| gmann | Office hour time.. | 01:00 |
|---|---|---|
| *** noonedeadpunk has quit IRC | 01:47 | |
| *** noonedeadpunk has joined #openstack-tc | 01:50 | |
| *** evrardjp has quit IRC | 04:33 | |
| *** evrardjp has joined #openstack-tc | 04:33 | |
| *** jaosorior has quit IRC | 06:14 | |
| *** ralonsoh has joined #openstack-tc | 06:35 | |
| *** slaweq has joined #openstack-tc | 06:54 | |
| *** e0ne has joined #openstack-tc | 07:15 | |
| *** tosky has joined #openstack-tc | 07:35 | |
| *** tosky has quit IRC | 08:05 | |
| *** tosky has joined #openstack-tc | 08:10 | |
| *** tosky has quit IRC | 08:14 | |
| *** tosky has joined #openstack-tc | 08:14 | |
| *** tosky has quit IRC | 09:39 | |
| *** tosky has joined #openstack-tc | 09:55 | |
| *** belmoreira has joined #openstack-tc | 10:28 | |
| *** jaosorior has joined #openstack-tc | 10:48 | |
| *** slaweq is now known as slaweq|lunch | 11:22 | |
| *** e0ne_ has joined #openstack-tc | 11:26 | |
| *** e0ne has quit IRC | 11:29 | |
| *** slaweq|lunch is now known as slaweq | 11:33 | |
| knikolla | o/ | 11:53 |
| mnaser | o/ | 11:55 |
| mnaser | belmoreira, spotz: can you add tech-commitee-chair to included groups in https://review.opendev.org/#/admin/groups/1453,members ? | 12:10 |
| mnaser | there's a bunch of leftover changes from months ago and i'm watching projects:openstack/governance and they're just sitting there idle | 12:10 |
| *** openstackgerrit has joined #openstack-tc | 12:14 | |
| openstackgerrit | Merged openstack/governance master: kolla-cli: deprecation - Mark as deprecated https://review.opendev.org/749694 | 12:14 |
| *** njohnston has joined #openstack-tc | 12:15 | |
| openstackgerrit | Merged openstack/governance master: Add openstack-ansible/os_senlin role https://review.opendev.org/748677 | 12:16 |
| *** redrobot has joined #openstack-tc | 12:18 | |
| *** e0ne_ has quit IRC | 12:21 | |
| openstackgerrit | Merged openstack/governance master: Add openstack-helm-deployments to openstack-helm https://review.opendev.org/748302 | 12:23 |
| *** e0ne has joined #openstack-tc | 12:43 | |
| *** Luzi has joined #openstack-tc | 12:55 | |
| *** beekneemech is now known as bnemec | 13:34 | |
| *** Luzi has quit IRC | 13:47 | |
| *** jaosorior has quit IRC | 14:16 | |
| *** dklyle has joined #openstack-tc | 14:38 | |
| gmann | o/ | 14:43 |
| *** tkajinam has quit IRC | 14:43 | |
| njohnston | o/ | 14:47 |
| *** diablo_rojo has joined #openstack-tc | 14:49 | |
| gmann | tc-members: any opinion on switching the base jobs to Focal in this week? - http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017060.html | 15:48 |
| gmann | few lib (already passed the freeze) need some work - http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017090.html | 15:48 |
| gmann | and few python client also | 15:48 |
| jungleboyj | o/ | 15:48 |
| gmann | smcginnis: ^^ release perspective? | 15:48 |
| smcginnis | Commenting here some of the discussion in the oslo channel... | 15:49 |
| gmann | I feel we switch it and failing one (which should not be many as most of the work for active projects are merged) can be fixed before release? | 15:50 |
| smcginnis | Release wise, we can do it and get requirements updates out. | 15:50 |
| jungleboyj | Think it is probably good to make the change sooner than later. | 15:50 |
| smcginnis | But I think we are way too late in the cycle to be switching this out now, and this close to the end of the cycle we should be work on stabilizing things, not potentially destabilizing them. | 15:50 |
| fungi | not switching means we're out of compliance with the platform we said we'd test on | 15:51 |
| gmann | yeah, i tried my best to do it as early (m-2 as planned) but less help from project side and I end up fixing a lot | 15:51 |
| smcginnis | We do still have tests running on focal. This would force all tests over though, right? | 15:51 |
| fungi | though this is why i was in favor of just ripping off the bandage at the start of the cycle and letting projects deal with all the broken preventing them from merging anything | 15:51 |
| gmann | no, Victoria test everything on Bionic | 15:51 |
| gmann | so no testing on Focal yet | 15:52 |
| smcginnis | fungi: For the record, that would be my preference going forward. | 15:52 |
| fungi | not switching means we end up having to keep support for the older platform until stable/victoria reaches eol rather than when stable/ussuri is gone | 15:52 |
| gmann | well, testing in advance is always good but Focal migartion has lot more changes | 15:53 |
| gmann | fungi: yes | 15:53 |
| fungi | or alternatively, means we will be unable to keep supporting stable/victoria for as long because the platform we proved it on has ceased to be viable | 15:53 |
| fungi | focal will have two years viability after bionic reaches eol | 15:54 |
| gmann | as per current testing and fixes, ~200 repo fixed/tested green ~30 repo fixes ready to merge ~100 repos failing or need some constraint update work. | 15:54 |
| clarkb | fungi: bionic has a 10 year lifespan iirc | 15:55 |
| fungi | clarkb: did they declare longer support duration for it? neat | 15:55 |
| clarkb | ya I think shuttleworth did that at an openstack summit too | 15:55 |
| fungi | gmann: in future, i don't think cooperation from projects is necessarily a prerequisite. it's fine to just break them and tell them "sorry, keeping to the claimed testing platform for the next release is part of being openstack, suck it up" | 15:56 |
| smcginnis | Totally agree! | 15:56 |
| gmann | integration testing has few open bugs so that can go later or in W cycle and switch tox base testing in Victoria ? | 15:56 |
| gmann | so that we can have unit/functional testing on Focalin Victoria itself | 15:57 |
| fungi | when trove tried to delay switching to a new ubuntu lts a cycle later than the rest of openstack, they were quickly broken by dependencies and integration testing merging changes which were incompatible with the older platform | 15:58 |
| gmann | fungi: i tried to avoid any failure but i think hard forcing it by m-2 and test/fix as much as possible before that. | 15:58 |
| clarkb | fungi: https://wiki.ubuntu.com/Releases they distinguish end of standard support and eol there | 15:58 |
| clarkb | not sure what the difference is, but bionic eol is 2028 | 15:59 |
| gmann | Focal migration delayed due to lot of incompatibly in deps etc | 15:59 |
| gmann | but anyways, doing it hard deadline in m-2 is what we should do in future | 15:59 |
| bnemec | Did I miss an email about this? Until today I wasn't aware that we needed to do anything. | 16:01 |
| bnemec | I guess I'm not surprised some lower constraints need bumping, but I didn't see anything that said what the project needed to do to determine this. | 16:02 |
| gmann | bnemec: '[all][tc][goals] Migrate CI/CD jobs to new Ubuntu LTS Focal' ML thread | 16:02 |
| gmann | i think i sent these udpates biweekly or so | 16:03 |
| *** belmoreira has quit IRC | 16:04 | |
| fungi | i think precise->trusty was the last time we actually switched on time, because the infra team just basically said one day (after repeated announcements that it would happen and how to test your stuff) "boop, all your jobs for the next release are running on trusty now" | 16:07 |
| fungi | and everyone revolted because they weren't ready | 16:07 |
| fungi | for trusty->xenial we tried to make it possible for projects to switch at their own pace, and lots of them didn't get around to it until nearly the release (and some didn't get around to it at all) and so stuff broke due to projects and unit vs functional vs integration jobs switching at different times, and it was a real mess | 16:08 |
| fungi | and after that the infra team decided they were done with people complaining and not wanting to cooperate, and declared that if the tc was making the guidelines the tc would need to enforce them from then on | 16:09 |
| fungi | and so xenial->bionic took two cycles to get done | 16:10 |
| fungi | that was really no fun | 16:10 |
| fungi | i'd rather wished the lesson learned from that experience was to just put on fireproof underpants and switch as early as possible | 16:11 |
| gmann | is it? it was done in Stein. i tested almost all the repo and due to less changes from Xenail to bionic it was good migration with Zero downtime (except cinder for 1 day) | 16:11 |
| fungi | yep, rocky should have been tested on bionic | 16:11 |
| fungi | but that didn't happen | 16:12 |
| *** ralonsoh has quit IRC | 16:12 | |
| gmann | http://lists.openstack.org/pipermail/openstack-discuss/2019-April/004647.html | 16:12 |
| *** ralonsoh has joined #openstack-tc | 16:12 | |
| gmann | i mean start and end of migration | 16:13 |
| fungi | ahh, barely made it for the release then. that was 5 days before stein released | 16:15 |
| gmann | I feel there is no issue on backporting also for project not going to finish before victoria is cut as these are mostly their gate fixes for lower constraints bump. | 16:16 |
| spotz | mnaser: looking | 16:16 |
| spotz | mnaser: done | 16:17 |
| *** e0ne has quit IRC | 16:17 | |
| *** ralonsoh has quit IRC | 16:39 | |
| *** ralonsoh has joined #openstack-tc | 16:40 | |
| clarkb | fungi: gmann: ya I think what experience has shown is that there is no perfect option for bumping up base platforms, but that ripping the bandaid off results in less pain long term | 16:41 |
| *** lpetrut has joined #openstack-tc | 16:46 | |
| *** bnemec has quit IRC | 16:52 | |
| *** bnemec has joined #openstack-tc | 17:02 | |
| *** ralonsoh has quit IRC | 17:23 | |
| *** e0ne has joined #openstack-tc | 17:25 | |
| *** e0ne has quit IRC | 17:39 | |
| *** lpetrut has quit IRC | 17:51 | |
| *** e0ne has joined #openstack-tc | 20:01 | |
| *** e0ne has quit IRC | 20:44 | |
| *** slaweq has quit IRC | 20:51 | |
| *** slaweq has joined #openstack-tc | 20:56 | |
| *** slaweq has quit IRC | 21:01 | |
| *** slaweq has joined #openstack-tc | 21:16 | |
| *** slaweq has quit IRC | 21:21 | |
| *** tosky has quit IRC | 22:55 | |
| *** tkajinam has joined #openstack-tc | 22:57 | |
| *** tkajinam has quit IRC | 22:57 | |
| *** tkajinam has joined #openstack-tc | 22:58 | |
| gmann | after fixing more today and as fixes will easy if any repo is pending, I think we are safe to switch tox based testing to Focal. - http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017104.html | 23:01 |
| gmann | and let's discuss about integration testing migration in tomorrow office hour, 3 blocker in that. | 23:01 |
| *** slaweq has joined #openstack-tc | 23:18 | |
| *** slaweq has quit IRC | 23:23 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!