18:00:19 #startmeeting tc 18:00:19 Meeting started Tue Mar 19 18:00:19 2024 UTC and is due to finish in 60 minutes. The chair is JayF. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:19 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:19 The meeting name has been set to 'tc' 18:00:25 Welcome to the weekly meeting of the OpenStack Technical Committee. A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct. 18:00:26 Today's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee. 18:00:29 #topic Roll Call 18:00:31 o/ 18:00:31 o/ 18:00:43 \o 18:00:45 o/ 18:00:48 o/ 18:00:51 o/ 18:01:01 o/ 18:01:27 o/ 18:01:39 I'm going to take this time to note that the election is over tomorrow; and this is the last meeting of the TC as currently constructed. Thanks to everyone who has been a part of TC over the last six months o/ 18:02:19 #info There are no tracked action items to follow up, agenda item skipped. 18:02:23 #topic Gate Health Check 18:02:27 How is the gate? 18:03:02 not great in nova land, there were a bunch of rechecks on trivial patches last week trying to get them in 18:03:13 a number of guest kernel crashes, which we have been seeing less of lately 18:03:24 I know for Ironic it's been a little rough; we had to disable a set of jobs because something broke in CI automation around building cirros partition images; we're going to be discussing at PTG how to get that build out of the CI flow 18:03:39 note there was an increase in zuul config errors from friday's centos-7 nodeset/label removal, but it mainly impacts very old branches (and a lot went away in subsequent unmaintained branch cleanup) 18:04:16 Thanks for the update Dan and fungi. Going to give another bit of time for additional info before moving on. 18:04:40 speaking about rechecks, I sent today email https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/7L6DSFN6GB65FBVHQATTLSQ7HWGBZJHM/ 18:05:01 I was looking at the reasons of rechecks in last 30 days and most of them are "unrelated failure" :/ 18:05:30 if only we could eliminate all those unrelated failures! 18:05:34 Thanks slaweq -- I'm glad you aggregated that. I'm a little concerned that 'unrelated failure' rechecks may indicate that we have contributors who are uncomfortable with how to troubleshoot CI. 18:05:38 so I asked there if people can be more verbose and give some more specific reasons, hopefully we may have some better data in few weeks 18:05:40 how unhelpful (the recheck comments i mean, your research is very helpful!) 18:05:48 and/or are so busy that they don't have time to further dig and are only focusing on the task in front of them 18:06:07 or get tired of typing long comments again and again 18:06:10 yeah the "recheck unrelated" has become the new bare recheck unfortunately 18:06:20 because nobody fixes the unrelated failures 18:06:38 the point is to get people looking at the why and not just assuming "must not be my problem" 18:06:53 there was one report recently of someone in nova re-re-re-rechecking things that were indeed failing because of stuff they broke 18:06:55 dansmith +100 to what You said :) 18:07:33 :) 18:07:33 ouch 18:08:02 dansmith: that's yet another sign of contributors who are focused on the change in front of them and/or are missing context on how to troubleshoot CI. I don't really know how to fix the first half of that, and I only know how to help with the latter via 1:1 mentoring 18:08:03 yeah, unrelated recheck is not helping at all 18:08:33 JayF: to be clear, their patch was broken, they weren't even seeing it was failing a test that covered the code they were doing 18:08:49 I don't think it was a failure to understand some cryptic CI failure 18:09:08 but if you mean that we've trained people to assume that things just fail a lot so they don't even look, that's definitely true 18:09:11 dansmith: ack; I will note that with extreme junior level contributors I've had to have that kinda conversation even in obvious context 18:09:23 is it because we have many job running in gate and a few of them failing and other passing make them of thinking that 'my code is ok as some job pass' ? 18:09:36 and why I say hoping they'll at least look instead of assume is the reason that asking people to cough up a reason is at least mildly helpful 18:10:33 obviously if we have people that aren't even going to click on a failed job and look at the testr results, then there's not much we can do there, other than 1-on-1 coaching, which is what happened in this case 18:10:38 I almost said "this is a good topic for PTG", but I'm not sure what could be said/discussed that hasn't been said/discussed before :/ 18:10:52 but I have seen more than a few seasoned should-know-better people doing it :) 18:11:00 and I poke them when I do 18:11:33 Yeah. It's a tough problem because even with some of the issues (the kernel panics on nova side; dnsmasq crashes on Ironic side), we've identified, escalated to people who should know and be able to help 18:11:36 and we end up stuck 18:11:58 yup, but at least checking and commenting helps slaweq's stats show where the hot spots are 18:12:09 ++ 18:12:14 I'm going to move on 18:12:17 if everything is "recheck unrelated" then even chatgpt (groan) can't summarize without hallucination :) 18:12:30 :) 18:12:37 "Write me a recheck comment describing the failure in these logs" ;) 18:13:02 #topic Implementation of Unmaintained Branch Statuses 18:13:15 JayF it actually may have more sense than just "recheck unrelated" ;) 18:13:27 hehe 18:13:32 I assume this is mostly on pause/moving slowly due to release activities for 2024.2? Is there an update beyond the issues noted in the etherpad from last week 18:13:49 not much happened due to rc1 stuff 18:14:00 maybe we can update here about 'unrelated recheck' or something like this is another bare recheck type https://docs.openstack.org/project-team-guide/testing.html#how-to-handle-test-failures 18:14:04 That's what I suspected, and thanks for all the work getting Caracal out o/ 18:14:28 there was a fair amount of branch cleanup at least? 18:14:31 gmann: I'd +1 such a change, depending on wording 18:14:45 fungi: nice, thank you 18:15:14 gmann I will propose some additional info to that document 18:15:19 i didn't do any of the cleanup, that was down to frickler and elodilles i think, just reporting it ;) 18:15:19 thx for pointing that out 18:15:58 slaweq: thanks 18:16:15 #topic TC vPTG 2024.2 18:16:30 #link https://etherpad.opendev.org/p/apr2024-ptg-os-tc 18:17:00 I suggest to please add any additional topic over the next week. 18:17:38 I have a few things I'm writing up as starters for PTG topics which I'll add over the next couple of days. 18:18:47 Going to give another minute in case there is input/comment on PTG before moving to open discussion. 18:19:08 Merged openstack/governance master: Fixing typo in 2024.2 testing runtime https://review.opendev.org/c/openstack/governance/+/913704 18:19:18 #topic Open Discussion 18:19:23 Anything for Open Dicsussion? 18:19:44 I'll reiterate that I will be landing the wsgi goal https://review.opendev.org/c/openstack/governance/+/902807 after the tc meeting today 18:19:51 reminder that TC voting closes in a bit over 24 hours 18:19:54 ++ 18:19:55 Can we please get the repos under the TC to count for AC status? 18:20:17 and we have other governance changes up for review: https://review.opendev.org/q/repo:%22openstack/governance%22+status:open 18:20:19 should be a relatively minor patch to the election tools to get them indexed 18:21:06 spotz[m]: I'd be OK with that, let me know if you get a patch up and I'll review it. 18:21:08 yes, many TC repo are not just governance part but a good contribution area for community for example p-t-g, doc repo etc 18:21:37 and we should include contributors of those repo in election AC status 18:21:45 Also, along the lines of rosmaita's comment: please evaluate if you want to be TC chair in next cycle. 18:21:45 <- !AC apparently due to repos not counting 18:21:56 yes, i think the fact that docs repos dropped from being sig-owned created a fairly large blindspot for contributions 18:22:23 we count SIG repo right? 18:22:45 yes 18:22:50 cool 18:23:02 but there's no longer a docs/tech writing sig 18:24:00 yeah, that was not good to exclude them bcz repo added under TC 18:24:50 does that need a resolution or is it really just a tooling issue? 18:25:28 the tc electorate is defined in the charter, it needs someone to double-check that the charter doesn't say they shouldn't count, at least 18:25:33 I honestly thought we had fixed it a few elections ago but when I asked where my ballot was this election found out it wasn't 18:25:51 I am thinking about the same 18:26:00 but unless adding them requires an adjustment to the charter for some reason, just do it 18:26:02 I know back when we had doc repos it counted 18:26:31 So Jayf can you retro make me an AC:) I can't vote but for the release anyways 18:26:31 the docs repos counted initially because they belonged to a project team, and then because they belonged to a sig and we added sigs repos 18:26:45 but now there is no sig, they belong to the tc directly 18:27:15 charter says 'who committed a change to a repository under any of the official OpenStack Project Teams (as defined in Projects) ' 18:27:15 https://governance.openstack.org/tc/reference/charter.html#voters-for-tc-seats-ac 18:27:29 spotz[m]: it is past the deadline, I don't think that'd be appropriate unless someone wants to rules-lawyer me to being convinced it's legal 18:27:36 spotz[m]: unsure if you were serious :) 18:27:42 So we would need to update the charter 18:27:49 I think we can explicitly say in charter that 'project repo', SIG repo, TC repo.. 18:28:19 Does someone want to take an action to revise the charter to correct this oversight? 18:28:33 I will 18:28:46 i can help if needed 18:29:08 let's do this once election are close just to avoid any confusion of changing it in between 18:29:20 also it will be decided by the new tc anyway 18:29:25 yeah 18:29:47 that way it will easy to count the vote etc also 18:30:26 #action spotz[m] Will propose an update to TC charter, to ensure documentation contributors are properly recognized automatically as ACs. 18:30:34 Thanks. Is there anything else we wanna hit on? 18:32:18 #endmeeting