13:00:08 <ttx> #startmeeting releaseteam 13:00:08 <opendevmeet> Meeting started Fri Sep 27 13:00:08 2024 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:08 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:08 <opendevmeet> The meeting name has been set to 'releaseteam' 13:00:14 <ttx> Ping list: release-team elod armstrong 13:00:29 <elodilles> o/ 13:00:34 <frickler> \o 13:00:43 <ttx> elodilles: o/ 13:00:57 <ttx> frickler: o/ 13:00:59 <ttx> #topic Review task completion 13:01:10 <ttx> Process (Very) late RC1 releases (all) 13:01:30 <ttx> So that was completed for Zun and Barbican team deliverables 13:01:42 <elodilles> \o/ 13:01:54 <ttx> The Kuryr deliverables that were not transferred to other teams remain blocked and should imho just not be released 13:02:11 <elodilles> ACK 13:02:13 <ttx> That is what https://review.opendev.org/c/openstack/releases/+/930554 clarified 13:02:37 <ttx> although we might still want to cut a stable/2024.2 branch for kuryr-kubernetes for Tacker testing purposes 13:02:53 <ttx> (we'll discuss that in open discussion) 13:03:00 <elodilles> ++ 13:03:06 <ttx> After all the projects enabled in devstack by default have been branched, we can engage with the (elod) 13:03:22 <elodilles> yepp, so 13:03:36 <elodilles> QA -> i've pinged the team, 13:03:49 <elodilles> devstack and grenade have branched 13:04:04 <elodilles> and the team started their release process 13:04:18 <elodilles> so we should be on track with this 13:05:01 <elodilles> I18n SIG -> i've sent a mail to I18n SIG chair (Ian) before the meeting 13:06:16 <elodilles> according to their process guide ( https://docs.openstack.org/i18n/latest/release_management.html#projects-affecting-stringfreezes ), 13:07:20 <elodilles> we probably are still OK, and the tooling can be updated for the new stable/2024.2 branch 13:07:50 <elodilles> and last but not least: 13:08:10 <elodilles> Requirements -> the branching patch has been proposed 13:08:38 <elodilles> #link https://review.opendev.org/c/openstack/releases/+/930429 13:08:40 <elodilles> and merged 13:08:47 <ttx> \o/ 13:08:57 <ttx> Process any remaining stable branching exception (ttx) 13:09:05 <ttx> That was also completed 13:09:23 <ttx> #link https://review.opendev.org/q/topic:%22dalmatian-missing-stable-branches%22 13:09:53 <elodilles> sorry, one more thing for the prior task o:) 13:10:07 <ttx> go for it 13:10:40 <elodilles> prometheanfire tonyb : fyi, this can be announced: the requirements freeze is lifted from master :) 13:11:11 <elodilles> ttx: that's all, sorry o:) 13:11:19 <ttx> Notify the TC that it should be safe to apply the process to create the new release series landing pages for docs.openstack.org (elod) 13:11:50 <elodilles> well, i created the patch series: https://review.opendev.org/q/topic:www-dalmatian-final 13:12:13 <elodilles> and the first patch already merged 13:12:16 <frickler> ah, still need to finish reviewing that 13:12:46 <elodilles> frickler: project data is updated, so zuul gave Verified+1 now :] 13:13:20 <frickler> yes, I wanted to check if there are other projects that can be dropped completely like openstack-chef 13:13:30 <elodilles> hmmm, now i remember that we still need a site-map update patch that i promised 13:13:47 <elodilles> frickler: good point, we can drop that from the template 13:13:56 <elodilles> frickler: so that next time it won't appear at all 13:14:18 <frickler> yes, that was my idea 13:14:25 <elodilles> +1 13:15:57 <elodilles> (that's all i think) 13:16:42 <ttx> On the day before the deadline for final release candidates, propose last-minute RCs where needed. (elod) 13:17:21 <elodilles> sorry, i was on PTO yesterday and forgot to do this earlier :/ 13:17:34 <elodilles> i've checked, and now i see that manila, 13:17:55 <elodilles> and kolla team's deliverables need rc2 releases 13:18:05 <elodilles> i'll generate them after the meeting 13:18:18 <frickler> kolla is trailing, there shouldn't be an rc1 even? 13:18:33 <elodilles> frickler: oh, right 13:18:47 <elodilles> frickler: that explains why i saw that many patches 13:18:50 <elodilles> o:) 13:18:51 <ttx> yeah, kolla can wait 13:18:52 <frickler> :) 13:19:10 <elodilles> so then only manila needs a 2nd rc 13:19:11 <ttx> so just manila? 13:19:28 <ttx> ok let's processs it real quick after meeting so that they are all up 13:19:43 <elodilles> ttx: ACK, will do 13:19:48 <ttx> Send countdown email (ttx) 13:19:53 <ttx> will do after meeting 13:20:11 <ttx> After the email is sent, use propose-final-releases to tag the existing most recent release candidates as the final release for projects using the cycle-with-rc model. (elod) 13:20:29 <ttx> we should do that once manila rc2 is processed 13:20:38 <elodilles> yepp \o/ 13:20:46 <elodilles> well 13:21:01 <elodilles> i guess i can generate that after you've sent the mail, 13:21:15 <elodilles> and just signal there that we need an update for manila's RC2 13:21:42 <ttx> elodilles: you could just build that patch on top of the manila rc2 patch 13:21:49 <ttx> that way it includes it 13:21:59 <elodilles> so that PTLs can see what is in progress 13:22:05 <elodilles> ttx: +1 13:22:13 <ttx> #topic Assign R+0 week tasks 13:22:21 <frickler> yes, otherwise we would lose reviews on the update 13:24:08 <ttx> I have several meetings (including press interviews about dalmatian) toward the end of Wednesday so I'd rather not be on the hook for the emails 13:24:26 <elodilles> ACK 13:24:40 <elodilles> btw, do we want to have a meeting next week on Friday? 13:24:46 <ttx> Press release is at 10am CT like usual 13:25:00 <ttx> (that's 17:00 CEST) 13:25:06 <elodilles> +1 13:25:27 <ttx> re: meetign, I think it's good to do a quick postmortem while things are still hot 13:25:41 <elodilles> +1 13:25:54 <elodilles> thanks ttx :) 13:26:37 <ttx> #topic Review countdown email for week R+0 13:26:44 <ttx> #link https://etherpad.opendev.org/p/relmgmt-weekly-emails 13:27:13 <elodilles> LGTM 13:27:24 <ttx> ok will sent shortly after 13:27:41 <ttx> #topic Open Discussion 13:28:09 <ttx> AFAICT the only remaining flag for the release is what to do with kurur-kubernetes 13:28:37 <fungi> we'll also need the temporary semaphore removal/revert changes, i can propose those unless someone else is already working on them (should that have been in the r+0 week tasks?) 13:28:41 <ttx> The discussion is happening on https://review.opendev.org/c/openstack/releases/+/930554 , which is the patch straightening out the various kuryr deliverables 13:29:23 <ttx> fungi: yes please 13:30:04 <ttx> kuryr-kubernetes gate is borked and nobody is working on that, but it's still used as a testing dependency for Tacker 13:30:32 <ttx> so stable/2024.2 testign of Tacker might fail if there is not kuryr-kubernetes stable/2024.2 branch 13:31:18 <ttx> If that is the case we can always manually create a 2024.2 branch for kuryr-kubernetes, outside of the deliverable file 13:31:49 <frickler> and never get that cleaned up again, no good idea IMO 13:31:51 <ttx> so I think we should do that if Tacker testing ends up broken 13:32:21 <ttx> frickler: what do you mean, never get that cleaned up again? 13:32:24 <fungi> tacker could just drop the affected tests on stable 13:32:46 <elodilles> i also added a comment: other option is to cut the branch from 'location' rather than from 'version' (like devstack and requirements branch case). and add a comment to the file why we don't have a release 13:32:49 <frickler> if it is not in the release file, there will be no transition to unmaintained, no eol 13:33:06 <frickler> s/release/deliverable/ 13:33:08 <elodilles> though i'm not 100% sure that our tooling will like that ^^^ 13:33:11 <ttx> ah, hm 13:33:35 <fungi> tacker could also switch stable branch tests to use a tagged version of kuryr-kubernetes if they don't want to drop the affected tests 13:34:26 <ttx> yeah, I feel like those questions are orthogonal to the release. We will not release kuryr-kubernetes for dalmatian, that's the only decision we need to make right now 13:35:01 <ttx> then Tacker has multiple ways to work around potentially-failing stable branch testing? 13:35:11 <frickler> +1 13:35:48 <ttx> But yes I agree in an ideal scenario we would not create anythign 2024.2 for kuryr-kubernetes 13:36:09 <ttx> especially not in a deliverables/dalmatian file 13:36:20 <fungi> if that test dependency is a dead end anyway, they'll need to do something about their choice to depend on it at some point 13:36:24 <ttx> because that would be very confusing 13:36:34 <frickler> they could also backport the "do not test with kuryr-kubernetes" fix from master once they finish it 13:36:38 <ttx> I think they are working on removing it already 13:37:27 <ttx> ok so I think we should merge https://review.opendev.org/c/openstack/releases/+/930554 asap to finalize the release content 13:37:46 <frickler> https://review.opendev.org/c/openstack/tacker/+/926504 looks like updated today 13:37:54 <ttx> otherwise we'll have alarms raised on every script due to unreleased stuff 13:38:54 <elodilles> :S 13:39:06 <elodilles> then i'm OK to merge that patch as it is 13:39:25 <ttx> It's also easy to revert 13:39:42 <ttx> but I don;t think anyone will convince me it's not the right path 13:39:51 <frickler> +2 13:40:50 <elodilles> +2+W'd 13:41:05 <ttx> I'll also abandon https://review.opendev.org/c/openstack/releases/+/928531 and https://review.opendev.org/c/openstack/releases/+/929530 13:41:17 <elodilles> ++ 13:41:27 <elodilles> let's abandon them 13:42:01 <ttx> ok done 13:42:21 <ttx> Dalmatian review board will soon be clean! 13:42:43 <ttx> anything else we need to cover? 13:42:51 <ttx> frickler: have a good time off! 13:42:52 * frickler just notes being on holiday for 3 weeks after today. also added to the etherpad, please carry over to the next cycle 13:43:01 <frickler> thx 13:43:18 <fungi> what time on wednesday are people planning to start mashing buttons? i'll set an alarm so i can be awake in case there are issues 13:43:45 <elodilles> frickler: yepp have a nice relaxing holiday o/ 13:44:09 <ttx> "On release day (around 10:00 UTC), approve the final release patch created earlier." 13:45:26 <fungi> and that's based on prior adjustments we made to start earlier due to publication delays and increased test turnaround time we hit in recent releases, right? 13:45:57 <fungi> i guess that gives us 5 hours to press release 13:46:03 <fungi> so should suffice 13:46:24 <ttx> I think so yes 13:46:31 <elodilles> yepp 13:46:57 <ttx> Anything else before we close? 13:47:19 <elodilles> - 13:47:25 <opendevreview> Merged openstack/releases master: Do not release Kuryr in Dalmatian https://review.opendev.org/c/openstack/releases/+/930554 13:47:30 * fungi has nothing 13:47:33 <ttx> woohoo! 13:47:37 <ttx> #endmeeting