15:00:06 #startmeeting tc 15:00:06 Meeting started Thu Dec 16 15:00:06 2021 UTC and is due to finish in 60 minutes. The chair is gmann. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:06 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:06 The meeting name has been set to 'tc' 15:00:14 #topic Roll call 15:00:15 o/ 15:01:01 we have 4 members listed in Absence section 15:01:03 Dan Smith on PTO until 2nd Jan 15:01:03 Radosław Piliszek - med appointment (2021-12-16) 15:01:03 Jay Bryant - illness (2021-12-16) 15:01:03 Belmiro Moreira - PTO 15:01:08 o/ 15:01:13 hoping other 5 are available 15:01:22 o/ 15:02:05 o/ 15:02:37 mnaser: ping, in case you are online 15:04:10 we need one more member for quorum but we do not do votes here and this is not the only meeting in month so we are good do do meeting as per process/bylaw 15:04:12 let's start 15:04:25 #link https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Agenda_Suggestions 15:04:28 today agenda ^^ 15:04:59 #topic Follow up on past action items 15:05:27 ricolin to check with Artem about championing the OSC community-wide goal 15:05:36 ricolin any updates? 15:05:56 didn't got chance to connect with him 15:06:08 will try to ping him later again 15:06:12 no prob 15:06:14 thanks 15:06:35 should we keep it as Action item or track in Yoga tracker itself? 15:07:10 IMO, let's keep it a action item, as didn't finish that action yet 15:07:16 sure 15:07:24 #action ricolin to check with Artem about championing the OSC community-wide goal 15:07:33 #topic Gate health check 15:08:06 I have not seen any frequent failure this week 15:08:24 The pypi stale indexes problem has popped up again 15:09:14 With some interesting debugging tactics... To be clear this is a known issue for at least a year. If pypi's CDN cannot get the results it wants from the primary backend it requests them from the backup backend. The backup backend is frequently out of date. Since openstack updates constraints frequently and wants specific new versions of software it often fails 15:09:19 this one? The conflict is caused by: 15:09:19 12:26 PM 15:09:19 keystone 20.1.0.dev19 depends on PyJWT>=1.6.1 15:09:19 12:26 PM 15:09:19 The user requested (constraint) pyjwt===2.3.0 15:09:32 Most pypi users don't notice these problems as they will just install an older version of the software 15:09:40 I observed that this two or three times and reported in requirement channel 15:09:43 gmann: yes that sort of error 15:10:00 but it is not happening now or is it? 15:10:01 also we' had some recent management event backlogs from bulk branch creation and deletion, but as of yesterday that's been fixed 15:10:17 yeah, i have not seen that since yesterday 15:10:20 gmann: it happens whenever the pypi CDN has a failure requested data to their primary backend 15:10:27 gmann: the pypi issue comes and goes, so it's hard to know whether it's fully subsided 15:10:32 its not a consistent thing as it depends entirely on the solar rays in their datacenters 15:10:52 yeah, let's monitor if it happen again or more frequent 15:11:00 Calling it out here in particular due to some of the crazy debugging steps that have been taken. Its a known issue for a long time. if you'd like to bring it up with pypa I think that would be appropriate 15:11:52 Most pypi users won't notice because they don't lock their dependencies to the latest of all deps 15:12:05 OpenStack is sort of at the forefront there I guess based on how we notice and no one else seems to 15:12:50 yeah, let's see how frequent it will be and based on that we can discuss to raise it 15:13:46 any other issue/observation on gate? 15:14:15 zuul config error has some progress but not all project have fixes up #link https://etherpad.opendev.org/p/zuul-config-error-openstack 15:14:27 We were down to 88 errors yesterday. Progress at least 15:14:42 yeah 15:14:57 ralonsoh brought up earlier today that swift's unit tests seem to have started failing on train in the past few days 15:15:13 fungi: there is a fix for that 15:15:18 it has to do wtih a dns lookup 15:15:23 oh, excellent, i'll let him know 15:15:32 https://review.opendev.org/c/openstack/swift/+/821759 I suspect 15:15:54 nice 15:16:04 https://review.opendev.org/c/openstack/swift/+/821762 is the train change 15:16:32 for unit testing, rpittau adding the py3.10 as non voting in yoga which sounds good as discussed in TC meeting 15:16:34 #link https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/821863 15:16:44 fungi: clarkb ^^ if you can have a look 15:17:11 anything else on gate health? 15:17:52 #topic Skyline as an official project 15:18:00 #link https://review.opendev.org/c/openstack/governance/+/814037 15:18:11 #link http://lists.openstack.org/pipermail/openstack-discuss/2021-December/026206.html 15:18:20 I think there was some really good discussion on the ML on this one 15:18:44 yeah, those were good points on skyline being consistent with OpenStack way 15:19:06 and Tc was waiting for skyline team to respond on gerrit or ML to confirm their plan 15:19:30 now they have responded (8 days back) that they will fix all those open points which seems good 15:19:41 Meanwhile they are waiting for us :) 15:20:00 I do not think so, we have replied to them since starting right? 15:20:05 I think we only need one more +1 Rollcall Vote to move forward? 15:20:13 fungi: has raised those question on ML as well on gerrit too 15:20:45 they have gotten concerned because they didn't realize the tc was waiting for them to answer more questions 15:20:53 gmann, they were. 15:21:07 humm, I think that was clear in my weekly reports too 15:21:12 Things are cleared up now though and nothing else is blocking so we really should get this merged. 15:21:20 but anyways checking gerrit review could be best thing 15:21:28 gmann, but they were not attending these meetings and might not have seen the email. 15:21:38 diablo_rojo: review? 15:21:55 anyways 15:22:04 gmann, agreed, but as they aren't an openstack project yet and not built within openstack, they are still learning the ways we communicate. 15:22:15 yes, i think they're still trying to learn about the openstack community and how to interface with us, and what our expectations are, so they often don't know where to watch for signs that people need their input 15:22:30 Exactly :) 15:22:48 * ricolin just done with review and +1ed 15:22:51 so the "long silence" came across as we had forgotten about them 15:22:53 Something they will learn the more they are brought into our family :) 15:22:54 I think if there is gerrit application review then they know the gerrit :) anyways 15:23:01 one good question is raised in ML whether we should have them to fix those consistent things first and then add it as official project or add them as official project and then help them to fix all. 15:23:37 I don't think those things are blocking. I think they should be added first and then they can fix it before the release. 15:23:40 fungi: no, I keep adding that in meeting open review and in my weekly summary email. it is just some miss-communication 15:24:01 They are more than happy to have direction on what to do once they are in. 15:24:11 They have confirmed that they will fix all and have those points in their roadmap 15:24:12 gmann, that assumes they know to read the summary emails. 15:24:34 Yeah it might be easier especially as any concerns that were tool or package related willl need to get fixed to work 15:24:46 diablo_rojo: I think they did in past right. replying there and PTG sumamry 15:24:47 spotz, thats a good point too. 15:24:52 Easier to be in first 15:24:56 anyways let's not discuss that and check next step 15:25:24 I think it as they agree to be in openstack way they are all good to e an offiical project 15:25:25 i can understand why they might be hesitant to invest a lot of time in the things we pointed out if there's no clear promise they'll be accepted as an official team once that's done. but also we've said in the past that we don't need functional, releaseable software to be able to form a team, and in fact we'd rather increase the amount of development done within openstack rather than 15:25:26 gmann, possibly, but they might not have seen they were staying on the agenda week over week. And since skyline wasn't being called out in the subject line, they might not have caught it. 15:25:27 expecting people to develop outside openstack and request addition afterward 15:25:56 We have the 5 +1 rollcall votes we need now, yes? 15:26:01 and we all have to help them to fix all those things 15:26:01 So it can be merged? 15:26:44 yeah, It can be. I will open it for today and if no objection then we can merge it tomorrow. 15:27:07 Yes please! 15:27:16 and as next step, we have to help them on setup the communication thigns like IRC etc. which i can do. did for venus project also 15:27:30 and monitor them to fix those open things in coming time 15:28:28 anything else on this topic? 15:28:38 reading the charter, it technically doesn't even need a strict majority, it just needs more positive than negative votes and at least one third of the tc participating, as well as 7 days since the last revision was posted, all of which seem to be met 15:29:33 one things is we as TC need to be more active in such cases where we might need to reachout to them in personal email or so. so that we fix such communication gap issue where they waited for TC and we waited for them 15:30:13 this is something we will take care in future 15:30:35 I can work with getting them to setup on IRC. 15:31:04 let's move 15:31:07 #topic #topic SIG i18n status check 15:31:11 +1 15:31:49 this SIG seems having less/no active maintainer 15:31:51 Xena translation missing 15:32:00 We also need to migrate tooling 15:32:03 #link http://lists.openstack.org/pipermail/openstack-discuss/2021-December/026244.html 15:32:20 Translation bug 15:32:23 #link https://review.opendev.org/c/openstack/contributor-guide/+/821371 15:32:52 ianychoi[m]: is chair of SIG 15:33:15 but we have not heard from him on ML. not sure if he is online 15:33:56 how can engage more people in this SIG? assuming transition is still needed 15:34:26 Maybe that is a good place to start? Who is relying on translations and are you able to help ensure the strings continue to be translated? 15:34:53 yes, to expand on what clarkb said, we're long overdue to move translation work off zanata (to something like weblate), red hat abandoned the software years ago and we're already working around security vulnerabilities in it which nobody is ever going to fix 15:35:07 yeah and we need to reachout to local user group for that so that they know the anwsers and people need it 15:35:33 Do we know what languages as well and if all are still needed? 15:35:40 and even Xena transition is not done that make it clear we need maitainer 15:36:03 spotz: all for Xena #link http://lists.openstack.org/pipermail/openstack-discuss/2021-December/026244.html 15:36:33 and ID transition has bug which is proposed for removal to unblock the contributors-gudie gate 15:36:41 #link https://review.opendev.org/c/openstack/contributor-guide/+/821371 15:36:46 I know way back when IBM was a big driver for translations but they aren't super involved anymore. But there may be other users/companies/groups that rely on them and they would be good candidates to help drive this along if they exist 15:37:06 yeah 15:37:32 let me reachout to ianychoi[m] first and what is his position to this SIG and they we can callout the help on ML 15:37:57 #action gmann to reach out to ianychoi[m] for SIG i18n status and then call for help on ML 15:38:12 anything else on this topic? 15:38:39 #topic Adjutant need PTLs and maintainers 15:39:09 #link http://lists.openstack.org/pipermail/openstack-discuss/2021-November/025786.html 15:40:00 still waiting from Braden response. I reached out in personal email also but still waiting for permission 15:40:06 so let's keep it open 15:40:18 #topic Open Reviews 15:40:40 we have 4 open reviews 15:40:43 #link https://review.opendev.org/q/projects:openstack/governance+is:open 15:40:49 skyline is all good 15:41:13 I will review the project health tool to confirm my comemnts 15:41:38 Retire js-openstack-lib need to wait for Adjutant PTL settled 15:42:24 ricolin: ack, can you please add that in comment and WIP until then? 15:42:37 sure 15:42:43 thanks 15:42:48 and this one need more TC feedback as we discussed in last TC meeting #link https://review.opendev.org/c/openstack/governance/+/816587 15:43:01 member present here, please review 15:43:25 one last things on meeting 15:43:43 who all will be present for next week meeting on 23rd Dec ? 15:43:51 not me:) 15:44:53 Dan and Belmiro are on PTO. so 6 TC available ? 15:45:02 ricolin: diablo_rojo ? 15:45:04 I'm available for sure 15:45:46 I am also available 15:47:18 anyways, let me check with other TC members also later and then we can decide to have meeting or cancel 15:47:33 that is all from my side, anything else to discuss ? 15:48:23 if nothing, let's close the meeting. thanks everyone for joining. 15:48:26 #endmeeting