17:59:12 #startmeeting tc 17:59:13 Meeting started Tue May 9 17:59:12 2023 UTC and is due to finish in 60 minutes. The chair is knikolla. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:59:13 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:59:13 The meeting name has been set to 'tc' 17:59:16 Hi all, welcome to the weekly meeting of the OpenStack Technical Committee 17:59:23 A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct 17:59:27 #topic Roll Call 17:59:29 o/ 17:59:31 o/ 17:59:32 o/ 17:59:36 o/ 17:59:40 o/ 17:59:41 o/ 17:59:43 Today's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee 17:59:43 o/ 18:00:36 🙋‍♂️ 18:01:23 o/ 18:01:25 There are no noted absences in the agenda. I'll give it 2 more minutes. 18:03:29 tc-members: meeting time :) 18:03:56 #topic Follow up on past action items 18:04:01 noonedeadpunk has submitted two governance proposal, to add py38 to PTI and to clarify general expectations on supported Python versions 18:04:08 #link https://review.opendev.org/c/openstack/governance/+/882165?usp=search 18:04:11 #link https://review.opendev.org/c/openstack/governance/+/882154?usp=search 18:04:17 We can discuss this item further on the appropriate meeting topic. 18:04:22 Thank you noonedeadpunk. 18:04:28 gmann has added python 3.8 back to the generic job template 18:04:33 #link https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/882175 18:04:33 o/ 18:04:35 yeah, all the reverts are merged #link https://review.opendev.org/q/topic:oslo-drop-py38+ 18:04:38 Thank you gmann. 18:04:41 Huzzah 18:04:45 and py3.8 is voting job now #link https://review.opendev.org/c/openstack/openstack-zuul-jobs/+/882175 18:04:54 ++ 18:04:55 *unit test job 18:04:57 We can discuss this item further on the appropriate meeting topic for py38. 18:05:21 Which I'm going to jump directly on :) 18:05:25 #topic Schedule of removing support for Python versions by libraries - how it should align with coordinated releases (tooz case) 18:05:52 Again thank you for submitting both sets of proposals and reverts. 18:06:12 #link https://review.opendev.org/c/openstack/governance/+/882165?usp=search 18:06:19 #link https://review.opendev.org/c/openstack/governance/+/882154?usp=search 18:06:47 There's some discussion on the first patch about whether to keep this for libs only or to extend it to all projects 18:07:05 yeah, IMO it make sense to keep it for all projects not just lib 18:07:27 if a project is imported, it becomes a lib 18:07:33 yup 18:07:35 we've done a fairly terrible job in the past of drawing the line between libs and not-libs and avoiding importing the latter directly 18:08:06 and also integrated projects even they are not imported, operators need to move version up if either of integrated project bump up the version 18:08:30 for example, nova drop py3.8 but neutron want to keep or vice versa 18:08:56 +1 18:09:37 in neutron case we can also ask if this is lib or not as it's still imported by many projects, like e.g. stadium ones 18:10:54 ++, for things to be coinstallable it helps to target the same python version 18:10:56 I've seen examples in my life where people were importing nova 18:11:11 Even if we don't support it 18:11:18 cinderlib imports cinder 18:11:39 it would be nice if that were actually allowed 18:11:45 would solve some problems 18:11:59 that is why it will be easy and consistent for everyone moving to new python at same time 18:12:14 Well Nova public APIs are strictly the REST ones 18:12:33 But sometimes it's simplier for ops to just import the top-level package 18:12:54 So they take the risk 18:13:10 Required xkcd https://xkcd.com/1172/ 18:13:45 The other proposal attempts to provide recommendations for which versions of Python to support, and mentions all not-EOL versions of Python 18:14:02 hehe 18:14:02 #link https://review.opendev.org/c/openstack/governance/+/882154 18:16:08 Unless there is something we want to discuss here, we can continue the discussion in the changes. Please take a look. 18:16:32 #topic Gate health check 18:16:54 This is also related to py38, so it's a good continuation of the topic in more general terms. 18:17:01 I think we got rid of the last job requiring focal last week 18:17:02 Anything to report? 18:17:06 so that's good and healthy 18:17:14 Ironic has been tracking a grenade issue for the last few days 18:17:21 ++ thanks again dansmith for working on it 18:17:26 ++ 18:17:36 also lots of volume tempest test fixes in the last two weeks which seem to be improving things across the board according to opensearch 18:17:46 Awesome, thank you dansmith 18:17:48 yeah 18:18:17 I'll note this down in the weekly mail summary. 18:18:52 JayF: anything more on the grenade issue? 18:19:12 It's being worked on; there's no real update other than "try things and wait to see if they have an impact" 18:20:05 in neutron gate we also see pretty many grenade failures 18:20:20 all caused by timeouts while doing requests to nova API to create security group rules 18:20:29 it's reported here https://bugs.launchpad.net/neutron/+bug/2015065 18:20:46 it can potentially be seen also in other projects probably 18:20:46 #link https://bugs.launchpad.net/neutron/+bug/2015065 18:21:00 there's been an uptick in node_error results for jobs relying on nested-virt nodes because we're down to only ovh supplying those and sometimes it takes >10 minutes for a server create call to switch from building to active there 18:21:02 JayF: does this sound familiar? 18:21:26 Looks like it's a different issue than the one being tracked by Ironic. 18:21:47 yup 18:22:02 fungi: :( 18:22:07 slaweq: are you seeing SG failure in OVS grenade only or general grenade job also ? 18:22:28 I don't think I've seen that security group thing myself 18:22:31 gmann I saw it in ovs grenade job only 18:22:55 but I think ykarel told us today he saw similar errors in other jobs as well 18:23:01 and he is going to investigate that 18:23:23 yeah, I have also not seen those 18:23:27 slaweq: ack 18:23:41 thanks slaweq 18:23:48 anything else on the topic? 18:24:56 #topic Broken docs due to inconsistent release naming 18:25:00 I have updated my previous patch to include redirects from docs.openstack.org/2023.1.antelope to /2023.1 (and same for bobcat) 18:25:06 #link https://review.opendev.org/c/openstack/openstack-manuals/+/881290 18:26:03 knikolla: thanks, will check it. 18:26:14 Thanks :) 18:26:36 #topic Recurring tasks check 18:26:42 Bare 'recheck' state 18:26:47 #link https://etherpad.opendev.org/p/recheck-weekly-summary 18:27:09 all good there 18:27:18 Awesome :) 18:27:39 #topic Open Reviews 18:27:45 #link https://review.opendev.org/q/projects:openstack/governance+is:open 18:28:10 We have 2 new changes proposing PTLs, and after that we're done, besides Winstackers. 18:28:17 #link https://review.opendev.org/c/openstack/governance/+/882374?usp=search 18:28:26 #link https://review.opendev.org/c/openstack/governance/+/882139?usp=search 18:28:36 Please take a look and review. 18:29:08 We also have a proposal from jamespage to introduce a new project Sunbeam 18:29:16 #link https://review.opendev.org/c/openstack/governance/+/881384?usp=search 18:29:41 I did see quite a few comments on the change, so thank you to everyone that did already take a look :) 18:30:09 indeed - I think I'm caught up on reponses and a few updates 18:31:14 Thanks jamespage for the quick responses. 18:31:58 jamespage: changes LGTM, hopefully Felipe will leave a comment soon 18:32:03 Unless there's anything else to be discussed, that's all for what I had prepared. 18:32:18 rosmaita: I shall nudge him again :) 18:32:19 dansmith: and I managed not to say PR the entire meeting :) 18:32:40 knikolla ups, You just did :P 18:32:45 womp womp 18:32:46 :) 18:32:47 how meta 18:33:07 you can say PR, you just can't say it in reference to something in gerrit :D 18:33:09 haha, you would have similarly called me out for euphemistically saying GitHub terminology :) 18:33:44 Please, don't speak about my nightmares :) 18:33:51 I think we're done here yes? 18:33:57 Thanks all for a quick and productive meeting. Giving you all back half-hour to use wisely. 18:34:01 #link https://etherpad.opendev.org/p/2023-05-board-openstack-sync May 2023 OpenInfra Board Sync with OpenStack (20:00 UTC tomorrow, connection details are in the pad) 18:34:03 \o/ 18:34:06 ++ 18:34:07 o/ 18:34:20 \o/ 18:34:22 thanks fungi for the reminder! 18:34:23 since there doesn't seem to be an "open discussion" topic 18:34:32 * bauzas cries at night when he dreams about OpenStack moving to GitLab or GH 18:34:35 i'll tack it onto the "open reviews" discussion 18:34:49 I've got a CentOS board meeting at the same time but if we end early I'll join 18:34:57 You should all have a calendar invite for the meeting tomorrow between TC and board. 18:35:13 yep I'll be attending 18:35:20 i sent a couple to the openstack-discuss ml and TheJulia also sent one to the board ml 18:35:35 Thanks fungi 18:35:35 feel free to add topics to discuss 18:35:50 ++ 18:36:10 There's an etherpad link in the email 18:36:14 #link https://etherpad.opendev.org/p/2023-05-board-openstack-sync 18:36:22 yep, that's what i just linked above 18:36:48 Awesome. Thanks all again. 18:36:49 #endmeeting