16:00:11 #startmeeting releaseteam 16:00:12 Meeting started Thu Aug 29 16:00:11 2019 UTC and is due to finish in 60 minutes. The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:13 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:15 The meeting name has been set to 'releaseteam' 16:01:03 #link https://etherpad.openstack.org/p/train-relmgt-tracking Agenda 16:01:13 Ping ttx dhellmann diablo_rojo hberaud evrardjp armstrong tonyb 16:01:15 ohai 16:01:18 o/ 16:01:43 o/ 16:02:39 Thank you ttx for the agenda again. 16:02:53 #topic Status update on recent release failures 16:03:02 #link http://lists.openstack.org/pipermail/release-job-failures/2019-August/thread.html 16:03:05 All the fails 16:03:09 yeah so I motices a bunch 16:03:15 damn I can't type anymore 16:03:23 noticed a bunch of fails 16:03:37 I *think* we discussed these in channel, but we can make sure. 16:03:43 The propose-update-constraints fails I did not investigate since i think somebody else did 16:04:06 They seem to be fixed, just unsure that we corrected the failed ones (or that they need to) 16:04:18 So would welcome someone double-checking 16:04:18 Those we decided were ok. One was manually proposed, the rest are getting picked up in the automatic updates. 16:04:24 ok 16:04:33 The tag-releases job fail I did investigate 16:04:43 while creating ocata-eol tag for all TripleO projects, as a post-job after https://review.opendev.org/#/c/677478/ merged, failed to push ocata-eol tag to openstack/os-cloud-config 16:04:44 aren't the triple O something to be expected? 16:04:53 I think we can ignore them yes 16:05:14 This is one that has come up in the past. A retired repo does not have a .gitreview file, so it will fail. 16:05:15 except that ocata-eol tag is still missing from openstack/os-cloud-config 16:05:41 ok, I'm fine with ignoring, not a big deal anyzay 16:05:43 I think, but not certain, that this should be the last series that will be affected by that. 16:06:00 It is retired and the current master README makes that very clear. 16:06:01 same for publish-release-notes ones 16:06:10 probably irrelevant 16:06:11 ttx: maybe we should just notify the ptl to do the EOL tag him/her-self? 16:06:18 well, also retired repos are usually set to read-only, so you can't push tags to them 16:06:25 evrardjp: they can't 16:06:28 It's still mwhahaha for now, right? 16:06:32 evrardjp: They would have to change ACLs for that. 16:06:37 fungi: that may be broken though 16:06:51 oh right I thought it was not mutually exclusive in the acl by default 16:06:52 ok 16:06:55 fungi: jeepyb ignores projects with the retired acl config so doesn't push the acl update to RO them 16:06:57 anyway, I think we can ignore it as noise 16:07:18 from running an EOL process on dead repos 16:07:22 clarkb: ahh, i guess we need to fix that when someone finds time, but i thought it got solved some months ago 16:07:40 * fungi would have to look back at jeepyb git history 16:07:42 just wanted to make sure everyone was aware 16:08:09 FWIW, we did add some clarifying text to the retirement process to make sure .gitreview was kept along with the README so we hopefully don't have this issue again. 16:08:09 I have nothing more on that topic 16:08:23 We could have fixed it for this repo, but it was considered not worth the effort. 16:08:31 "We'll just remember this for next time" :D 16:08:38 cary ove 16:08:40 r 16:08:44 #topic Final release period start 16:08:44 r 16:08:56 Hi! me again 16:09:12 #link https://releases.openstack.org/reference/process.html#final-library-release-week-before-milestone-3 16:09:18 So I did spend some time this morning planning the end of the cycle on the tracking doc 16:09:21 and copying the tasks 16:09:30 It should all be set and ready 16:09:35 Thanks! 16:09:59 Including reminders for things we should check in weekly emails 16:10:26 One thing that would be cool is if you could add any availability issue 16:10:46 Thanks for the reminder. I do have some travel coming up. 16:10:55 so that the other ones are aware they will have to care for those tasks 16:12:03 Then we need to assign next week tasks 16:12:32 would be good if someone else did the library autoreleases this time 16:12:51 It's a bit manual, but someone might get bored enough to build automation/tools 16:13:29 We should really get that proposed at the beginning of the week too. 16:13:40 Or even tomorrow or the weekend to make sure there is time by the actual deadline. 16:13:55 I can take that unless someone else wants to grab it. 16:14:10 I think Monday should be enough... Deadline is Thursday EOD 16:14:28 I'll take the weekly post draft 16:15:00 No other takers on proposing the releases? 16:15:01 I cannot say 100% I will work on this, so I am preferring to let someone do it, but I will actively review, and help 16:15:23 I just cannot commit yet 16:15:31 OK, I will take that, but if someone finds the desire and time to do it before me, I will not be upset. ;) 16:15:36 evrardjp: could you take the last task? 16:15:38 ofc :) 16:15:55 I'll let you add your name to it 16:16:03 the ones with the examples? 16:16:09 yes 16:16:36 OK we seem to be good 16:16:47 Note that starting now we have meetings every week and weekly emails 16:16:54 that would be sad if I couldn't do one liners 16:16:58 Crunch time. 16:17:26 #topic Notify the Infrastructure team to generate an artifact signing key (but not replace the current one yet), and begin the attestation process. 16:17:29 That is all I had on this topic. You can scroll down that doc and see what the future looks like now 16:17:40 fungi: ^^ Notified? 16:17:43 thanks smcginnis, i'm adding that to my to do list now! 16:17:49 Thanks! 16:17:59 #topic R-6 email content final review 16:18:06 #link https://etherpad.openstack.org/p/relmgmt-weekly-emails 16:18:07 I did a final pass on that one 16:18:12 fungi: I would be happy to know how things are done, if you don't mind filling me in on the topic after the meeting or at another moment 16:18:16 Also removed confusing double R mentions 16:18:32 evrardjp: you bet, it's a thoroughly documented process 16:18:39 ttx: Double R? 16:18:39 The R-6 email is sent at the end of the R-7 week, but i use a date to point to that 16:19:09 the "send by" in the etyherpad should make it a bit less confusing 16:19:23 ++ 16:19:32 So the email to send is ready for you around line 350 16:19:37 more readable indeed 16:19:44 Just re-read through it and I think it looks good. 16:20:27 agreed 16:20:49 {o} 16:21:33 #topic Open discussion 16:21:53 So... PTL nominations are open. 16:21:56 smcginnis: Are you running for U? 16:22:08 Unless we have someone that wants to, I was going to just go again. 16:22:38 ++ 16:22:40 thanks smcginnis! 16:22:48 once more unto the breach! 16:23:04 on another note: 16:23:06 I'll wait a bit in case someone else steps forward, but I guess I should probably get that drafted. 16:24:59 Anything else? 16:25:01 so yeah my other note... hum... how can I say that 16:25:08 I have a bad news: I won't be there to help you fine folks doing the final release (and the week before). I am a little sad, but family first. 16:25:30 Don't feel bad about putting your family first. 16:26:02 evrardjp: well, it's not as bad as you think. Peak activity is actually the two weeks just before 16:26:03 I imagined joining you and having a party at my home blowing in the nice wooden whistle for the great pleasure of my neighbours 16:26:07 but it's not gonna happen 16:26:14 Last week we just drink 16:26:18 :) 16:26:43 ttx: that's reassuring, but makes me even sader 16:26:48 even more sad* 16:27:06 anyway 16:27:07 Don't worry, we can find some other excuse to drink together to make up for it. ;) 16:27:11 I will fill that in the etherpad 16:27:11 yeah, release week is "push the button, and then cross your fingers and hope you don't need the infra team" 16:27:27 smcginnis: btw I did push the "post-RC1 tasks" to the week after RC1 week since that is when we actually did those for all previous releases 16:27:29 We should actually add that to the process doc ^ 16:27:49 ttx: OK, that makes sense. 16:27:53 i don;t expect us having all our ducks^WRC1s in order before well past end of week anyway 16:28:04 like always 16:28:12 Yeah 16:28:18 also spreads out load nicely 16:28:43 Very true. 16:28:49 but yeah, R-2 is actually the heaviest week, and *I* won't be around that week much 16:28:57 so apologies for that :) 16:29:19 * smcginnis notices ttx has been planning vacations around busy weeks 16:29:20 :P 16:29:32 It's not vacation, it's OID Nordics 16:29:46 that's a big event 16:29:48 I know it looks a lot like vacation but my wife says it's not 16:29:50 That should be a good one. Wish it was closer for me. 16:29:51 ..I will also be there? 16:29:57 diablo_rojo: I hope so 16:30:12 OK, anything else we should discuss in-meeting? 16:30:20 I wanted to go, but I decided to stay to help releases. ahem ahem 16:30:24 diablo_rojo: you just are better than me at getting work done on the road 16:30:32 smcginnis: nothing from me 16:30:40 ttx, if you say so lol I do try. 16:30:40 evrardjp: I'm not sure I buy that 16:30:45 :D 16:30:52 oh, also in case it didn't get mentioned visibly enough, the opendev sysadmins are performing a batch of project renames (including the kayobe deliverable namespace moves) on monday september 16 16:31:23 yes I left a note on that releases review to put it on hold 16:31:33 it's cycle-trailing so it should be fine 16:31:36 fungi: Thanks for pointing that out here. Good to get that captured in meeting notes. 16:31:50 #link http://eavesdrop.openstack.org/meetings/infra/2019/infra.2019-08-27-19.01.log.html#l-228 16:32:39 OK, if nothing else, my coffee cup needs a refill. :) 16:32:49 thanks smcginnis!!! 16:33:01 Thanks everyone. And thanks again ttx for managing the agenda and getting the tasks planned out. 16:33:02 thanks smcginnis 16:33:09 #endmeeting