May 2019 - Starlingx-discuss - lists.starlingx.io

[Starlingx-discuss] Package extraction to new repos
by Dean Troyer 17 May '19

17 May '19

In the versioning thread Saul started yesterday[0] we started talking about extracting more of the components of StarlingX, in particular clients and other common dependences, in order to more easily build and version them independently. Let's keep the discussion of what versions look like there and talk about the extraction stuff here... I have done a little investigating starting with extracting cgts-client from the config repo [1] (also renaming it in the process) using the same process we used to split up the original monorepo; the extraction process preserving git history is mostly automated. That POC also renamed the client to get a feel for how much work that is in that context (without touching all of the references to it, much more work!). I have created an ethercalc[2] to explore/identify some initial targets and set an overall direction for this work. This is also a good time to rename things and set a common format. However, spreadsheets and notepads are not always the best forums for discussing the questions. Here are my suggestions to start things off: * Extract clients and their internal dependencies and publish them to PyPI to simplify installing them in places apart from the controller/worker nodes. * Prefix StarlingX-specific things published to PyPI with 'stx-' to avoid collisions. * Use '-lib' or '-common' in names for libraries. (We have -common already in a number of places but for fm-common that is not the python module name, maybe those should match?) * The Python bit of platform-util looks to be all about the license manager, is that still needed? Is this just a script library otherwise? * I renamed cgts-client to system-client in my POC, we should find a better name than that, should the command be renamed too?[3] There are of course a number of process things to work out like coordination and overlap between using the prior "embedded" code and switching to the new "independent" code. I have started with a story[4] but expect to write a spec covering the process and conclusions we derive here. dt [0] http://lists.starlingx.io/pipermail/starlingx-discuss/2019-May/004554.html [1] https://github.com/dtroyer/stx-system-client [2] https://ethercalc.openstack.org/stx-repo-org [3] Full-disclosure: I have the intention to create a StarlingXClient similar to OpenStackClient that would follow the same command format rules and use the command name 'stxc'. (https://github.com/dtroyer/starlingxclient, https://github.com/dtroyer/stxc-faultmanagement) [4] https://storyboard.openstack.org/#!/story/2005495 -- Dean Troyer dtroyer(a)gmail.com

2 1

[Starlingx-discuss] Request for Cloud:StarlingX:2.0 project on OBS
by Saul Wold 17 May '19

17 May '19

Hi Dirk, Not sure what the offical process is to request a project on OBS, but we would like to have the Cloud:StarlingX project added to OBS. We would have 1 tarball associated with different top level packages with one specfile / package. So instead of how I initially did it with _multibuild, we will create packages for existing specfile in centos: fm-common fm-api fm-doc fm-mgr fm-rest-api python-fmclient snmp-audittrail snmp-ext And as we start on other parts of StarlingX we will create more packages and add the tarball from the git repo to those new packages. We will have somewhere around 100 packages to start with and that might grow a little larger. Thanks Sau!

3 2

[Starlingx-discuss] DUPLEX: UI Problem: Failed to create network
by Marcel Schaible 17 May '19

17 May '19

Hi, I have the following problem with the UI at port 31000. 1. Only the external Networks and subnets are visible in the UI. 2. When we try to create a new network we get the error: Error: Failed to create network "testnet". Unable to create the Network. No tenant Network is available for allocation. Neutron Server Returns request ids … We are using the image from 20190429/Duplex/bm. What is the current state of the UI and can we update just the UI? Thanks Marcel openstack network list +--------------------------------------+---------------+--------------------------------------+ | ID | Name | Subnets | +--------------------------------------+---------------+--------------------------------------+ | 1848fe83-bf0a-40d3-a78a-2a66d4765cbb | internal-net1 | 9b468e0f-4171-4aa5-8c99-b08cd60748f7 | | 1ea8a6ad-f826-4550-8e2f-a576977fd7c2 | external-net1 | 082089ee-ef7e-4843-bceb-a536cdec3685 | | d296e412-1eec-4cac-bff6-18653ca2b5cc | public-net1 | 86096834-322d-4b87-8a40-1f84932466a3 | +--------------------------------------+---------------+--------------------------------------+ controller-1:~# !45 openstack router show public-router0 No Router found for public-router0 controller-1:~# openstack subnet list +--------------------------------------+------------------+--------------------------------------+------------------+ | ID | Name | Network | Subnet | +--------------------------------------+------------------+--------------------------------------+------------------+ | 082089ee-ef7e-4843-bceb-a536cdec3685 | external-subnet1 | 1ea8a6ad-f826-4550-8e2f-a576977fd7c2 | 192.168.51.0/24 | | 86096834-322d-4b87-8a40-1f84932466a3 | public-subnet1 | d296e412-1eec-4cac-bff6-18653ca2b5cc | 192.168.101.0/24 | | 9b468e0f-4171-4aa5-8c99-b08cd60748f7 | internal-subnet1 | 1848fe83-bf0a-40d3-a78a-2a66d4765cbb | 10.1.1.0/24 | +--------------------------------------+------------------+--------------------------------------+------------------+ controller-1:~# openstack router list +--------------------------------------+-----------------+--------+-------+----------------------------------+-------------+-------+ | ID | Name | Status | State | Project | Distributed | HA | +--------------------------------------+-----------------+--------+-------+----------------------------------+-------------+-------+ | 1ca1acfe-299f-41ae-9a7e-157f867ed374 | public-router1 | ACTIVE | UP | f52d0bd6b5164ad2a3f9f9de56f3d071 | False | False | | eb1716cf-519f-4a58-bae8-c5d036579473 | private-router1 | ACTIVE | UP | f52d0bd6b5164ad2a3f9f9de56f3d071 | False | False | +--------------------------------------+-----------------+--------+-------+----------------------------------+-------------+-------+ controller-1:~#

6 8

[Starlingx-discuss] [Containers] Sanity Test - ISO 20190516
by Perez Ibarra, Maria G 16 May '19

16 May '19

Status of the Sanity Test for last CENGN ISO: bootimage.iso from 2019-MAY-16 (link<http://mirror.starlingx.cengn.ca/mirror/starlingx/master/centos/20190516T16…>) Status: YELLOW ====================== Bare Metal environment ====================== AIO - Simplex: Setup 03 TCs Provision-Containers 01 TCs Sanity-OpenStack 49 TCs | 24 TCs FAIL Sanity-Platform 11 TCs ------------------------------ TOTAL: 64 TCs AIO - Duplex: Setup 03 TCs Provision-Containers 01 TCs Sanity-OpenStack 52 TCs | 29 TCs FAIL Sanity-Platform 09 TCs ------------------------------ TOTAL: 65 TCs Standard - Local Storage (2+2): Setup 03 TCs Provision-Containers 01 TCs Sanity-OpenStack 52 TCs | 30 TCs FAIL Sanity-Platform 09 TCs ------------------------------ TOTAL: 65 TCs Standard - External Storage (2+2+2): Setup 03 TCs Provision-Containers 01 TCs Sanity-OpenStack 52 TCs | 34 TCs FAIL Sanity-Platform 05 TCs ------------------------------ TOTAL: 65 TCs cinder volume create failed https://bugs.launchpad.net/starlingx/+bug/1828059 For more detail of the tests: https://wiki.openstack.org/wiki/StarlingX/Test/SanityTests#Sanity-OpenStack Regards Maria G.

1 0

[Starlingx-discuss] May 20 -- holiday in Canada
by Khalil, Ghada 16 May '19

16 May '19

FYI. Monday May 20th is a statutory holiday in Canada. There will be no containerization meeting next Monday. A number of core reviewers will be on holiday. Regards, Ghada

1 0

Re: [Starlingx-discuss] UPDATE: Limited code merges until sanity results improve
by Khalil, Ghada 16 May '19

16 May '19

** limited merges are still in effect ** The following sanity issues are still open: - https://bugs.launchpad.net/starlingx/+bug/1828562 - issue was reproduced and re-opened - https://bugs.launchpad.net/starlingx/+bug/1828059 - issue was reproduced and re-opened - https://bugs.launchpad.net/starlingx/+bug/1829299 - new issue in the 20190515T220331Z load - https://bugs.launchpad.net/starlingx/+bug/1829433 - new issue in the 20190515T220331Z load Note: As of now, there are no additional fixes merged since the 20190515T220331Z cengn load. Core reviewers, please only merge fixes for critical/high sanity issues. From: Khalil, Ghada Sent: Wednesday, May 15, 2019 6:09 PM To: starlingx-discuss(a)lists.starlingx.io Subject: UPDATE: Limited code merges until sanity results improve ** limited merges are still in effect ** Update: The following sanity issues have fixes merged in stx master: - https://bugs.launchpad.net/starlingx/+bug/1829260 - https://bugs.launchpad.net/starlingx/+bug/1829060 - https://bugs.launchpad.net/starlingx/+bug/1828896 - https://bugs.launchpad.net/starlingx/+bug/1828877 - https://bugs.launchpad.net/starlingx/+bug/1827521 - https://bugs.launchpad.net/starlingx/+bug/1828059 (dup) - https://bugs.launchpad.net/starlingx/+bug/1828899 (dup) A build has been triggered on cengn. ** Test team, please start sanity on the new build ** Regards, Ghada From: Khalil, Ghada [mailto:Ghada.Khalil@windriver.com] Sent: Wednesday, May 15, 2019 11:09 AM To: starlingx-discuss(a)lists.starlingx.io<mailto:starlingx-discuss@lists.starlingx.io> Subject: [Starlingx-discuss] Limited code merges until sanity results improve Importance: High Hello all, As discussed in the community meeting today, effective immediately, we ask the core reviewers to only merge code for sanity gating issues and critical/high priority bugs only. This is in an effort to get some stability in the load and address the sanity issues recently reported. A new label, stx.sanity, will be introduced to easily help the community identify issues affecting sanity moving forward. https://bugs.launchpad.net/starlingx/+bugs?field.tag=stx.sanity (tagging of the current issues is in progress) For more details on the current issues, see today's meeting minutes on the community etherpad: https://etherpad.openstack.org/p/stx-status Another email will be sent once the limited merge is lifted. We appreciate your patience as we work together to get stable loads. Regards, Ghada PS: Developers can continue to post reviews as per the normal process. Core reviewers can continue to review code as well, but not to merge them unless they meet the criteria stated above.

1 0

[Starlingx-discuss] [Docs] First round of Intro docs posted
by Jones, Bruce E 16 May '19

16 May '19

I've written some of the documents for the Introduction section of the docs, and posted them for review. Comments and feedback welcome! Brucej https://review.opendev.org/#/c/659668/

1 0

[Starlingx-discuss] Six in a box - Meetings and Status + Packet
by Martinez Landa, Hayde 16 May '19

16 May '19

Hi All, We had our bi-weekly meeting Today, So far the script is able to create the VM without keyboard interaction, Guillermo was working on this, but due to his new responsibilities he transferred all to me. The repo is based on the automation suite from the GDC QA team, they are planning to release the code, When they do, we will release our code too. Also, due to release gating priorities, this project is being paused, therefore the meetings will be canceled. The meeting info was removed from the Wiki, but updates will still be posted on the project etherpad [0]. We discussed that having a script that provides an installation for testing was step 1, but we can have a step 0 Which could be a Packet instance, on this subject we discussed Today with Curtis about the current implementation For Packet; we would like to know: * How can we help on this effort? * Is there something already being work on or where are we right now about the Packet initiative? [0] https://etherpad.openstack.org/p/stx-inabox Best, Haydé

2 1

[Starlingx-discuss] UPDATE: Limited code merges until sanity results improve
by Khalil, Ghada 16 May '19

16 May '19

** limited merges are still in effect ** Update: The following sanity issues have fixes merged in stx master: - https://bugs.launchpad.net/starlingx/+bug/1829260 - https://bugs.launchpad.net/starlingx/+bug/1829060 - https://bugs.launchpad.net/starlingx/+bug/1828896 - https://bugs.launchpad.net/starlingx/+bug/1828877 - https://bugs.launchpad.net/starlingx/+bug/1827521 - https://bugs.launchpad.net/starlingx/+bug/1828059 (dup) - https://bugs.launchpad.net/starlingx/+bug/1828899 (dup) A build has been triggered on cengn. ** Test team, please start sanity on the new build ** Regards, Ghada From: Khalil, Ghada [mailto:Ghada.Khalil@windriver.com] Sent: Wednesday, May 15, 2019 11:09 AM To: starlingx-discuss(a)lists.starlingx.io Subject: [Starlingx-discuss] Limited code merges until sanity results improve Importance: High Hello all, As discussed in the community meeting today, effective immediately, we ask the core reviewers to only merge code for sanity gating issues and critical/high priority bugs only. This is in an effort to get some stability in the load and address the sanity issues recently reported. A new label, stx.sanity, will be introduced to easily help the community identify issues affecting sanity moving forward. https://bugs.launchpad.net/starlingx/+bugs?field.tag=stx.sanity (tagging of the current issues is in progress) For more details on the current issues, see today's meeting minutes on the community etherpad: https://etherpad.openstack.org/p/stx-status Another email will be sent once the limited merge is lifted. We appreciate your patience as we work together to get stable loads. Regards, Ghada PS: Developers can continue to post reviews as per the normal process. Core reviewers can continue to review code as well, but not to merge them unless they meet the criteria stated above.

2 5

[Starlingx-discuss] [ANSIBLE] Werec onfigurations values are stored
by Perez Carranza, Jose 16 May '19

16 May '19

Hi Previously when used "config_controller" we were able to cosult the configuration applied values at ` /opt/platform/config/<sw_version>/cgcs_config`, do you know where those values are of the applied configuration are stored with ANSIBLE strategy? Regards, José

2 1