15:00:56 #startmeeting releaseteam 15:00:57 Meeting started Fri Aug 31 15:00:56 2018 UTC and is due to finish in 60 minutes. The chair is smcginnis. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:58 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:01:00 The meeting name has been set to 'releaseteam' 15:01:07 o/ 15:01:07 o/ 15:01:33 o/ 15:01:57 I see Sigyn here, so not pinging the list. 15:02:06 wise 15:02:12 #link https://etherpad.openstack.org/p/rocky-relmgt-tracking Agenda 15:02:20 All the way down to the bottom! 15:02:50 #topic Remaining release week process 15:02:54 heh 15:03:14 #link https://review.openstack.org/#/c/598621/ Marking Stein as under development 15:03:34 I think that's the last step in getting the releases.o.o up to date. ^ 15:03:45 yep 15:04:04 ttx: Thanks for getting that one up there. I updated Rocky but didn't cover it all. 15:04:14 o/ 15:04:35 The other item listed on the etherpad was to notify openstack-dev. I took care of that one a couple hours ago. 15:05:10 Overall I think the release went well. We ended up with good timing with release jobs finishing right in time for annabelleB's webinar. 15:05:49 Thanks everyone for helping get things through. 15:05:57 Thank YOU 15:06:00 Thanks to our fearless leader! 15:06:02 #topic 404s on Rocky release page 15:06:11 #link https://etherpad.openstack.org/p/vPBa2EDSdD 15:06:36 I transcripted the list on the meeting agenda 15:06:41 for clarity 15:06:56 OK, good. 15:07:14 #link https://review.openstack.org/598952 Skipping release-test for missing-releases 15:07:22 bgpvpn is the only "urgent" one 15:07:25 We still have a few without signatures. 15:07:35 the others we can add to teh Stein team plan 15:07:38 bgpvpn is fixed now, right? 15:07:50 Yep, I believe so. 15:07:57 * ttx checks 15:07:58 After switching the job, that one passed. 15:08:00 we switched jobs and reenqeuued the tag 15:08:15 And my run of missing-releases this morning did not complain about anything. 15:08:32 i think it was after wine-thirty in france though so you missed the exciting conclusion 15:08:33 ok yes solved 15:08:37 They still have the dependency issues in their code, but the releases can now happen despite that. 15:08:54 Missing signatures for neutron-tempest-plugin ? 15:09:15 can/should we generate signature files? or is it better to just leave those? 15:09:15 #link https://review.openstack.org/#/c/596907/ fixes tripleo-ui signing 15:09:22 it was running the python3 job which was missing the signing role. let me find the fix we merged 15:09:35 neutron-tempest-plugin i mean 15:09:50 ok 15:10:29 dhellmann: I'd say leave them out, and maybe have a the same skip-link flag be able to act on signatures only 15:10:34 I think dhellmann fixed up the jobs that were missing the signing. 15:10:59 I fixed 2 jobs; I don't know if that was all of them 15:11:06 ttx: yeah, that makes sense 15:11:13 At least all we've identified so far I think. 15:11:33 I will restore the story for that work 15:11:49 dhellmann: thx 15:11:51 huh, i'm not finding the patch for the python3 job 15:11:51 I see the notes for https://storyboard.openstack.org/#!/story/2001689 show it was closed as invalid. 15:12:18 dhellmann: OK, so you'll restore and we do want to list without linking? 15:12:34 "We can't remember why we want to do this instead of just removing the entry for the broken release, so I am going to close this as invalid." 15:12:35 smcginnis : yeah 15:12:46 OK, that sounds reasonable to me. 15:12:48 Now I remember why it's better 15:13:06 please leave a note on the story so we don't forget again :-) 15:13:08 which is that the history should match the tags, and we just unlink to things we could not generate 15:13:28 i.e. that version still exists, it just failed to generate artifacts 15:13:43 Makes sense. 15:14:22 I think dhellmann's comment is clear enough 15:14:41 aha, found. linked in the pad now 15:15:00 Just added a quick comment 15:15:19 fungi: Thanks 15:15:39 Missing old releases - how old are these? 15:15:41 ok, all set then 15:15:52 They are just not the current one 15:15:56 OK 15:16:20 #topic AOB 15:16:37 Ocata is now extended maintenance per the timeframe that was set for it. 15:17:01 Maybe that's something tonyb should announce officially? 15:17:09 Or I can. 15:17:17 yes I was thinking it needs to be communicated beyond taht yaml change 15:17:28 Seems like there should be some public declaration other than the releases page quietly updating. 15:17:32 if only to kick ext-maint people into next gear hopefully 15:17:32 Yeah 15:17:54 Ops list and such 15:17:55 not much happened on that front publicly lately 15:18:12 annabelleB: Good point, should probably be -dev and -ops. 15:18:18 While we have two lists, right fungi? :) 15:18:24 * ttx sighs briefly 15:18:34 heh 15:18:44 cross-post-em while you got em 15:18:54 Maybe -announce too? 15:19:03 smcginnis: hmm, probably not 15:19:08 OK 15:19:22 It's not that big of a change, it's just the occasion to promote ext-maint a bit 15:19:38 you take the action ? 15:19:49 How about this - I'll wait and get input from tonyb, then one of us can decide to send out the announcement. 15:19:54 ++ 15:20:09 let's not approve that change in the meantime ? 15:20:45 Will anything change? I think that it is going to e-m is a done deal. Just a matter of who should communicate that. 15:20:45 added a note on it 15:20:58 It's just a convenient way to remember to do it :) 15:20:59 But we could hold until tonyb has a chance to ack it. 15:21:27 OK, meeting next week. 15:21:34 ttx: I assume you'll be traveling at that point. 15:21:44 I'll leave the next morning 15:21:52 Could use the extra time for packing 15:22:01 unless we expect to have something to discuss 15:22:03 And with the start of the new cycle and travel prep, it's a quiet time for activity. 15:22:13 seems like a good week to skip 15:22:16 I doubt we'll have anything, so I'm for skipping. 15:22:25 +1 15:22:30 wfm 15:22:34 * smcginnis hits gavel 15:22:37 No meeting next week. 15:22:40 oh, in other (forward-looking) news, the artifact signing key for stein has been generated and published and attested to by a the rocky signing key as well as a few infra sysadmins (with hopefully more to come). need to decide when we want it swapped into production 15:22:42 * dhellmann wants a gavel 15:22:45 #link https://sks-keyservers.net/pks/lookup?op=vindex&search=0x27023b1ffccd8e3ae9a5ce95d943d5d270273ada&fingerprint=on Stein Cycle Artifact Signing Key 15:23:01 fungi: Oh good. 15:23:08 Don't we usually swap around this point? 15:23:25 since there's no immediate end to cycle-trailing stuff any more, maybe we don't delay it significantly? 15:23:38 previously we waited a couple weeks for teh cycle-trailing releases to get through 15:23:46 Yeah, I think it would be a challenge now waiting for the trailing stuff. 15:24:14 since they have 3 months to do those 15:24:15 We stated we would give two months, so that already puts us well into stein activities. 15:24:17 okay, i'll whip up the patch to the releases repo and do the remainnig steps in our key replacement process 15:24:22 2 or 3? Either way. 15:24:27 6? 15:24:29 fungi: Thanks 15:24:31 Hah! 15:25:08 Oh, last meeting for the rocky tracking etherpad. We should switch to the stein one now. 15:25:16 https://etherpad.openstack.org/p/stein-relmgt-tracking 15:25:17 hhhhrm 15:25:49 The release deadline for projects using the release:cycle-trailing model that follow the main release cycle is set to November 28, 2018. 15:25:50 fungi: I see either etherpad hasn't been upgraded or the undocumented highlighting feature is still there. 15:26:04 so... 3 15:26:06 ttx: OK, then it is three months. 15:26:13 smcginnis: it's been upgraded. when did you see it happen? 15:26:17 Could have sworn it was 2, but either is fine. 15:26:19 except that for Stein I think we got it wrong 15:26:20 i restarted the service late last night 15:26:35 "The release deadline for projects using the release:cycle-trailing model that follow the main release cycle is set to 13 June, 2019." 15:26:40 fungi: I just appended CANCELLED to the meeting line in the stein etherpad and it changed the whole line's highlighting. 15:26:42 that is 2 montsh 15:27:02 huh, okay i wonder if something failed to apply the upgraded version 15:27:03 we might want to double-check which is right and fix 15:27:27 3 months according to https://releases.openstack.org/reference/release_models.html#cycle-trailing 15:27:29 Should it be 2 or 3? I saw another place I missed the Wednesday vs Thursday final deadline, so I need to update the stein schedule anyway. I could fix that if it needs fixin. 15:27:34 ttx: yeah, the docs say 3 15:27:37 OK, I'll update it. 15:28:18 Anything else we need to cover today? 15:28:26 I wonder if those 2 months are hardcoded in the schedule generator or somethnig 15:28:53 I think I probably put it there thinking we had decided 2. 15:28:59 ah! 15:29:07 PEBKAC then 15:29:35 Yep 15:29:41 Where the problem usually lies. 15:30:39 OK, if that's all we can wrap up. 15:30:51 Thanks!! 15:30:57 smcginnis: any chance you had that etherpad already up in your browser and simply reconnected rather than reloading the tab 15:30:59 Thanks again everyone for all your help over the last six months. I think it went really well. 15:31:00 ? 15:31:22 fungi: I don't think so, but maybe there's some caching issue and I need a hard refresh. 15:31:33 i'll keep digging, thanks! 15:31:33 OK, thanks everyone. Have a good weekend. 15:31:43 #endmeeting