18:00:16 #startmeeting tc 18:00:16 Meeting started Tue Sep 12 18:00:16 2023 UTC and is due to finish in 60 minutes. The chair is knikolla. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:16 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:16 The meeting name has been set to 'tc' 18:00:22 #topic Roll Call 18:00:27 o/ 18:00:28 Hi all, welcome to the weekly meeting of the OpenStack Technical Committee 18:00:29 o/ 18:00:32 A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct 18:00:37 Today's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee 18:00:38 o/ 18:00:41 o/ 18:00:56 We have 2 noted absences (noonedeadpunk and jamespage) and one late (gmann) 18:01:04 o/ 18:01:07 hi diablo_rojo :) 18:01:32 o/ 18:02:34 i'll give it 2 more mins and we can get started at 4 minutes past the hour. 18:04:20 #topic Follow up on past action items 18:04:24 o/ 18:04:43 We have one action item noted down 18:04:45 gmann to ensure that a call for extra-ACs is included in future election/governance operational documentation to ensure it's difficult to miss in the future. 18:04:50 He'll be late to the meeting but he was awesome and provided an update before it. 18:04:57 gmann proposed this and we received some push back about the narrow focus of the change and the need for broader guidelines related to preparing people to vote, rather than just adding extra AC. 18:05:13 For example the number of extra ATCs (now ACs) is much smaller than the number of non-foundation members (community members) or gerrit submitters who did not join in openinfra.dev. 18:05:20 So more general guidelines would go a longer way towards helping the electorate. 18:05:28 We can talk more about this item during the election topic, which is part of today's agenda. 18:06:45 Ah, we have one more related action item 18:06:48 JayF to contact i18n SIG about extra-ACs 18:07:07 I emailed Ian Choi and got no response; but I did see i18n sig has some docs about setting up extra ACs. 18:07:40 I believe this can be grouped in with the other extra AC work; there is clearly need for us to be more active with ensuring this happens in a timely manner. 18:08:14 ++, thanks JayF 18:08:25 you can also try ianychoi[m] in #openstack-i18n 18:08:27 we can circle back to the election item during it's agenda topic. 18:08:27 #topic Gate health check 18:08:38 Any updates on the state of the gate? 18:09:12 frickler: I doubt there's a real action to take, especially in light of the commentary on/around election changes. 18:09:21 tbh I haven't been sending push through the gate in the past week so I'm not sure, 18:09:38 I've seen oom failures in cinder tempest jobs today 18:09:39 but I have seen plenty of things landing, so I take that as a good sign that it hasn't significantly regressed 18:09:55 seems to be blocking them quite a bit 18:10:18 I know they're looking to back down the concurrency on their jobs, probably as a result of that 18:10:40 are they jobs that have more stuff running than normal? 18:10:45 more services I mean 18:11:14 possibly c-bak in addition to the usual 18:11:15 just the usual integrated projects afair 18:11:31 okay 18:12:09 rosmaita: yeah, why does c-bak take more memory? 18:12:24 i don't think it does 18:12:26 https://review.opendev.org/c/openstack/cinder/+/887081 is an example with a lot of retries already and no success 18:12:36 rosmaita: we have it disabled specifically because it does, IIRC 18:12:41 c-bak historically was one of the worst memory hogs 18:12:46 yeah that ^ 18:13:07 i suspect that if you look at performance.json you will find that not to be the case 18:13:16 for quite a while now 18:13:20 well, the oom was for some python3 process, but I didn't check which one 18:13:34 frickler: oh really? normally all OOMs get charged against mysql, 18:13:48 so if it was a python process that's highly unusual, IMHO, so I'd want to know which one for sure 18:15:41 I'm not sure which job I looked at, but you can check the above patch yourself 18:15:42 knikolla: nothing else from me on gate though, maybe if frickler can dig up the one in question we can chat in -qa later 18:15:54 thanks dansmith 18:16:04 #topic OpenStack TC Charter Updates for OIF Foundation Simplification 18:16:10 As discussed in the previous TC+Board syncup, the OpenInfra Foundation bylaws contain a lot of OpenStack specific language. 18:16:35 Me, gmann, and rosmaita have been meeting with Allison from the board to study the necessary changes to move that language to the charter and free up the bylaws for OpenStack references. 18:16:46 Bylaws changes and what all to go in the TC charter have been shared in an email which you all should have received. TC needs to provide the feedback before Sept 19. 18:17:32 If you have any other questions or concerns, please feel free to bring them up now 18:17:58 I wanna thank the folks on the email thread who commented on them and all, I had no feedback but I appreciated the discussion there 18:18:21 is there a reason that this is not a public discussion? 18:18:53 That's a good question. I'll bring it up in the thread and will share the PDF in openstack-discuss if I get a go ahead. 18:19:18 I don't think it's not not a public discussion 18:19:43 there's too many negations there and my head now hurts, haha 18:19:51 hehe 18:20:34 in part, the initial pass is a back-and-forth with the foundation's legal counsel, and we don't want to have them spending billable hours answering community questions while the proposal is being solidified 18:20:36 I'm pretty sure fungi has emailed the sync info to the dev list each time and the bylaw changes are public as well 18:21:25 I'll go ahead and share the TC charter changes without the bylaws bit. 18:22:17 if parts of the proposal that require an individual member vote go ahead, then obviously there's public discussion with the individual membership that has to go on anyway 18:22:32 That part should be safe to share and will be more relevant for the OpenStack community. 18:22:46 ++, the sept 19 deadline is before it goes to legal, IIRC. 18:23:04 But it's not like these changes are going to take effect without a long public phase, both for the tc charter and oif bylaws. 18:23:15 my concern is more about TC actions being as open as possible, but I guess that's ok-ish this way 18:23:56 I understand and share your concern, so thank you for raising it. 18:24:30 Anything else on the topic? 18:24:54 for parts that require tc approval, there will also be a public vote, i'm certain 18:25:17 (because the tc has no other way to record consensus) 18:26:19 Yes, everything needs to go through Gerrit, and I tend to keep changes open for longer just in case. 18:26:45 (It's totally not because I don't have executive skills ;) 18:27:36 #topic OpenStack Elections 18:27:56 We're one week away from the end of the voting period. 18:28:32 We already discussed a bit about this topic during the action items phase of the meeting. 18:28:53 is this the first cycle we've accepted votes for two weeks instead of one? 18:29:12 No, IIRC 18:29:12 I was going to ask 18:29:15 I'm fairly certain we did it last cycle as well 18:29:32 thanks, i couldn't remember if it was last cycle or this one 18:30:00 i do recall it being intentional because people complained they missed the voting period due to being on vacation 18:30:08 well this cycle there was no TC vote at least, iirc none at all even 18:30:23 last cycle? 18:30:33 oh last election, right 18:30:39 last election cycle, yes 18:30:45 Maybe I should say I'm fairly certain *we changed the rules* before last election 18:30:52 I do not recall if we actually had an election 18:30:55 this cycle we have 6 candidates for tc 18:31:07 makes sense, this may be the first time we've exercised the two-week voting period then 18:31:13 thank you frickler for volunteering :) 18:31:47 yes 18:32:10 I believe that is correct 18:32:31 many thanks to everyone, new and returning, who cast their candidacies for tc 18:32:32 #link https://review.opendev.org/c/openstack/election/+/893810 18:32:43 There was some good feedback on the patch above 18:33:05 especially on the necessity to provide guidelines on people for how to prepare to vote 18:33:58 I'm not yet sure about the best way to go about that, as people may just ignore templated emails 18:34:32 I was thinking some kind of actionable PTL end of cycle checklist 18:35:25 reaching ptls to remind them of periodic tasks (release, ptg, etc) has been a perpetual challenge 18:35:57 Yes, take the current release process. Three ptls didn't approve their client library patches. 18:36:21 We need to reach a lot of incumbent PTLs after the election cycle to even remind them to run again. 18:36:34 It seems an area where being more proactive will go a long way. 18:36:54 also, extra-ac refreshes aren't clearly covered by any specific liaison position in the dpl model 18:37:04 that might be a gap worth closing 18:37:22 Ohhh that is a good point too. 18:37:29 I'll note that overall our election participation is extremely low which may also drive down importance of following up on extra-acs. 18:37:53 Given that currently the only real tangible benefit of being an extra-ac is the ability to vote (or be PTL/TC) 18:37:56 *and* extra-acs may not be associated with any project team (the i18n sig for example is not a project team) 18:38:08 or increase it since every single vote may be more valuable? 18:38:11 The extra-acs being collected we found were forgotten this cycle as well 18:38:39 Currently that is the only value in being an extra-ac. Historically there have been more benefits. 18:38:55 frickler: My point is more that if 90% of our contributors don't care to vote that would imply there may be a % of PTLs who don't see value in it either (and de-prioritizes it in favor of the overgrown todo lists we all have) 18:38:59 Oh and I guess being listed in the release marketing materials 18:39:26 Both are a signal of the lack of importance that elections have had in the past few cycles 18:39:28 I'm just saying, there are multiple approaches; proactive reminders are good, but if we can provide more incentive it might help as well 18:39:34 are attributed * 18:39:37 technically we list contributors in the release info even if they aren't able to vote (but we include extra-acs in the list, yes) 18:40:02 Voting is important when there are choices, which we scarcely provide. 18:40:15 fungi: good clarification :) 18:40:49 knikolla: exactly. Its been quite some time since we've actually had a poll for more than one team/project I think. 18:41:24 this is sort of related to the thing spotz[m] wants to talk about later too, i think 18:41:36 yep yep 18:41:37 So the floor is yours spotz 18:43:35 So earlier today the D&I WG met and we brought up a topic we had discussed back in the Spring. Basically a leadership shadowing program where folks in the OpenStack projects and the larger OpnInfra projects could see what the leadership does, get some experience and visibility and hopefully run in the future 18:44:09 We are starting to put possibility ideas in an etherpad 18:44:11 o/ 18:44:12 #link https://etherpad.opendev.org/p/diversity-leadership 18:44:58 One I've seen used by some projects is a non-lader running the bug reports, sending out emails, and possibly running a bug meeting 18:45:20 One thing we do in RDO is take volunteers to run the next meeting so everyone can get experience doing that 18:45:49 I do want to sign up the TC for the pilot and have a few ideas :) 18:46:11 This is similar to what we've historically done in Ironic of finding someone who has not yet been PTL; teaching them what needs to be done then doing a handoff at the election period ... but I know, at least for Ironic, we don't have a lot of interested potential leaders breaking down the doors 18:46:22 so my main question with this is where to the mentees/shadowers come from? 18:46:22 Obviously pending TC consensus and assuming I won't get voted out of chair this cycle. 18:46:44 Please add them to the etherpad. I think if we have some good examples of things people can do we'll have more buy-in from leaders and potential leaders 18:47:36 JayF: i think that's one of the questions we hope to find answers for 18:47:45 Shadowers will come from current community members who show an interest and ask to take part. This isn't welcome to the community type mentorship so they should already be involved 18:48:17 Hence I think the TC would be a great fit. 18:48:19 though it is also something that folks like the first-contact sig can set the tone for 18:48:27 And that means for StarlingX/Kata/etc they already contribute there 18:48:40 spotz: have you looked at the k8s shadow program? They might have some advice/best practices to help advise too 18:49:19 we did discuss some of that during one of the openstack/kubernetes cross-community leadership get-togethers right? 18:49:31 yeah 18:49:47 Nope I haven't we've just been talking about this between ourselves, we delayed it thinking we'd have an in-person PTG which would have been a great place to kick it off cause they could literally shadow someone 18:50:06 fungi: a little, but I think there are docs somewhere we can look at 18:50:29 spotz: we could also invite the k8s release folks or their tsc to come talk about the programs 18:50:35 (to the ptg) 18:50:45 I think we've discussed regular mentoring before. I know we switch to co-horts because that's what K8s was doing and that really didn't work for us and kinda finished off out mentoring program 18:51:35 Yeah we can definitely schedule some time specifically for this at the PTG as we get better attendance there then our meetings 18:51:54 this is good topic for TC+leaders interaction sessions too 18:51:55 I think all projects are open to volunteers doing things, our problem has been getting people to volunteer ... this sounds like another effort for us to build something in the hope that people will show up 18:51:55 i think it would be better if the D&I WG did something like direct reachout to people telling them to show up at project meetings or in irc channels and volunteer, and then projects will handle the mentoring 18:52:06 Yeah they have been really successful with their release team shadow program 18:52:09 in particular 18:52:13 I think k8s was also recently revamping their shadowing program, IIRC. 18:52:14 rosmaita++ 18:52:25 rosmaita: agree 18:52:40 yep 18:53:14 rosmaita: yes, i heard clarkb mention something similar. basically the suggestion is that we might get more traction if there's someone seeking out and identifying potential future community leaders and encouraging them 18:53:23 rosmaita: that's a great point. I think this is more targeted to people who have an interest but are struggling to find a footing due to uncertainty regarding time commitment or lack of skills. 18:53:40 So providing a softer landing to try leadership roles 18:53:48 Without the full commitment and risk associated with it. 18:54:20 frickler: knikolla: on the bylaws change discussion, it is public one only. we call out the interested members in TC+board syncup Aug call who can join the discussion of bylaw changes. after we got the interested members then we started the discussion. 18:54:48 once that group make changes in better shape it will be voted public only via board, individual member, tc charter in gerrit etc 18:55:05 Or they're just not sure what the role entails. So it's a bit of everything folks are mentioning. fungi and I can't be everywhere in OpenStack let alone all the projects so if you want us to reach out to people to encourage them someone has to give us their names 18:55:34 if anyone still interested in the discussion of getting changes in mroe readable/shape form, let me know and i will get you included in the next call 18:56:29 gmann if you can include me I'll join, the normal syncs have coincided with the CentOS Board meetings which is why I haven't been able to attend 18:56:46 spotz[m]: not sure what you mean about names 18:56:50 and this is etherpad where discussion is captured publicly #link https://etherpad.opendev.org/p/openstack-tc-charter%2Bbylaws-scratchpad 18:57:03 spotz[m]: noted, will do 18:57:04 #topic Open Discussion 18:57:30 I'll be unavailable tomorrow as I'm flying back to the US 18:57:45 and boy will your arms be tired 18:57:59 rosmaita: thanks DAD 18:58:02 * fungi apologizes in advance for the tsa experience 18:58:23 haha 18:58:24 I would like to land https://review.opendev.org/c/openstack/project-config/+/894814/ to remove a current job dependency for monasca from zuul 18:58:52 there are two motiviations for this. The first is elastic/kibana has a significant number of branches in it which causes zuul to do many queries to sort of branch details which leads to us hitting api rate limits 18:59:07 second is kibana is not open source anymore and I'd rather we not integrate with that in our CI system 18:59:29 but it will break monasca's integration job and sounds like broken CI has impacted ability to get contributors/ptl candidates 18:59:32 fungi: i'll be flying through istanbul, and they do some weird gate security check for all flights towards the USA. 18:59:33 @rosmaita someone mentioned the D&I WG should reach out to potential shadows and we can definitely email them to encourage them but someone would have to tell us WHO to contact 18:59:37 frickler: let me know if you wanted to be part of that discussion and I can include, it is schedule next at Tue Sep 19, 2023 4pm - 5pm (EDT) 18:59:50 I don't think this will make things significantly worse so if I don't ehar otherwise I'll try to land that soonish but please give me your feedback if you think that is an issue 19:00:01 also infra related (meant to bring it up during state-of-the-gate), we dropped fedora node labels in zuul/nodepool yesterday. no impact was anticipated nor observed 19:00:01 We're almost out of time. Thanks all! 19:00:02 clarkb: as you might know, there are lack in PTL and maintainers there so this break is good indication 19:00:08 #endmeeting