| frickler | elodilles: I think we would rather abandon the remaining patches in https://review.opendev.org/q/topic:%22bobcat-stable%22+status:open than force-merge them? (except ceilometer now) | 10:19 |
|---|---|---|
| frickler | gmaan_pto: did you intentionally change your nick and email address? I think it is the conflict with the governance data that makes e.g. https://review.opendev.org/c/openstack/releases/+/948214 not get the PTL-Approved flag | 11:47 |
| opendevreview | Merged openstack/releases master: eol the bobcat release of watcher deliverables https://review.opendev.org/c/openstack/releases/+/948118 | 11:49 |
| * frickler assumes https://review.opendev.org/c/x/stackalytics/+/946012 is related | 11:55 | |
| opendevreview | Merged openstack/releases master: [tool] Fix get-contacts --ptl and --liaisons https://review.opendev.org/c/openstack/releases/+/946648 | 12:31 |
| opendevreview | Merged openstack/releases master: [tool] Fix get-contacts in case of retired team https://review.opendev.org/c/openstack/releases/+/947958 | 12:31 |
| opendevreview | Merged openstack/releases master: [tool] Fix unreleased-since listing https://review.opendev.org/c/openstack/releases/+/947965 | 12:52 |
| opendevreview | Merged openstack/releases master: [tool] Skip tempest plugins when listing unreleased and open changes https://review.opendev.org/c/openstack/releases/+/947942 | 12:54 |
| opendevreview | Merged openstack/releases master: [tool] Fix missing-releases command https://review.opendev.org/c/openstack/releases/+/945630 | 13:00 |
| opendevreview | Merged openstack/releases master: [ec2-api] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948202 | 14:07 |
| opendevreview | Elod Illes proposed openstack/releases master: [OpenStackSDK] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948218 | 14:17 |
| opendevreview | Merged openstack/releases master: [keystone] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948230 | 14:32 |
| opendevreview | Merged openstack/releases master: [venus] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948242 | 14:32 |
| opendevreview | Merged openstack/releases master: [tacker] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948241 | 14:45 |
| opendevreview | Merged openstack/releases master: [storlets] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948207 | 14:50 |
| opendevreview | Brian Rosmaita proposed openstack/releases master: Final cinder project 2023.2 (Bobcat) releases https://review.opendev.org/c/openstack/releases/+/948332 | 14:51 |
| opendevreview | Merged openstack/releases master: [kuryr] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948210 | 14:55 |
| opendevreview | Merged openstack/releases master: [adjutant] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948221 | 14:55 |
| *** gmaan_pto is now known as gmaan | 14:59 | |
| opendevreview | Merged openstack/releases master: Release ceilometer for stable/2023.2 https://review.opendev.org/c/openstack/releases/+/941381 | 15:11 |
| croelandt | Hello! I'm the new Glance PTL and I'm wondering whether a patch of ours is backportable to 2024.2/2024.1. It is a bugfix that... sort of requires a new feature. We work around slowness while sorting image locations by adding a new weighing mechanism to the glance store. As a result, we get two patches: https://review.opendev.org/c/openstack/glance/+/940528 and | 15:12 |
| croelandt | https://review.opendev.org/c/openstack/glance_store/+/940529 . While we're not editing requirements files in these patches, we definitely need a newer version of glance_store for the glance patch to work. Is that ok to merge, or should we give up on fixing this issue in stable branches? | 15:12 |
| opendevreview | Merged openstack/releases master: [blazar] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948223 | 15:14 |
| opendevreview | Merged openstack/releases master: [ironic] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948209 | 15:14 |
| opendevreview | Merged openstack/releases master: [designate] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948225 | 15:14 |
| opendevreview | Merged openstack/releases master: [solum] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948208 | 15:14 |
| elodille1 | croelandt: let me try to answer :) | 15:15 |
| elodille1 | croelandt: according to stable policy no feature patches should be backported. though if it kind of solves some serious issue, then a team can decide to backport anyway. but these kind of backports should be avoided as much as possible | 15:16 |
| elodille1 | and only grant exception for that kind of backport if that does not cause any regression at all and upgrade should not cause any issue or behaviour change on already deployed environments | 15:18 |
| elodille1 | regarding requirements.txt update: that is one thing that should be also avoided. though oslo and other openstack library typed deliverables' version can be bumped | 15:19 |
| elodille1 | (if those libraries have a new stable/202*.* releases as well) | 15:20 |
| croelandt | yes, we started the backport with the glance_store patch for 2024.2 and ended up having a new stable/2024.2 release for glance_store. I can imagine a situation where an admin ends up updating glance without updating glance_store though, and in that case, things are not going to go well | 15:21 |
| elodille1 | croelandt: well, a release note can be added to glance release to *only* *upgrade* together with glance_store. | 15:24 |
| elodille1 | so that sounds feasible to me, but if glance team decides that it's too risky and operators may fail with the upgrade together glance & glance_store and it could cause somewhat serious issues, then probably best is to not backport these | 15:26 |
| croelandt | yeah, it's easy to overlook the release note | 15:29 |
| elodille1 | croelandt: so I say it's not a strong YES or NO, but the team should carefully consider which is more safe / beneficial from operator point of view | 15:29 |
| croelandt | Thanks a lot for the detailed answer, it helps a lot! | 15:30 |
| elodille1 | croelandt: no problem, i hope it makes sense what i said and gives some direction how to proceed | 15:31 |
| *** elodille1 is now known as elodilles | 15:31 | |
| croelandt | yep, this is definitely going to help us figure out what to do with those patches! | 15:34 |
| elodilles | fungi clarkb : is it safe now if i run EOL branch clean up ? is any activity planned on infra in the coming hour? | 16:52 |
| clarkb | elodilles: it should be safe. I am trying to do a noop update to gitea but ran into problems and plan to shutdown the old gerrit server today as part of its eventual removal. But I wouldn't expect either of those to impact you | 16:53 |
| elodilles | clarkb: ACK, thanks, i'm starting the script now | 16:53 |
| clarkb | elodilles: one thing is we have a very low percentage of failures for job log uploads | 16:53 |
| clarkb | elodilles: not sure how important those jobs / logs are for EOL branch cleanup | 16:54 |
| clarkb | we don't produce things like release artifacts so we can probably just reneqneue any jobs that have problems | 16:54 |
| elodilles | hmmm, actually, the script does not cleanup 2023.2 bobcat branches as it is not yet marked as EOL in releases.o.o , so i'd rather wait 2-3 days with the cleanup | 16:58 |
| clarkb | wfm I don't expect any big changes to zuul or gerrit this week. I'm trying to finalize the gerrit server swap which at this point means cleaning up the old stuff that is shutdown/unused | 16:59 |
| elodilles | clarkb: i don't think missing job logs would cause any issue when deleting EOL'd branches, but i'm postponing the cleanup anyway ^^^ | 16:59 |
| elodilles | sorry for the noise o:) | 16:59 |
| elodilles | clarkb: ACK, I'll try to run the cleanup this week (planned deadline is April 30th, but there is a slight chance some teams will miss that o:)) | 17:00 |
| clarkb | and the noop gitea update is a fix to how we build gitea images due to a change in one of the gitea module dependencies hosting. We're just rebuilding the same version of everything and deploying that to prove the image build updates work | 17:01 |
| elodilles | +1 | 17:02 |
| opendevreview | Brian Rosmaita proposed openstack/releases master: Final cinder project 2023.2 (Bobcat) releases https://review.opendev.org/c/openstack/releases/+/948332 | 17:32 |
| opendevreview | Brian Rosmaita proposed openstack/releases master: [cinder] Transition 2023.2 Bobcat to End of Life https://review.opendev.org/c/openstack/releases/+/948224 | 17:32 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!