[Starlingx-discuss] Placement could not be started issue fix!

Rowsell, Brent Brent.Rowsell at windriver.com
Thu Jun 20 11:21:47 UTC 2019


Zhipeng,

See inline.

Brent

-----Original Message-----
From: Liu, ZhipengS [mailto:zhipengs.liu at intel.com] 
Sent: Thursday, June 20, 2019 4:16 AM
To: Khalil, Ghada <Ghada.Khalil at windriver.com>; Dean Troyer <dtroyer at gmail.com>; starlingx <starlingx-discuss at lists.starlingx.io>
Subject: [Starlingx-discuss] Placement could not be started issue fix!

Hi Ghada and all,

I see placement already merged today. 
However, I found an obvious issue that we have to fix it before starting daily build sanity test!!
I raised a LP and related fix as below
https://bugs.launchpad.net/starlingx/+bug/1833497 placement po will not be started https://review.opendev.org/#/c/666491/

Root cause is in below patch
https://review.opendev.org/#/c/653932/
It overrides openstack-compute-kit group which do not include openstack-placement.

I have done basic verification (simplex deploy and vm creation), Please help get it merged soon, thanks!

BTW, it is really not a good design. If other guy add new chart in compute-kit,  he have to Update ovs and ironic related helm file as well.
sysinv/helm/ironic.py
sysinv/helm/openvswitch.py
[BR] Please open a LP for this.

Thanks!
Zhipeng

-----Original Message-----
From: Khalil, Ghada [mailto:Ghada.Khalil at windriver.com]
Sent: 2019年6月20日 7:26
To: Dean Troyer <dtroyer at gmail.com>; starlingx <starlingx-discuss at lists.starlingx.io>
Subject: Re: [Starlingx-discuss] [stx-nova] Stein branch rebase

FYI. The external placement service code merged earlier today:
https://review.opendev.org/#/c/662614/
https://review.opendev.org/#/c/662371/

Cheers,
Ghada

-----Original Message-----
From: Dean Troyer [mailto:dtroyer at gmail.com]
Sent: Wednesday, June 19, 2019 6:53 PM
To: starlingx
Subject: [Starlingx-discuss] [stx-nova] Stein branch rebase

I have done a preliminary rebase of the stx-nova Stein branch into
stx/stein.2 [0].  It is passing unit and functional tests but since some changes were required from the current upstream reviews it really needs to be checked out further.

* I started with upstream stable/stein acd2daa9 (current as of yesterday noon-ish)
* Artom rebased the upstream NUMA patches in master so I pulled the current patchset of those
  * There is a missing import in 635229 that is causing the test failures, I inserted the  commit adding that inline in the PR
  * There were conflicts in 634605 and 634606 due to ongoing development in master since stable/stein was branched.  I made the obvious corrections, there may be more required that someone who is not familiar with this code (me) would likely miss.

The final pep, unit and functional jobs are running under
https://review.opendev.org/#/c/656065/8 and I expect them to pass.

I do believe this requires the extraced placement to be merged so we may not be able to test it in StarlingX until that is complete.  I am hoping this can be tested with just replacing the Nova docker image but I do not have the time to run through that.

dt

[0] PR: https://github.com/starlingx-staging/stx-nova/pull/25

--
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
_______________________________________________
Starlingx-discuss mailing list
Starlingx-discuss at lists.starlingx.io
http://lists.starlingx.io/cgi-bin/mailman/listinfo/starlingx-discuss


More information about the Starlingx-discuss mailing list