Tuesday, 2021-09-28

*** ianw is now known as ianw_pto07:21
clarkbAnyone else here for the meeting? I expect it may be a thin group today. If that is the case I'll hit highlights and get us through the agenda quickly19:00
clarkb#startmeeting infra19:01
opendevmeetMeeting started Tue Sep 28 19:01:08 2021 UTC and is due to finish in 60 minutes.  The chair is clarkb. Information about MeetBot at http://wiki.debian.org/MeetBot.19:01
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.19:01
opendevmeetThe meeting name has been set to 'infra'19:01
clarkb#link http://lists.opendev.org/pipermail/service-discuss/2021-September/000286.html Our Agenda19:01
clarkb#topic Announcements19:01
clarkbThe openstack xena release is starting to solidify and should be out the door next week. For this reason we should attempt to be slushy19:02
clarkbI know we've got one zuul bug we want to restart for but otherwise we should probably err on the side of caution at this point19:02
clarkbAlso I'm going to be half here Friday and Monday as I attempt to have a mini weekend get away19:02
clarkbI should be around early friday and late monday19:02
clarkb#topic Actions from last meeting19:03
clarkb#link http://eavesdrop.openstack.org/meetings/infra/2021/infra.2021-09-21-19.01.txt minutes from last meeting19:03
clarkbThe one action was for all y'all to weigh in on the prometheus spec and a couple of you have. Looks like we're leaning towards node exporter but would prefer we avoid docker and use one of their shipped binaries19:03
clarkbThinking on it I like that approach a bit more if we host copies of the binary ourselves and verify shas and all that19:04
clarkbI'll try to update the spec to propose that method in the near future19:04
clarkb#topic Specs19:04
clarkbI think I've covered the prometheus spec already. Do still leave your thoughts if you have them19:04
clarkb#link https://review.opendev.org/c/opendev/infra-specs/+/804122 Prometheus Cacti replacement19:05
clarkbThe other spec we should be looking at is fungi's mailman 3 spec19:05
clarkb#link https://review.opendev.org/810990 Mailman 3 spec19:05
clarkbThis is hot off the presses and I haven't had a look yet but I am excited.19:05
clarkbPlease take a look if you can and we can start refining this and preparing to make the mailing list server and services more sustainable going forward19:05
clarkb#topic Topics19:06
clarkb#topic Improving OpenDev's CD throughput19:07
fungiyeah, i'm sure i've missed things, and i'm not the most up to speed on what we should be looking for in their current container designs19:07
clarkbianw mentioned he wouldn't make it today but says the stack of changes here are basically ready for review and left some notes. I'll post the links and the notes he gave us momentarily19:07
clarkb#link https://review.opendev.org/c/opendev/system-config/+/80767219:07
clarkbTHis change should list dependencies for all jobs. And zuul doesn't trigger jobs for this change and he is wondering how to make this mergeable (maybe a noop change to a readme or something will do it)19:08
clarkb#link https://review.opendev.org/c/opendev/base-jobs/+/80780719:08
clarkbCurrently every executor adds keys for bridge, then logs in and clones system-config before running playbooks. This change makes split jobs to do this. however, production remains the same as both are called.19:08
clarkb#link https://review.opendev.org/c/opendev/system-config/+/80780819:08
clarkbThis is a follow-on that adds a base job to clone system-config, and stops the other production jobs re-cloning.19:08
clarkbThis job must run first, but then all other jobs can run in parallel, as they are all in the same buildset and using the same "view" of system-config for that particular run19:08
clarkbFrom the notes I expect this will allow us to go step by step and carefully ensure things are as expected. I'll be attempting to review them soon and offer feedback. Much appreciated if others can do the same19:09
clarkbAnd thank you ianw for digging into this19:09
clarkb#topic Gerrit account cleanups19:09
clarkbThis has become my low priority thing with everything else going on the last week or two with zuul and gerrit and openstack19:09
clarkbI'd like to get back to this but haven't had time.19:10
clarkbMostly just an apology here to say that :)19:10
clarkb#topic OpenDev logo hosting19:10
clarkbAs of last night relative to my timezone I think this effort is basically done19:10
clarkbgitea, paste, and gerrit are all using the opendev logo via self hosted locations19:10
clarkbThis means that we can now upgrade gitea to 1.15.x safely without worrying we'll break random theming in other places19:10
clarkb#topic Gitea 1.15.3 upgrade19:11
clarkb#link https://review.opendev.org/c/opendev/system-config/+/803231/19:11
clarkbI think this change is as ready as it will be. I have a held instance for verification at https://198.72.124.104:3081/opendev/system-config19:11
clarkbThat said due to my previous comments about being slushy maybe we wait until next week after the openstack release is done?19:11
clarkbOur testing is pretty good, but with all the random stuff that has happend with zuul, apache, devstack zuul configs and so on I'd like to not be worrying about more potentail problems for the release:)19:12
clarkbAnyway please check that held instance and review the change19:12
clarkb#topic Scheduling Gerrit project renames19:12
clarkbPenciled in October 11-15. We should pick a specific day next week.19:13
clarkbI don't expect the gitea upgrade to impact this as our testing tests renames in gitea now.19:13
clarkbOne thing we should look at is running the full project metadata update against renamed projects as a subsequent step to the renames19:13
clarkbI think we can update the test jobs to do this as well19:13
clarkb#topic Open Discussion19:14
clarkbIf there is anything else I'll give it a few minuets to capture those items. Otherwise as expected not much involvement today (and that is fine I'm recording this for people who watch logs)19:14
fungiyeah i didn't have anything beyond what you've covered, thanks!19:15
clarkbSounds like that was it. Thank you for listening (or reading the logs)19:19
clarkb#endmeeting19:19
opendevmeetMeeting ended Tue Sep 28 19:19:19 2021 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)19:19
opendevmeetMinutes:        https://meetings.opendev.org/meetings/infra/2021/infra.2021-09-28-19.01.html19:19
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/infra/2021/infra.2021-09-28-19.01.txt19:19
opendevmeetLog:            https://meetings.opendev.org/meetings/infra/2021/infra.2021-09-28-19.01.log.html19:19

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!