[Starlingx-discuss] Distro.openstack meeting 9 / 10 / 2019

[Jones, Bruce E]

09/10/2019 Meeting
Openstack Train release Oct 16, : https://releases.openstack.org/train/schedule.html#t-final
*         TL/PL election
*         Transition to OpenStack master / Train - how/when do we do this?

  *   When - as soon as possible.  We can run off master.

  *   Besides those projects which could be directly fetch from upstream to generate docker images, we might need additional efforts on following projects:
?  0. Create a manifest to pull from Train/master, do a private build and test it (Sanity)
?  1. stx-keystone on host: which was created from a SRPM (e.g. Source/openstack-keystone-15.0.0-0.2.0rc2.el7.src.rpm) + patches
?  2. starlingx-dashboard (plugin into horizon)
?  3. Port stx-nova branches with the latest NUMA live migration patches (which still haven't merged) - pull using the repo tool directly from gerrit (due to the dynamic changes, for example, in Stein, Dean made 2 branches stein.1 and stein.2 in "starlingx-staging") .  Latest NUMA patches depend on new features in Train making backporting to Stein hard.
?  4. ironic (having configs in stx-config and patches in openstack-helm, might need efforts to adapt Ironic on Train)
?  4.5. Decide if we need to rebase openstack-helm and when.  If needed rebase to the new version
?  5. Verify proper operation of helm overrides-  Helm itself may lag the overall OpenStack release.   We may find ourselves at the leading edge of Helm Train.
*         Live migration patch status

  *   https://review.opendev.org/#/q/topic:bp/numa-aware-live-migration+(status:open+OR+status:merged)

  *   We need these patches tested on Train - test Nova/Train with the patches applied and verify correct behavior - Ya at 99 to handle.
*         Upstream activities

  *   Nova placement helm chart https://review.opendev.org/#/c/662229/
?  Zhipeng: the dependent patch is still waiting for W+1.

  *   Orphan instance cleanup: https://review.openstack.org/#/c/627765/
?  Yongli: Witt concern the safe of this patch. Hard to tell if it's safe for all code flow.  Not update due to handle topology's cli code.

  *   NUMA topology: https://review.openstack.org/#/c/621476/
?  Yongli: 9/10 This patch is merged.
?  python-novaclient patch is merged: https://review.opendev.org/#/c/670790/
?  python-openstackclient patch is reviewing: https://review.opendev.org/#/c/680928/

  *   Nova Upstream Activities from 99Cloud:
?  Wang Ya: https://review.opendev.org/#/q/branch:master+topic:bp/expose-auto-converge-post-copy
*         targeted for U
*         https://review.opendev.org/674523
*         https://review.opendev.org/674525
?  Wang Ya: https://review.opendev.org/#/q/project:openstack/nova+branch:master+topic:bp/cpu-model-selection
*         targeted for Train
*         https://review.opendev.org/670298
*         https://review.opendev.org/670300
*         3 patches have got one +2, but still need more +2. Ya is pinging core reviewers.
?  99Cloud to update other patch updates
*         Defect status / updates

  *   Two of our open High defects are believed to be caused by https://review.opendev.org/#/c/674719/ which was going to be reverted by Zhipeng, but the revert review has been abandoned.  Yong to check with Zhipeng.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.starlingx.io/pipermail/starlingx-discuss/attachments/20190910/4888d9ec/attachment-0001.html>