| *** pojadhav|afk is now known as pojadhav | 03:16 | |
| yoctozepto | gmann: ok, we will do it on irc; I wish you and your family good health | 07:44 |
|---|---|---|
| *** pojadhav is now known as pojadhav|brb | 11:57 | |
| *** pojadhav|brb is now known as pojadhav | 12:10 | |
| *** akahat is now known as akahat|rover | 13:45 | |
| *** akahat|rover is now known as akahat | 13:56 | |
| yoctozepto | tc-members: it's time to meet | 15:00 |
| yoctozepto | #startmeeting tc | 15:00 |
| opendevmeet | Meeting started Thu Feb 3 15:00:15 2022 UTC and is due to finish in 60 minutes. The chair is yoctozepto. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
| opendevmeet | The meeting name has been set to 'tc' | 15:00 |
| yoctozepto | #topic Roll call | 15:00 |
| yoctozepto | \o/ | 15:00 |
| dansmith | o/ | 15:00 |
| spotz | o/ sorta | 15:00 |
| jungleboyj | o/ | 15:00 |
| yoctozepto | there is an apologies entry for spotz and ricolin but spotz is here nonetheless, welcome! | 15:01 |
| yoctozepto | counting 4 members | 15:01 |
| yoctozepto | will there be quorum today | 15:01 |
| yoctozepto | belmoreira seemingly not around | 15:02 |
| jungleboyj | Thought we were supposed to do a video meeting today. | 15:02 |
| yoctozepto | mnaser is on the openinfra live | 15:02 |
| * jungleboyj not that I am complaining | 15:02 | |
| yoctozepto | seemingly along with belmoreira | 15:02 |
| mnaser | yoctozepto: belmoreiro is hosting with me yes :p | 15:02 |
| mnaser | (sorry) | 15:02 |
| yoctozepto | jungleboyj: yeah, we (with gmann) changed to make it more accessible | 15:03 |
| jungleboyj | Awesome. | 15:03 |
| yoctozepto | mnaser: yeah, no problem, that time conflict is something to resolve in the future | 15:03 |
| dansmith | so, quorum or no? | 15:04 |
| yoctozepto | seems no quorum | 15:04 |
| spotz | It's handy when we're on the show as a group but sucks when it's 1 here and 1 there | 15:05 |
| yoctozepto | spotz: indeed | 15:05 |
| jungleboyj | We can quarrel. | 15:05 |
| yoctozepto | we can continue with no decision powers I guess | 15:05 |
| yoctozepto | lol | 15:05 |
| yoctozepto | diablo_rojo_phone: around? | 15:06 |
| yoctozepto | _phone suggests not | 15:06 |
| yoctozepto | ok, let's continue for the sake of having discussions today | 15:06 |
| yoctozepto | #topic Follow up on past action items | 15:07 |
| yoctozepto | but there were none | 15:07 |
| yoctozepto | #topic Gate health check | 15:07 |
| yoctozepto | ** Fixing Zuul config error in OpenStack | 15:07 |
| yoctozepto | *** https://etherpad.opendev.org/p/zuul-config-error-openstack | 15:07 |
| dansmith | the qos thing si resolved? | 15:08 |
| dansmith | that was biting several of us for a few days | 15:08 |
| yoctozepto | dansmith: no idea, I know nova-next is practically broken | 15:08 |
| yoctozepto | (read about that on the ml) | 15:08 |
| dansmith | yeah but being worked on, that only affects nova though IIRC | 15:08 |
| yoctozepto | and devstack gate | 15:08 |
| diablo_rojo_phone | Hosting open infra live so I'm a bit spotty | 15:08 |
| dansmith | the qos one was breaking lots of people | 15:08 |
| diablo_rojo_phone | Sorry! | 15:08 |
| yoctozepto | diablo_rojo_phone: ack, enjoy it | 15:09 |
| dansmith | throughput has been good, except monday (or tuesday?) it was somewhat backed up, | 15:09 |
| dansmith | but no complaints from me.. seems pretty decent lately | 15:09 |
| yoctozepto | (seems 3 members captured by openinfra live, one holidays, one sick) | 15:09 |
| yoctozepto | dansmith: ack, thanks | 15:09 |
| jungleboyj | We had some issues with the Stable gates this week but I think those have been worked out as well. | 15:10 |
| yoctozepto | great, thanks for reporting | 15:10 |
| yoctozepto | so | 15:11 |
| yoctozepto | on the zuul config errors | 15:11 |
| yoctozepto | seems some projects working diligently on fixing them | 15:11 |
| yoctozepto | others not so much | 15:11 |
| dansmith | I don't even know what the issue is? | 15:11 |
| yoctozepto | dansmith: infra asked us to clear up https://zuul.opendev.org/t/openstack/config-errors | 15:12 |
| yoctozepto | it's ridiculously long | 15:12 |
| yoctozepto | the progress (along with details) is tracked here: https://etherpad.opendev.org/p/zuul-config-error-openstack | 15:12 |
| dansmith | man that's a list | 15:12 |
| yoctozepto | yup | 15:12 |
| jungleboyj | Yikes. | 15:13 |
| dansmith | seems mostly networking? | 15:13 |
| yoctozepto | I think we really need a person from TC to watch over this progress | 15:13 |
| yoctozepto | dansmith: hard to tell for sure | 15:13 |
| yoctozepto | but looks like it | 15:13 |
| dansmith | can we make a list of all the repos those jobs are in? | 15:14 |
| yoctozepto | https://review.opendev.org/c/openstack/releases/+/818207 | 15:14 |
| yoctozepto | they seemed to have eoled these branches | 15:14 |
| yoctozepto | so no idea why this bugs the zuul still | 15:14 |
| dansmith | because it's hard to tell, and that would surely be a smaller list | 15:14 |
| dansmith | ah, I see the etherpad has a smaller list of projects | 15:14 |
| yoctozepto | yeah, it's been parsed | 15:14 |
| yoctozepto | fungi, clarkb: you around? :-) it seems eoling old neutron/networking branches did not help with zuul errors | 15:15 |
| yoctozepto | I mean, it is hard to tell which are still affected | 15:16 |
| yoctozepto | guessing eoled ones are not | 15:16 |
| fungi | eol'ing branches without eol'ing the branches of other projects using things from them is what creates a lot of those errors | 15:16 |
| yoctozepto | ah, that makes sense | 15:17 |
| fungi | we used to eol across the board in every repo | 15:17 |
| fungi | these days it's on-demand and some projects are asking to have branches eol'd which contain things used by other projects who aren't asking to have the same branches eol'd | 15:17 |
| yoctozepto | messy | 15:18 |
| fungi | so those branches are effectively broken and untestable, but nobody's paying attention to them anyway | 15:18 |
| fungi | in short, better cleanup would help | 15:18 |
| dansmith | does this actually hurt anything? other than maybe making that config errors page less useful for actual errors? | 15:19 |
| fungi | pretty much that | 15:20 |
| dansmith | ack | 15:20 |
| fungi | and it's an indication of essentially abandoned projects/branches in openstack | 15:20 |
| dansmith | well, looks like neutron and ironic are probably the biggest of the ones I know how to contact | 15:21 |
| yoctozepto | well, I guess the TC does not have the bandwidth to go and fix it for the projects; the projects are aware and... that's it I guess? | 15:22 |
| dansmith | well, I was just thinking about a little tree-shaking, | 15:22 |
| dansmith | and that would be the most helpful for the biggest offenders | 15:22 |
| dansmith | oh I guess only two from ironic | 15:24 |
| yoctozepto | neutron/networking has 33 entries | 15:24 |
| dansmith | yeah, neutron is the big one | 15:24 |
| yoctozepto | ironic 4 entries | 15:24 |
| yoctozepto | manila 1 | 15:25 |
| dansmith | I only see two from ironic, what are the others? | 15:25 |
| dansmith | nm I see | 15:26 |
| * gouthamr takes note on the manila job | 15:26 | |
| gouthamr | #link https://review.opendev.org/c/openstack/python-manilaclient/+/825068 | 15:26 |
| gouthamr | will be merged soon | 15:26 |
| fungi | i think one of the big cases for netron was that a project (driver?) was retired but not removed from job configs | 15:26 |
| fungi | er, neutron. though netron is a humorously appropriate typo | 15:27 |
| dansmith | yoctozepto: we should probably do this async to the meeting | 15:27 |
| dansmith | yoctozepto: I rang the ironic bell, so hopefully we'll get some action there.. four entries should be not too bad | 15:28 |
| yoctozepto | ok, I finished with the current list | 15:28 |
| yoctozepto | but there seems to be more projects affected | 15:28 |
| dansmith | they were was blissfully ignorant as I was until the top of this hour :) | 15:28 |
| yoctozepto | as there are 79 entires in total | 15:28 |
| yoctozepto | dansmith: haha | 15:28 |
| yoctozepto | thanks for letting them know | 15:28 |
| yoctozepto | thanks gouthamr | 15:29 |
| yoctozepto | ok, we made some progress | 15:29 |
| yoctozepto | even without the quorum | 15:29 |
| yoctozepto | :-) | 15:29 |
| yoctozepto | let's switch to the next topic | 15:29 |
| yoctozepto | #topic Progress check on Yoga tracker | 15:30 |
| yoctozepto | ** https://etherpad.opendev.org/p/tc-yoga-tracker | 15:30 |
| yoctozepto | technical guidelines seem to have stalled | 15:31 |
| yoctozepto | is that right, spotz? | 15:31 |
| dansmith | that was my deal actually | 15:31 |
| dansmith | and yeah, I suck | 15:31 |
| spotz | yoctozepto: Yeah I did the original ask but then we semmed to change what we wanted and it made sense to just put it on hold | 15:31 |
| jungleboyj | Wow, that took a turn. | 15:31 |
| yoctozepto | yeah | 15:32 |
| yoctozepto | point 2 is gate health check that we hvae done | 15:32 |
| yoctozepto | point 3 is what I'm working atm | 15:33 |
| yoctozepto | there is one issue to be aware of - the releases team is relying on one of the tags (stable policy) | 15:33 |
| yoctozepto | it's well documented so I need to restore that tag for the time being | 15:33 |
| yoctozepto | I have nothing else to report, there should be a new rev of the change proposed today | 15:34 |
| yoctozepto | mergable this time | 15:34 |
| yoctozepto | folks for 4,5,6 not present | 15:35 |
| yoctozepto | point 7 is at jungleboyj | 15:35 |
| jungleboyj | I should be able to get to the Survey feedback next week hopefully. | 15:35 |
| yoctozepto | please speak up, Tarzan | 15:36 |
| yoctozepto | thanks jungleboyj | 15:36 |
| jungleboyj | yoctozepto: ha ha. | 15:36 |
| yoctozepto | ok, we have reviewed the progress on the yoga tracker | 15:37 |
| yoctozepto | #topic Z Release Cycle Name | 15:37 |
| yoctozepto | ** Election results: | 15:37 |
| yoctozepto | https://civs1.civs.us/cgi-bin/results.pl?id=E_693bd8bbe63ca52f | 15:37 |
| yoctozepto | ** Next step, the foundation has started the trademark checks. | 15:37 |
| jungleboyj | Yay! | 15:37 |
| yoctozepto | I *really* hope Zen will pass | 15:37 |
| yoctozepto | as I will deny to work on a Zombie project | 15:38 |
| yoctozepto | I guess nothing more to say on that! | 15:39 |
| yoctozepto | #topic Z cycle Technical Elections | 15:39 |
| yoctozepto | ** Proposed Election schedule, need to merge it if no objection. | 15:39 |
| yoctozepto | *** https://review.opendev.org/c/openstack/election/+/825017 | 15:39 |
| yoctozepto | I'm not an election official | 15:40 |
| yoctozepto | so can't +2 | 15:40 |
| yoctozepto | but I guess there are no objections | 15:40 |
| yoctozepto | jungleboyj, spotz: can you comment on that? | 15:40 |
| yoctozepto | this is prime time it gets merged | 15:40 |
| spotz | yoctozepto: I can +w if diablo_rojo_phone doesn't have anything on it | 15:41 |
| jungleboyj | Do we need to wait for the release name? That was the thing I thought was holding it up. | 15:42 |
| jungleboyj | Otherwise, I am fine with it. | 15:42 |
| spotz | Oh good question not sure | 15:42 |
| yoctozepto | I did not understand it like this | 15:42 |
| yoctozepto | fungi: may you help? you know pretty much everything related to these processes | 15:42 |
| jungleboyj | He he. | 15:43 |
| fungi | the q is whether we need a release name before we approve an election schedule? | 15:44 |
| yoctozepto | fungi: yeah | 15:45 |
| yoctozepto | I think we can fix it later | 15:45 |
| fungi | it's all a matter of communications, yes | 15:45 |
| yoctozepto | both there and in comms | 15:45 |
| yoctozepto | ++ | 15:45 |
| fungi | we can say "z cycle" and use a placeholder | 15:45 |
| yoctozepto | thanks fungi | 15:45 |
| jungleboyj | Ok. Then we should probably merge that. | 15:45 |
| fungi | it gets harder if we need to fix it after we start accepting nominations | 15:45 |
| yoctozepto | my thoughts exactly | 15:45 |
| fungi | but this early it's not critical yet | 15:46 |
| yoctozepto | diablo_rojo_phone: no objections against the schedule? | 15:46 |
| yoctozepto | guess she's too busy being the live hostess | 15:48 |
| yoctozepto | no problem, we can do it after the meeting | 15:48 |
| yoctozepto | #topic Open Reviews | 15:48 |
| yoctozepto | https://review.opendev.org/q/projects:openstack/governance+is:open | 15:49 |
| yoctozepto | my WIP will change later today, please feel very welcome to review then | 15:49 |
| yoctozepto | other than that | 15:49 |
| yoctozepto | we have two from ricolin that have stalled | 15:49 |
| yoctozepto | and ricolin not around today | 15:50 |
| yoctozepto | and one test change | 15:50 |
| yoctozepto | from lourot | 15:50 |
| yoctozepto | so nothing to review atm | 15:50 |
| yoctozepto | good job | 15:50 |
| yoctozepto | #topic Open discussion | 15:51 |
| yoctozepto | we have ~9 minutes if you want to discuss something in general | 15:51 |
| yoctozepto | otherwise I will close the meeting | 15:51 |
| dansmith | doeeet! | 15:52 |
| yoctozepto | #endmeeting | 15:53 |
| opendevmeet | Meeting ended Thu Feb 3 15:53:08 2022 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:53 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/tc/2022/tc.2022-02-03-15.00.html | 15:53 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/tc/2022/tc.2022-02-03-15.00.txt | 15:53 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/tc/2022/tc.2022-02-03-15.00.log.html | 15:53 |
| yoctozepto | thanks everyone | 15:53 |
| jungleboyj | Thank you yoctozepto ! | 15:53 |
| clarkb | fungi: dansmith: re zuul errors the other consideration is that zuul may not be doing what projects expect when they have errors (running too few jobs or older configs) | 15:54 |
| dansmith | okay I figured any jobs with config errors wouldn't run at all | 15:55 |
| clarkb | I think that is the case. But also any that may have depended on those jobs as well | 15:55 |
| fungi | the jobs with config errors won't run, or depending on the nature of the error no jobs may run for a particular change (making it unmergeable) | 15:55 |
| fungi | but also having jobs not run means you may end up not testing something you think is tested and not noticing | 15:56 |
| dansmith | I guess that only matters for non-voting jobs? | 15:57 |
| dansmith | but yeah, fair point | 15:57 |
| fungi | well, voting jobs which configuration errors not getting run could mean errors you thought would be caught aren't and don't prevent the change from merging | 15:58 |
| dansmith | eh? I figured a voting job that failed to run for config reasons leads to a -1 from zuul? | 15:59 |
| fungi | not if it can't figure out it's a job which should be run | 16:04 |
| fungi | job selection is dynamic | 16:05 |
| dansmith | hmm, okay | 16:06 |
| fungi | whether a job is expected to be triggered for a given change is determined in part by the job's definition | 16:06 |
| dansmith | every time I mess up a job config zuul is harsh with its rebuke | 16:06 |
| fungi | yes, normally zuul won't merge changes which introduce config errors | 16:07 |
| fungi | but config errors can come from outside changes like repo renames or branch deletions | 16:07 |
| dansmith | ack | 16:07 |
| fungi | or (less often) changes merging to a project in another tenant which has config reused in this tenant but isn't gated by it | 16:11 |
| jungleboyj | Is the release cadence meeting happening? | 16:12 |
| fungi | in that case zuul at least gives us feedback letting us know what we'll be breaking though (e.g. centos-8 nodeset removal change in the opendev/base-jobs repo) | 16:12 |
| jungleboyj | If so, what is the connection info? | 16:12 |
| *** whoami-rajat__ is now known as whoami-rajat | 16:12 | |
| fungi | oh, i probably should have brought centos-8 removal up in the gate issues part of the meeting | 16:13 |
| dansmith | jungleboyj: gmann said he'd rescheulde | 16:13 |
| jungleboyj | Ah. Ok. Thanks! | 16:13 |
| fungi | for those who weren't paying attention, centos linux 8 was removed from the official centos mirrors (they "moved it to the vault") and so our mirrors pulled that deletion before we noticed | 16:15 |
| fungi | we took that as a cue to go ahead and rip out the centos-8 nodeset from opendev/base-jobs, but didn't consider that doing so could make fixing job definitions impossible because their configs were effectively deadlocked | 16:17 |
| fungi | after considering options, we temporarily reintroduced the centos-8 nodeset but based it on centos-8-stream node labels, so jobs which say they want centos-8 will be effectively running on centos-8-stream instead now, until their maintainers have a chance to correct them | 16:18 |
| spotz | Thanks for that fungi | 16:55 |
| fungi | wait to thank me until it blows up in your face ;) | 16:58 |
| spotz | hehe | 16:58 |
| opendevreview | Radosław Piliszek proposed openstack/governance master: Remove the tags framework (part 1) https://review.opendev.org/c/openstack/governance/+/822900 | 17:54 |
| opendevreview | Radosław Piliszek proposed openstack/governance master: Remove the tags framework (part 1) https://review.opendev.org/c/openstack/governance/+/822900 | 18:31 |
| opendevreview | Radosław Piliszek proposed openstack/governance master: Remove the tags framework (part 1) https://review.opendev.org/c/openstack/governance/+/822900 | 18:32 |
| *** diablo_rojo is now known as Guest1688 | 21:57 | |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!