14:00:02 #startmeeting releaseteam 14:00:02 Meeting started Fri Mar 24 14:00:02 2023 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:03 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:03 The meeting name has been set to 'releaseteam' 14:00:08 o/ 14:00:09 Ping list: hberaud armstrong elodilles 14:00:19 o/ 14:01:01 #topic Review task completion / Release postmortem 14:01:12 We did complete all the last week tasks in time for release on March 22 14:01:55 The only issue was that duplication of release notes that frickler spotted? 14:02:13 and as far as i understand that is not a new problem, 14:02:30 and it does not affect many deliverables 14:02:33 also we ran a little over time waiting for the final changes to merge 14:02:36 all in all, not bad given the impact of playing with release names in the middle of the cycle 14:03:02 fungi: was there a change that took too long, or just us (me) not being around to push buttons? 14:03:03 fungi: yes, i think it took a bit longer than at zed release 14:03:21 and there was a release name/number ordering problem in the openstack-manuals repo which clarkb uploaded a last-minute fix to correct 14:03:59 right 14:04:19 and sorry, that was my fault ^^^ :/ i did not notice at proposal time that the ordering had an issue :/ 14:04:22 Nothing that really points to an issue on our process though? 14:04:59 rough timeline is that the main releases change was approved at 11z and took an hour in the gate pipeline to validate all the new releases, then the tag-releases job started at approximately 12z and took roughly an hour to push all the tags and for the resulting tag/release pipeline jobs to complete 14:05:36 so as of 13z we were at the double-check, merge the openstack-manuals change and then the releases final change 14:06:14 as i remember the final change merge took longer than at zed case (mentioned that above) 14:06:18 those took some time (also the schedule said not to approve them until 14z), and so the releases.o.o site didn't update until about 15:10z 14:06:54 probably better not to have the schedule say to wait until an hour before press release for those steps 14:07:02 ok so maybe we need to fix that 14:07:51 starting the whole process a little earlier would also have given us a bit more breathing room to deal with the last-minute docs site issue, but still that was corrected on the published version before 16z 14:08:10 ok noted 14:08:12 i agree with fungi, maybe we don't need to wait (or we should wait less) 14:08:16 (under things to change) 14:08:39 related, we merged a change the day before the release to temporarily remove release publication job semaphores and then merged a revert of that once the tag pipeline cleared out 14:09:01 should we add that to the process? 14:09:05 so instead of taking all day to update release notes for projects, they were done before the packages were even finished uploading 14:09:11 it would be a good idea to add, tes 14:09:14 s/tes/yes/ 14:09:33 ok notes too 14:09:34 +1 14:09:35 noted 14:10:24 as for the wait-till-14z in the schedule, we did go ahead about 20 minutes early because i was worried things could run long (and they did) 14:10:36 alright, anything else on release postmortem? 14:11:23 fewest issues of any release i can remember, even though we had totally new release designation schema 14:11:30 so great work everyone! 14:11:43 yeah, let's not change that schema again please 14:11:55 could have bitten us more deeply 14:11:55 :) 14:12:10 thanks too everyone \o/ 14:12:42 #topic Assign next week tasks 14:13:00 We have a new etherpad at https://etherpad.opendev.org/p/bobcat-relmgt-tracking 14:13:33 We need to assign R-27 tasks but also chairs for meetings until milestone-1 14:14:00 Also confirmig meeting skips 14:14:05 i'll finish the etherpad today, but feel free to fix typos, missed tasks, other things later on 14:15:21 elodilles: can you chait next week meetings? I took the one after that 14:15:25 chair* 14:15:42 ttx: yepp, i can. 14:15:55 ttx: though, do we have a PTG session? 14:16:59 (let's not mark the etherpad as 'done' as it still needs updates & reviews afterwards o:)) 14:18:41 Hmm I don;t think we booked a PTG session. Let me doublecheck 14:18:43 i don't see a track for it 14:18:50 confirmed no 14:18:51 i can add one if you want it 14:18:53 We could make our usual meeting a video session 14:19:06 there is still room on the Friday 14:19:07 ++ 14:19:26 OK let's turn that meetign into a PTG session 14:19:32 book it into folsom-FriB2 since that's where os-tc is starting the next hour 14:19:47 good idea, I will once you add the track 14:19:55 adding now 14:20:30 it's added, but i see you're not in #openinfra-events 14:20:55 os-release is the track 14:20:56 I am 14:21:37 usually it was relmgt or rel-mgt? 14:21:59 the other openstack non-projects are in there as os-tc, os-security, et cetera 14:22:10 so as not to conflict with similar tracks other non-openstack projects might want 14:22:18 ack 14:22:27 so that's 4pm CET next Friday 14:23:03 also i guess ttx was in the channel via a bridge, because irc wasn't showing him in there and then he suddenly joined when he sent something to the channel, hence my confusion 14:23:16 elodilles: I will take the PTL email task if you take the countdown task 14:24:04 ttx: deal! 14:24:22 thanks! 14:24:34 been a while since I last took it :) 14:24:59 #topic Review countdown email 14:25:03 I don;t think tehre is one? 14:25:35 no, we don't have directly after release :) 14:25:40 #topic Open Discussion 14:25:55 I wanted to raise PTG, but we already discussed that 14:26:03 \o/ 14:26:17 #agreed next meeting will be done as part of PTG 14:26:25 anything else? 14:26:54 nothing else from me 14:27:00 On Bobcat etherpad, do y'all agree with the meeting skips between now and milestone-1? 14:27:03 nope 14:27:08 yeah 14:27:32 yeah release team downtime 14:28:42 ++ 14:28:54 alright, all set 14:28:59 #endmeeting