NOT OK (MISSING KEYS: (MD5) PGP#596bea5d) The team is already taking care of missing packages and manifest update. Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtroyer at gmail.com Mon Oct 22 21:59:58 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 22 Oct 2018 16:59:58 -0500 Subject: [Starlingx-discuss] Some PR will be created for python3 support of stx-cinder/stx-glance In-Reply-To: References: Message-ID: On Mon, Oct 22, 2018 at 11:51 AM Fang, Liang A wrote: > Just let you know, in the following days, some patches will be backported from upstream repo to stx repo(stx-cinder/stx-glance) for python3 support. So you may see some PR of stx-cinder and stx-glance. And thanks for review these. Thanks for the heads-up for those. Please make sure the commit message includes pointers back to the upstream commit that is being backported in addition to the original commit message. I would suggest we also backport the python 3 testing from master or rocky if it not already in pike to ensure at least unit tests are executed. At the moment this is a manual task. dt -- Dean Troyer dtroyer at gmail.com From ada.cabrales at intel.com Tue Oct 23 01:48:47 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Tue, 23 Oct 2018 01:48:47 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-22-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C356525@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-22-32-r-2018.10.iso 31 TCs PASS, 0 TCs FAIL. - "Error on system restore attempt" [0] created today, happened on duplex bare metal configuration. However this was not reproduced in other configurations. Will go into deeper debug. - "Cannot launch an instance with local storage" [1] seen in multi node with local storage, both bare metal and virtual. Assigned for debugging. - Execution complete on Simplex Bare metal, Duplex bare metal and virtual, multinode local storage bare metal and virtual, multinode external storage bare metal. - Execution on Simplex virtual began today. - Testing report for this release WIP (failures, verification of those, decisions made, etc). - Execution progress (Fail/Pass/Pending/Total) Simplex bare metal - 2/4/0/76 - done Simplex virtual - 1/11/55/67 Duplex bare metal - 5/93/0/98 - done Duplex virtual - 3/100/0/103 - done Multinode local storage bare metal - 5/113/0/118 - done Multinode local storage virtual - 11/115/0/126 - done Multinode external storage bare metal - 3/112/0/115 - done Multinode external storage virtual - 6/107/5/118 Full report in the wiki [2] [0] https://bugs.launchpad.net/starlingx/+bug/1799271 [1] https://bugs.launchpad.net/starlingx/+bug/1799312 [2] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary From cindy.xie at intel.com Tue Oct 23 07:33:51 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Tue, 23 Oct 2018 07:33:51 +0000 Subject: [Starlingx-discuss] Agenda: Weekly StarlingX non-OpenStack Distro meeting, 10/24 Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35D8E14E@SHSMSX104.ccr.corp.intel.com> Please let me know if any specific topics you'd like to discuss on 10/24. Below are the agendas I have: Agenda for 10/24 meeting: - bug review for non-openstack distro (Cindy) - non-Openstack patch refactor status (Zhipeng) - non-Openstack patch upstream status (Saul) - Ceph upgrade status (Vivian/Dehao) - Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; 'Ildiko Vancsa'; Wold, Saul; Chen, Yan; Somerville, Jim; Wang, Yi C; Khalil, Ghada; starlingx-discuss at lists.starlingx.io; Sun, Austin; Rowsell, Brent; Lin, Shuicheng Cc: Hu, Yong; Martinez Monroy, Elio; 'Leo Xu'; 'Young, Ken'; Gomez, Juan P; Shang, Dehao; 'Eslimi, Dariush'; Qi, Mingyuan; Hernandez Gonzalez, Fernando; 'Waines, Greg'; Perez Rodriguez, Humberto I; Zhu, Vivian; Arce Moreno, Abraham; 'Chen, Jacky'; Jones, Bruce E; Cobbley, David A; 'Komiyama, Takeo'; Hu, Wei W; 'Seiler, Glenn' Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 24, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other From huifeng.le at intel.com Tue Oct 23 07:40:08 2018 From: huifeng.le at intel.com (Le, Huifeng) Date: Tue, 23 Oct 2018 07:40:08 +0000 Subject: [Starlingx-discuss] Questions about patch upstreaming for e490b5a and 8e7249 Message-ID: <76647BD697F40748B1FA4F56DA02AA0B4D52B5D6@SHSMSX104.ccr.corp.intel.com> Hi Matt, We're looking into the below patches, could you please help to clarify below questions? Thanks much! * e490b5a: CGTS-3239: add process synchronization around segment allocation >From the comments "in upstream they attempted to address this by changing the segmentation allocation to be random so that the probability of 2 processes accessing the same record is lessened but it does not completely eliminate the possibility. In our environment we allocate sequentially so the likelyhood of this occurring is even greater. We allocate sequentially to facilitate predictable network setup order." Do you think this issue still need to be addressed when STX adopt upstream type driver since (1) the possibility is low and (2) the patch introduce a global lock which may have side effect on performance? * 8e7249: GTS-6926: wsgi: prevent accepting socket without a greenthread >From the comments "This is really a bug (IMHO) in the eventlet code but to test the waters and make sure this works properly I am pushing this to neutron first before deciding whether to make the same/equivalent change at a lower level. Neutron is planning on moving away from the eventlet.wsgi.server so hopefully in the future the API will be run by a proper webserver which won't be susceptible to this type of failure." By reviewing the code, I think Allain is correct and it make more sense to fix this issue in low level code (maybe in eventlet.wsgi.Server, check the status for available working threads then determine whether to process the request or return 'server busy' error directly; by applying the patch in neutron, the client may get "server unavailable" error). And since eventlet.wsgi.server is planned to be removed by neutron, do you think hold on this fix for now make sense? Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From chenjie.xu at intel.com Tue Oct 23 08:19:11 2018 From: chenjie.xu at intel.com (Xu, Chenjie) Date: Tue, 23 Oct 2018 08:19:11 +0000 Subject: [Starlingx-discuss] Request use cases for RFE(patch 9f926a5) for StartlingX upstreaming Message-ID: Hi Matt, For patch 9f926a5, the RFE "Add l2pop support for floating ip resources" has been drafted and is under internal reviewing for now. This RFE adds (host, mac_address, floating_ip_address) to l2pop fdb. 1. For non-DVR floating ip resources, the mac_address belongs to router_gateway port which belongs to the router related to the specific floating ip. As following: mac_address: router_gateway port's mac_address --------------------- router -------- floating ip 2. For DVR floating ip resources, the mac_address belongs to floatingip_agent_gateway port which belongs to the router related to the specific floating ip. As following: mac_address: floatingip_agent_gateway port's mac_address -------- router -------- floating ip I'm wondering in what use case these information will be used to avoid the broadcast in mac learning and ARP resolution . Could you please provide some use cases? Best Regards, Xu, Chenjie -------------- next part -------------- An HTML attachment was scrubbed... URL: From liang.a.fang at intel.com Tue Oct 23 08:30:12 2018 From: liang.a.fang at intel.com (Fang, Liang A) Date: Tue, 23 Oct 2018 08:30:12 +0000 Subject: [Starlingx-discuss] Some PR will be created for python3 support of stx-cinder/stx-glance In-Reply-To: References: Message-ID: Good tips, thanks Dean. Regards Liang -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Tuesday, October 23, 2018 6:00 AM To: Fang, Liang A Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Some PR will be created for python3 support of stx-cinder/stx-glance On Mon, Oct 22, 2018 at 11:51 AM Fang, Liang A wrote: > Just let you know, in the following days, some patches will be backported from upstream repo to stx repo(stx-cinder/stx-glance) for python3 support. So you may see some PR of stx-cinder and stx-glance. And thanks for review these. Thanks for the heads-up for those. Please make sure the commit message includes pointers back to the upstream commit that is being backported in addition to the original commit message. I would suggest we also backport the python 3 testing from master or rocky if it not already in pike to ensure at least unit tests are executed. At the moment this is a manual task. dt -- Dean Troyer dtroyer at gmail.com From Allain.Legacy at windriver.com Tue Oct 23 13:17:26 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Tue, 23 Oct 2018 13:17:26 +0000 Subject: [Starlingx-discuss] Request use cases for RFE(patch 9f926a5) for StartlingX upstreaming In-Reply-To: References: Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC16390@ALA-MBD.corp.ad.wrs.com> The primary reason for this patch is to be able to export the FIP address information via BGP eVPN. Our BGP eVPN extensions make use of the L2POP framework to export MAC/IP information and this patch ensures that the FIP addresses are included in the exported information. In addition to the BGP eVPN functionality, this patch also avoids broadcast for scenarios when two VM instances, residing on different networks, communicate via their respective FIP addresses. Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Xu, Chenjie [mailto:chenjie.xu at intel.com] Sent: Tuesday, October 23, 2018 4:19 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Request use cases for RFE(patch 9f926a5) for StartlingX upstreaming Hi Matt, For patch 9f926a5, the RFE "Add l2pop support for floating ip resources" has been drafted and is under internal reviewing for now. This RFE adds (host, mac_address, floating_ip_address) to l2pop fdb. 1. For non-DVR floating ip resources, the mac_address belongs to router_gateway port which belongs to the router related to the specific floating ip. As following: mac_address: router_gateway port's mac_address --------------------- router -------- floating ip 2. For DVR floating ip resources, the mac_address belongs to floatingip_agent_gateway port which belongs to the router related to the specific floating ip. As following: mac_address: floatingip_agent_gateway port's mac_address -------- router -------- floating ip I'm wondering in what use case these information will be used to avoid the broadcast in mac learning and ARP resolution . Could you please provide some use cases? Best Regards, Xu, Chenjie -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From Allain.Legacy at windriver.com Tue Oct 23 13:29:00 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Tue, 23 Oct 2018 13:29:00 +0000 Subject: [Starlingx-discuss] Questions about patch upstreaming for e490b5a and 8e7249 In-Reply-To: <76647BD697F40748B1FA4F56DA02AA0B4D52B5D6@SHSMSX104.ccr.corp.intel.com> References: <76647BD697F40748B1FA4F56DA02AA0B4D52B5D6@SHSMSX104.ccr.corp.intel.com> Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC163B2@ALA-MBD.corp.ad.wrs.com> [WIND] From: Le, Huifeng [mailto:huifeng.le at intel.com] Sent: Tuesday, October 23, 2018 3:40 AM To: Peters, Matt Cc: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Questions about patch upstreaming for e490b5a and 8e7249 Hi Matt, We're looking into the below patches, could you please help to clarify below questions? Thanks much! * e490b5a: CGTS-3239: add process synchronization around segment allocation >From the comments "in upstream they attempted to address this by changing the segmentation allocation to be random so that the probability of 2 processes accessing the same record is lessened but it does not completely eliminate the possibility. In our environment we allocate sequentially so the likelyhood of this occurring is even greater. We allocate sequentially to facilitate predictable network setup order." Do you think this issue still need to be addressed when STX adopt upstream type driver since (1) the possibility is low and (2) the patch introduce a global lock which may have side effect on performance? [AL] My opinion is that the upstream method of avoiding the issue is not acceptable and a solution is required. Any code that relies on the low likelihood of a failure occurring rather than coding a solution that works 100% of the time needs to be fixed. * 8e7249: GTS-6926: wsgi: prevent accepting socket without a greenthread >From the comments "This is really a bug (IMHO) in the eventlet code but to test the waters and make sure this works properly I am pushing this to neutron first before deciding whether to make the same/equivalent change at a lower level. Neutron is planning on moving away from the eventlet.wsgi.server so hopefully in the future the API will be run by a proper webserver which won't be susceptible to this type of failure." By reviewing the code, I think Allain is correct and it make more sense to fix this issue in low level code (maybe in eventlet.wsgi.Server, check the status for available working threads then determine whether to process the request or return 'server busy' error directly; by applying the patch in neutron, the client may get "server unavailable" error). And since eventlet.wsgi.server is planned to be removed by neutron, do you think hold on this fix for now make sense? [AL] Until Neutron adopts a strategy that uses something other than eventlet.wsgi.server we need to apply this patch to avoid the issue that it solves. We should attempt to upstream this fix to Neutron to highlight that this issue exists. Thank much! Best Regards, Le, Huifeng -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From scott.little at windriver.com Tue Oct 23 13:46:08 2018 From: scott.little at windriver.com (Scott Little) Date: Tue, 23 Oct 2018 09:46:08 -0400 Subject: [Starlingx-discuss] CENGN mirror spec Message-ID: <06d062bc-f0ab-97f9-ccc6-0d6570c1a23b@windriver.com> StarlingX Community: A spec for the CENGN package mirror has been posted for review [1].  Feedback from interested community members would be welcome and appreciated. [1]: https://review.openstack.org/#/c/612480/ Scott Little -------------- next part -------------- An HTML attachment was scrubbed... URL: From ildiko.vancsa at gmail.com Tue Oct 23 16:45:40 2018 From: ildiko.vancsa at gmail.com (Ildiko Vancsa) Date: Tue, 23 Oct 2018 18:45:40 +0200 Subject: [Starlingx-discuss] Reminder, October 24 StarlingX On-Boarding Webinar Message-ID: Hi, The first release of StarlingX is quickly approaching! Thank you to everyone who has spent time working on this first release and helping shape our community; it’s been great work to be a part of and I’m excited to see this community progress. To spread the word about StarlingX we have the project featured on the next occasion of the OSF Community Meeting series on October 24, 8am PDT / 1500 UTC. Please join the call to get an overview about the project, learn more about the technology and ways to get involved with the community. We also invite you to follow @StarlingX on Twitter for announcements and news moving forward. Thanks, Ildikó -------------- next part -------------- A non-text attachment was scrubbed... Name: Mail Attachment.ics Type: text/calendar Size: 3160 bytes Desc: not available URL: -------------- next part -------------- From David.Sullivan at windriver.com Tue Oct 23 21:28:54 2018 From: David.Sullivan at windriver.com (Sullivan, David) Date: Tue, 23 Oct 2018 21:28:54 +0000 Subject: [Starlingx-discuss] question about system upgrade In-Reply-To: <9700A18779F35F49AF027300A49E7C765FE12EE8@SHSMSX101.ccr.corp.intel.com> References: <9700A18779F35F49AF027300A49E7C765FE12EE8@SHSMSX101.ccr.corp.intel.com> Message-ID: Hi Shuicheng, Which load are you using to import and what are you trying to test? Upgrades are for one release to the next so typically we'd need a second release to test. David From: Lin, Shuicheng [mailto:shuicheng.lin at intel.com] Sent: Thursday, October 18, 2018 3:05 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] question about system upgrade Hi all, I try to execute "system load-import" in controller node, and it shows "Metadata file not found". And from code, it is due to there is no "/upgrades/metadata.xml" in the ISO. Anybody know how to generate ISO with this file? Thanks. Best Regards Shuicheng -------------- next part -------------- An HTML attachment was scrubbed... URL: From Rajesh.Jha at rsystems.com Tue Oct 23 06:40:35 2018 From: Rajesh.Jha at rsystems.com (Rajesh Jha) Date: Tue, 23 Oct 2018 12:10:35 +0530 Subject: [Starlingx-discuss] Updation of wiki link for Build Message-ID: <2F61A2101B4FD447A683EF7E362CF9421C63E217DE@RSI-NOD-EXMDB6.india.rsystems.com> Hi, This is just to inform the community that, there are two items which are missing from the wiki build pages which is essential for making starlingX iso on Ubuntu 16.04 host system. Under section Installation Requirements and Dependencies on https://wiki.openstack.org/wiki/StarlingX/Developer_Guide url create a user $ useradd -m -d /home/builder builder $ sudo sh -c "echo 'builder ALL=(ALL:ALL) ALL' >> /etc/sudoers" under section Docker CE, after doker installation is done. Add the following $ groupadd -aG docker builder Best Regards, Rajesh Jha The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. From michael.l.tullis at intel.com Tue Oct 23 23:08:13 2018 From: michael.l.tullis at intel.com (Tullis, Michael L) Date: Tue, 23 Oct 2018 23:08:13 +0000 Subject: [Starlingx-discuss] Updation of wiki link for Build In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217DE@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217DE@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: <3808363B39586544A6839C76CF81445EA1A42B49@FMSMSX151.amr.corp.intel.com> Note: The latest versions of the developer guide and installation guide are now on https://docs.starlingx.io/. That should be considered the master going forward for contributions. A few changes are pending: https://review.openstack.org/#/c/611978/ -- Mike -----Original Message----- From: Rajesh Jha [mailto:Rajesh.Jha at rsystems.com] Sent: Tuesday, October 23, 2018 12:41 AM To: starlingx-discuss at lists.starlingx.io Cc: Avanish Yadav ; Sumit Agarwal ; Paras Agarwal ; Sachin Alok ; NK Garg Subject: [Starlingx-discuss] Updation of wiki link for Build Hi, This is just to inform the community that, there are two items which are missing from the wiki build pages which is essential for making starlingX iso on Ubuntu 16.04 host system. Under section Installation Requirements and Dependencies on https://wiki.openstack.org/wiki/StarlingX/Developer_Guide url create a user $ useradd -m -d /home/builder builder $ sudo sh -c "echo 'builder ALL=(ALL:ALL) ALL' >> /etc/sudoers" under section Docker CE, after doker installation is done. Add the following $ groupadd -aG docker builder Best Regards, Rajesh Jha The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Rajesh.Jha at rsystems.com Tue Oct 23 06:41:56 2018 From: Rajesh.Jha at rsystems.com (Rajesh Jha) Date: Tue, 23 Oct 2018 12:11:56 +0530 Subject: [Starlingx-discuss] make file updation Message-ID: <2F61A2101B4FD447A683EF7E362CF9421C63E217DF@RSI-NOD-EXMDB6.india.rsystems.com> Hi, This is an observation which we made while doing the build for an iso image for StarlingX, which we think can be looked into and improved upon. The observation is given below. $ make build can we look at the make file and ensure that it doesn't look for components which are not yet part of official StarlingX release. Best Regards, Rajesh Jha The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. From bruce.e.jones at intel.com Tue Oct 23 23:45:21 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Tue, 23 Oct 2018 23:45:21 +0000 Subject: [Starlingx-discuss] Updation of wiki link for Build In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217DE@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217DE@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: <9A85D2917C58154C960D95352B22818BB1E82EFF@ORSMSX158.amr.corp.intel.com> Thank you! This has been filed as defect https://bugs.launchpad.net/starlingx/+bug/1799611 for the Docs team to address. brucej -----Original Message----- From: Rajesh Jha [mailto:Rajesh.Jha at rsystems.com] Sent: Monday, October 22, 2018 11:41 PM To: starlingx-discuss at lists.starlingx.io Cc: Avanish Yadav ; Sumit Agarwal ; Paras Agarwal ; Sachin Alok ; NK Garg Subject: [Starlingx-discuss] Updation of wiki link for Build Hi, This is just to inform the community that, there are two items which are missing from the wiki build pages which is essential for making starlingX iso on Ubuntu 16.04 host system. Under section Installation Requirements and Dependencies on https://wiki.openstack.org/wiki/StarlingX/Developer_Guide url create a user $ useradd -m -d /home/builder builder $ sudo sh -c "echo 'builder ALL=(ALL:ALL) ALL' >> /etc/sudoers" under section Docker CE, after doker installation is done. Add the following $ groupadd -aG docker builder Best Regards, Rajesh Jha The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From ada.cabrales at intel.com Wed Oct 24 00:17:03 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Wed, 24 Oct 2018 00:17:03 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-23-18 Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C356EAA@fmsmsx104.amr.corp.intel.com> Sanity on stx-2018-10-23-33-r-2018.10.iso 31 TCs PASS, 0 TCs FAIL. 4 Launchpads created: - "hw:mem_page_size=1GB" Metadata can not be assigned to a flavors with cli commands" [0] - seems to be a configuration issue, checking - "1G hugepages for memory allocation is not enable in Virtual external storage config" [1] - seems to be a configuration issue, checking - "Process 'neutron-l3-agent' does not work" [2] - "$ ip netns does not display anything" [3] - The only config not finished is Simplex virtual. We estimate can be finished by Monday next week. - working on verifying some failures we have, that might be false positives. - Execution progress (Fail/Pass/Pending/Total) Simplex virtual - 1/29/37/67 Multinode external storage virtual - 8/110/0/118 - done Full report in the wiki [4] [0] https://bugs.launchpad.net/starlingx/+bug/1799517 [1] https://bugs.launchpad.net/starlingx/+bug/1799539 [2] https://bugs.launchpad.net/starlingx/+bug/1799590 [3] https://bugs.launchpad.net/starlingx/+bug/1799591 [4] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary From juan.carlos.alonso at intel.com Wed Oct 24 00:51:37 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Wed, 24 Oct 2018 00:51:37 +0000 Subject: [Starlingx-discuss] Ping from controller to an instance Message-ID: <8557B550001AFB46A43A0CCC314BF85153C43802@fmsmsx101.amr.corp.intel.com> Hi, Is it possible to ping or ssh to an instance from an active controller via namespaces? Or should be done only from a compute? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From abraham.arce.moreno at intel.com Wed Oct 24 02:35:18 2018 From: abraham.arce.moreno at intel.com (Arce Moreno, Abraham) Date: Wed, 24 Oct 2018 02:35:18 +0000 Subject: [Starlingx-discuss] Updation of wiki link for Build In-Reply-To: <9A85D2917C58154C960D95352B22818BB1E82EFF@ORSMSX158.amr.corp.intel.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217DE@RSI-NOD-EXMDB6.india.rsystems.com> <9A85D2917C58154C960D95352B22818BB1E82EFF@ORSMSX158.amr.corp.intel.com> Message-ID: Thanks Rajesh, Mike, Bruce! > This is just to inform the community that, there are two items which are > missing from the wiki build pages which is essential for making starlingX iso > on Ubuntu 16.04 host system. > This has been filed as defect > https://bugs.launchpad.net/starlingx/+bug/1799611 for the Docs team to > address. > Note: The latest versions of the developer guide and installation guide are > now on https://docs.starlingx.io/. That should be considered the master > going forward for contributions. A few changes are pending: > https://review.openstack.org/#/c/611978/ We are incorporating the proposal of fix for the missing items you have identified under this review: https://review.openstack.org/#/c/612826 From Allain.Legacy at windriver.com Wed Oct 24 11:20:59 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Wed, 24 Oct 2018 11:20:59 +0000 Subject: [Starlingx-discuss] Ping from controller to an instance In-Reply-To: <8557B550001AFB46A43A0CCC314BF85153C43802@fmsmsx101.amr.corp.intel.com> References: <8557B550001AFB46A43A0CCC314BF85153C43802@fmsmsx101.amr.corp.intel.com> Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC171DB@ALA-MBD.corp.ad.wrs.com> This question was raised in a Launchpad bug. Namespaces should not be used for end to end testing. Please see the explanation in https://bugs.launchpad.net/starlingx/+bug/1797217. Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Alonso, Juan Carlos [mailto:juan.carlos.alonso at intel.com] Sent: Tuesday, October 23, 2018 8:52 PM To: 'starlingx-discuss at lists.starlingx.io' Subject: [Starlingx-discuss] Ping from controller to an instance Hi, Is it possible to ping or ssh to an instance from an active controller via namespaces? Or should be done only from a compute? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From Ghada.Khalil at windriver.com Wed Oct 24 11:58:56 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Wed, 24 Oct 2018 11:58:56 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-23-18 In-Reply-To: <4F6AACE4B0F173488D033B02A8BB5B7E7C356EAA@fmsmsx104.amr.corp.intel.com> References: <4F6AACE4B0F173488D033B02A8BB5B7E7C356EAA@fmsmsx104.amr.corp.intel.com> Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46F1EB@ALA-MBD.corp.ad.wrs.com> Hi Ada, I've put comments in the 4 new bugs reported below. None of them are gating stx.2018.10 (3 are invalid, 1 is lower priority). See Launchpad for more details. Regards, Ghada -----Original Message----- From: Cabrales, Ada [mailto:ada.cabrales at intel.com] Sent: Tuesday, October 23, 2018 8:17 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] [ Test ] October release status - 10-23-18 Sanity on stx-2018-10-23-33-r-2018.10.iso 31 TCs PASS, 0 TCs FAIL. 4 Launchpads created: - "hw:mem_page_size=1GB" Metadata can not be assigned to a flavors with cli commands" [0] - seems to be a configuration issue, checking - "1G hugepages for memory allocation is not enable in Virtual external storage config" [1] - seems to be a configuration issue, checking - "Process 'neutron-l3-agent' does not work" [2] - "$ ip netns does not display anything" [3] - The only config not finished is Simplex virtual. We estimate can be finished by Monday next week. - working on verifying some failures we have, that might be false positives. - Execution progress (Fail/Pass/Pending/Total) Simplex virtual - 1/29/37/67 Multinode external storage virtual - 8/110/0/118 - done Full report in the wiki [4] [0] https://bugs.launchpad.net/starlingx/+bug/1799517 [1] https://bugs.launchpad.net/starlingx/+bug/1799539 [2] https://bugs.launchpad.net/starlingx/+bug/1799590 [3] https://bugs.launchpad.net/starlingx/+bug/1799591 [4] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Wed Oct 24 13:48:54 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Wed, 24 Oct 2018 13:48:54 +0000 Subject: [Starlingx-discuss] Minites: Weekly StarlingX non-OpenStack Distro meeting, 10/24 Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35DA4922@SHSMSX104.ccr.corp.intel.com> Agenda & Notes for 10/24 meeting: - bug review for non-openstack distro (Cindy) Lshell removal issue: Shuicheng to contact Numan Waheed (numan.waheed at windriver.com) - WR PV test prime - to get a list of regression test-cases related to ldap users to execute. This is required before code submission. process for the bug assigned to WR (Dariush Eslimi), please get the patch owner answer the questions and assign back to Cindy so that we can continue working on that. - non-Openstack patch refactor status (Zhipeng) Master story: https://storyboard.openstack.org/#!/story/2003768 23 tasks has been creatd and spec patch has been submitted. End of Nov is the tentative target date. around 4 sRPM can be replaced by RPM. ~50 patches can be removed - non-Openstack patch upstream status (Saul) Saul working on spec reviews and no update on upstream topic for this week. - Ceph upgrade status (Vivian/Dehao) check Puppet conflict w/ Ceph v13, no conflict found so far. Deploy the system on StarlingX on new Ceph, and use Puppet to config? Didn't run StarlingX deployment yet. AR: Dehao please do the starlingX deployment and also do Ceph regression testing, and Puppet config testing before the patch can be merged. We have a list of patches on Ceph v10, need to patch migration from Ceph v10 to v13 progress dashboard. More packages needs to be downloaded w/ new Ceph, and the build steps will needed to be modified. AR: Send PR this week to review. - Opens (All) None -----Original Appointment----- From: Xie, Cindy Sent: Friday, August 24, 2018 9:38 AM To: Xie, Cindy; 'Ildiko Vancsa'; Wold, Saul; Chen, Yan; Somerville, Jim; Wang, Yi C; Khalil, Ghada; starlingx-discuss at lists.starlingx.io; Sun, Austin; Rowsell, Brent; Lin, Shuicheng Cc: Hu, Yong; Martinez Monroy, Elio; 'Leo Xu'; 'Young, Ken'; Gomez, Juan P; Shang, Dehao; 'Eslimi, Dariush'; Qi, Mingyuan; Hernandez Gonzalez, Fernando; 'Waines, Greg'; Perez Rodriguez, Humberto I; Zhu, Vivian; Arce Moreno, Abraham; 'Chen, Jacky'; Jones, Bruce E; Cobbley, David A; 'Komiyama, Takeo'; Hu, Wei W; 'Seiler, Glenn' Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 24, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 . Cadence and time slot: o Wednesday 9AM EDT (9PM China time) . Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ . Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other From ricardo.o.perez at intel.com Wed Oct 24 13:57:44 2018 From: ricardo.o.perez at intel.com (Perez, Ricardo O) Date: Wed, 24 Oct 2018 13:57:44 +0000 Subject: [Starlingx-discuss] Weekly StarlingX non-OpenStack Distro meeting Message-ID: Hello list, About the work that is being performed by Dehao, about CEPH migration, here at GDC we can provide help with the deployment and testing of the new CEPH version. Just provide us the ISO or the patches that can be applied flawlessly to build the StarlingX image. By some reason you didn't hear me on the meeting, maybe it was my mic. Thanks -Ricardo. -----Original Appointment----- From: Xie, Cindy [mailto:cindy.xie at intel.com] Sent: Thursday, August 23, 2018 8:38 PM To: Xie, Cindy; Khalil, Ghada; Wold, Saul; Rowsell, Brent; Sun, Austin; Wang, Yi C; Lin, Shuicheng; Chen, Yan; Somerville, Jim; Ildiko Vancsa; starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 24, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 * Cadence and time slot: * Wednesday 9AM EDT (9PM China time) * Call Details: * Zoom link: https://zoom.us/j/342730236 * Dialing in from phone: * Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 * Meeting ID: 342 730 236 * International numbers available: https://zoom.us/u/ed95sU7aQ * Meeting Agenda and Minutes: * https://etherpad.openstack.org/p/stx-distro-other << File: ATT00001.txt >> -------------- next part -------------- An HTML attachment was scrubbed... URL: From ada.cabrales at intel.com Wed Oct 24 14:21:07 2018 From: ada.cabrales at intel.com (Cabrales, Ada) Date: Wed, 24 Oct 2018 14:21:07 +0000 Subject: [Starlingx-discuss] [ Test ] October release status - 10-23-18 In-Reply-To: <151EE31B9FCCA54397A757BC674650F0BA46F1EB@ALA-MBD.corp.ad.wrs.com> References: <4F6AACE4B0F173488D033B02A8BB5B7E7C356EAA@fmsmsx104.amr.corp.intel.com> <151EE31B9FCCA54397A757BC674650F0BA46F1EB@ALA-MBD.corp.ad.wrs.com> Message-ID: <4F6AACE4B0F173488D033B02A8BB5B7E7C357070@fmsmsx104.amr.corp.intel.com> Agree on that, Ghada Thank you so much for your diligence on reviewing and adding comments. Ada > -----Original Message----- > From: Khalil, Ghada [mailto:Ghada.Khalil at windriver.com] > Sent: Wednesday, October 24, 2018 6:59 AM > To: Cabrales, Ada ; starlingx- > discuss at lists.starlingx.io > Subject: RE: [ Test ] October release status - 10-23-18 > > Hi Ada, > I've put comments in the 4 new bugs reported below. None of them are > gating stx.2018.10 (3 are invalid, 1 is lower priority). > See Launchpad for more details. > > Regards, > Ghada > > -----Original Message----- > From: Cabrales, Ada [mailto:ada.cabrales at intel.com] > Sent: Tuesday, October 23, 2018 8:17 PM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] [ Test ] October release status - 10-23-18 > > Sanity on stx-2018-10-23-33-r-2018.10.iso 31 TCs PASS, 0 TCs FAIL. > > 4 Launchpads created: > - "hw:mem_page_size=1GB" Metadata can not be assigned to a flavors with > cli commands" [0] - seems to be a configuration issue, checking > - "1G hugepages for memory allocation is not enable in Virtual external > storage config" [1] - seems to be a configuration issue, checking > - "Process 'neutron-l3-agent' does not work" [2] > - "$ ip netns does not display anything" [3] > > - The only config not finished is Simplex virtual. We estimate can be finished > by Monday next week. > - working on verifying some failures we have, that might be false positives. > > - Execution progress (Fail/Pass/Pending/Total) > Simplex virtual - 1/29/37/67 > Multinode external storage virtual - 8/110/0/118 - done > > Full report in the wiki [4] > > [0] https://bugs.launchpad.net/starlingx/+bug/1799517 > [1] https://bugs.launchpad.net/starlingx/+bug/1799539 > [2] https://bugs.launchpad.net/starlingx/+bug/1799590 > [3] https://bugs.launchpad.net/starlingx/+bug/1799591 > [4] https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_TestingSummary > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From austin.sun at intel.com Wed Oct 24 14:45:12 2018 From: austin.sun at intel.com (Sun, Austin) Date: Wed, 24 Oct 2018 14:45:12 +0000 Subject: [Starlingx-discuss] flake8 error issue Message-ID: Hi All: It seems Openstack community upgrade flake8 from 3.5.0 to 3.6.0, which brings some new static code check . it might cause current code base Zuul openstack-tox-pep8 job fail. I have force for flake8 to 3.5.0 for stx-ha , https://review.openstack.org/#/c/613026/. This might cause other repos zuul pep8 job fail. We can merge this first and provide official fix later , or wait provide fix directly. Thanks Bailey good suggestion for fix. Thanks. BR Austin Sun. From Frank.Miller at windriver.com Wed Oct 24 14:48:47 2018 From: Frank.Miller at windriver.com (Miller, Frank) Date: Wed, 24 Oct 2018 14:48:47 +0000 Subject: [Starlingx-discuss] flake8 error issue In-Reply-To: References: Message-ID: This issue should be tracked by a Launchpad. I created https://bugs.launchpad.net/starlingx/+bug/1799721 - please use this LP # to link your gerrit review. The same change is required for all repos where zuul is configured to run flake8. Frank -----Original Message----- From: Sun, Austin [mailto:austin.sun at intel.com] Sent: Wednesday, October 24, 2018 10:45 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] flake8 error issue Hi All: It seems Openstack community upgrade flake8 from 3.5.0 to 3.6.0, which brings some new static code check . it might cause current code base Zuul openstack-tox-pep8 job fail. I have force for flake8 to 3.5.0 for stx-ha , https://review.openstack.org/#/c/613026/. This might cause other repos zuul pep8 job fail. We can merge this first and provide official fix later , or wait provide fix directly. Thanks Bailey good suggestion for fix. Thanks. BR Austin Sun. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Barton.Wensley at windriver.com Wed Oct 24 14:49:55 2018 From: Barton.Wensley at windriver.com (Wensley, Barton) Date: Wed, 24 Oct 2018 14:49:55 +0000 Subject: [Starlingx-discuss] flake8 error issue In-Reply-To: References: Message-ID: <5CDBBEDBFFF82E4C9E004A2C0F42FE58BAA1BD19@ALA-MBD.corp.ad.wrs.com> We have a bug open for this now: https://bugs.launchpad.net/starlingx/+bug/1799721 You should add a Partial-Bug line to your commit to reference this bug. I will fix all the other stx repos in the next hour or two. Bart -----Original Message----- From: Sun, Austin [mailto:austin.sun at intel.com] Sent: October 24, 2018 10:45 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] flake8 error issue Hi All: It seems Openstack community upgrade flake8 from 3.5.0 to 3.6.0, which brings some new static code check . it might cause current code base Zuul openstack-tox-pep8 job fail. I have force for flake8 to 3.5.0 for stx-ha , https://review.openstack.org/#/c/613026/. This might cause other repos zuul pep8 job fail. We can merge this first and provide official fix later , or wait provide fix directly. Thanks Bailey good suggestion for fix. Thanks. BR Austin Sun. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From sgw at linux.intel.com Wed Oct 24 16:59:32 2018 From: sgw at linux.intel.com (Saul Wold) Date: Wed, 24 Oct 2018 09:59:32 -0700 Subject: [Starlingx-discuss] usage of ldapusersetup? Message-ID: <65c2076e-50d4-7ea1-0c61-198797af7305@linux.intel.com> Dariush, Shuichen: As I was reviewing the lshell removal patch I saw that it is changes the way a new script (ldapusersetup) is added to ldapscripts. I started to dig deeper to see what uses the the ldapusersetup script. The ldapusersetup appears to be a convenience script that could be moved to either a standalone package for StarlingX or if we have an existing package for such scripts. I don't see this script actually being used by any other tools, so does it have documented usage somewhere? See existing Gerrit Review and comments: https://review.openstack.org/#/c/610761/7 Sau! From Dariush.Eslimi at windriver.com Wed Oct 24 17:20:28 2018 From: Dariush.Eslimi at windriver.com (Eslimi, Dariush) Date: Wed, 24 Oct 2018 17:20:28 +0000 Subject: [Starlingx-discuss] usage of ldapusersetup? In-Reply-To: <65c2076e-50d4-7ea1-0c61-198797af7305@linux.intel.com> References: <65c2076e-50d4-7ea1-0c61-198797af7305@linux.intel.com> Message-ID: Saul, The script is helper script for end users, you can move it. Dariush -----Original Message----- From: Saul Wold [mailto:sgw at linux.intel.com] Sent: October-24-18 1:00 PM To: starlingx-discuss at lists.starlingx.io; Eslimi, Dariush; shuicheng.lin at intel.com Subject: usage of ldapusersetup? Dariush, Shuichen: As I was reviewing the lshell removal patch I saw that it is changes the way a new script (ldapusersetup) is added to ldapscripts. I started to dig deeper to see what uses the the ldapusersetup script. The ldapusersetup appears to be a convenience script that could be moved to either a standalone package for StarlingX or if we have an existing package for such scripts. I don't see this script actually being used by any other tools, so does it have documented usage somewhere? See existing Gerrit Review and comments: https://review.openstack.org/#/c/610761/7 Sau! From ricardo.o.perez at intel.com Wed Oct 24 17:39:39 2018 From: ricardo.o.perez at intel.com (Perez, Ricardo O) Date: Wed, 24 Oct 2018 17:39:39 +0000 Subject: [Starlingx-discuss] Minites: Weekly StarlingX non-OpenStack Distro meeting, 10/24 In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F35DA4922@SHSMSX104.ccr.corp.intel.com> References: <2FD5DDB5A04D264C80D42CA35194914F35DA4922@SHSMSX104.ccr.corp.intel.com> Message-ID: Hi Cindy, We at GDC can provide some help testing the new CEPH version, if you have the ISO or the patches that can be applied flawlessly to build our ISO, just let us know. Thanks in advance -Ricardo > -----Original Message----- > From: Xie, Cindy [mailto:cindy.xie at intel.com] > Sent: Wednesday, October 24, 2018 8:49 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Minites: Weekly StarlingX non-OpenStack Distro > meeting, 10/24 > > Agenda & Notes for 10/24 meeting: > - bug review for non-openstack distro (Cindy) > Lshell removal issue: Shuicheng to contact Numan Waheed > (numan.waheed at windriver.com) - WR PV test prime - to get a list of regression > test-cases related to ldap users to execute. This is required before code > submission. > process for the bug assigned to WR (Dariush Eslimi), please get the patch > owner answer the questions and assign back to Cindy so that we can continue > working on that. > > - non-Openstack patch refactor status (Zhipeng) > Master story: https://storyboard.openstack.org/#!/story/2003768 > 23 tasks has been creatd and spec patch has been submitted. End of Nov is the > tentative target date. > around 4 sRPM can be replaced by RPM. > ~50 patches can be removed > > - non-Openstack patch upstream status (Saul) > Saul working on spec reviews and no update on upstream topic for this week. > > - Ceph upgrade status (Vivian/Dehao) > check Puppet conflict w/ Ceph v13, no conflict found so far. Deploy the > system on StarlingX on new Ceph, and use Puppet to config? Didn't run StarlingX > deployment yet. > AR: Dehao please do the starlingX deployment and also do Ceph regression > testing, and Puppet config testing before the patch can be merged. > We have a list of patches on Ceph v10, need to patch migration from Ceph > v10 to v13 progress dashboard. > More packages needs to be downloaded w/ new Ceph, and the build steps will > needed to be modified. AR: Send PR this week to review. > > - Opens (All) > None > > -----Original Appointment----- > From: Xie, Cindy > Sent: Friday, August 24, 2018 9:38 AM > To: Xie, Cindy; 'Ildiko Vancsa'; Wold, Saul; Chen, Yan; Somerville, Jim; Wang, Yi > C; Khalil, Ghada; starlingx-discuss at lists.starlingx.io; Sun, Austin; Rowsell, Brent; > Lin, Shuicheng > Cc: Hu, Yong; Martinez Monroy, Elio; 'Leo Xu'; 'Young, Ken'; Gomez, Juan P; > Shang, Dehao; 'Eslimi, Dariush'; Qi, Mingyuan; Hernandez Gonzalez, Fernando; > 'Waines, Greg'; Perez Rodriguez, Humberto I; Zhu, Vivian; Arce Moreno, > Abraham; 'Chen, Jacky'; Jones, Bruce E; Cobbley, David A; 'Komiyama, Takeo'; > Hu, Wei W; 'Seiler, Glenn' > Subject: Weekly StarlingX non-OpenStack Distro meeting > When: Wednesday, October 24, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, > Chongqing, Hong Kong, Urumqi. > Where: https://zoom.us/j/342730236 > > > . Cadence and time slot: > o Wednesday 9AM EDT (9PM China time) > . Call Details: > o Zoom link: https://zoom.us/j/342730236 > o Dialing in from phone: > o Dial(for higher quality, dial a number based on your current location): US: +1 > 669 900 6833 or +1 646 876 9923 > o Meeting ID: 342 730 236 > o International numbers available: https://zoom.us/u/ed95sU7aQ > . Meeting Agenda and Minutes: > o https://etherpad.openstack.org/p/stx-distro-other > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ghada.Khalil at windriver.com Wed Oct 24 18:30:38 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Wed, 24 Oct 2018 18:30:38 +0000 Subject: [Starlingx-discuss] [Build] Recent Build issue reported Message-ID: <151EE31B9FCCA54397A757BC674650F0BA46F5C5@ALA-MBD.corp.ad.wrs.com> Hello all, Can someone from the StarlingX build team look at this build issue that was reported a couple of hours ago? ...Reported by a new starlingx user https://bugs.launchpad.net/starlingx/+bug/1799735 Thanks, Ghada -------------- next part -------------- An HTML attachment was scrubbed... URL: From hayde.martinez.landa at intel.com Wed Oct 24 18:31:08 2018 From: hayde.martinez.landa at intel.com (Martinez Landa, Hayde) Date: Wed, 24 Oct 2018 18:31:08 +0000 Subject: [Starlingx-discuss] Problems building on master Message-ID: While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: ‘/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch’ -> ‘/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch’ 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Don.Penney at windriver.com Wed Oct 24 18:38:00 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Wed, 24 Oct 2018 18:38:00 +0000 Subject: [Starlingx-discuss] Problems building on master In-Reply-To: References: Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com> I think you meant this update: https://review.openstack.org/#/c/612517/ I don’t understand why you’d be seeing an issue, though. It built fine in my environment, and it shouldn’t impact any %if in the spec From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:31 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Problems building on master While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: ‘/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch’ -> ‘/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch’ 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From hayde.martinez.landa at intel.com Wed Oct 24 18:48:59 2018 From: hayde.martinez.landa at intel.com (Martinez Landa, Hayde) Date: Wed, 24 Oct 2018 18:48:59 +0000 Subject: [Starlingx-discuss] Problems building on master In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com> Message-ID: <278D5915-C22B-47A4-AC07-3034E4CB75F7@intel.com> Hi Don, %if %{with_python2} CFLAGS="$RPM_OPT_FLAGS" %{__python2} setup.py build %py2_build_wheel Well I tried running it removing the line with the macro “py2_build_wheel” and it worked But whit this line on, it fails with the error message: libvirt-python.spec:131: bad %if condition From: "Penney, Don" Date: Wednesday, October 24, 2018 at 1:38 PM To: "Martinez Landa, Hayde" , "starlingx-discuss at lists.starlingx.io" Subject: RE: Problems building on master I think you meant this update: https://review.openstack.org/#/c/612517/ I don’t understand why you’d be seeing an issue, though. It built fine in my environment, and it shouldn’t impact any %if in the spec From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:31 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Problems building on master While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: ‘/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch’ -> ‘/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch’ 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Irina.Mihai at windriver.com Wed Oct 24 18:53:16 2018 From: Irina.Mihai at windriver.com (Mihai, Irina) Date: Wed, 24 Oct 2018 18:53:16 +0000 Subject: [Starlingx-discuss] Problems building on master In-Reply-To: <278D5915-C22B-47A4-AC07-3034E4CB75F7@intel.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com>, <278D5915-C22B-47A4-AC07-3034E4CB75F7@intel.com> Message-ID: Hey, For what's worth, I also saw that today in my build env. I'm building in a centos lxc container. Irina ________________________________ From: Martinez Landa, Hayde [hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:48 PM To: Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hi Don, %if %{with_python2} CFLAGS="$RPM_OPT_FLAGS" %{__python2} setup.py build %py2_build_wheel Well I tried running it removing the line with the macro “py2_build_wheel” and it worked But whit this line on, it fails with the error message: libvirt-python.spec:131: bad %if condition From: "Penney, Don" Date: Wednesday, October 24, 2018 at 1:38 PM To: "Martinez Landa, Hayde" , "starlingx-discuss at lists.starlingx.io" Subject: RE: Problems building on master I think you meant this update: https://review.openstack.org/#/c/612517/ I don’t understand why you’d be seeing an issue, though. It built fine in my environment, and it shouldn’t impact any %if in the spec From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:31 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Problems building on master While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: ‘/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch’ -> ‘/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch’ 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From ildiko.vancsa at gmail.com Wed Oct 24 19:32:22 2018 From: ildiko.vancsa at gmail.com (Ildiko Vancsa) Date: Wed, 24 Oct 2018 21:32:22 +0200 Subject: [Starlingx-discuss] Announcing the First Release of StarlingX, an open source edge computing platform Message-ID: As you most probably already know, StarlingX[1] is a new independent, top-level, open source pilot project that's supported by the OpenStack Foundation. StarlingX joins other pilot projects hosted at OpenStack Foundation[2], including Airship, Kata Containers and Zuul. Today the first release of StarlingX is here! Thank you to everyone who has spent time working on this first release and helping shape our community; it’s been great work to be a part of and we're excited to see this community progress. We invite you to participate in getting the word out that the release is ready and that we’re eager to welcome more contributors to this project. Learn more about it: • Read more about the project at starlingx.io • Listen to a recording of the onboarding webinar[3] • On-boarding slide deck[4] • Overview document[5] Some things you can share: • A blog on starlingx.io[6] • Social sharing: Announcements on Twitter[7] Want to get involved in the community? • Mailing Lists[8] • Weekly Calls[9] • Freenode IRC: #starlingx channel[10] Ready to dive into the code? • You can get download the first release at git.starlingx.io • StarlingX Install Guide[11] • StarlingX Developer Guide[12] If you’re at the Berlin Summit November 13-15[13]: Tuesday 11/13 • StarlingX – Project update – 6 months in the life of a new Open Source project with Brent Rowsell & Dean Troyer • StarlingX CI, from zero to Zuul with Hazzim Anaya & Elio Martinez Wednesday 11/14 • Keynote spotlight on the main stage with Ian Jolliffe & Dean Troyer • MVP (Minimum Viable Product) architecture for edge - Forum session • "Ask Me Anything" about StarlingX - Forum session • StarlingX Enhancements for Edge Networking presentation with Kailun Qin, Ruijing Guo & Dan Chen • Project Onboarding session with Greg Waines • Integrating IOT Device Management with the Edge Cloud - Forum session Thursday 11/15 • Containerized Applications' Requirements on Kubernetes Cluster at the Edge - Forum session Thank you again to everyone who has been hard at work on the first release, and hope to see many of you in Berlin! Ildikó [1] https://www.starlingx.io/ [2] https://www.openstack.org/foundation/ [3] https://www.youtube.com/watch?v=G9uwGnKD6tM&t=232s [4] https://www.starlingx.io/collateral/StarlingX-Onboarding-Deck-Web.pdf [5] https://www.starlingx.io/collateral/StarlingX_OnePager_Web-102318pdf/ [6] https://www.starlingx.io/blog/starlingx-initial-release.html [7] https://twitter.com/starlingx [8] http://lists.starlingx.io/cgi-bin/mailman/listinfo [9] https://wiki.openstack.org/wiki/Starlingx/Meetings [10] https://freenode.net/ [11] https://docs.starlingx.io/installation_guide/index.html [12] https://docs.starlingx.io/developer_guide/index.html [13] https://www.openstack.org/summit/berlin-2018 From Al.Bailey at windriver.com Wed Oct 24 19:55:31 2018 From: Al.Bailey at windriver.com (Bailey, Henry Albert (Al)) Date: Wed, 24 Oct 2018 19:55:31 +0000 Subject: [Starlingx-discuss] Problems building on master In-Reply-To: References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com>, <278D5915-C22B-47A4-AC07-3034E4CB75F7@intel.com> Message-ID: I pushed a fix for review. The build passed in my env. Others can cherry pick and try in theirs while waiting for Zuul. https://review.openstack.org/#/c/613148/ Al From: Mihai, Irina [mailto:Irina.Mihai at windriver.com] Sent: Wednesday, October 24, 2018 2:53 PM To: Martinez Landa, Hayde; Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hey, For what's worth, I also saw that today in my build env. I'm building in a centos lxc container. Irina ________________________________ From: Martinez Landa, Hayde [hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:48 PM To: Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hi Don, %if %{with_python2} CFLAGS="$RPM_OPT_FLAGS" %{__python2} setup.py build %py2_build_wheel Well I tried running it removing the line with the macro "py2_build_wheel" and it worked But whit this line on, it fails with the error message: libvirt-python.spec:131: bad %if condition From: "Penney, Don" Date: Wednesday, October 24, 2018 at 1:38 PM To: "Martinez Landa, Hayde" , "starlingx-discuss at lists.starlingx.io" Subject: RE: Problems building on master I think you meant this update: https://review.openstack.org/#/c/612517/ I don't understand why you'd be seeing an issue, though. It built fine in my environment, and it shouldn't impact any %if in the spec From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:31 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Problems building on master While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: '/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch' -> '/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch' 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Irina.Mihai at windriver.com Wed Oct 24 20:00:02 2018 From: Irina.Mihai at windriver.com (Mihai, Irina) Date: Wed, 24 Oct 2018 20:00:02 +0000 Subject: [Starlingx-discuss] Problems building on master In-Reply-To: References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com>, <278D5915-C22B-47A4-AC07-3034E4CB75F7@intel.com> , Message-ID: It works for me :) Thanks, Al! Irina ________________________________ From: Bailey, Henry Albert (Al) Sent: Wednesday, October 24, 2018 3:55 PM To: Mihai, Irina; Martinez Landa, Hayde; Penney, Don; starlingx-discuss at lists.starlingx.io Subject: RE: Problems building on master I pushed a fix for review. The build passed in my env. Others can cherry pick and try in theirs while waiting for Zuul. https://review.openstack.org/#/c/613148/ Al From: Mihai, Irina [mailto:Irina.Mihai at windriver.com] Sent: Wednesday, October 24, 2018 2:53 PM To: Martinez Landa, Hayde; Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hey, For what's worth, I also saw that today in my build env. I'm building in a centos lxc container. Irina ________________________________ From: Martinez Landa, Hayde [hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:48 PM To: Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hi Don, %if %{with_python2} CFLAGS="$RPM_OPT_FLAGS" %{__python2} setup.py build %py2_build_wheel Well I tried running it removing the line with the macro “py2_build_wheel” and it worked But whit this line on, it fails with the error message: libvirt-python.spec:131: bad %if condition From: "Penney, Don" Date: Wednesday, October 24, 2018 at 1:38 PM To: "Martinez Landa, Hayde" , "starlingx-discuss at lists.starlingx.io" Subject: RE: Problems building on master I think you meant this update: https://review.openstack.org/#/c/612517/ I don’t understand why you’d be seeing an issue, though. It built fine in my environment, and it shouldn’t impact any %if in the spec From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:31 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Problems building on master While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: ‘/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch’ -> ‘/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch’ 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From hayde.martinez.landa at intel.com Wed Oct 24 20:05:14 2018 From: hayde.martinez.landa at intel.com (Martinez Landa, Hayde) Date: Wed, 24 Oct 2018 20:05:14 +0000 Subject: [Starlingx-discuss] Problems building on master In-Reply-To: References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3D0546@ALA-MBD.corp.ad.wrs.com> <278D5915-C22B-47A4-AC07-3034E4CB75F7@intel.com> Message-ID: <1449DABA-1EEF-4526-895D-9972832B1300@intel.com> Thanks! Worked for me 2 : ) From: "Mihai, Irina" Date: Wednesday, October 24, 2018 at 3:01 PM To: "Bailey, Henry Albert (Al)" , "Martinez Landa, Hayde" , "Penney, Don" , "starlingx-discuss at lists.starlingx.io" Subject: RE: Problems building on master It works for me :) Thanks, Al! Irina ________________________________ From: Bailey, Henry Albert (Al) Sent: Wednesday, October 24, 2018 3:55 PM To: Mihai, Irina; Martinez Landa, Hayde; Penney, Don; starlingx-discuss at lists.starlingx.io Subject: RE: Problems building on master I pushed a fix for review. The build passed in my env. Others can cherry pick and try in theirs while waiting for Zuul. https://review.openstack.org/#/c/613148/ Al From: Mihai, Irina [mailto:Irina.Mihai at windriver.com] Sent: Wednesday, October 24, 2018 2:53 PM To: Martinez Landa, Hayde; Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hey, For what's worth, I also saw that today in my build env. I'm building in a centos lxc container. Irina ________________________________ From: Martinez Landa, Hayde [hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:48 PM To: Penney, Don; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Problems building on master Hi Don, %if %{with_python2} CFLAGS="$RPM_OPT_FLAGS" %{__python2} setup.py build %py2_build_wheel Well I tried running it removing the line with the macro “py2_build_wheel” and it worked But whit this line on, it fails with the error message: libvirt-python.spec:131: bad %if condition From: "Penney, Don" Date: Wednesday, October 24, 2018 at 1:38 PM To: "Martinez Landa, Hayde" , "starlingx-discuss at lists.starlingx.io" Subject: RE: Problems building on master I think you meant this update: https://review.openstack.org/#/c/612517/ I don’t understand why you’d be seeing an issue, though. It built fine in my environment, and it shouldn’t impact any %if in the spec From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Wednesday, October 24, 2018 2:31 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Problems building on master While trying to build on master I encountered the following issue: 17:09:42 b4: patching file SPECS/libvirt-python.spec 17:09:42 b4: ‘/localdisk/designer/haydemtzl/starlingx/cgcs-root/stx/stx-integ/virt/libvirt-python/centos/patches/0001-Fix-setup-import-to-support-building-wheel.patch’ -> ‘/localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SOURCES/./0001-Fix-setup-import-to-support-building-wheel.patch’ 17:09:42 b4: SPEC file: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec 17:09:42 b4: SRPM build directory: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild 17:09:42 b4: TIS_PATCH_VER: 2 17:09:42 b4: error: parse error in expression 17:09:42 b4: error: /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec:131: bad %if condition 17:09:42 b4: ERROR: srpm_assemble (1296): rpmbuild failed: rpmbuild -bs /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild/SPECS/libvirt-python.spec --define='%_topdir /localdisk/loadbuild/haydemtzl/starlingx/std/srpm_assemble/libvirt-python/rpmbuild' --define='_tis_dist .tis' --undefine=dist 17:09:42 b4: ERROR: build_dir_srpm (802): failed to assemble srpm for 'libvirt-python-4.7.0-1.fc28' 17:09:42 ERROR: reaper (1302): Failed to build src.rpm from source at 'b4' 17:09:42 ######## Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Wed Oct 24 17:09:42 UTC 2018: build-srpm-parallel --std failed with rc=1 Whit the help of Erich for debugging this, We think it come from this change [0] that affects the specific line mentioned on the error above: [0] https://review.openstack.org/#/c/612519/ I was able to reproduce this in 3 diff environments, including our Jenkins automation setup. -------------- next part -------------- An HTML attachment was scrubbed... URL: From dtroyer at gmail.com Wed Oct 24 21:40:49 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Wed, 24 Oct 2018 16:40:49 -0500 Subject: [Starlingx-discuss] StarlingX Release 2018.10.0 Message-ID: I am pleased to announce the final tagging of the StarlingX 2018.10.0 release. The source is available from https://git.starlingx.io/cgit. This release is simply a tag in the Git repositories of '2018.10.0' on the r/2018.10 branch, there are no additional release artifacts such as tar archive files. Guidelines for building this release may be found in https://docs.starlingx.io/developer_guide/index.html. As this is the first public release of StarlingX there are no traditional release notes describing changes from the previous release. However there are highlights of recent development for each project in the Release Notes documentation at https://docs.starlingx.io/releasenotes/index.html. On behalf of the StarlingX Technical Steering Committee I would like to thank everyone who has contributed to the project and helped us reach this landmark. This is truly a team and community effort! dt -- Dean Troyer dtroyer at gmail.com From Rajesh.Jha at rsystems.com Wed Oct 24 07:54:16 2018 From: Rajesh.Jha at rsystems.com (Rajesh Jha) Date: Wed, 24 Oct 2018 13:24:16 +0530 Subject: [Starlingx-discuss] Build of StarlingX ISO on CentOs System Message-ID: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> Hi, Since the developers guide has the steps followed in making an iso image of StarlingX taking Ubuntu 16.04 as the host OS. Therefore all scripts need to be updated to reflect the corresponding centOS commands in the script. So is there any scripts available for CentOS development environment as well so we can refer and run it. So shall we go ahead with changes in script written for Ubuntu to work on CentOS. Any suggestion will be very helpful. Best Regards, Rajesh Jha The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. From dtroyer at gmail.com Wed Oct 24 23:09:40 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Wed, 24 Oct 2018 18:09:40 -0500 Subject: [Starlingx-discuss] Build of StarlingX ISO on CentOs System In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: On Wed, Oct 24, 2018 at 4:58 PM Rajesh Jha wrote: > Since the developers guide has the steps followed in making an iso image of StarlingX taking Ubuntu 16.04 as the host OS. Therefore all scripts need to be updated to reflect the corresponding centOS commands in the script. So is there any scripts available for CentOS development environment as well so we can refer and run it. We (Brian and I at Intel) initially developed the CentOS container approach to building StarlingX because what we had available to do builds were Ubuntu 16.04 systems. You could conceivably just run the build directly on CentOS systems without the container. This in fact is how many of the developers at Wind River work. > So shall we go ahead with changes in script written for Ubuntu to work on CentOS. Any suggestion will be very helpful. I can see where using the container build process may be desirable even on a CentOS host, keeping things neat and clean, or if the version requirements do not exactly match. Generalizing the Ubuntu scripts to work on other distros is desirable but just not something that anyone else done before now. We would welcome contributions that broaden the availability of StarlingX. dt -- Dean Troyer dtroyer at gmail.com From erich.cordoba.malibran at intel.com Wed Oct 24 23:35:10 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Wed, 24 Oct 2018 23:35:10 +0000 Subject: [Starlingx-discuss] Build of StarlingX ISO on CentOs System In-Reply-To: References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: On Wed, 2018-10-24 at 18:09 -0500, Dean Troyer wrote: > On Wed, Oct 24, 2018 at 4:58 PM Rajesh Jha > wrote: > > Since the developers guide has the steps followed in making an iso > > image of StarlingX taking Ubuntu 16.04 as the host OS. Therefore > > all scripts need to be updated to reflect the corresponding centOS > > commands in the script. So is there any scripts available for > > CentOS development environment as well so we can refer and run it. > > We (Brian and I at Intel) initially developed the CentOS container > approach to building StarlingX because what we had available to do > builds were Ubuntu 16.04 systems. You could conceivably just run the > build directly on CentOS systems without the container. This in fact > is how many of the developers at Wind River work. > I use Fedora 28 for my daily work and the only issue I faced was this one: https://storyboard.openstack.org/#!/story/2002560 So, the only OS specific steps are to create the builder container, which are the `make base-build` and `make build` commands. Once you have your container, then everything after the `./tb.sh exec` command should work. We can try to build the container on CentOS 7 to see is the steps work and update the documentation if it's needed. > > So shall we go ahead with changes in script written for Ubuntu to > > work on CentOS. Any suggestion will be very helpful. > > I can see where using the container build process may be desirable > even on a CentOS host, keeping things neat and clean, or if the > version requirements do not exactly match. Generalizing the Ubuntu > scripts to work on other distros is desirable but just not something > that anyone else done before now. > > We would welcome contributions that broaden the availability of > StarlingX. > > dt > From Rajesh.Jha at rsystems.com Wed Oct 24 11:09:45 2018 From: Rajesh.Jha at rsystems.com (Rajesh Jha) Date: Wed, 24 Oct 2018 16:39:45 +0530 Subject: [Starlingx-discuss] Query on Nove and Network test case In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> Hi, While testing nova test cases we are able to make the Nova online successfully. However controller-0 is unable to communicate with compute-0(nova) and not allowing to add data interface on controller-0. Due to this issue we are not able to execute the remaining test cases for the compute (nova) and network can be run from controller-0.we also observe sub-functions are missing from the controller-0. We are following below link 1.https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions 2.https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Simplex running command: "$system host-show 1 | grep subfunctions" does not return - "| subfunctions | controller,compute" Thank you, Rajesh Jha The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. From Rajesh.Jha at rsystems.com Thu Oct 25 10:10:45 2018 From: Rajesh.Jha at rsystems.com (Rajesh Jha) Date: Thu, 25 Oct 2018 15:40:45 +0530 Subject: [Starlingx-discuss] unexpected behaviour in duplex mode in virtual env In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com>, <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: <2F61A2101B4FD447A683EF7E362CF9421C63E217F0@RSI-NOD-EXMDB6.india.rsystems.com> Hi, We have observed an unstable behaviour while running below scenario Setup- Virtual environment testing using kvm Mode "Duplex" (One active and one standby controller script setup_allinone.sh) Steps- 1. Followed steps in document "https://docs.starlingx.io/installation_guide/duplex.html#duplex" 2. Able to proceed till section "system host-unlock controller-0" command. After issuing this command system get reboot as per expectation but once it comes up availability status changed to "degrade" and alarm raised (screen shot attached). 3. Then we execute "destroy_allinone.sh" and "destroy_network.sh" command to cleanup the previous setup to build new one 4. After following same step in above documents till "sudo config_controller" command which was successful and after run "source /etc/nova/openrc" system get reboot again and again and nothing happening after instead of rebooting. Screenshot is attached below. CLI is no more available to the controller Thank you, Rajesh The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. From longqian.zhao at intel.com Thu Oct 25 12:24:04 2018 From: longqian.zhao at intel.com (Zhao, Longqian) Date: Thu, 25 Oct 2018 12:24:04 +0000 Subject: [Starlingx-discuss] compute-0 failed configuring Message-ID: <6325970F810676468E0EC5D8ED76D2190B7C6DF5@SHSMSX101.ccr.corp.intel.com> Hi, I have a problem as follows: My problems: Compute-0 failed configuring Problem Description: After I install and deploy controller-0... 1. Power on compute-0, The console prints the following information: +++++++++++++++++++++++++++++++++++++++++++++++++++++++ Booting from Hard Disk... Boot failed: not a bootable disk Booting from DVD/CD... Boot failed: Could not read from CDROM (code 0003) Booting from ROM... iPXE (PCI 02:02:0) starting execution...ok iPXE initializing devices...ok iPXE 1.0.0+git-20150424.a25a16d-1ubuntu1.2 - Open Source Network Boot Firmware --http://ipxe.org Features: Omitted here Configuring(net1 52:54:00:21:22:66)......................................Error 0x040ee119 (http://ipxe.org/040ee119) No more network devices No bootable device. ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 2. I configured my controller-0 according to the information provided by this url: https://bugs.launchpad.net/starlingx/+bug/1790716 a. $ source /etc/nova/openrc b. ~(keystone_admin)$ system host-add -n compute-0 compute -m 52:54:00:21:22:66 c. ~(keystone_admin)$ system host-show compute-0 | grep install The console prints the following information: ++++++++++++++++++++++++++++++++++++++++++++++++++++++ ~(keystone_admin)$ system host-show compute-0 | grep install | install_output | text | | install_state | None | | install_state_info | None| +++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3. I use this ISO (http://starlingx-hub.zpn.intel.com/ISO/stx-2018-10-23-33-r-2018.10.iso) to install. Please give me a tip or help, thank you very much... Thanks again, Longqian. -------------- next part -------------- An HTML attachment was scrubbed... URL: From michel.thebeau at windriver.com Thu Oct 25 12:55:08 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Thu, 25 Oct 2018 08:55:08 -0400 Subject: [Starlingx-discuss] compute-0 failed configuring In-Reply-To: <6325970F810676468E0EC5D8ED76D2190B7C6DF5@SHSMSX101.ccr.corp.intel.com> References: <6325970F810676468E0EC5D8ED76D2190B7C6DF5@SHSMSX101.ccr.corp.intel.com> Message-ID: Hi Longqian, The pxe boot attempt showing in the compute's console says it didn't get a response for pxe. Configuring(net1 52:54:00:21:22:66)......................................Error 0x040ee119 (http://ipxe.org/040ee119) No more network devices So one would first make sure the network was correctly plugged. Could you describe what networks are plugged to which interfaces? Is the network interface that is plugged to managment network configured to try boot?  Which network is interface "net1 52:54:00:21:22:66" plugged to? Is it the case that you were able to install and configure controller-1? The mac address starting with 52 is in the range of locally administered mac addresses, which implies you are running virtual servers.  Are you using the libvirt deployment method found in stx-tools git repo and referenced on the wiki https://wiki.openstack.org/wiki/StarlingX/Installation_Guide If you are following that wiki and using stx-tools, may I inquire: what prompted you to use https://bugs.launchpad.net/starlingx/+bug/1790716 as a source of information for boot, and use to "system host-add" as opposed to "system host-update"? M On 2018-10-25 8:24 a.m., Zhao, Longqian wrote: > Hi, > > I have a problem as follows: > > My problems: > > Compute-0 failed configuring > > Problem Description: > After I install and deploy controller-0... > > > 1. Power on compute-0, The console prints the following information: > > > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > Booting from Hard Disk... > > Boot failed: not a bootable disk > > > > Booting from DVD/CD... > > Boot failed: Could not read from CDROM (code 0003) > > Booting from ROM... > > iPXE (PCI 02:02:0) starting execution...ok > > iPXE initializing devices...ok > > > > iPXE 1.0.0+git-20150424.a25a16d-1ubuntu1.2 - Open Source Network Boot Firmware > --http://ipxe.org > Features: Omitted here > Configuring(net1 52:54:00:21:22:66)......................................Error 0x040ee119 (http://ipxe.org/040ee119) > No more network devices > > No bootable device. > ++++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > 2. I configured my controller-0 according to the information provided by this url: https://bugs.launchpad.net/starlingx/+bug/1790716 > > a. $ source /etc/nova/openrc > > b. ~(keystone_admin)$ system host-add -n compute-0 compute -m 52:54:00:21:22:66 > > c. ~(keystone_admin)$ system host-show compute-0 | grep install > > > > The console prints the following information: > > ++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > ~(keystone_admin)$ system host-show compute-0 | grep install > > | install_output | text | > > | install_state | None | > > | install_state_info | None| > > +++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > > 3. I use this ISO (http://starlingx-hub.zpn.intel.com/ISO/stx-2018-10-23-33-r-2018.10.iso) to install. > > Please give me a tip or help, thank you very much... > > Thanks again, > Longqian. > > > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Thu Oct 25 13:08:53 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Thu, 25 Oct 2018 13:08:53 +0000 Subject: [Starlingx-discuss] Minites: Weekly StarlingX non-OpenStack Distro meeting, 10/24 In-Reply-To: References: <2FD5DDB5A04D264C80D42CA35194914F35DA4922@SHSMSX104.ccr.corp.intel.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35DA5DC0@SHSMSX104.ccr.corp.intel.com> This is good to know! Thanks Ricardo! + Vivian and Dehao. @Dehao, please let me & Ricardo know when you generate the ISO w/ new Ceph. Thx. - cindy -----Original Message----- From: Perez, Ricardo O Sent: Thursday, October 25, 2018 1:40 AM To: Xie, Cindy ; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Minites: Weekly StarlingX non-OpenStack Distro meeting, 10/24 Hi Cindy, We at GDC can provide some help testing the new CEPH version, if you have the ISO or the patches that can be applied flawlessly to build our ISO, just let us know. Thanks in advance -Ricardo > -----Original Message----- > From: Xie, Cindy [mailto:cindy.xie at intel.com] > Sent: Wednesday, October 24, 2018 8:49 AM > To: starlingx-discuss at lists.starlingx.io > Subject: [Starlingx-discuss] Minites: Weekly StarlingX non-OpenStack > Distro meeting, 10/24 > > Agenda & Notes for 10/24 meeting: > - bug review for non-openstack distro (Cindy) > Lshell removal issue: Shuicheng to contact Numan Waheed > (numan.waheed at windriver.com) - WR PV test prime - to get a list of > regression test-cases related to ldap users to execute. This is > required before code submission. > process for the bug assigned to WR (Dariush Eslimi), please get > the patch owner answer the questions and assign back to Cindy so that > we can continue working on that. > > - non-Openstack patch refactor status (Zhipeng) > Master story: https://storyboard.openstack.org/#!/story/2003768 > 23 tasks has been creatd and spec patch has been submitted. End of > Nov is the tentative target date. > around 4 sRPM can be replaced by RPM. > ~50 patches can be removed > > - non-Openstack patch upstream status (Saul) > Saul working on spec reviews and no update on upstream topic for this week. > > - Ceph upgrade status (Vivian/Dehao) > check Puppet conflict w/ Ceph v13, no conflict found so far. > Deploy the system on StarlingX on new Ceph, and use Puppet to config? > Didn't run StarlingX deployment yet. > AR: Dehao please do the starlingX deployment and also do Ceph > regression testing, and Puppet config testing before the patch can be merged. > We have a list of patches on Ceph v10, need to patch migration > from Ceph > v10 to v13 progress dashboard. > More packages needs to be downloaded w/ new Ceph, and the build > steps will needed to be modified. AR: Send PR this week to review. > > - Opens (All) > None > > -----Original Appointment----- > From: Xie, Cindy > Sent: Friday, August 24, 2018 9:38 AM > To: Xie, Cindy; 'Ildiko Vancsa'; Wold, Saul; Chen, Yan; Somerville, > Jim; Wang, Yi C; Khalil, Ghada; starlingx-discuss at lists.starlingx.io; > Sun, Austin; Rowsell, Brent; Lin, Shuicheng > Cc: Hu, Yong; Martinez Monroy, Elio; 'Leo Xu'; 'Young, Ken'; Gomez, > Juan P; Shang, Dehao; 'Eslimi, Dariush'; Qi, Mingyuan; Hernandez > Gonzalez, Fernando; 'Waines, Greg'; Perez Rodriguez, Humberto I; Zhu, > Vivian; Arce Moreno, Abraham; 'Chen, Jacky'; Jones, Bruce E; Cobbley, > David A; 'Komiyama, Takeo'; Hu, Wei W; 'Seiler, Glenn' > Subject: Weekly StarlingX non-OpenStack Distro meeting > When: Wednesday, October 24, 2018 9:00 PM-10:00 PM (UTC+08:00) > Beijing, Chongqing, Hong Kong, Urumqi. > Where: https://zoom.us/j/342730236 > > > . Cadence and time slot: > o Wednesday 9AM EDT (9PM China time) > . Call Details: > o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: > o Dial(for higher quality, dial a number based on your current > location): US: +1 > 669 900 6833 or +1 646 876 9923 > o Meeting ID: 342 730 236 > o International numbers available: https://zoom.us/u/ed95sU7aQ . > Meeting Agenda and Minutes: > o https://etherpad.openstack.org/p/stx-distro-other > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From andy.ning at windriver.com Thu Oct 25 13:53:49 2018 From: andy.ning at windriver.com (Andy Ning) Date: Thu, 25 Oct 2018 09:53:49 -0400 Subject: [Starlingx-discuss] usage of ldapusersetup? In-Reply-To: References: <65c2076e-50d4-7ea1-0c61-198797af7305@linux.intel.com> Message-ID: Yes ldapusersetup is our addition to ldapscripts as one more helper utility. It's using other scripts in the ldapscripts package. ldapusersetup together with the original scripts in ldapscripts package are all wrappers of the utilities provided by openldap-clients. So we could move it to a separate package. Or, we could push to ldapscripts upstream as a nice addition. Andy On 2018-10-24 01:20 PM, Eslimi, Dariush wrote: > Saul, > > The script is helper script for end users, you can move it. > > Dariush > > -----Original Message----- > From: Saul Wold [mailto:sgw at linux.intel.com] > Sent: October-24-18 1:00 PM > To: starlingx-discuss at lists.starlingx.io; Eslimi, Dariush; shuicheng.lin at intel.com > Subject: usage of ldapusersetup? > > > Dariush, Shuichen: > > As I was reviewing the lshell removal patch I saw that it is changes the way a new script (ldapusersetup) is added to ldapscripts. I started to dig deeper to see what uses the the ldapusersetup script. The ldapusersetup appears to be a convenience script that could be moved to either a standalone package for StarlingX or if we have an existing package for such scripts. > > I don't see this script actually being used by any other tools, so does it have documented usage somewhere? > > See existing Gerrit Review and comments: > https://review.openstack.org/#/c/610761/7 > > Sau! > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -- Andy Ning Cube: 3071 Tel: 613-9631408 (int: 4408) Skype: andy.ning.wr From cesar.lara at intel.com Thu Oct 25 14:07:43 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Thu, 25 Oct 2018 14:07:43 +0000 Subject: [Starlingx-discuss] [build][meetings] Agenda for 10/25/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471053DAE9@fmsmsx104.amr.corp.intel.com> Build team meeting 10/25/2018 Agenda - Cengn mirror status - Cengn hosting spec discussion - Erich - Koji walkthrough - Victor R - Koji Docker file - Chuy - Plans for next release - opens Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Wed Oct 24 17:27:39 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Wed, 24 Oct 2018 17:27:39 +0000 Subject: [Starlingx-discuss] Community call Oct 24 notes Message-ID: <9A85D2917C58154C960D95352B22818BB1E83970@ORSMSX158.amr.corp.intel.com> Agenda and Notes for the Oct 24th call * Let's do a release!!! o Our new web page is live! o Bug status ? Only one stx.2018.10 bug still open: https://bugs.launchpad.net/starlingx/+bug/1797234 * Notes say that the issue is not reproducible, so this does not hold up stx.2018.10 - removing the label and recommending closure ? A quick review of the new bugs in the last couple of days does not reveal in any issues that would hold up the release ? Erich to file a new launchpad for the latest build issues on master - potential build blocking issues in the mirror script * RT kernel package location - to be discussed over email... * https://review.openstack.org/#/c/612822/ o Testing status o Docs status ? Release note status o Backporting policy??? ? Changes that are candidates for backporting should be sent to the [Release] team in email. ? Release team will review and decide if a backport is required. o Release tagging is on hold until EOD today, with a strong desire to get the build changes in first. * Priorities for r/2019.03 o See https://etherpad.openstack.org/p/stx-2019.03-priorities o Strongly encourage sub-projects TL/PLs to put together their lists/plans and update the etherpad asap. * When to post code for review? Early and often? After testing? o https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines o We should consider clarifying the early review needed to catch design issues between the final code inspection that gates code acceptance o The Intel team in particular would benefit from early/often review o The goal is to help close the gap on detailed design approaches where the spec may not cover that level of detail, but the developer needs input on the approach before being able to test. o There are mechanisms that the developer should use to make it clear that they are requesting early feedback by including WIP or RFC in the first line of the commit message and/or Workflow -1 by the review submitter o Core reviewer workload is high but we can offload that to Contributors who are all able to review code. * How to test complex changes? o Leverage the virutal environment where possible e.g. in the current Ceph changes that need testing can be done in a virtual duplex config o Other changes that require real hardware should be discussed with the Test team. * Thank you to everyone for all your work on the release! * The new website is live! * @StarlingX on twitter is live, please re-tweet. * Please join the webinar at 8AM PT today. -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Thu Oct 25 16:54:48 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Thu, 25 Oct 2018 16:54:48 +0000 Subject: [Starlingx-discuss] TSC meeting notes Oct 25th 2018 Message-ID: <9A85D2917C58154C960D95352B22818BB1E84EED@ORSMSX158.amr.corp.intel.com> https://etherpad.openstack.org/p/stx-cores 10/25/18 * TSC member update - two new members, one is pending submission of a candidate name o Welcome Curtis Collicutt and Shuquan Huang! * Final governance approval and election schedule to be determined/approved by the TSC once it reaches 7 members * Review any pending stx-gov reviews o Please review https://review.openstack.org/#/c/611185/ https://review.openstack.org/#/c/610446/ * Enforcing core mappings to subprojects/repos (Brent) o RIght now if you are a Core for one project you are a Core for everything. Gerrit groups are being used. Brent to publish the mapping and the team can work with Ildiko to limit Cores to the projects they are Cores for. * Revist spec approval process (Brent) o We have reviewed a couple of specs now and some have gotten to the +2 step, then someone came along later with a -1 because they had a question. Not all -1's are equal. What are the community norms for reviewing specs? Sometimes people may feel like their questions are important and might be ignored without a -1. But we don't want to block forward progress on specs. -1 votes are cleared for new patch sets but not for -2s which must be manually approved. o We would like the community to use -1 on specs for important issues with specs. Or important questions. It's a judgment call. This also applies to code reviews. * Review and update Project Priorities: https://etherpad.openstack.org/p/stx-2019.03-priorities o We want to finish this quickly - March is coming! o Bruce to turn the etherpad into a more structured document that can be used to describe the work in more detail, show which projects are effected, etc... * Review any pending stx-specs reviews o Please help review https://review.openstack.org/#/c/611725/ * Retirement of more staging trees, general discussion about strategy and plans * Review submission process (Brent) * We had a review self-approved that went into the release yesterday that went in the day before the cut-off. Ghada to follow up. * IA Features sub-project - Bruce o https://wiki.openstack.org/wiki/StarlingX/IA_Features o We will be upstream first for new features e.g. Nova. o We should discuss this as part of an overall Test strategy o Making this a sub-project is probably not the right approach. Intel may be very interested but what if we had an ARM vendor feature list? Vendors will assign resources to things that they think are important. o Bruce to track on the wiki but not as a sub-project. * Presentation at the joint leadership meeting in Berlin (Nov 12) o Brent to cover, other project members are welcome to attend -------------- next part -------------- An HTML attachment was scrubbed... URL: From Barton.Wensley at windriver.com Thu Oct 25 18:47:23 2018 From: Barton.Wensley at windriver.com (Wensley, Barton) Date: Thu, 25 Oct 2018 18:47:23 +0000 Subject: [Starlingx-discuss] Community call Oct 24 notes In-Reply-To: <9A85D2917C58154C960D95352B22818BB1E83970@ORSMSX158.amr.corp.intel.com> References: <9A85D2917C58154C960D95352B22818BB1E83970@ORSMSX158.amr.corp.intel.com> Message-ID: <5CDBBEDBFFF82E4C9E004A2C0F42FE58BAA1C826@ALA-MBD.corp.ad.wrs.com> Bruce (and others), As a core reviewer, I'd like to give some comments on "when to post code for review". I have been seeing a significant number of commits put up for review with zero testing and in some cases, the code is so broken that it won't even compile. This is a complete waste of everyone's time. For simple changes (this is the vast majority of changes), the author should be completing their testing before posting it for review. This should include: · Code compiles successfully. · Code can be built into RPMs and an ISO. · Tox tests (flake8, py27, etc...) all pass. These can all be run manually and should be run before launching a review. · Existing unit tests updated and new unit tests added where applicable. · Functional testing done on a live system to ensure code functions correctly. If the author does not know how to test the code they are changing, then they should be asking questions on the mailing list - before posting the review. When a review is updated with a new patch set, the above testing needs to be re-done before the new patch set is posted. I agree that in specific cases, we may want the author to post the code before testing it. But in those cases, the author should immediately do a Workflow -1 and add a comment to the review explaining why it is being posted without testing. This is not happening, so reviewers have no idea what (if any) testing has been done and are often forced to ask (multiple times) before getting a response from the author. Our goal should be to make everyone productive (both authors and reviewers). By putting in the upfront effort to test changes before reviews, we will have more productive reviews, less iterations and better quality code. Bart From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: October 24, 2018 1:28 PM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Community call Oct 24 notes Agenda and Notes for the Oct 24th call * Let's do a release!!! o Our new web page is live! o Bug status * Only one stx.2018.10 bug still open: https://bugs.launchpad.net/starlingx/+bug/1797234 * Notes say that the issue is not reproducible, so this does not hold up stx.2018.10 - removing the label and recommending closure * A quick review of the new bugs in the last couple of days does not reveal in any issues that would hold up the release * Erich to file a new launchpad for the latest build issues on master - potential build blocking issues in the mirror script * RT kernel package location - to be discussed over email... * https://review.openstack.org/#/c/612822/ o Testing status o Docs status * Release note status o Backporting policy??? * Changes that are candidates for backporting should be sent to the [Release] team in email. * Release team will review and decide if a backport is required. o Release tagging is on hold until EOD today, with a strong desire to get the build changes in first. * Priorities for r/2019.03 o See https://etherpad.openstack.org/p/stx-2019.03-priorities o Strongly encourage sub-projects TL/PLs to put together their lists/plans and update the etherpad asap. * When to post code for review? Early and often? After testing? o https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines o We should consider clarifying the early review needed to catch design issues between the final code inspection that gates code acceptance o The Intel team in particular would benefit from early/often review o The goal is to help close the gap on detailed design approaches where the spec may not cover that level of detail, but the developer needs input on the approach before being able to test. o There are mechanisms that the developer should use to make it clear that they are requesting early feedback by including WIP or RFC in the first line of the commit message and/or Workflow -1 by the review submitter o Core reviewer workload is high but we can offload that to Contributors who are all able to review code. * How to test complex changes? o Leverage the virutal environment where possible e.g. in the current Ceph changes that need testing can be done in a virtual duplex config o Other changes that require real hardware should be discussed with the Test team. * Thank you to everyone for all your work on the release! * The new website is live! * @StarlingX on twitter is live, please re-tweet. * Please join the webinar at 8AM PT today. -------------- next part -------------- An HTML attachment was scrubbed... URL: From sgw at linux.intel.com Thu Oct 25 19:27:08 2018 From: sgw at linux.intel.com (Saul Wold) Date: Thu, 25 Oct 2018 12:27:08 -0700 Subject: [Starlingx-discuss] Makefile Standardization Spec Message-ID: <254ae887-e7d8-16be-4330-0af1d3fe77ad@linux.intel.com> Ian & reset of TSC: Erich just pushed a fixed version of the Makefile spec (https://review.openstack.org/#/c/610672/) , there was no content change, just the update to fix Tox Lint issues and moving it to the 2019.03 directory. Will this restart the 2 day timers? Thanks Sau! From sgw at linux.intel.com Thu Oct 25 19:31:03 2018 From: sgw at linux.intel.com (Saul Wold) Date: Thu, 25 Oct 2018 12:31:03 -0700 Subject: [Starlingx-discuss] usage of ldapusersetup? In-Reply-To: References: <65c2076e-50d4-7ea1-0c61-198797af7305@linux.intel.com> Message-ID: On 10/25/2018 06:53 AM, Andy Ning wrote: > Yes ldapusersetup is our addition to ldapscripts as one more helper > utility. It's using other scripts in the ldapscripts package. > ldapusersetup together with the original scripts in ldapscripts package > are all wrappers of the utilities provided by openldap-clients. > > So we could move it to a separate package. Or, we could push to > ldapscripts upstream as a nice addition. > Andy We will start with a separate package. Are you the original author; do you want to work with the ldapscripts upstream to get the ldapusersetup upstreamed? Thanks Sau! > Andy > > > On 2018-10-24 01:20 PM, Eslimi, Dariush wrote: >> Saul, >> >> The script is helper script for end users, you can move it. >> >> Dariush >> >> -----Original Message----- >> From: Saul Wold [mailto:sgw at linux.intel.com] >> Sent: October-24-18 1:00 PM >> To: starlingx-discuss at lists.starlingx.io; Eslimi, Dariush; >> shuicheng.lin at intel.com >> Subject: usage of ldapusersetup? >> >> >> Dariush, Shuichen: >> >> As I was reviewing the lshell removal patch I saw that it is changes >> the way a new script (ldapusersetup) is added to ldapscripts. I >> started to dig deeper to see what uses the the ldapusersetup script. >> The ldapusersetup appears to be a convenience script that could be >> moved to either a standalone package for StarlingX or if we have an >> existing package for such scripts. >> >> I don't see this script actually being used by any other tools, so >> does it have documented usage somewhere? >> >> See existing Gerrit Review and comments: >> https://review.openstack.org/#/c/610761/7 >> >> Sau! >> _______________________________________________ >> Starlingx-discuss mailing list >> Starlingx-discuss at lists.starlingx.io >> http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss > From dtroyer at gmail.com Thu Oct 25 20:24:15 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Thu, 25 Oct 2018 15:24:15 -0500 Subject: [Starlingx-discuss] Makefile Standardization Spec In-Reply-To: <254ae887-e7d8-16be-4330-0af1d3fe77ad@linux.intel.com> References: <254ae887-e7d8-16be-4330-0af1d3fe77ad@linux.intel.com> Message-ID: On Thu, Oct 25, 2018 at 2:27 PM Saul Wold wrote: > Erich just pushed a fixed version of the Makefile spec > (https://review.openstack.org/#/c/610672/) , there was no content > change, just the update to fix Tox Lint issues and moving it to the > 2019.03 directory. > > Will this restart the 2 day timers? I don't think it should need to but at least one of us should verify that it was only format-type changes then +W it again. If content changed then I would think the review cycle starts over. dt -- Dean Troyer dtroyer at gmail.com From jesus.ornelas.aguayo at intel.com Thu Oct 25 20:30:22 2018 From: jesus.ornelas.aguayo at intel.com (Ornelas Aguayo, Jesus) Date: Thu, 25 Oct 2018 20:30:22 +0000 Subject: [Starlingx-discuss] Koji discovery Message-ID: Hi, As mentioned in the starlingx build meeting today, you can start exploring koji by running the next steps using a centos clean environment installed with the CentOS-7-x86_64-Minimal-1804.iso. curl -O https://raw.githubusercontent.com/chuyd/koji-dojo/baremetal/baremetal.sh chmod +x ./baremetal.sh sudo ./baremetal To test your access please run: koji hello Note: If you want to use the koji-cli in a different machine, please use the certificates that are stored in /opt/koji-certificates. For more information about the koji usage please observe: https://docs.pagure.org/koji/using_the_koji_build_system/ https://fedoraproject.org/wiki/Using_the_Koji_build_system https://fedoraproject.org/wiki/Koji From sgw at linux.intel.com Thu Oct 25 20:30:38 2018 From: sgw at linux.intel.com (Saul Wold) Date: Thu, 25 Oct 2018 13:30:38 -0700 Subject: [Starlingx-discuss] Makefile Standardization Spec In-Reply-To: References: <254ae887-e7d8-16be-4330-0af1d3fe77ad@linux.intel.com> Message-ID: On 10/25/2018 01:24 PM, Dean Troyer wrote: > On Thu, Oct 25, 2018 at 2:27 PM Saul Wold wrote: >> Erich just pushed a fixed version of the Makefile spec >> (https://review.openstack.org/#/c/610672/) , there was no content >> change, just the update to fix Tox Lint issues and moving it to the >> 2019.03 directory. >> >> Will this restart the 2 day timers? > > I don't think it should need to but at least one of us should verify > that it was only format-type changes then +W it again. If content > changed then I would think the review cycle starts over. > It's white space and fixing a couple of URL's to the stx-*, no actual content changes. Sau! > dt > From Ian.Jolliffe at windriver.com Thu Oct 25 20:35:26 2018 From: Ian.Jolliffe at windriver.com (Jolliffe, Ian) Date: Thu, 25 Oct 2018 20:35:26 +0000 Subject: [Starlingx-discuss] Makefile Standardization Spec In-Reply-To: References: <254ae887-e7d8-16be-4330-0af1d3fe77ad@linux.intel.com> Message-ID: On 10/25/2018 01:24 PM, Dean Troyer wrote: > On Thu, Oct 25, 2018 at 2:27 PM Saul Wold wrote: >> Erich just pushed a fixed version of the Makefile spec >> (https://review.openstack.org/#/c/610672/) , there was no content >> change, just the update to fix Tox Lint issues and moving it to the >> 2019.03 directory. >> >> Will this restart the 2 day timers? > > I don't think it should need to but at least one of us should verify > that it was only format-type changes then +W it again. If content > changed then I would think the review cycle starts over. > It's white space and fixing a couple of URL's to the stx-*, no actual content changes. Sau! I will have a look and push it through for formatting only Ian > dt > From cesar.lara at intel.com Thu Oct 25 20:49:21 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Thu, 25 Oct 2018 20:49:21 +0000 Subject: [Starlingx-discuss] [build][meetings] Meeting minutes 10/25/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB471053E18D@fmsmsx104.amr.corp.intel.com> Build team meeting 10/25/2018 Attendees Ken, Scott, Jason, Saul, Humberto, Mario, Chuy(Jesus), Victor, Cesar Agenda - Cengn mirror status - Ken - Cengn hosting spec discussion - Erich - Koji walkthrough - Victor R - Koji Docker file - Chuy Deferred for next week - Plans for next release - opens Notes - Cengn mirror status Server is alive and available for access, mirror is being updated starting with 3rd party content. Hosted mirror will also have HTTP access available. Team is working on a few details on the implementation of this new mirror, the plan is to have it fully functional by November 5th 2018, just before the OpenStack summit. Cengn mirror should be the authoritative place for content download. Update of packages in Cengn mirror stills WIP ISOs and other artifacts are going to be out of scope for this mirror and November 5th plan. - Cengn hosting spec discussion All comments were addressed and spec is now +1 and approved, we as a team are all for working through specs for new build requirements in the future. - Presentation & Demo for Koji AR - Cesar to post slides and Chuy to post link to his repo for Koji setup to ML Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From erich.cordoba.malibran at intel.com Thu Oct 25 20:54:59 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Thu, 25 Oct 2018 20:54:59 +0000 Subject: [Starlingx-discuss] Makefile Standardization Spec In-Reply-To: References: <254ae887-e7d8-16be-4330-0af1d3fe77ad@linux.intel.com> Message-ID: <881b26e3f16c6c8f177b154e5ad14f29bd3e26ad.camel@intel.com> Thank you all :) On Thu, 2018-10-25 at 20:35 +0000, Jolliffe, Ian wrote: > On 10/25/2018 01:24 PM, Dean Troyer wrote: > > On Thu, Oct 25, 2018 at 2:27 PM Saul Wold > wrote: > >> Erich just pushed a fixed version of the Makefile spec > >> (https://review.openstack.org/#/c/610672/) , there was no > content > >> change, just the update to fix Tox Lint issues and moving it > to the > >> 2019.03 directory. > >> > >> Will this restart the 2 day timers? > > > > I don't think it should need to but at least one of us should > verify > > that it was only format-type changes then +W it again. If > content > > changed then I would think the review cycle starts over. > > > It's white space and fixing a couple of URL's to the stx-*, no > actual > content changes. > > Sau! > > I will have a look and push it through for formatting only > > Ian > > > > dt > > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ghada.Khalil at windriver.com Thu Oct 25 22:17:12 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Thu, 25 Oct 2018 22:17:12 +0000 Subject: [Starlingx-discuss] [Process] Updates to the StarlingX Code Submission Guidelines Message-ID: <151EE31B9FCCA54397A757BC674650F0BA47010E@ALA-MBD.corp.ad.wrs.com> Hello all, There have been several discussions in community meetings and on the mailing lists related to reviews / testing / early feedback. I've updated the StarlingX Code Submission wiki to capture the key recommendations. https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines Please have a look. Any feedback is more than welcome. Regards, Ghada From yong.hu at intel.com Fri Oct 26 02:46:08 2018 From: yong.hu at intel.com (Hu, Yong) Date: Fri, 26 Oct 2018 02:46:08 +0000 Subject: [Starlingx-discuss] Koji discovery Message-ID: Hi Chuy, Is this online build system for StarlingX packages and ISO? On 26/10/2018, 4:30 AM, "Ornelas Aguayo, Jesus" wrote: Hi, As mentioned in the starlingx build meeting today, you can start exploring koji by running the next steps using a centos clean environment installed with the CentOS-7-x86_64-Minimal-1804.iso. curl -O https://raw.githubusercontent.com/chuyd/koji-dojo/baremetal/baremetal.sh chmod +x ./baremetal.sh sudo ./baremetal To test your access please run: koji hello Note: If you want to use the koji-cli in a different machine, please use the certificates that are stored in /opt/koji-certificates. For more information about the koji usage please observe: https://docs.pagure.org/koji/using_the_koji_build_system/ https://fedoraproject.org/wiki/Using_the_Koji_build_system https://fedoraproject.org/wiki/Koji _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From huang.shuquan at 99cloud.net Fri Oct 26 08:43:27 2018 From: huang.shuquan at 99cloud.net (Shuquan Huang) Date: Fri, 26 Oct 2018 16:43:27 +0800 Subject: [Starlingx-discuss] TSC meeting notes Oct 25th 2018 Message-ID: <765F1484-A3C0-4134-B67D-72AD909FD44B@99cloud.net> Hello all, It’s my honor to join StarlingX TSC. I’m very excited and looking forward to working with you guys on StarlingX. ☺ From: on behalf of "Jones, Bruce E" Date: Friday, October 26, 2018 at 1:17 AM To: "starlingx-discuss at lists.starlingx.io" Subject: [Starlingx-discuss] TSC meeting notes Oct 25th 2018 https://etherpad.openstack.org/p/stx-cores 10/25/18 TSC member update - two new members, one is pending submission of a candidate name Welcome Curtis Collicutt and Shuquan Huang! Final governance approval and election schedule to be determined/approved by the TSC once it reaches 7 members Review any pending stx-gov reviews Please review https://review.openstack.org/#/c/611185/ https://review.openstack.org/#/c/610446/ Enforcing core mappings to subprojects/repos (Brent) RIght now if you are a Core for one project you are a Core for everything. Gerrit groups are being used. Brent to publish the mapping and the team can work with Ildiko to limit Cores to the projects they are Cores for. Revist spec approval process (Brent) o We have reviewed a couple of specs now and some have gotten to the +2 step, then someone came along later with a -1 because they had a question. Not all -1's are equal. What are the community norms for reviewing specs? Sometimes people may feel like their questions are important and might be ignored without a -1. But we don't want to block forward progress on specs. -1 votes are cleared for new patch sets but not for -2s which must be manually approved. o We would like the community to use -1 on specs for important issues with specs. Or important questions. It's a judgment call. This also applies to code reviews. Review and update Project Priorities: https://etherpad.openstack.org/p/stx-2019.03-priorities We want to finish this quickly - March is coming! Bruce to turn the etherpad into a more structured document that can be used to describe the work in more detail, show which projects are effected, etc... Review any pending stx-specs reviews Please help review https://review.openstack.org/#/c/611725/ Retirement of more staging trees, general discussion about strategy and plans Review submission process (Brent) We had a review self-approved that went into the release yesterday that went in the day before the cut-off. Ghada to follow up. IA Features sub-project - Bruce o https://wiki.openstack.org/wiki/StarlingX/IA_Features o We will be upstream first for new features e.g. Nova. o We should discuss this as part of an overall Test strategy o Making this a sub-project is probably not the right approach. Intel may be very interested but what if we had an ARM vendor feature list? Vendors will assign resources to things that they think are important. o Bruce to track on the wiki but not as a sub-project. Presentation at the joint leadership meeting in Berlin (Nov 12) Brent to cover, other project members are welcome to attend _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From vm.rod25 at gmail.com Fri Oct 26 16:07:13 2018 From: vm.rod25 at gmail.com (Victor Rodriguez) Date: Fri, 26 Oct 2018 11:07:13 -0500 Subject: [Starlingx-discuss] [build][meetings] Meeting minutes 10/25/2018 In-Reply-To: <0B566C62EC792145B40E29EFEBF1AB471053E18D@fmsmsx104.amr.corp.intel.com> References: <0B566C62EC792145B40E29EFEBF1AB471053E18D@fmsmsx104.amr.corp.intel.com> Message-ID: On Thu, Oct 25, 2018 at 3:49 PM Lara, Cesar wrote: > > Build team meeting 10/25/2018 > > Attendees > > Ken, Scott, Jason, Saul, Humberto, Mario, Chuy(Jesus), Victor, Cesar > > > > Agenda > > - Cengn mirror status – Ken > > - Cengn hosting spec discussion – Erich > > - Koji walkthrough - Victor R > > - Koji Docker file – Chuy > > > > Deferred for next week > > - Plans for next release > > - opens > > > > Notes > > - Cengn mirror status > > Server is alive and available for access, mirror is being updated starting with 3rd party content. > > Hosted mirror will also have HTTP access available. > > Team is working on a few details on the implementation of this new mirror, the plan is to have it fully functional by November 5th 2018, just before the OpenStack summit. > > Cengn mirror should be the authoritative place for content download. > > Update of packages in Cengn mirror stills WIP > > ISOs and other artifacts are going to be out of scope for this mirror and November 5th plan. > > > > - Cengn hosting spec discussion > > All comments were addressed and spec is now +1 and approved, we as a team are all for working through specs for new build requirements in the future. > > > > - Presentation & Demo for Koji > > AR – Cesar to post slides and Chuy to post link to his repo for Koji setup to ML > > > > Regards > > > > Cesar Lara > > Software Engineering Manager > > OpenSource Technology Center > > > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss Hi team Attached are the slides from the koji presentation If you have any question about this topic, please let us know Best regards Victor Rodriguez -------------- next part -------------- A non-text attachment was scrubbed... Name: Koji 101.pdf Type: application/pdf Size: 1492759 bytes Desc: not available URL: From hayde.martinez.landa at intel.com Fri Oct 26 16:18:36 2018 From: hayde.martinez.landa at intel.com (Martinez Landa, Hayde) Date: Fri, 26 Oct 2018 16:18:36 +0000 Subject: [Starlingx-discuss] General question about docs instruction Message-ID: Hi Team, I was reviewing the developer guide [0] and on step 2 under “Download Source Code Repositories” section I see: $ repo init -u https://git.starlingx.io/stx-manifest -m default.xml This specific line brings the source code from master, but if someone wants to use the branch r/2018.10 The line has to be: $ repo init -u https://git.starlingx.io/stx-manifest.git -m default.xml -b r/2018.10 Shouldn't the docs include this above line maybe as a suggestion for users to correctly download from release branch? [0] https://docs.starlingx.io/developer_guide/index.html#download-source-code-repositories From michael.l.tullis at intel.com Fri Oct 26 16:23:19 2018 From: michael.l.tullis at intel.com (Tullis, Michael L) Date: Fri, 26 Oct 2018 16:23:19 +0000 Subject: [Starlingx-discuss] General question about docs instruction In-Reply-To: References: Message-ID: <3808363B39586544A6839C76CF81445EA1A4DB73@ORSMSX104.amr.corp.intel.com> Thanks for the feedback Hayde. I'll raise this issue ASAP with the full docs team, and we'll update this example in the developer guide as needed. Thx. -- Mike -----Original Message----- From: Martinez Landa, Hayde [mailto:hayde.martinez.landa at intel.com] Sent: Friday, October 26, 2018 10:19 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] General question about docs instruction Hi Team, I was reviewing the developer guide [0] and on step 2 under “Download Source Code Repositories” section I see: $ repo init -u https://git.starlingx.io/stx-manifest -m default.xml This specific line brings the source code from master, but if someone wants to use the branch r/2018.10 The line has to be: $ repo init -u https://git.starlingx.io/stx-manifest.git -m default.xml -b r/2018.10 Shouldn't the docs include this above line maybe as a suggestion for users to correctly download from release branch? [0] https://docs.starlingx.io/developer_guide/index.html#download-source-code-repositories _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From humberto.i.perez.rodriguez at intel.com Fri Oct 26 16:24:59 2018 From: humberto.i.perez.rodriguez at intel.com (Perez Rodriguez, Humberto I) Date: Fri, 26 Oct 2018 16:24:59 +0000 Subject: [Starlingx-discuss] General question about docs instruction Message-ID: <46942C29-842B-4960-AB53-1E9B5A6A67F3@intel.com> I agree with Hayde ... On 10/26/18, 11:18 AM, "Martinez Landa, Hayde" wrote: Hi Team, I was reviewing the developer guide [0] and on step 2 under “Download Source Code Repositories” section I see: $ repo init -u https://git.starlingx.io/stx-manifest -m default.xml This specific line brings the source code from master, but if someone wants to use the branch r/2018.10 The line has to be: $ repo init -u https://git.starlingx.io/stx-manifest.git -m default.xml -b r/2018.10 Shouldn't the docs include this above line maybe as a suggestion for users to correctly download from release branch? [0] https://docs.starlingx.io/developer_guide/index.html#download-source-code-repositories _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From bruce.e.jones at intel.com Fri Oct 26 21:53:13 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Fri, 26 Oct 2018 21:53:13 +0000 Subject: [Starlingx-discuss] AR from TSC - Project Priorities spreadsheet Message-ID: <9A85D2917C58154C960D95352B22818BB1E85D57@ORSMSX158.amr.corp.intel.com> I took at AR at the last TSC meeting to turn our current working Project Priorities Etherpad [0] into a spreadsheet. It's done. You can find it in an Ethercalc [1]. I copied work items over from the Etherpad, and added a brief description and justification for each. I gave each a Priority which right now is just the order that the items are listed by in the Etherpad. I took the liberty of adding a few items at the bottom that I've been thinking about. The spreadsheet includes placeholders for each team. I'm going to ask at the community meeting next Wednesday that the TLs and PLs for each team get together, make their wish list for the March release, and update the Ethercalc. Feedback / comments / changes welcomed graciously! Brucej [0] https://etherpad.openstack.org/p/stx-2019.03-priorities [1] https://ethercalc.openstack.org/fafyo2729fnr -------------- next part -------------- An HTML attachment was scrubbed... URL: From jesus.ornelas.aguayo at intel.com Fri Oct 26 22:10:15 2018 From: jesus.ornelas.aguayo at intel.com (Ornelas Aguayo, Jesus) Date: Fri, 26 Oct 2018 22:10:15 +0000 Subject: [Starlingx-discuss] Koji discovery In-Reply-To: References: Message-ID: Hi Yong, The steps mentioned in the first mail describes how to install your own koji, the instructions were tested using the http://isoredirect.centos.org/centos/7/isos/x86_64/CentOS-7-x86_64-Minimal-1804.iso on a clean environment. -----Original Message----- From: "Hu, Yong" Date: Thursday, October 25, 2018 at 9:46 PM To: "Ornelas Aguayo, Jesus" , "starlingx-discuss at lists.starlingx.io" Subject: Re: [Starlingx-discuss] Koji discovery Hi Chuy, Is this online build system for StarlingX packages and ISO? On 26/10/2018, 4:30 AM, "Ornelas Aguayo, Jesus" wrote: Hi, As mentioned in the starlingx build meeting today, you can start exploring koji by running the next steps using a centos clean environment installed with the CentOS-7-x86_64-Minimal-1804.iso. curl -O https://raw.githubusercontent.com/chuyd/koji-dojo/baremetal/baremetal.sh chmod +x ./baremetal.sh sudo ./baremetal To test your access please run: koji hello Note: If you want to use the koji-cli in a different machine, please use the certificates that are stored in /opt/koji-certificates. For more information about the koji usage please observe: https://docs.pagure.org/koji/using_the_koji_build_system/ https://fedoraproject.org/wiki/Using_the_Koji_build_system https://fedoraproject.org/wiki/Koji _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From juan.p.gomez at intel.com Fri Oct 26 23:31:11 2018 From: juan.p.gomez at intel.com (Gomez, Juan P) Date: Fri, 26 Oct 2018 23:31:11 +0000 Subject: [Starlingx-discuss] unexpected behaviour in duplex mode in virtual env In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217F0@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com>, <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> <2F61A2101B4FD447A683EF7E362CF9421C63E217F0@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: <0483622846A57742B81A944248DD690425B16AEE@fmsmsx101.amr.corp.intel.com> Hi Rajesh, There have been some changes recently merged I suggest to build again your ISO with the latest changes There were not screen shot attached, could you share them again? Could you verify the output from the cmd: nova service-list, verify the nova-compute services is up and enable? What is the output from system host-list? Best Regards, JP -----Original Message----- From: Rajesh Jha [mailto:Rajesh.Jha at rsystems.com] Sent: Thursday, October 25, 2018 5:11 AM To: starlingx-discuss at lists.starlingx.io Cc: Sachin Alok ; Sumit Agarwal ; Avanish Yadav ; Ajay Tomar ; NK Garg Subject: [Starlingx-discuss] unexpected behaviour in duplex mode in virtual env Hi, We have observed an unstable behaviour while running below scenario Setup- Virtual environment testing using kvm Mode "Duplex" (One active and one standby controller script setup_allinone.sh) Steps- 1. Followed steps in document "https://docs.starlingx.io/installation_guide/duplex.html#duplex" 2. Able to proceed till section "system host-unlock controller-0" command. After issuing this command system get reboot as per expectation but once it comes up availability status changed to "degrade" and alarm raised (screen shot attached). 3. Then we execute "destroy_allinone.sh" and "destroy_network.sh" command to cleanup the previous setup to build new one 4. After following same step in above documents till "sudo config_controller" command which was successful and after run "source /etc/nova/openrc" system get reboot again and again and nothing happening after instead of rebooting. Screenshot is attached below. CLI is no more available to the controller Thank you, Rajesh The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Sun Oct 28 03:12:30 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Sun, 28 Oct 2018 03:12:30 +0000 Subject: [Starlingx-discuss] Weekly StarlingX non-OpenStack Distro meeting, 10/31 Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35DA98A3@SHSMSX104.ccr.corp.intel.com> All, I have regular agenda for this Wed's call but I am OOP this week and not able to join/host the meeting. Yong Hu from my team will drive the meeting. If you have any open agenda want to add, please reply to the ML. Thx. - cindy Agenda for 10/31 meeting: - bug review for non-openStack distro (Yong) - non-Openstack patch refactor status (Zhipeng/Shuicheng) - non-Openstack patch upstream status (Saul) - Ceph upgrade status (Vivian/Dehao) - Opens (All) -----Original Appointment----- From: Xie, Cindy Sent: Thursday, August 23, 2018 6:38 PM To: Xie, Cindy; 'Ildiko Vancsa'; Wold, Saul; Chen, Yan; Somerville, Jim; Wang, Yi C; Khalil, Ghada; starlingx-discuss at lists.starlingx.io; Sun, Austin; Rowsell, Brent; Lin, Shuicheng Cc: Hu, Yong; Martinez Monroy, Elio; 'Leo Xu'; 'Young, Ken'; Gomez, Juan P; Shang, Dehao; 'Eslimi, Dariush'; Qi, Mingyuan; Hernandez Gonzalez, Fernando; 'Waines, Greg'; Perez Rodriguez, Humberto I; Zhu, Vivian; Arce Moreno, Abraham; 'Chen, Jacky'; Jones, Bruce E; Cobbley, David A; 'Komiyama, Takeo'; Hu, Wei W; 'Seiler, Glenn' Subject: Weekly StarlingX non-OpenStack Distro meeting When: Wednesday, October 31, 2018 9:00 PM-10:00 PM (UTC+08:00) Beijing, Chongqing, Hong Kong, Urumqi. Where: https://zoom.us/j/342730236 * Cadence and time slot: o Wednesday 9AM EDT (9PM China time) * Call Details: o Zoom link: https://zoom.us/j/342730236 o Dialing in from phone: o Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 876 9923 o Meeting ID: 342 730 236 o International numbers available: https://zoom.us/u/ed95sU7aQ * Meeting Agenda and Minutes: o https://etherpad.openstack.org/p/stx-distro-other -------------- next part -------------- An HTML attachment was scrubbed... URL: From Rajesh.Jha at rsystems.com Sun Oct 28 10:12:55 2018 From: Rajesh.Jha at rsystems.com (Rajesh Jha) Date: Sun, 28 Oct 2018 15:42:55 +0530 Subject: [Starlingx-discuss] unexpected behaviour in duplex mode in virtual env In-Reply-To: <0483622846A57742B81A944248DD690425B16AEE@fmsmsx101.amr.corp.intel.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com>, <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> <2F61A2101B4FD447A683EF7E362CF9421C63E217F0@RSI-NOD-EXMDB6.india.rsystems.com>, <0483622846A57742B81A944248DD690425B16AEE@fmsmsx101.amr.corp.intel.com> Message-ID: <2F61A2101B4FD447A683EF7E362CF9421C63E21801@RSI-NOD-EXMDB6.india.rsystems.com> Hi Juan, Thanks for the update. I will be making the iso build and retest the scenario . Will send the screens shot in testing this time. Best Regards, Rajesh Jha ________________________________________ From: Gomez, Juan P [juan.p.gomez at intel.com] Sent: Saturday, October 27, 2018 5:01 AM To: Rajesh Jha; starlingx-discuss at lists.starlingx.io Cc: Sachin Alok; Sumit Agarwal; Avanish Yadav; Ajay Tomar; NK Garg Subject: RE: [Starlingx-discuss] unexpected behaviour in duplex mode in virtual env Hi Rajesh, There have been some changes recently merged I suggest to build again your ISO with the latest changes There were not screen shot attached, could you share them again? Could you verify the output from the cmd: nova service-list, verify the nova-compute services is up and enable? What is the output from system host-list? Best Regards, JP -----Original Message----- From: Rajesh Jha [mailto:Rajesh.Jha at rsystems.com] Sent: Thursday, October 25, 2018 5:11 AM To: starlingx-discuss at lists.starlingx.io Cc: Sachin Alok ; Sumit Agarwal ; Avanish Yadav ; Ajay Tomar ; NK Garg Subject: [Starlingx-discuss] unexpected behaviour in duplex mode in virtual env Hi, We have observed an unstable behaviour while running below scenario Setup- Virtual environment testing using kvm Mode "Duplex" (One active and one standby controller script setup_allinone.sh) Steps- 1. Followed steps in document "https://docs.starlingx.io/installation_guide/duplex.html#duplex" 2. Able to proceed till section "system host-unlock controller-0" command. After issuing this command system get reboot as per expectation but once it comes up availability status changed to "degrade" and alarm raised (screen shot attached). 3. Then we execute "destroy_allinone.sh" and "destroy_network.sh" command to cleanup the previous setup to build new one 4. After following same step in above documents till "sudo config_controller" command which was successful and after run "source /etc/nova/openrc" system get reboot again and again and nothing happening after instead of rebooting. Screenshot is attached below. CLI is no more available to the controller Thank you, Rajesh The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. From zhipengs.liu at intel.com Mon Oct 29 01:40:17 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Mon, 29 Oct 2018 01:40:17 +0000 Subject: [Starlingx-discuss] Can we remove the ghostscript patch of netpbm and use RPM instead of SRPM Message-ID: <93814834B4855241994F290E959305C752FEE99F@SHSMSX104.ccr.corp.intel.com> Hi all, We have a ticket as below https://storyboard.openstack.org/#!/story/2003758 netpbm: Test what happens when we remove the ghostscript patch For package build, it can work after remove the patch. Furthermore, I saw that this package is not packaged to iso. After remove this package and use below 2 RPM instead of SRPM, iso can be built out and deployment test pass. netpbm-10.79.00-7.el7.x86_64.rpm netpbm-progs-10.79.00-7.el7.x86_64.rpm. Any concern? If it is OK, I will raise a patch to finish it. Thanks! Zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From shuicheng.lin at intel.com Mon Oct 29 02:19:48 2018 From: shuicheng.lin at intel.com (Lin, Shuicheng) Date: Mon, 29 Oct 2018 02:19:48 +0000 Subject: [Starlingx-discuss] Can we remove the ghostscript patch of netpbm and use RPM instead of SRPM In-Reply-To: <93814834B4855241994F290E959305C752FEE99F@SHSMSX104.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FEE99F@SHSMSX104.ccr.corp.intel.com> Message-ID: <9700A18779F35F49AF027300A49E7C765FE3859C@SHSMSX101.ccr.corp.intel.com> Hi Zhipeng, Since netpbm is not included in ISO, and no other package depends on it, I think we could just remove it. Replace with RPM is not needed. Best Regards Shuicheng From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: Monday, October 29, 2018 9:40 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Can we remove the ghostscript patch of netpbm and use RPM instead of SRPM Hi all, We have a ticket as below https://storyboard.openstack.org/#!/story/2003758 netpbm: Test what happens when we remove the ghostscript patch For package build, it can work after remove the patch. Furthermore, I saw that this package is not packaged to iso. After remove this package and use below 2 RPM instead of SRPM, iso can be built out and deployment test pass. netpbm-10.79.00-7.el7.x86_64.rpm netpbm-progs-10.79.00-7.el7.x86_64.rpm. Any concern? If it is OK, I will raise a patch to finish it. Thanks! Zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From zhipengs.liu at intel.com Mon Oct 29 05:10:23 2018 From: zhipengs.liu at intel.com (Liu, ZhipengS) Date: Mon, 29 Oct 2018 05:10:23 +0000 Subject: [Starlingx-discuss] Can we remove the ghostscript patch of netpbm and use RPM instead of SRPM In-Reply-To: <9700A18779F35F49AF027300A49E7C765FE3859C@SHSMSX101.ccr.corp.intel.com> References: <93814834B4855241994F290E959305C752FEE99F@SHSMSX104.ccr.corp.intel.com> <9700A18779F35F49AF027300A49E7C765FE3859C@SHSMSX101.ccr.corp.intel.com> Message-ID: <93814834B4855241994F290E959305C752FEEBAB@SHSMSX104.ccr.corp.intel.com> Hi Water, There are some build dependency on netpbm. Thanks! Zhipeng From: Lin, Shuicheng Sent: 2018年10月29日 10:20 To: Liu, ZhipengS ; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] Can we remove the ghostscript patch of netpbm and use RPM instead of SRPM Hi Zhipeng, Since netpbm is not included in ISO, and no other package depends on it, I think we could just remove it. Replace with RPM is not needed. Best Regards Shuicheng From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] Sent: Monday, October 29, 2018 9:40 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Can we remove the ghostscript patch of netpbm and use RPM instead of SRPM Hi all, We have a ticket as below https://storyboard.openstack.org/#!/story/2003758 netpbm: Test what happens when we remove the ghostscript patch For package build, it can work after remove the patch. Furthermore, I saw that this package is not packaged to iso. After remove this package and use below 2 RPM instead of SRPM, iso can be built out and deployment test pass. netpbm-10.79.00-7.el7.x86_64.rpm netpbm-progs-10.79.00-7.el7.x86_64.rpm. Any concern? If it is OK, I will raise a patch to finish it. Thanks! Zhipeng -------------- next part -------------- An HTML attachment was scrubbed... URL: From longqian.zhao at intel.com Mon Oct 29 07:55:00 2018 From: longqian.zhao at intel.com (Zhao, Longqian) Date: Mon, 29 Oct 2018 07:55:00 +0000 Subject: [Starlingx-discuss] How to slove "Interface not found: host compute-0 if eth1000." this problem Message-ID: <6325970F810676468E0EC5D8ED76D2190B7C8066@SHSMSX101.ccr.corp.intel.com> Hi, My question: when I run following command: [wrsroot at controller-0 ~(keystone_admin)]$ system host-if-modify -p providernet-a -c data compute-0 eth1000 The console printf following info: Interface not found: host compute-0 if eth1000. ========================================================================================================================== My current environment: +++++++++++++++++++++++++++++++++++++++++++++++++++ In compute-0: compute-0:~$ ifconfig eth1000: 192.168.204.23 lo: 127.0.0.1 ++++++++++++++++++++++++++++++++++++++++++++++++++++ In controller-0: controller-0:~$ ifconfig enp2s1: 10.10.10.3 enp2s2: 192.168.204.3 lo: 127.0.0.1 +++++++++++++++++++++++++++++++++++++++++++++++++++ ===================================================== By the way, I have tested that compute-0 and controller-0 can communicate with each other. I unlock the compute-0 nodes because of this problem. Please give me some tips or help Thanks again, Longqian. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Allain.Legacy at windriver.com Mon Oct 29 12:14:47 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Mon, 29 Oct 2018 12:14:47 +0000 Subject: [Starlingx-discuss] How to slove "Interface not found: host compute-0 if eth1000." this problem In-Reply-To: <6325970F810676468E0EC5D8ED76D2190B7C8066@SHSMSX101.ccr.corp.intel.com> References: <6325970F810676468E0EC5D8ED76D2190B7C8066@SHSMSX101.ccr.corp.intel.com> Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC24BB3@ALA-MBD.corp.ad.wrs.com> Use the "system host-if-list compute-0" command to determine the list of known interfaces and then use the name listed there to issue commands requiring an interface name. Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Zhao, Longqian [mailto:longqian.zhao at intel.com] Sent: Monday, October 29, 2018 3:55 AM To: 'starlingx-discuss at lists.starlingx.io' Cc: Zhao, Longqian Subject: [Starlingx-discuss] How to slove "Interface not found: host compute-0 if eth1000." this problem Hi, My question: when I run following command: [wrsroot at controller-0 ~(keystone_admin)]$ system host-if-modify -p providernet-a -c data compute-0 eth1000 The console printf following info: Interface not found: host compute-0 if eth1000. ========================================================================================================================== My current environment: +++++++++++++++++++++++++++++++++++++++++++++++++++ In compute-0: compute-0:~$ ifconfig eth1000: 192.168.204.23 lo: 127.0.0.1 ++++++++++++++++++++++++++++++++++++++++++++++++++++ In controller-0: controller-0:~$ ifconfig enp2s1: 10.10.10.3 enp2s2: 192.168.204.3 lo: 127.0.0.1 +++++++++++++++++++++++++++++++++++++++++++++++++++ ===================================================== By the way, I have tested that compute-0 and controller-0 can communicate with each other. I unlock the compute-0 nodes because of this problem. Please give me some tips or help Thanks again, Longqian. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From michel.thebeau at windriver.com Mon Oct 29 12:32:25 2018 From: michel.thebeau at windriver.com (Michel Thebeau) Date: Mon, 29 Oct 2018 08:32:25 -0400 Subject: [Starlingx-discuss] Query on Nove and Network test case In-Reply-To: <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> References: <2F61A2101B4FD447A683EF7E362CF9421C63E217E5@RSI-NOD-EXMDB6.india.rsystems.com> <2F61A2101B4FD447A683EF7E362CF9421C63E217E8@RSI-NOD-EXMDB6.india.rsystems.com> Message-ID: <52d10914-ca68-94b7-dc68-5c3fcf88b2a9@windriver.com> On 2018-10-24 7:09 a.m., Rajesh Jha wrote: > Hi, > > While testing nova test cases we are able to make the Nova online successfully. However controller-0 is unable to communicate > with compute-0(nova) and not allowing to add data interface on controller-0. Due to this issue we are not able to execute the > remaining test cases for the compute (nova) and network can be run from controller-0.we also observe sub-functions are missing > from the controller-0. Hi Rajesh, May I assume that your looking at test case 245, "Add up to 4 compute nodes to a duplex deployment", which adds additional compute nodes to a all-in-one duplex install? M > We are following below link > > 1.https://wiki.openstack.org/wiki/StarlingX/stx.2018.10_Testplan_Instructions > 2.https://wiki.openstack.org/wiki/StarlingX/Installation_Guide_Virtual_Environment/Simplex > > running command: "$system host-show 1 | grep subfunctions" does not return - > "| subfunctions | controller,compute" > > Thank you, > Rajesh Jha > > The information contained in this e-mail and any accompanying documents may contain information that is confidential or otherwise protected from disclosure. If you are not the intended recipient of this message, please immediately alert the sender by reply e-mail and then delete this message. Any dissemination, distribution or other use of the contents of this message by anyone other than the intended recipient is strictly prohibited. R Systems has taken every reasonable precaution to minimize the risk, but is not liable for any damage you may sustain as a result of any virus in this e-mail. You should carry out your own virus checks before opening the e-mail or attachment. > > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Ghada.Khalil at windriver.com Mon Oct 29 13:48:12 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Mon, 29 Oct 2018 13:48:12 +0000 Subject: [Starlingx-discuss] [Issue] config_controller failing on stx master loads from 10-26 Message-ID: <151EE31B9FCCA54397A757BC674650F0BA47A06B@ALA-MBD.corp.ad.wrs.com> This is to let everyone know that there is an issue with config_controller on stx master - loads built from October 26 and later. The issue is tracked via Launchpad: https://bugs.launchpad.net/starlingx/+bug/1800466 and is under investigation. Regards, Ghada -------------- next part -------------- An HTML attachment was scrubbed... URL: From austin.sun at intel.com Mon Oct 29 14:20:28 2018 From: austin.sun at intel.com (Sun, Austin) Date: Mon, 29 Oct 2018 14:20:28 +0000 Subject: [Starlingx-discuss] How flat provision network access external network Message-ID: Hi All: In my set up , compute-0 eno1 is connected external network as data network. Is it possible VM on compute-0 can access external network directly ? I tried below commands , but it did not work. :( openstack providernet create providernet.a --type flat openstack network create ext_net --provider-network-type flat --provider-physical-network providernet.a system host-if-modify -p providernet.a -c data compute-0 eno1 openstack server create --flavor vm1.flavor --image cirros --nic net-id= ext_net vm1.tiny Thanks. BR Austin Sun. -------------- next part -------------- An HTML attachment was scrubbed... URL: From Volker.Hoesslin at swsn.de Mon Oct 29 15:09:51 2018 From: Volker.Hoesslin at swsn.de (von Hoesslin, Volker) Date: Mon, 29 Oct 2018 15:09:51 +0000 Subject: [Starlingx-discuss] compute node: lvg without pv ? Message-ID: hi, i have been working on the installation of a starlingX openstack cluster for a few days. now i'm almost at the end of the installation guide and fail due to the connection between nova and ceph :( since my compute nodes only have one HDD (hardware raid1 with 2x SSD) i can't add another HDD (pv) to the lvg. is there a way to work directly on the ceph storage without having to go through a local hard drive? -------------- next part -------------- An HTML attachment was scrubbed... URL: From Allain.Legacy at windriver.com Mon Oct 29 15:22:35 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Mon, 29 Oct 2018 15:22:35 +0000 Subject: [Starlingx-discuss] How flat provision network access external network In-Reply-To: References: Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC25D4B@ALA-MBD.corp.ad.wrs.com> Your question is not clear enough for us to determine what you were trying to do, what you expected to happen and what exactly didn't work. 1. You start by stating that you wanted to use an "external network", but the network create command you used lacks the "--external" option which means you only created a regular tenant network. 2. You did not add a "subnet" to the network so the VM will not be assigned an IP address 3. The server create command listed in your email has a syntax error which would cause it to not run at all (extra space after the net-id=). Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Sun, Austin [mailto:austin.sun at intel.com] Sent: Monday, October 29, 2018 10:20 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] How flat provision network access external network Hi All: In my set up , compute-0 eno1 is connected external network as data network. Is it possible VM on compute-0 can access external network directly ? I tried below commands , but it did not work. :( openstack providernet create providernet.a --type flat openstack network create ext_net --provider-network-type flat --provider-physical-network providernet.a system host-if-modify -p providernet.a -c data compute-0 eno1 openstack server create --flavor vm1.flavor --image cirros --nic net-id= ext_net vm1.tiny Thanks. BR Austin Sun. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From Volker.Hoesslin at swsn.de Mon Oct 29 15:43:58 2018 From: Volker.Hoesslin at swsn.de (von Hoesslin, Volker) Date: Mon, 29 Oct 2018 15:43:58 +0000 Subject: [Starlingx-discuss] move storage node in replication group Message-ID: hello, i have 3 storage nodes and after provisioning nodes 1+2 are in replication group 0 and node 3 are in replication group 1?! how can i move node 3 to group 0? greetings, volker -------------- next part -------------- An HTML attachment was scrubbed... URL: From ildiko.vancsa at gmail.com Mon Oct 29 18:35:50 2018 From: ildiko.vancsa at gmail.com (Ildiko Vancsa) Date: Mon, 29 Oct 2018 19:35:50 +0100 Subject: [Starlingx-discuss] Switching to winter time Message-ID: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> Hi, As we’ve reached the end of DST for this year in Europe already and on Sunday the USA will switch over to winter time as well I would like to ask the preference on the meeting times. Would there be a preference to stick to the UTC time slots for the meetings or we should keep the time slots specified in Pacific Time? Thanks, Ildikó From bruce.e.jones at intel.com Mon Oct 29 19:32:58 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Mon, 29 Oct 2018 19:32:58 +0000 Subject: [Starlingx-discuss] Switching to winter time In-Reply-To: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> References: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> Message-ID: <9A85D2917C58154C960D95352B22818BB1E94B82@fmsmsx117.amr.corp.intel.com> Let's discuss on the Wednesday community call. brucej -----Original Message----- From: Ildiko Vancsa [mailto:ildiko.vancsa at gmail.com] Sent: Monday, October 29, 2018 11:36 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] Switching to winter time Hi, As we’ve reached the end of DST for this year in Europe already and on Sunday the USA will switch over to winter time as well I would like to ask the preference on the meeting times. Would there be a preference to stick to the UTC time slots for the meetings or we should keep the time slots specified in Pacific Time? Thanks, Ildikó _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From dtroyer at gmail.com Mon Oct 29 19:47:24 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Mon, 29 Oct 2018 14:47:24 -0500 Subject: [Starlingx-discuss] Switching to winter time In-Reply-To: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> References: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> Message-ID: On Mon, Oct 29, 2018 at 1:36 PM Ildiko Vancsa wrote: > As we’ve reached the end of DST for this year in Europe already and on Sunday the USA will switch over to winter time as well I would like to ask the preference on the meeting times. > > Would there be a preference to stick to the UTC time slots for the meetings or we should keep the time slots specified in Pacific Time? I have a preference for sticking to UTC time, my experiences doing that for nearly everything OpenStack-related these years has been overall positive. Generally everyone knows how to get from UTC to their local time, it avoids needs to remember when the US changes for those whose local DST rules are not aligned with the US for example. dt -- Dean Troyer dtroyer at gmail.com From erich.cordoba.malibran at intel.com Mon Oct 29 20:04:37 2018 From: erich.cordoba.malibran at intel.com (Cordoba Malibran, Erich) Date: Mon, 29 Oct 2018 20:04:37 +0000 Subject: [Starlingx-discuss] Purpose of wrs-remote-clients package? Message-ID: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> Hi all, I was checking the specfile of wrs-remote-clients package[0] and I can't understand what's the purpose of this package and when is used. It doesn't seems to be installed in the final ISO and according to it's content it looks like a devel package. Also, it has a curious structure because the rpm only installs a compressed file (.tgz) that contains a set of installer scripts and other .tgz files that were grabbed from another rpms. I hope someone can help me to understand the purpose of remote-clients. Thanks in advance. -Erich [0] http://git.starlingx.io/cgit/stx-clients/tree/remote-clients/centos/remote-clients.spec From ildiko.vancsa at gmail.com Mon Oct 29 20:10:18 2018 From: ildiko.vancsa at gmail.com (Ildiko Vancsa) Date: Mon, 29 Oct 2018 21:10:18 +0100 Subject: [Starlingx-discuss] Switching to winter time In-Reply-To: References: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> Message-ID: <2F47C197-D29E-4F23-8C09-CF0D9C204134@gmail.com> Hi, I created a poll to collect data on what the community prefers until the community call on Wednesday: https://doodle.com/poll/pw7mds8qms9d989y Thanks, Ildikó > On 2018. Oct 29., at 20:47, Dean Troyer wrote: > > On Mon, Oct 29, 2018 at 1:36 PM Ildiko Vancsa wrote: >> As we’ve reached the end of DST for this year in Europe already and on Sunday the USA will switch over to winter time as well I would like to ask the preference on the meeting times. >> >> Would there be a preference to stick to the UTC time slots for the meetings or we should keep the time slots specified in Pacific Time? > > I have a preference for sticking to UTC time, my experiences doing > that for nearly everything OpenStack-related these years has been > overall positive. Generally everyone knows how to get from UTC to > their local time, it avoids needs to remember when the US changes for > those whose local DST rules are not aligned with the US for example. +1 While I like the convenience of always knowing when the meeting is, one issue with it is to have collisions with other weekly meetings that do not stick to UTC. In OPNFV the community decided to deal with the switch and not sticking to UTC. > > dt > > -- > > Dean Troyer > dtroyer at gmail.com From andy.ning at windriver.com Mon Oct 29 21:01:43 2018 From: andy.ning at windriver.com (Andy Ning) Date: Mon, 29 Oct 2018 17:01:43 -0400 Subject: [Starlingx-discuss] Purpose of wrs-remote-clients package? In-Reply-To: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> References: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> Message-ID: It's part of the SDK. Refer to Winder River Titanium Cloud Software Development Kit. Andy On 2018-10-29 04:04 PM, Cordoba Malibran, Erich wrote: > Hi all, > > I was checking the specfile of wrs-remote-clients package[0] and I > can't understand what's the purpose of this package and when is used. > > It doesn't seems to be installed in the final ISO and according to it's > content it looks like a devel package. Also, it has a curious structure > because the rpm only installs a compressed file (.tgz) that contains a > set of installer scripts and other .tgz files that were grabbed from > another rpms. > > > I hope someone can help me to understand the purpose of remote-clients. > > Thanks in advance. > > -Erich > > > [0] http://git.starlingx.io/cgit/stx-clients/tree/remote-clients/centos/remote-clients.spec > _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -- Andy Ning Cube: 3071 Tel: 613-9631408 (int: 4408) Skype: andy.ning.wr From bruce.e.jones at intel.com Mon Oct 29 21:13:48 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Mon, 29 Oct 2018 21:13:48 +0000 Subject: [Starlingx-discuss] Purpose of wrs-remote-clients package? In-Reply-To: References: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> Message-ID: <9A85D2917C58154C960D95352B22818BB1E99BC6@fmsmsx117.amr.corp.intel.com> StarlingX isn't Titanium Cloud. Is the SDK part of StarlingX? brucej -----Original Message----- From: Andy Ning [mailto:andy.ning at windriver.com] Sent: Monday, October 29, 2018 2:02 PM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? It's part of the SDK. Refer to Winder River Titanium Cloud Software Development Kit. Andy On 2018-10-29 04:04 PM, Cordoba Malibran, Erich wrote: > Hi all, > > I was checking the specfile of wrs-remote-clients package[0] and I > can't understand what's the purpose of this package and when is used. > > It doesn't seems to be installed in the final ISO and according to > it's content it looks like a devel package. Also, it has a curious > structure because the rpm only installs a compressed file (.tgz) that > contains a set of installer scripts and other .tgz files that were > grabbed from another rpms. > > > I hope someone can help me to understand the purpose of remote-clients. > > Thanks in advance. > > -Erich > > > [0] > http://git.starlingx.io/cgit/stx-clients/tree/remote-clients/centos/re > mote-clients.spec _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -- Andy Ning Cube: 3071 Tel: 613-9631408 (int: 4408) Skype: andy.ning.wr _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From cindy.xie at intel.com Mon Oct 29 23:18:32 2018 From: cindy.xie at intel.com (Xie, Cindy) Date: Mon, 29 Oct 2018 23:18:32 +0000 Subject: [Starlingx-discuss] Purpose of wrs-remote-clients package? In-Reply-To: <9A85D2917C58154C960D95352B22818BB1E99BC6@fmsmsx117.amr.corp.intel.com> References: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> <9A85D2917C58154C960D95352B22818BB1E99BC6@fmsmsx117.amr.corp.intel.com> Message-ID: <2FD5DDB5A04D264C80D42CA35194914F35DAD23B@SHSMSX104.ccr.corp.intel.com> Bruce, I believe this is part of SDK for the tenant OS, part of HA design to get tenant heart-beat. Thx. - cindy -----Original Message----- From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Monday, October 29, 2018 2:14 PM To: Andy Ning ; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? StarlingX isn't Titanium Cloud. Is the SDK part of StarlingX? brucej -----Original Message----- From: Andy Ning [mailto:andy.ning at windriver.com] Sent: Monday, October 29, 2018 2:02 PM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? It's part of the SDK. Refer to Winder River Titanium Cloud Software Development Kit. Andy On 2018-10-29 04:04 PM, Cordoba Malibran, Erich wrote: > Hi all, > > I was checking the specfile of wrs-remote-clients package[0] and I > can't understand what's the purpose of this package and when is used. > > It doesn't seems to be installed in the final ISO and according to > it's content it looks like a devel package. Also, it has a curious > structure because the rpm only installs a compressed file (.tgz) that > contains a set of installer scripts and other .tgz files that were > grabbed from another rpms. > > > I hope someone can help me to understand the purpose of remote-clients. > > Thanks in advance. > > -Erich > > > [0] > http://git.starlingx.io/cgit/stx-clients/tree/remote-clients/centos/re > mote-clients.spec _______________________________________________ > Starlingx-discuss mailing list > Starlingx-discuss at lists.starlingx.io > http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -- Andy Ning Cube: 3071 Tel: 613-9631408 (int: 4408) Skype: andy.ning.wr _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Stefan.Dinescu at windriver.com Tue Oct 30 09:34:27 2018 From: Stefan.Dinescu at windriver.com (Dinescu, Stefan) Date: Tue, 30 Oct 2018 09:34:27 +0000 Subject: [Starlingx-discuss] compute node: lvg without pv ? In-Reply-To: References: Message-ID: Hello, You have to have a PV for the nova-local LVG, but if you use remote backing, the size of that PV can be really small as it is not used. If your raid1 config has enough free space, you can create partition on your root disk and assign that partition to the LVG using the partition UUID instead of the disk UUID. Then change the nova local backing to remote and you should be fine. ________________________________ From: von Hoesslin, Volker [Volker.Hoesslin at swsn.de] Sent: Monday, October 29, 2018 5:09 PM To: 'starlingx-discuss at lists.starlingx.io' Subject: [Starlingx-discuss] compute node: lvg without pv ? hi, i have been working on the installation of a starlingX openstack cluster for a few days. now i'm almost at the end of the installation guide and fail due to the connection between nova and ceph :( since my compute nodes only have one HDD (hardware raid1 with 2x SSD) i can't add another HDD (pv) to the lvg. is there a way to work directly on the ceph storage without having to go through a local hard drive? -------------- next part -------------- An HTML attachment was scrubbed... URL: From Stefan.Dinescu at windriver.com Tue Oct 30 09:48:03 2018 From: Stefan.Dinescu at windriver.com (Dinescu, Stefan) Date: Tue, 30 Oct 2018 09:48:03 +0000 Subject: [Starlingx-discuss] move storage node in replication group In-Reply-To: References: Message-ID: Hello, The default replication of ceph is 2. Nodes are assigned to each group automatically and cannot be manually changed. if you want replication 3, you need you set that before installing storage nodes. ________________________________ From: von Hoesslin, Volker [Volker.Hoesslin at swsn.de] Sent: Monday, October 29, 2018 5:43 PM To: 'starlingx-discuss at lists.starlingx.io' Subject: [Starlingx-discuss] move storage node in replication group hello, i have 3 storage nodes and after provisioning nodes 1+2 are in replication group 0 and node 3 are in replication group 1?! how can i move node 3 to group 0? greetings, volker -------------- next part -------------- An HTML attachment was scrubbed... URL: From Greg.Waines at windriver.com Tue Oct 30 09:54:58 2018 From: Greg.Waines at windriver.com (Waines, Greg) Date: Tue, 30 Oct 2018 09:54:58 +0000 Subject: [Starlingx-discuss] Purpose of wrs-remote-clients package? In-Reply-To: <2FD5DDB5A04D264C80D42CA35194914F35DAD23B@SHSMSX104.ccr.corp.intel.com> References: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> <9A85D2917C58154C960D95352B22818BB1E99BC6@fmsmsx117.amr.corp.intel.com> <2FD5DDB5A04D264C80D42CA35194914F35DAD23B@SHSMSX104.ccr.corp.intel.com> Message-ID: Actually the HA VM Heartbeat functionality is a different SDK Module. The remote clients SDK Module packages up the CLIs and their Clients such that they can be installed on a remote workstation and enable the CLIs to be used on a remote workstation ... i.e. CLI --> Clients --> REST APIs --> StarlingX In Titanium, because we had made non-upstreamed changes to some OpenStack Services’ CLIs, we packaged up an SDK Module containing * Sysinv CLI (i.e. ‘system ...’) + client * All OpenStack Services’ CLIs & Clients where we had made changes, and * An install script that installed the CLIs & Clients from the SDK Module ... and pulled any unchanged OpenStack Services’ CLIs & Clients from public repos using pip So in StarlingX ... I suppose we need to just make sure that the StarlingX CLIs & Clients can be pulled & installed from public repos using pip ... basically in the same way that other OpenStack Services’ CLIs & Clients can be installed. ... and don’t need the Remote Clients SDK Module. Greg. From: "Xie, Cindy" Date: Monday, October 29, 2018 at 7:18 PM To: "Jones, Bruce E" , "Ning, Antai (Andy)" , "starlingx-discuss at lists.starlingx.io" Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? Bruce, I believe this is part of SDK for the tenant OS, part of HA design to get tenant heart-beat. Thx. - cindy -----Original Message----- From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Monday, October 29, 2018 2:14 PM To: Andy Ning >; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? StarlingX isn't Titanium Cloud. Is the SDK part of StarlingX? brucej -----Original Message----- From: Andy Ning [mailto:andy.ning at windriver.com] Sent: Monday, October 29, 2018 2:02 PM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? It's part of the SDK. Refer to Winder River Titanium Cloud Software Development Kit. Andy On 2018-10-29 04:04 PM, Cordoba Malibran, Erich wrote: Hi all, I was checking the specfile of wrs-remote-clients package[0] and I can't understand what's the purpose of this package and when is used. It doesn't seems to be installed in the final ISO and according to it's content it looks like a devel package. Also, it has a curious structure because the rpm only installs a compressed file (.tgz) that contains a set of installer scripts and other .tgz files that were grabbed from another rpms. I hope someone can help me to understand the purpose of remote-clients. Thanks in advance. -Erich [0] http://git.starlingx.io/cgit/stx-clients/tree/remote-clients/centos/re mote-clients.spec _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -- Andy Ning Cube: 3071 Tel: 613-9631408 (int: 4408) Skype: andy.ning.wr _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From Volker.Hoesslin at swsn.de Tue Oct 30 11:34:07 2018 From: Volker.Hoesslin at swsn.de (von Hoesslin, Volker) Date: Tue, 30 Oct 2018 11:34:07 +0000 Subject: [Starlingx-discuss] move storage node in replication group In-Reply-To: References: Message-ID: Ok, I understand. how can I define this when creating a storage node? neither cli nor gui can I find corresponding options. how are the "replication-groups" to be understood? is it desirable to put all three storage-nodes into one group or should each storage-node have its own group? greez & thx, volker... Von: Dinescu, Stefan [mailto:Stefan.Dinescu at windriver.com] Gesendet: Dienstag, 30. Oktober 2018 10:48 An: von Hoesslin, Volker; 'starlingx-discuss at lists.starlingx.io' Betreff: RE: move storage node in replication group Hello, The default replication of ceph is 2. Nodes are assigned to each group automatically and cannot be manually changed. if you want replication 3, you need you set that before installing storage nodes. ________________________________ From: von Hoesslin, Volker [Volker.Hoesslin at swsn.de] Sent: Monday, October 29, 2018 5:43 PM To: 'starlingx-discuss at lists.starlingx.io' Subject: [Starlingx-discuss] move storage node in replication group hello, i have 3 storage nodes and after provisioning nodes 1+2 are in replication group 0 and node 3 are in replication group 1?! how can i move node 3 to group 0? greetings, volker -------------- next part -------------- An HTML attachment was scrubbed... URL: From Stefan.Dinescu at windriver.com Tue Oct 30 12:02:06 2018 From: Stefan.Dinescu at windriver.com (Dinescu, Stefan) Date: Tue, 30 Oct 2018 12:02:06 +0000 Subject: [Starlingx-discuss] move storage node in replication group In-Reply-To: References: , Message-ID: Check the "capabilities" parameter for adding the ceph backend [1]. you need to set replication to 3 (min_replication should automatically be set to 2). I think the "system" command for it should be: system storage-backend-add ceph -s cinder,glance,swift,nova replication=3 --confirmed The help of the command states: $ system help storage-backend-add usage: system storage-backend-add [-s ] [-n ] [-t ] [-c ] [--confirmed] [--ceph-mon-gib ] [ [ ...]] Replication should be given as a . Storage nodes inside a replication group all share the same (replicated) data. So if one node fails for any reason, the data is still being served by the remaining nodes. As, I said, nodes are assigned automatically to a group based on replication, so you cannot manually assign each node to a different group. It is recommended that your number of storage nodes be a multiple of your replication factor. So for replication 2, you should use 2,4,6.... storage nodes and for replication 3 you should use 3,6,9... storage nodes. [1]: https://docs.starlingx.io/api-ref/stx-config/api-ref-sysinv-v1-config.html?expanded=add-a-ceph-storage-backend-detail#add-a-ceph-storage-backend ________________________________ From: von Hoesslin, Volker [Volker.Hoesslin at swsn.de] Sent: Tuesday, October 30, 2018 1:34 PM To: 'starlingx-discuss at lists.starlingx.io' Subject: Re: [Starlingx-discuss] move storage node in replication group Ok, I understand. how can I define this when creating a storage node? neither cli nor gui can I find corresponding options. how are the "replication-groups" to be understood? is it desirable to put all three storage-nodes into one group or should each storage-node have its own group? greez & thx, volker... Von: Dinescu, Stefan [mailto:Stefan.Dinescu at windriver.com] Gesendet: Dienstag, 30. Oktober 2018 10:48 An: von Hoesslin, Volker; 'starlingx-discuss at lists.starlingx.io' Betreff: RE: move storage node in replication group Hello, The default replication of ceph is 2. Nodes are assigned to each group automatically and cannot be manually changed. if you want replication 3, you need you set that before installing storage nodes. ________________________________ From: von Hoesslin, Volker [Volker.Hoesslin at swsn.de] Sent: Monday, October 29, 2018 5:43 PM To: 'starlingx-discuss at lists.starlingx.io' Subject: [Starlingx-discuss] move storage node in replication group hello, i have 3 storage nodes and after provisioning nodes 1+2 are in replication group 0 and node 3 are in replication group 1?! how can i move node 3 to group 0? greetings, volker -------------- next part -------------- An HTML attachment was scrubbed... URL: From austin.sun at intel.com Tue Oct 30 13:09:14 2018 From: austin.sun at intel.com (Sun, Austin) Date: Tue, 30 Oct 2018 13:09:14 +0000 Subject: [Starlingx-discuss] How flat provision network access external network In-Reply-To: <70A7408C6E1BFB41B192A929744D8523BAC25D4B@ALA-MBD.corp.ad.wrs.com> References: <70A7408C6E1BFB41B192A929744D8523BAC25D4B@ALA-MBD.corp.ad.wrs.com> Message-ID: Hi Allain: Sorry , I did not state clearly. In current my env, There is one compute node with 2 NIC, eno2 is used for Management network, eno1 is used for Data network and is connected to company intranet. "VM access external network " means "VM can access and be accessed in intranet". VM can get intranet IP , but can not ping other computer in intranet. 1. You start by stating that you wanted to use an "external network", but the network create command you used lacks the "--external" option which means you only created a regular tenant network. [Austin] I tried with "-external" parameters before, it is same result as no "-external" parameter , so I did not list this parameter. 2. You did not add a "subnet" to the network so the VM will not be assigned an IP address [Austin] Because there is DHCP in intranet , subnet is not created for data network. 3. The server create command listed in your email has a syntax error which would cause it to not run at all (extra space after the net-id=). [Austin] Thanks you. This is typo. Thanks. BR Austin Sun. From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Monday, October 29, 2018 11:23 PM To: Sun, Austin ; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] How flat provision network access external network Your question is not clear enough for us to determine what you were trying to do, what you expected to happen and what exactly didn't work. 1. You start by stating that you wanted to use an "external network", but the network create command you used lacks the "--external" option which means you only created a regular tenant network. 2. You did not add a "subnet" to the network so the VM will not be assigned an IP address 3. The server create command listed in your email has a syntax error which would cause it to not run at all (extra space after the net-id=). Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Sun, Austin [mailto:austin.sun at intel.com] Sent: Monday, October 29, 2018 10:20 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] How flat provision network access external network Hi All: In my set up , compute-0 eno1 is connected external network as data network. Is it possible VM on compute-0 can access external network directly ? I tried below commands , but it did not work. :( openstack providernet create providernet.a --type flat openstack network create ext_net --provider-network-type flat --provider-physical-network providernet.a system host-if-modify -p providernet.a -c data compute-0 eno1 openstack server create --flavor vm1.flavor --image cirros --nic net-id= ext_net vm1.tiny Thanks. BR Austin Sun. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From Ghada.Khalil at windriver.com Tue Oct 30 18:47:21 2018 From: Ghada.Khalil at windriver.com (Khalil, Ghada) Date: Tue, 30 Oct 2018 18:47:21 +0000 Subject: [Starlingx-discuss] Purpose of wrs-remote-clients package? In-Reply-To: References: <8aea8e1d656fbbf8ee5fb2ba4249fdbf8eee3af9.camel@intel.com> <9A85D2917C58154C960D95352B22818BB1E99BC6@fmsmsx117.amr.corp.intel.com> <2FD5DDB5A04D264C80D42CA35194914F35DAD23B@SHSMSX104.ccr.corp.intel.com> Message-ID: <151EE31B9FCCA54397A757BC674650F0BA47C626@ALA-MBD.corp.ad.wrs.com> To answer Bruce’s question: “ Is the SDK part of StarlingX? “ The answer is yes. All elements of the SDK are present in the StarlingX code base and are buildable. The SDK provided value to previous users based on our experience. We need to align as a StarlingX community on whether to better document/publish its use or phase it out. As Greg points out the remote-clients package is intended to allow users to remotely access the REST API (including the extensions present in StarlingX). Given that StarlingX continues to keep these extensions for the time-being, I would advise against removing this package at this time. Package removal should be considered in line with the continuing overall efforts for patch elimination from StarlingX (i.e. we shouldn’t remove the package until we remove the backend extensions). Regards, Ghada From: Waines, Greg [mailto:Greg.Waines at windriver.com] Sent: Tuesday, October 30, 2018 5:55 AM To: Xie, Cindy; Jones, Bruce E; Ning, Antai (Andy); starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? Actually the HA VM Heartbeat functionality is a different SDK Module. The remote clients SDK Module packages up the CLIs and their Clients such that they can be installed on a remote workstation and enable the CLIs to be used on a remote workstation ... i.e. CLI --> Clients --> REST APIs --> StarlingX In Titanium, because we had made non-upstreamed changes to some OpenStack Services’ CLIs, we packaged up an SDK Module containing * Sysinv CLI (i.e. ‘system ...’) + client * All OpenStack Services’ CLIs & Clients where we had made changes, and * An install script that installed the CLIs & Clients from the SDK Module ... and pulled any unchanged OpenStack Services’ CLIs & Clients from public repos using pip So in StarlingX ... I suppose we need to just make sure that the StarlingX CLIs & Clients can be pulled & installed from public repos using pip ... basically in the same way that other OpenStack Services’ CLIs & Clients can be installed. ... and don’t need the Remote Clients SDK Module. Greg. From: "Xie, Cindy" > Date: Monday, October 29, 2018 at 7:18 PM To: "Jones, Bruce E" >, "Ning, Antai (Andy)" >, "starlingx-discuss at lists.starlingx.io" > Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? Bruce, I believe this is part of SDK for the tenant OS, part of HA design to get tenant heart-beat. Thx. - cindy -----Original Message----- From: Jones, Bruce E [mailto:bruce.e.jones at intel.com] Sent: Monday, October 29, 2018 2:14 PM To: Andy Ning >; starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? StarlingX isn't Titanium Cloud. Is the SDK part of StarlingX? brucej -----Original Message----- From: Andy Ning [mailto:andy.ning at windriver.com] Sent: Monday, October 29, 2018 2:02 PM To: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] Purpose of wrs-remote-clients package? It's part of the SDK. Refer to Winder River Titanium Cloud Software Development Kit. Andy On 2018-10-29 04:04 PM, Cordoba Malibran, Erich wrote: Hi all, I was checking the specfile of wrs-remote-clients package[0] and I can't understand what's the purpose of this package and when is used. It doesn't seems to be installed in the final ISO and according to it's content it looks like a devel package. Also, it has a curious structure because the rpm only installs a compressed file (.tgz) that contains a set of installer scripts and other .tgz files that were grabbed from another rpms. I hope someone can help me to understand the purpose of remote-clients. Thanks in advance. -Erich [0] http://git.starlingx.io/cgit/stx-clients/tree/remote-clients/centos/re mote-clients.spec _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -- Andy Ning Cube: 3071 Tel: 613-9631408 (int: 4408) Skype: andy.ning.wr _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss -------------- next part -------------- An HTML attachment was scrubbed... URL: From ildiko.vancsa at gmail.com Tue Oct 30 19:42:10 2018 From: ildiko.vancsa at gmail.com (Ildiko Vancsa) Date: Tue, 30 Oct 2018 20:42:10 +0100 Subject: [Starlingx-discuss] Switching to winter time In-Reply-To: <2F47C197-D29E-4F23-8C09-CF0D9C204134@gmail.com> References: <3C0F1E8C-F034-43E3-A88A-608D712B949D@gmail.com> <2F47C197-D29E-4F23-8C09-CF0D9C204134@gmail.com> Message-ID: Hi, This is a friendly reminder to take a quick look at the Doodle poll and vote on your preference on whether or not to stick to the UTC time slot for the weekly meetings: https://doodle.com/poll/pw7mds8qms9d989y We are planning to discuss this topic on the community call tomorrow. Thanks, Ildikó > On 2018. Oct 29., at 21:10, Ildiko Vancsa wrote: > > Hi, > > I created a poll to collect data on what the community prefers until the community call on Wednesday: https://doodle.com/poll/pw7mds8qms9d989y > > Thanks, > Ildikó > > >> On 2018. Oct 29., at 20:47, Dean Troyer wrote: >> >> On Mon, Oct 29, 2018 at 1:36 PM Ildiko Vancsa wrote: >>> As we’ve reached the end of DST for this year in Europe already and on Sunday the USA will switch over to winter time as well I would like to ask the preference on the meeting times. >>> >>> Would there be a preference to stick to the UTC time slots for the meetings or we should keep the time slots specified in Pacific Time? >> >> I have a preference for sticking to UTC time, my experiences doing >> that for nearly everything OpenStack-related these years has been >> overall positive. Generally everyone knows how to get from UTC to >> their local time, it avoids needs to remember when the US changes for >> those whose local DST rules are not aligned with the US for example. > > +1 > > While I like the convenience of always knowing when the meeting is, one issue with it is to have collisions with other weekly meetings that do not stick to UTC. In OPNFV the community decided to deal with the switch and not sticking to UTC. > >> >> dt >> >> -- >> >> Dean Troyer >> dtroyer at gmail.com > From juan.carlos.alonso at intel.com Tue Oct 30 21:52:33 2018 From: juan.carlos.alonso at intel.com (Alonso, Juan Carlos) Date: Tue, 30 Oct 2018 21:52:33 +0000 Subject: [Starlingx-discuss] Provisioning of Controller-0 fails in Simplex Message-ID: <8557B550001AFB46A43A0CCC314BF85153C58C26@FMSMSX108.amr.corp.intel.com> Hi, Has someone tried to install and provision a Simplex system with today's ISO: stx-2018-10-30-171-master.iso? Based on the wiki (https://docs.starlingx.io/installation_guide/simplex.html#simplex ), the last step is Verifying Controller-0 Configuration. When executing system service-list, should appear "oam-ip" and "management-ip", but "oam-ip" does not appear. Besides, the status of Controller-0 starts as "Enabling Compute Service" and then change to "Critical failure, Auto-recovery, re-enabling", and connection lost. Has someone faced the same issue? Regards. Juan Carlos Alonso -------------- next part -------------- An HTML attachment was scrubbed... URL: From dehao.shang at intel.com Wed Oct 31 08:39:51 2018 From: dehao.shang at intel.com (Shang, Dehao) Date: Wed, 31 Oct 2018 08:39:51 +0000 Subject: [Starlingx-discuss] ceph upgrading patch Message-ID: <71AECFE5078153419EB7B8DBE0644B263CBF3EC3@shsmsx102.ccr.corp.intel.com> hi: In order to integrate ceph13.2.0 upgrading into current building flow, i have send two patches and one PR. Patch 1: Because ceph13.2.0 depend on some new packages, need to download these packages at generating mirror process. Patch 2: Based on stx-ceph 13.2.0 source code, some preparation works need to be done at creating ceph-13.2.0.src.rpm package. So, I modify some building codes of default_build_srpm scrib. PR: For stx-ceph 13.2.0 source code repo, create new ceph.spce file and modify make-dist scrib. After we finish transferring patches from stx-ceph10.2.0 to stx-ceph13.2.0, we will modify commit id at stc-integ repo Please review our patch and upgrading methods. Thanks Dehao -------------- next part -------------- An HTML attachment was scrubbed... URL: From vivian.zhu at intel.com Wed Oct 31 08:40:41 2018 From: vivian.zhu at intel.com (Zhu, Vivian) Date: Wed, 31 Oct 2018 08:40:41 +0000 Subject: [Starlingx-discuss] ceph upgrading patch In-Reply-To: <71AECFE5078153419EB7B8DBE0644B263CBF3EC3@shsmsx102.ccr.corp.intel.com> References: <71AECFE5078153419EB7B8DBE0644B263CBF3EC3@shsmsx102.ccr.corp.intel.com> Message-ID: <371DF9A763E9F44F924F4A821FC070264C54C84E@SHSMSX104.ccr.corp.intel.com> Hi Brent, Please help the review. Thanks! - Vivian SSG OTC NST Storage Tel: (8621)61167437 From: Shang, Dehao Sent: Wednesday, October 31, 2018 4:40 PM To: starlingx-discuss at lists.starlingx.io; Brent.Rowsell at windriver.com Cc: Zhu, Vivian ; Liu, Changcheng ; Ma, Jianpeng Subject: ceph upgrading patch hi: In order to integrate ceph13.2.0 upgrading into current building flow, i have send two patches and one PR. Patch 1: Because ceph13.2.0 depend on some new packages, need to download these packages at generating mirror process. Patch 2: Based on stx-ceph 13.2.0 source code, some preparation works need to be done at creating ceph-13.2.0.src.rpm package. So, I modify some building codes of default_build_srpm scrib. PR: For stx-ceph 13.2.0 source code repo, create new ceph.spce file and modify make-dist scrib. After we finish transferring patches from stx-ceph10.2.0 to stx-ceph13.2.0, we will modify commit id at stc-integ repo Please review our patch and upgrading methods. Thanks Dehao -------------- next part -------------- An HTML attachment was scrubbed... URL: From Don.Penney at windriver.com Wed Oct 31 10:47:34 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Wed, 31 Oct 2018 10:47:34 +0000 Subject: [Starlingx-discuss] bashate failing in stx-integ in some reviews? Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1E65@ALA-MBD.corp.ad.wrs.com> Hi folks, I'm seeing a couple of reviews in stx-integ failing the tox linter job on a bashate error, while most others are successful. Happy in this test: 2018-10-31 03:06:25.270134 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh 2018-10-31 03:06:25.270348 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/initscripts/centos/files/mountnfs.sh But failing in this one: 2018-10-31 08:21:56.690432 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh 2018-10-31 08:21:56.690706 | ubuntu-xenial | /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline 2018-10-31 08:21:56.690799 | ubuntu-xenial | 1 bashate error(s) found The custom.sh script is definitely missing the EOL character, but that has been the case since it was created two years ago. So I guess we just fix the script and rebase any failing reviews? It seems very odd that bashate would not always be reporting this error. The tox.ini bashate invocation has not changed recently, either. Cheers, Don. Don Penney, Developer, Wind River -------------- next part -------------- An HTML attachment was scrubbed... URL: From thierry at openstack.org Wed Oct 31 13:24:04 2018 From: thierry at openstack.org (Thierry Carrez) Date: Wed, 31 Oct 2018 14:24:04 +0100 Subject: [Starlingx-discuss] [tsc] Discussion between OSF projects in Berlin Message-ID: Hi! For those who will be in Berlin for the Summit, please mark your calendars for the following Forum session: https://www.openstack.org/summit/berlin-2018/summit-schedule/events/22815/cross-technical-leadership-session-openstack-kata-starlingx-airship-zuul This session is an opportunity for the technical leadership of all OSF projects to meet, discuss pain points, compare experiences and explore synergies. We expect to have members of the OpenStack, Kata Containers, StarlingX, Airship, and Zuul communities around. It will be the occasion, among other things, to compare the initial governance structures for the pilot projects. If you can make it, please join us ! -- Thierry Carrez (ttx) From dtroyer at gmail.com Wed Oct 31 13:46:44 2018 From: dtroyer at gmail.com (Dean Troyer) Date: Wed, 31 Oct 2018 08:46:44 -0500 Subject: [Starlingx-discuss] bashate failing in stx-integ in some reviews? In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1E65@ALA-MBD.corp.ad.wrs.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1E65@ALA-MBD.corp.ad.wrs.com> Message-ID: On Wed, Oct 31, 2018 at 5:47 AM Penney, Don wrote: > I’m seeing a couple of reviews in stx-integ failing the tox linter job on a bashate error, while most others are successful. > Happy in this test: > 2018-10-31 03:06:25.270134 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > > But failing in this one: > 2018-10-31 08:21:56.690432 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > 2018-10-31 08:21:56.690706 | ubuntu-xenial | /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline > > So I guess we just fix the script and rebase any failing reviews? It seems very odd that bashate would not always be reporting this error. The tox.ini bashate invocation has not changed recently, either. Yeah, that is odd. Can you send links tot he reviews those came from? I think it would be interesting to see what is different between the job runs, ie cloud provider and such. There _shouldn't_ be anything that matters different... dt -- Dean Troyer dtroyer at gmail.com From Don.Penney at windriver.com Wed Oct 31 13:51:45 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Wed, 31 Oct 2018 13:51:45 +0000 Subject: [Starlingx-discuss] bashate failing in stx-integ in some reviews? In-Reply-To: References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1E65@ALA-MBD.corp.ad.wrs.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1FED@ALA-MBD.corp.ad.wrs.com> Failure case: http://logs.openstack.org/84/614384/5/check/openstack-tox-linters/70d0ace/job-output.txt.gz Success case: http://logs.openstack.org/23/613823/7/check/openstack-tox-linters/03e4b93/job-output.txt.gz -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Wednesday, October 31, 2018 9:47 AM To: Penney, Don Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] bashate failing in stx-integ in some reviews? On Wed, Oct 31, 2018 at 5:47 AM Penney, Don wrote: > I’m seeing a couple of reviews in stx-integ failing the tox linter job on a bashate error, while most others are successful. > Happy in this test: > 2018-10-31 03:06:25.270134 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > > But failing in this one: > 2018-10-31 08:21:56.690432 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > 2018-10-31 08:21:56.690706 | ubuntu-xenial | /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline > > So I guess we just fix the script and rebase any failing reviews? It seems very odd that bashate would not always be reporting this error. The tox.ini bashate invocation has not changed recently, either. Yeah, that is odd. Can you send links tot he reviews those came from? I think it would be interesting to see what is different between the job runs, ie cloud provider and such. There _shouldn't_ be anything that matters different... dt -- Dean Troyer dtroyer at gmail.com From Don.Penney at windriver.com Wed Oct 31 14:09:02 2018 From: Don.Penney at windriver.com (Penney, Don) Date: Wed, 31 Oct 2018 14:09:02 +0000 Subject: [Starlingx-discuss] bashate failing in stx-integ in some reviews? In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1FED@ALA-MBD.corp.ad.wrs.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1E65@ALA-MBD.corp.ad.wrs.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1FED@ALA-MBD.corp.ad.wrs.com> Message-ID: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E203D@ALA-MBD.corp.ad.wrs.com> Al noticed the failure is only reported if it's in the last file processed by bashate: Sure enough: $ bashate -v ../stx-integ/base/setup/files/prompt.sh ../stx-integ/base/setup/files/custom.sh Running bashate on ../stx-integ/base/setup/files/prompt.sh Running bashate on ../stx-integ/base/setup/files/custom.sh ../stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline 1 bashate error(s) found $ bashate -v ../stx-integ/base/setup/files/custom.sh ../stx-integ/base/setup/files/prompt.sh Running bashate on ../stx-integ/base/setup/files/custom.sh Running bashate on ../stx-integ/base/setup/files/prompt.sh -----Original Message----- From: Penney, Don [mailto:Don.Penney at windriver.com] Sent: Wednesday, October 31, 2018 9:52 AM To: Dean Troyer Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] bashate failing in stx-integ in some reviews? Failure case: http://logs.openstack.org/84/614384/5/check/openstack-tox-linters/70d0ace/job-output.txt.gz Success case: http://logs.openstack.org/23/613823/7/check/openstack-tox-linters/03e4b93/job-output.txt.gz -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Wednesday, October 31, 2018 9:47 AM To: Penney, Don Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] bashate failing in stx-integ in some reviews? On Wed, Oct 31, 2018 at 5:47 AM Penney, Don wrote: > I’m seeing a couple of reviews in stx-integ failing the tox linter job on a bashate error, while most others are successful. > Happy in this test: > 2018-10-31 03:06:25.270134 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > > But failing in this one: > 2018-10-31 08:21:56.690432 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > 2018-10-31 08:21:56.690706 | ubuntu-xenial | /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline > > So I guess we just fix the script and rebase any failing reviews? It seems very odd that bashate would not always be reporting this error. The tox.ini bashate invocation has not changed recently, either. Yeah, that is odd. Can you send links tot he reviews those came from? I think it would be interesting to see what is different between the job runs, ie cloud provider and such. There _shouldn't_ be anything that matters different... dt -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Al.Bailey at windriver.com Wed Oct 31 15:20:06 2018 From: Al.Bailey at windriver.com (Bailey, Henry Albert (Al)) Date: Wed, 31 Oct 2018 15:20:06 +0000 Subject: [Starlingx-discuss] bashate failing in stx-integ in some reviews? In-Reply-To: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E203D@ALA-MBD.corp.ad.wrs.com> References: <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1E65@ALA-MBD.corp.ad.wrs.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3E1FED@ALA-MBD.corp.ad.wrs.com> <6703202FD9FDFF4A8DA9ACF104AE129FBA3E203D@ALA-MBD.corp.ad.wrs.com> Message-ID: Bug: https://bugs.launchpad.net/starlingx/+bug/1800865 Review: https://review.openstack.org/#/c/614542/ Once this gets merged, the other reviews that are randomly failing zuul should be able to pass Al -----Original Message----- From: Penney, Don [mailto:Don.Penney at windriver.com] Sent: Wednesday, October 31, 2018 10:09 AM To: Penney, Don; Dean Troyer Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] bashate failing in stx-integ in some reviews? Al noticed the failure is only reported if it's in the last file processed by bashate: Sure enough: $ bashate -v ../stx-integ/base/setup/files/prompt.sh ../stx-integ/base/setup/files/custom.sh Running bashate on ../stx-integ/base/setup/files/prompt.sh Running bashate on ../stx-integ/base/setup/files/custom.sh ../stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline 1 bashate error(s) found $ bashate -v ../stx-integ/base/setup/files/custom.sh ../stx-integ/base/setup/files/prompt.sh Running bashate on ../stx-integ/base/setup/files/custom.sh Running bashate on ../stx-integ/base/setup/files/prompt.sh -----Original Message----- From: Penney, Don [mailto:Don.Penney at windriver.com] Sent: Wednesday, October 31, 2018 9:52 AM To: Dean Troyer Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] bashate failing in stx-integ in some reviews? Failure case: http://logs.openstack.org/84/614384/5/check/openstack-tox-linters/70d0ace/job-output.txt.gz Success case: http://logs.openstack.org/23/613823/7/check/openstack-tox-linters/03e4b93/job-output.txt.gz -----Original Message----- From: Dean Troyer [mailto:dtroyer at gmail.com] Sent: Wednesday, October 31, 2018 9:47 AM To: Penney, Don Cc: starlingx-discuss at lists.starlingx.io Subject: Re: [Starlingx-discuss] bashate failing in stx-integ in some reviews? On Wed, Oct 31, 2018 at 5:47 AM Penney, Don wrote: > I’m seeing a couple of reviews in stx-integ failing the tox linter job on a bashate error, while most others are successful. > Happy in this test: > 2018-10-31 03:06:25.270134 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > > But failing in this one: > 2018-10-31 08:21:56.690432 | ubuntu-xenial | Running bashate on /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh > 2018-10-31 08:21:56.690706 | ubuntu-xenial | /home/zuul/src/git.openstack.org/openstack/stx-integ/base/setup/files/custom.sh:1:1: E004 File did not end with a newline > > So I guess we just fix the script and rebase any failing reviews? It seems very odd that bashate would not always be reporting this error. The tox.ini bashate invocation has not changed recently, either. Yeah, that is odd. Can you send links tot he reviews those came from? I think it would be interesting to see what is different between the job runs, ie cloud provider and such. There _shouldn't_ be anything that matters different... dt -- Dean Troyer dtroyer at gmail.com _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss _______________________________________________ Starlingx-discuss mailing list Starlingx-discuss at lists.starlingx.io http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss From Allain.Legacy at windriver.com Wed Oct 31 15:21:45 2018 From: Allain.Legacy at windriver.com (Legacy, Allain) Date: Wed, 31 Oct 2018 15:21:45 +0000 Subject: [Starlingx-discuss] How flat provision network access external network In-Reply-To: References: <70A7408C6E1BFB41B192A929744D8523BAC25D4B@ALA-MBD.corp.ad.wrs.com> Message-ID: <70A7408C6E1BFB41B192A929744D8523BAC2B492@ALA-MBD.corp.ad.wrs.com> If the VM is able to get an IP address from your intranet IP address then at least you know that connectivity to your intranet is physically possible. You will need to do further debugging to understand which parts of your ping request are making it out (and back in). For example, is ARP resolving for the IP address you are trying to ping? If not, do you see the ARP request on your physical network? If ARP is resolving, then do you see the ICMP request on your physical network? etc... you need to break the problem down into its smallest parts to understand where exactly the problem is occurring. Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Sun, Austin [mailto:austin.sun at intel.com] Sent: Tuesday, October 30, 2018 9:09 AM To: Legacy, Allain; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] How flat provision network access external network Hi Allain: Sorry , I did not state clearly. In current my env, There is one compute node with 2 NIC, eno2 is used for Management network, eno1 is used for Data network and is connected to company intranet. "VM access external network " means "VM can access and be accessed in intranet". VM can get intranet IP , but can not ping other computer in intranet. 1. You start by stating that you wanted to use an "external network", but the network create command you used lacks the "--external" option which means you only created a regular tenant network. [Austin] I tried with "-external" parameters before, it is same result as no "-external" parameter , so I did not list this parameter. 2. You did not add a "subnet" to the network so the VM will not be assigned an IP address [Austin] Because there is DHCP in intranet , subnet is not created for data network. 3. The server create command listed in your email has a syntax error which would cause it to not run at all (extra space after the net-id=). [Austin] Thanks you. This is typo. Thanks. BR Austin Sun. From: Legacy, Allain [mailto:Allain.Legacy at windriver.com] Sent: Monday, October 29, 2018 11:23 PM To: Sun, Austin >; starlingx-discuss at lists.starlingx.io Subject: RE: [Starlingx-discuss] How flat provision network access external network Your question is not clear enough for us to determine what you were trying to do, what you expected to happen and what exactly didn't work. 1. You start by stating that you wanted to use an "external network", but the network create command you used lacks the "--external" option which means you only created a regular tenant network. 2. You did not add a "subnet" to the network so the VM will not be assigned an IP address 3. The server create command listed in your email has a syntax error which would cause it to not run at all (extra space after the net-id=). Regards, Allain Allain Legacy, Software Developer, Wind River direct 613.270.2279 fax 613.492.7870 skype allain.legacy 350 Terry Fox Drive, Suite 200, Ottawa, Ontario, K2K 2W5 [WIND] From: Sun, Austin [mailto:austin.sun at intel.com] Sent: Monday, October 29, 2018 10:20 AM To: starlingx-discuss at lists.starlingx.io Subject: [Starlingx-discuss] How flat provision network access external network Hi All: In my set up , compute-0 eno1 is connected external network as data network. Is it possible VM on compute-0 can access external network directly ? I tried below commands , but it did not work. :( openstack providernet create providernet.a --type flat openstack network create ext_net --provider-network-type flat --provider-physical-network providernet.a system host-if-modify -p providernet.a -c data compute-0 eno1 openstack server create --flavor vm1.flavor --image cirros --nic net-id= ext_net vm1.tiny Thanks. BR Austin Sun. -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.png Type: image/png Size: 1807 bytes Desc: image001.png URL: From ildiko.vancsa at gmail.com Wed Oct 31 16:15:35 2018 From: ildiko.vancsa at gmail.com (Ildiko Vancsa) Date: Wed, 31 Oct 2018 17:15:35 +0100 Subject: [Starlingx-discuss] Use Pacific Time for the meetings Message-ID: <625564A9-4B5C-4159-8DD3-46A37AFB9808@gmail.com> Hi, On today’s community call we agreed to stick with the Pacific Time slots for the various weekly calls as opposed to UTC. This means that starting next week the UTC times will shift to an hour later. The main reason for the decision was that the community has a couple of meetings which are currently start at 6am Pacific Time and sticking with the UTC slots would mean to have these meetings at 5am in the morning which the community voted against. As opposed to moving these meetings we will just keep the Pacific Time slots as reference and be conscious of the daylight savings time switches twice a year. Please refer to the wiki for the meeting times: https://wiki.openstack.org/wiki/Starlingx/Meetings Thanks and Best Regards, Ildikó From cesar.lara at intel.com Wed Oct 31 17:44:01 2018 From: cesar.lara at intel.com (Lara, Cesar) Date: Wed, 31 Oct 2018 17:44:01 +0000 Subject: [Starlingx-discuss] [Build][meetings] Agenda for 11/1/2018 Message-ID: <0B566C62EC792145B40E29EFEBF1AB47105547ED@fmsmsx104.amr.corp.intel.com> Build team meeting 11/1/2018 Agenda - Cengn mirror status - Ken - Koji discussion follow up 0- Victor R - Mirror script refactor - Marcela - opens Regards Cesar Lara Software Engineering Manager OpenSource Technology Center -------------- next part -------------- An HTML attachment was scrubbed... URL: From bruce.e.jones at intel.com Wed Oct 31 19:09:11 2018 From: bruce.e.jones at intel.com (Jones, Bruce E) Date: Wed, 31 Oct 2018 19:09:11 +0000 Subject: [Starlingx-discuss] Community meeting minutes Oct 31 2018 Message-ID: <9A85D2917C58154C960D95352B22818BB1E9BDB5@fmsmsx117.amr.corp.intel.com> Agenda and Notes for the Oct 31st call * How to handle DST changes? Use UTC timeslots or keep Pacific time? * Remote client SDK module - remove it? Keep it? Do we keep the TiC SDK? If so, should we support it and document it? * Brent: We can remove the SDK. WR to handle. * Updated Code Submission Guidelines: https://wiki.openstack.org/wiki/StarlingX/CodeSubmissionGuidelines * Still seeing reviews that haven't been tested or compiled. Please review the updated guidelines and post with WF -1 and WIP if you're looking for early review comments * March release planning * Project priority call for submissions * Each sub-team should be thinking about what they want to do for the release ? Can/should we schedule the teams to come into this meeting and present their plans? ? Each team to update the Ethercalc with their high priority work items * https://ethercalc.openstack.org/fafyo2729fnr ? Each team to come to the community call next week to briefly review ? TSC to review globally at or before Berlin. * Release team to consider how/when/if to establish some additional milestones * Can we please have some reviewers look into this Ceph upgrade PR: * https://github.com/starlingx-staging/stx-ceph/pull/5 * There are no notifications for new PRs so please send email out to the mailing list when you post one * TSC members - now that we've grown the TSC we need more +1/+2 on specs to approve specs * "Clear the decks over the next few days to approve specs in flight" * Meeting time switch options * Keep meeting times as per UTC - 6AM meetings would move to 5AM * Move meeting times around to keep 5AM meetings off US calendars * Poll: https://doodle.com/poll/pw7mds8qms9d989y * Meetings: https://wiki.openstack.org/wiki/Starlingx/Meetings * Decision - move meeting times with US DST and keep the 6AM meetings at 6am -------------- next part -------------- An HTML attachment was scrubbed... URL: