14:00:25 #startmeeting qa 14:00:26 Meeting started Tue May 25 14:00:25 2021 UTC and is due to finish in 60 minutes. The chair is kopecmartin. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:27 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:29 The meeting name has been set to 'qa' 14:00:43 #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting 14:00:47 ^^^^^^ today's agenda 14:00:54 hi all , who is here today? 14:01:07 o/ 14:01:08 o/ 14:01:09 Hello, I'm here 14:01:11 (this week the meeting bot works as expected :) ) 14:01:18 \o/ 14:01:27 o/ 14:01:46 kopecmartin: because we've been nice this week 14:01:49 so bot likes us 14:02:13 and we haven't been before? :D 14:02:22 hi o/ 14:02:36 #topic Xena Priority Items progress 14:02:41 #link https://etherpad.opendev.org/p/qa-xena-priority 14:02:48 any updates? 14:02:56 none from my side unfortunately 14:03:06 kopecmartin: perhaps :D 14:03:12 none from me here 14:03:20 (that was re being nice) 14:03:31 nothing from my side. busy in other things 14:04:17 #topic Announcement and Action Item 14:04:39 i don't have anything special 14:05:13 i see one action item though 14:05:19 "kopecmartin to check subunit2sql status" 14:05:32 i haven't done that, we can discuss later in the meeting 14:05:48 #topic OpenStack Events Updates and Planning 14:05:55 nothing here, we can skip 14:06:09 #topic Gate Status Checks 14:06:15 anything to share regarding gates? 14:06:41 hello 14:07:09 gates rather happy, 14:07:15 or issues hiding from yoctozepto 14:07:25 soniya29: hi 14:07:41 that's good news (not that the bugs are hiding though) :D 14:07:42 yeah, nothing here from me too. it was green 14:07:56 #topic Periodic jobs Status Checks 14:08:02 #link https://zuul.openstack.org/builds?job_name=tempest-full-victoria-py3&job_name=tempest-full-ussuri-py3&job_name=tempest-full-train-py3&pipeline=periodic-stable 14:08:06 #link https://zuul.openstack.org/builds?job_name=tempest-all&job_name=tempest-full-oslo-master&pipeline=periodic 14:08:43 i see one recent failure 14:08:44 #link https://zuul.openstack.org/build/c21ddcecda2d4fe18d7d431e4a222c9f 14:09:06 not sure atm if it's gonna cost us more problems or it's just a random or already fixed thing 14:09:32 seems like req conflict 14:09:32 I've seen these issues randomly popping around 14:09:46 yeah, it's a conflict but it looks like it should be passing 14:09:55 I think the error message is wrong 14:10:05 and the issue is with pypi index 14:10:06 yeah 2.25.1 is > 2.18.0 14:10:09 (temporary) 14:10:17 yeah 14:10:20 like, some metadata is wrong; or files are missing 14:10:31 I've seen this with different packages 14:10:37 and in jobs like unit tests or docs 14:10:54 it's not crippling gates atm so I don't bother 14:11:00 but if we see it getting problematic 14:11:05 we might need to investigate 14:11:26 ack, let's continue 14:11:28 #topic Sub Teams highlights 14:11:39 #link https://review.openstack.org/#/q/project:openstack/tempest+status:open 14:11:39 #link https://review.openstack.org/#/q/project:openstack/patrole+status:open 14:11:39 #link https://review.openstack.org/#/q/project:openstack/devstack+status:open 14:11:41 #link https://review.openstack.org/#/q/project:openstack/grenade+status:open 14:11:43 #link https://review.opendev.org/#/q/project:openstack/hacking+status:open 14:11:53 let's merge that periodic on devstack 14:12:04 #link https://review.opendev.org/c/openstack/devstack/+/741807 14:12:47 yoctozepto: +2 14:13:24 I'll check after the meeting 14:13:57 let's see Changes with Review-Priority == +1 14:14:01 https://review.opendev.org/q/label:Review-Priority%253D%252B1+status:open+(project:openstack/tempest+OR+project:openstack/patrole+OR+project:openstack/devstack+OR+project:openstack/grenade+OR+project:openstack/hacking) 14:14:09 #link https://review.opendev.org/q/label:Review-Priority%253D%252B1+status:open+(project:openstack/tempest+OR+project:openstack/patrole+OR+project:openstack/devstack+OR+project:openstack/grenade+OR+project:openstack/hacking) 14:14:24 2 more than the one mentioned already 14:14:35 one of them is actually approved already 14:15:23 #topic Gate Blocker Fix / Urgent Change 14:15:30 #link https://review.opendev.org/q/label:Review-Priority%253D%252B2+status:open+(project:openstack/tempest+OR+project:openstack/patrole+OR+project:openstack/devstack+OR+project:openstack/grenade+OR+project:openstack/hacking) 14:15:49 oh, merge conflict 14:16:21 i removed the RP +2 from it as we reverted the original change 14:16:35 should we abandon then? 14:16:38 but revert of revert with fixes was ready too? 14:17:04 brr, no idea, I was not in the flow 14:17:14 this one and this too merged https://review.opendev.org/c/openstack/devstack/+/791436 14:17:52 yeah, but I have no idea whether this does not have that issue 14:17:59 i need to check if we still need to add neutron_plugins/ovs_source ? 14:18:05 yeah 14:18:27 I was suppose to check log in 791436 for this file source but then forget 14:18:44 will do today 14:18:49 superb 14:18:55 gmann: thanks 14:19:00 #topic Open Discussion 14:19:11 Shutdown of ELK, Health dashboard, subunit2sql services 14:19:48 how much effort do we need to take in order to maintain this? 14:19:56 yeah we still not decided on o-h dashbaord 14:20:27 o-h need JS skill which i do not have so cannot help there much 14:20:49 there are multiple things needed, like os update and stuff, right? similar to the changes i had to do with refstack server 14:21:07 i don't have that much free time as i had with refstack server, so can't contribute much here 14:21:08 kopecmartin: may be we need to maintain subunit2sql too which is in infra now. but we can check on that ML thread about what all we need to maintain in QA to keep o-h dashbaord up 14:21:09 I have js skill but I don't know if I have enough time to handle more responsibilities 14:21:20 k 14:21:40 I mean, I could probably fix something here and there but not anything heroic 14:21:46 honestly saying i used to see o-h dashboard in past but i have not checked it even in last 6-7 months to so 14:22:09 gmann: do you have a link for the dashborad? 14:22:12 which means i am not using it much so I am fine on shutdown 14:22:28 can we learn how popular it is? 14:22:29 #link http://status.openstack.org/openstack-health/#/ 14:22:33 kopecmartin: ^^ 14:22:40 I have not used it either 14:22:59 perhaps nobody cares about these whistles 14:23:01 but mtreinish can feedback too as he was there when it was started/created 14:23:06 or masayukig ^^ 14:23:29 .. well I've used the openstack-health page couple of times in the past year 14:24:15 but no that much I would like to invest a lot of time into the maintenance 14:24:47 should we discuss https://review.opendev.org/c/openstack/devstack/+/755554 14:24:51 it has +2 from kopecmartin 14:24:56 yet -1 from me and frickler 14:25:02 and it seems to be popular among non-cores ;d 14:25:36 may be after Open Discussion items ? :) 14:25:59 oh, sorry, I missed we have more 14:26:14 yeah 14:26:31 QA feedback on Freenode situation 14:26:33 so Freenode now? 14:26:33 is the next topic 14:26:34 ok 14:26:43 so on o-h let's wait for mtreinish and masayukig feedback. (for ref http://lists.openstack.org/pipermail/openstack-discuss/2021-May/022359.html) 14:27:01 yeah 14:27:18 for freenode, i sent email to collect project feedback #link http://lists.openstack.org/pipermail/openstack-discuss/2021-May/022675.html 14:27:38 as QA team we need to add our choice in the mentioned therpad 14:27:41 etherpad 14:27:54 you know my personal preference 14:28:03 - start planning the move ;-) 14:28:10 I personally prefer to wait and monitor the situation than move ASAP 14:28:37 I'd wait as well 14:29:04 and would invest time to choose a good platform to use 14:29:10 Where would the move go to? 14:29:25 something where I will not need a znc (or other bounce) server maybe O:) 14:29:55 jparoly: that is not yet decided too 14:30:20 kopecmartin: that is going into the future ;-) 14:30:25 we are talking about moving to OFTC really 14:30:46 as weird things happen on freenode these days 14:31:32 something that looks modern? has features... sorry if thats not popular 14:31:58 kopecmartin: e.g., http://lists.openstack.org/pipermail/openstack-discuss/2021-May/022670.html 14:32:21 jparoly: the future plan is most likely going matrix 14:32:25 https://matrix.org 14:32:39 though we should really evaluate whether China likes it 14:32:54 and it's a different discussion to this wait/not-wait on Freenode 14:33:15 yeah, that is one of the reason i like 'wait and see' instead of 'move ASAP' as we do not have new platofrm finalized yet 14:33:57 it is big move in term of usage also so choosing long term stable platform is must before we say move from current one 14:34:01 gmann: o-h won't work without subunit2sql or ES (although it can be configured to run without ES available and only use subunit2sql) 14:34:23 mtreinish: yeah subunit2sql is also proposed to retire 14:34:28 i would like to avoid rush decisions in a case like this, let's think about it, gather data in order to do an informative decision , not an emotional one 14:34:35 the resources required to run the o-h api server are minimal compared to the backing data store. But if elk and subunnit2sql go away we can't run it anymore 14:35:17 mtreinish: exactly we (in TC) are looking for elk help as critical concern but not for subunnit2sql 14:35:28 kopecmartin: +1 14:35:49 mtreinish: and no one raised hand to help subunnit2sql yet 14:35:50 well the main of subunit2sql is much less than ELK. You just need a big enough sql db somewhere (the current db has about 500GB of data) 14:36:03 kopecmartin: the drama's been going for a month already, many other communities migrated already; it's far from emotional at the moment 14:36:18 and the workers to add data to the db are basically identical to the workers that push logs to ES 14:36:59 gmann: but opendev is really preparing for move to oftc - you want to skip that and decide on another platform instead? 14:37:16 I think we see this very differently :-) 14:37:26 yoctozepto: it is not yet decided. opendev waiting for openstack and other infra project to decide it 14:37:55 gmann: yeah, I know it's not decided 14:38:09 but it does not mean people don't have plans already 14:38:12 oftc is favorite choice for sure but until we decide it as final option 14:38:14 or visions 14:38:18 whatever to call it 14:38:24 Hmm, it's 8 to 4 in favour of the option2 in the etherpad now 14:38:24 they have options not exact plan 14:38:45 kopecmartin: we have 50 projects and they need to vote 14:38:47 gmann: well, fungi is working on technical aspects of opendev's integration 14:38:58 we, as TC, need a community-wise plan for sure 14:39:02 yeah, definitely .. 14:39:45 yeah we can just add QA preference there 14:40:34 I think QA has a mixed preference atm ;p 14:40:50 mtreinish: so we can wait for help on maintaining subunit2sql and if no one does then we retire o-h dashboard too (as pre deps) 14:41:03 ok, anyone from QA, put your opinion to the etherpad .. I'll write it too and then based on them we decide for whole QA 14:41:04 #link https://etherpad.opendev.org/p/feedback-on-freenode 14:41:22 kopecmartin: I will say we decide here and then you put the final option 14:41:42 otherwise it is difficult for me to conclude that from TC perspective :) 14:41:59 ah, i see 14:42:11 gmann: if you're going to come up with a maint solution for elk I would probably do subunit2sql (it's super easy and basically the same: https://docs.huihoo.com/openstack/docs.openstack.org/infra/system-config/logstash.html) 14:42:16 assuming there are users interested in it 14:42:53 mtreinish: ohk, sounds great. so if we are able to save ELK then we can save most of it. ack. 14:43:34 umm what is TC? 14:43:47 my take is we've waited enough and should start planning the move 14:44:02 the weird situation is only getting weirder, not better 14:44:11 jparoly: Technical Committee 14:44:19 thanks 14:44:25 the folks actually driving the OpenStack governance 14:44:44 gmann is our chair and I'm vice-chair 14:45:13 https://governance.openstack.org/tc/ 14:45:39 hmm, thinking about it , it's Option2 if the move is safe from tech perspective (it's planned and clear) and most of the projects will want to move right away 14:46:07 I just want to avoid confusion around where the discussions of that or that team are happening 14:46:17 OFTC is very safe and we wanted it before already ;d 14:46:43 yeah, we need to think about how to carry it out user-wise 14:46:50 optin1 is not stay at freenode, it is take more time and then choose best platform 14:47:05 how much time is more time? 14:47:08 it there a deadline? 14:47:12 the tech stuff is doable and fungi doing it 14:47:37 precisely, there is no deadline and we've waited already, with observing no better effects on freenode 14:47:51 we've been keeping all the same channels registered on oftc for over 7 years, since we first started recommending to move off freenode 14:48:15 i'm just making adjustments to some of the bots to handle variations in things like the services syntax 14:48:41 kopecmartin: not deadline but we need to decide that in TC meeting or so 14:48:56 \ and then it's a click-of-a-button for bots integration \ 14:49:02 fungi: what if a new contributor connects to the old freenode network, will there be a warning or a message shown somehow informing about the move? 14:49:27 but the impact is on people coming to freenode 14:49:29 precisely 14:49:55 the issue is with new policy freenode can ban our approach with "showing new place" on freenode 14:49:58 kopecmartin: we can set the channel topic, but it's hard to guarantee the new regime on freenode won't undo those, there's already been evidence of them erasing forwarding address topics from other channels to make them look like they're still in use 14:50:02 and it has done already 14:50:07 precisely 14:50:33 but we can squat here 14:50:38 for some time 14:50:44 and we direct people using docs 14:50:46 until they kickban our nicks at least ;) 14:50:57 yeah, they have to force us to leave 14:50:58 :D 14:51:00 yeah, my thinking too, we can monitor it from the beginning and keep setting the topic 14:51:20 we dropped into TC discussion now though 14:51:21 so 14:51:25 QA is Option2 14:51:28 and we can move on 14:51:36 ok, I'm for Option2 .. it's also about sending a message i would say :) 14:51:42 precisely 14:51:57 I'm overusing that adverb 14:51:58 my vote in option1 but as you would like to go 14:52:36 on the other hand, what "right now means"? :D 14:52:40 Option2: We want to move away from Freenode right now. 14:52:51 though on the matrix front, several folks have done some positive experiments with matrix bridging to irc in the past few days and it sounds like the newcomer experience might be very simple to document for connecting through matrix rather than using a traditional irc client/bouncer 14:53:21 (but using/recommending matrix is a separate topic from where to do irc) 14:53:45 kopecmartin: asap. 14:54:01 ah, i'm at the beginning again 14:54:16 kopecmartin: which means we need to choose platofrm asap and so does migration 14:55:03 i don't like when things are done asap , .. let's go with option1 however with a deadline 14:55:20 what's a reasonable deadline? 2 weeks? 14:55:38 milestone 2 14:55:41 kopecmartin: my take is we are after deadline 14:55:46 and should start *planning* 14:56:18 we need to choose deadline and i added this topic in TC meeting on thursday and where we can discuss on deadline or plan 14:56:29 yeah, if we wanna start planning then we need the option2 14:56:56 gmann: good, so until then it's option1 from QA, let's get TC feedback first 14:57:12 kopecmartin: this is actually TC asking for our feedback 14:57:28 I guess we should work on the question asked as it seems to be confusing 14:57:30 ;d 14:58:38 let me rephrase, let's wait for TC to choose a deadline which is a topic for the thursday's meeting, is that right? 14:59:04 so option1 until next QA meeting? 14:59:14 fine for me 14:59:15 then we reevaluate 14:59:39 good, i'll try to write a summary of this discussion to the etherpad 14:59:52 anything else for the open discussion? 14:59:56 that i what option1 is actually, 'wait and evaluate '. I am bad in writing it on etherpad may be 15:00:25 encryption-type patch I will review but may be tomorrow 15:00:31 gmann: ty 15:00:35 that will be fine 15:02:37 ok, let's close the office hour 15:02:44 we're over our time anyway 15:02:54 thank you all for the participation! 15:02:55 #endmeeting