[Starlingx-discuss] Distro.openstack meeting notes for May 21 2019
5/21 meeting: * How are we going to handle the changes for Placement / Nova? * See email from Chris Dent: ? > Do we lose anything without monkey patch on placement? ? No, placement doesn't need eventlet monkey patching and would prefer the package not be imported at all. ? I suspect that the reason that this problem has not been widely noticed is because in Stein placement is extracted to its own repo and project and primarily tested using the extracted placement (the nova functional tests use the extracted placement). ? nova-placement-api remains around in Stein to ease upgrades (that is you can upgrade everything to Stein, keeping placement in nova, and then once that is done, switch to the extracted placement). Docs for some of that at: https://docs.openstack.org/placement/latest/upgrade/to-stein.html ? Placement doesn't intentionally use eventlet anywhere. The placement-in- nova inherits monkey patching as a result of package hierarchies. In the extracted placement that goes away. ? Therefore the cleanest, but not necessarily quickest, solution to this problem would be to switch to the extracted placement, which is going to need to happen at some point anyway. Now might be the time to pay the cost? ? If there are questions or issues with making that happen, please as for help on the openstack-discuss list with a [placement] tag and several folk will leap to your aid. * Dean: in a perfect world we would switch to the Stein extracted Placement now, which would ease upgrading. The code in our stable Nova branch is functionally equivalent (intentionally). Switching would not impact us functionally. We believe Helm does _not_ have a Placement container and is consuming the Nova version. * Frank: If we don't do this work now, we will have to do a database migration to do. Which might not be that bad. * If we stay with the current code, there might be a subtle issue due to how eventlet can be blocking due to interactions with the Whiskey app. In fact, we are already hitting issues. * If we decide to switch, we have to create our own placement container. If we don't, we have to remember to do a database migration for upgrades. * AR Bruce to find someone who can build the Docker image for Placement. Worst case if the work does not land for 2.0 it's OK. ? New story https://storyboard.openstack.org/#!/story/2005750 created ? Will likely fix https://bugs.launchpad.net/bugs/1827692 * We are (will be) switching to our own stable branch for Nova. When / if do we re-sync with upstream stable stein? We have changes from Gerry pending review against the repo. AR Dean to check to see if we even need to rebase. * PCI Affinity review 2 merged! Fix for simplex submitted for review * Alex (Nova Core) reports: * Just FYI, Kashyap ping me last night, he abandoned the spec https://review.opendev.org/642030. I guess you need someone else take that over. * 99Cloud to take over this spec. * Scrub other open bugs https://bugs.launchpad.net/starlingx/+bugs?field.tag=stx.distro.openstack * Yongli please check https://bugs.launchpad.net/starlingx/+bug/1824858 it might be related to the Orphan instance * Bruce to get help for bugs assigned to him * One of the two pending reviews for the rbd disks issue (640271) has merged. 642667 is still pending.
Hi bruce, · One of the two pending reviews for the rbd disks issue (640271)[1] has merged. 642667[2] is still pending. In fact, the patch(642667) aims to optimize the lvm image backend like rbd. So the rbd issue of starlingx[3] has been finished after the patch(640271) has been merged. We can remove the link of 640271 from rbd issue[4] and set this issue as 'Done'. [1] https://review.opendev.org/#/c/640271/ [2] https://review.opendev.org/#/c/642667/ [3] https://docs.google.com/spreadsheets/d/1udAtEpQljV2JZVs-525UhWyx-5ePOaSSkKD1... [4] rbd disks: convert from source format to raw Intel to fix upstram Best Regards -- Boxiang 发件人："Jones, Bruce E" <bruce.e.jones@intel.com> 发送日期：2019-05-22 00:25:57 收件人："starlingx-discuss@lists.starlingx.io" <starlingx-discuss@lists.starlingx.io> 主题：[Starlingx-discuss] Distro.openstack meeting notes for May 21 2019 5/21 meeting: · How are we going to handle the changes for Placement / Nova? See email from Chris Dent:§ > Do we lose anything without monkey patch on placement? § No, placement doesn't need eventlet monkey patching and would prefer the package not be imported at all. § I suspect that the reason that this problem has not been widely noticed is because in Stein placement is extracted to its own repo and project and primarily tested using the extracted placement (the nova functional tests use the extracted placement). § nova-placement-api remains around in Stein to ease upgrades (that is you can upgrade everything to Stein, keeping placement in nova, and then once that is done, switch to the extracted placement). Docs for some of that at: https://docs.openstack.org/placement/latest/upgrade/to-stein.html § Placement doesn't intentionally use eventlet anywhere. The placement-in- nova inherits monkey patching as a result of package hierarchies. In the extracted placement that goes away. § Therefore the cleanest, but not necessarily quickest, solution to this problem would be to switch to the extracted placement, which is going to need to happen at some point anyway. Now might be the time to pay the cost? § If there are questions or issues with making that happen, please as for help on the openstack-discuss list with a [placement] tag and several folk will leap to your aid. Dean: in a perfect world we would switch to the Stein extracted Placement now, which would ease upgrading. The code in our stable Nova branch is functionally equivalent (intentionally). Switching would not impact us functionally. We believe Helm does _not_ have a Placement container and is consuming the Nova version. Frank: If we don't do this work now, we will have to do a database migration to do. Which might not be that bad.If we stay with the current code, there might be a subtle issue due to how eventlet can be blocking due to interactions with the Whiskey app. In fact, we are already hitting issues.If we decide to switch, we have to create our own placement container. If we don't, we have to remember to do a database migration for upgrades.AR Bruce to find someone who can build the Docker image for Placement. Worst case if the work does not land for 2.0 it's OK.§ New story https://storyboard.openstack.org/#!/story/2005750 created § Will likely fix https://bugs.launchpad.net/bugs/1827692 · We are (will be) switching to our own stable branch for Nova. When / if do we re-sync with upstream stable stein? We have changes from Gerry pending review against the repo. AR Dean to check to see if we even need to rebase. · PCI Affinity review 2 merged! Fix for simplex submitted for review · Alex (Nova Core) reports: Just FYI, Kashyap ping me last night, he abandoned the spec https://review.opendev.org/642030. I guess you need someone else take that over.99Cloud to take over this spec.· Scrub other open bugs https://bugs.launchpad.net/starlingx/+bugs?field.tag=stx.distro.openstack Yongli please check https://bugs.launchpad.net/starlingx/+bug/1824858 it might be related to the Orphan instanceBruce to get help for bugs assigned to him· One of the two pending reviews for the rbd disks issue (640271) has merged. 642667 is still pending.
participants (2)
-
Jones, Bruce E
-
朱博祥