18:00:07 <gouthamr> #startmeeting tc 18:00:07 <opendevmeet> Meeting started Tue May 21 18:00:07 2024 UTC and is due to finish in 60 minutes. The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot. 18:00:07 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 18:00:07 <opendevmeet> The meeting name has been set to 'tc' 18:00:20 <opendevreview> Ghanshyam proposed openstack/governance master: Remove retired project from Inactive project list https://review.opendev.org/c/openstack/governance/+/920146 18:00:20 <gouthamr> Welcome to the weekly meeting of the OpenStack Technical Committee. A reminder that this meeting is held under the OpenInfra Code of Conduct available at https://openinfra.dev/legal/code-of-conduct. 18:00:20 <gouthamr> Today's meeting agenda can be found at https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee 18:00:26 <gouthamr> #chair frickler 18:00:26 <opendevmeet> Current chairs: frickler gouthamr 18:00:29 <gouthamr> #topic Roll Call 18:00:31 <gmann> o/ 18:00:35 <slaweq> o/ 18:00:37 <spotz[m]> o/ 18:00:40 <dansmith> o/ 18:01:40 <frickler> \o 18:01:46 <gtema> o/ 18:02:01 <gouthamr> noted absence: noonedeadpunk 18:02:19 <JayF> o/ but distracted 18:02:37 <gouthamr> awesome; we have quorum 18:02:39 <gouthamr> lets get started 18:02:47 <gouthamr> #topic AIs from last week 18:03:05 <gouthamr> PyPi maintainers cleanup lists (gouthamr) 18:03:22 <gouthamr> on this, we identified a loooong list of non-openstackci maintainers that we flagged for cleanup. 18:03:27 <gouthamr> #link https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup#L52 (PyPi maintainers flagged for cleanup) 18:03:30 <gouthamr> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/64IDNMRS6AUC7T3NBAHF3A7RUCXZUFAG/ ([ptl][tc] OpenStack packages PyPi additional external maintainers audit & cleanup) 18:03:42 <gouthamr> This is a large task - we're cleaning up 207 maintainers off of 146 packages. We can't seem to automate this because of a lack of an API 18:03:54 <gouthamr> We'll keep you posted on how this goes; at best, we'll share responsibilities if fungi is okay so we can spread the risk of RSI between willing TC members 18:04:42 <clarkb> gouthamr: that first link is a list of projects where we do have ownership and can clean things up directly right? How big was the other list where we don't have ownership? 18:04:47 <fungi> yes, the goal was to ask those people to self-remove, but if there's nobody responding them we can discuss how we'll go about doing all the clicking 18:04:58 <clarkb> oh I see its below the first list 18:04:58 <fungi> s/them/then/ 18:05:00 <gmann> ++ on doing it on share basis 18:05:39 <gouthamr> clarkb yes, we have that list - sadly, this is long because we did ask folks to self-remove a bunch of times and hasn't happened 18:06:06 <fungi> to be clear on what the ui is like, it's several clicks from the list of projects to get to the collaborators list, and then when you click remove you need to type or paste the username you're removing for confirmation 18:06:20 <gouthamr> and yes, fungi identified some packages where we cant bump maintainers 18:06:29 <fungi> also loading delays, because pypi is a bit sluggish 18:06:32 <gouthamr> packages where we can't remove maintainers: eventlet, kuryr-lib, pymod2pkg, pbrx, git-nit, certbot-dns-openstack, rally-runners, networking-ovs-dpdk, keystoneclient, keystoneauth3, keystoneauth2, prep_source_repos, solum-infra-guestagent, reviewday 18:06:43 <gouthamr> ^ not all of them are under openstack governance 18:06:58 <gmann> because there openstackci is not owner right? 18:07:06 <fungi> right. that was from the full list of ~800 packages that openstackci is a collaborator on 18:07:12 <gmann> i see 18:07:23 <JayF> eventlet, for instance, we wouldn't even want to remove maintainers -- we should probably limit scope for what TC is handling to openstack namespace projects, yeah? 18:07:40 <gmann> one question, should openstackci should be maintainers in openstack governance projects only? 18:07:45 <fungi> and yes, that's the subset where openstackci is a maintainer rather than an owner 18:07:54 <gouthamr> JayF: yes, this list was pulled from the UI 18:07:57 <JayF> gmann: openstackci is the automation user used for all opendev pypi publishing, aiui 18:08:14 <JayF> gmann: I suspect in this case, the naming is just an artifact of it existing prior to opendev's name 18:08:16 <gmann> JayF: ohk then name is confusing. 18:08:27 <fungi> we'd need separate accounts for teh non-openstack projects who are uploading releases to pypi by tagging them in gerrit 18:08:34 <gmann> ++ 18:08:36 <fungi> if we switched that one to be openstack-only 18:08:58 <spotz[m]> Weird about the keystone ones? 18:09:13 <fungi> and yes, that account has been in continuous use for basically as long as we've been uploading releases to pypi from our ci/cd systems 18:09:55 <gouthamr> spotz[m]: true; on that, i will reach out to d34dh0r53 and the maintainer 18:09:56 <fungi> spotz[m]: not all that weird when you consider that some of those where either early experiments in openstack that were never realized or were deleted from openstack 18:10:14 <gouthamr> ^ oh, that i didn't know 18:10:15 <gmann> separating is good so that openstack TC can help in cleanup without risk of other projects impact if any 18:10:18 <fungi> i would consider deleting those projects if they have no releases with files 18:10:32 <spotz[m]> If still in use I think we care, but if not yeah it won't matter 18:10:49 <fungi> "back in the day" you had to reserve a project on pypi before you could upload releases, while today it's teh reverse 18:11:29 <fungi> so quite a few projects were reserved on the idea that we'd start developing/releasing them in openstack but then we didn't for various reasons 18:12:41 <fungi> the release team should, i think, have or tell us how to generate a list of the packages we're uploading as openstack deliverable artifacts, and then we can limit the cleanup effort to those 18:12:55 <fungi> should be able to tell us, i mean 18:13:06 <fungi> (not that i'm demanding they do so) 18:14:26 <gouthamr> i see; i can follow up on this 18:14:33 <frickler> mapping deliverables repos to pypi pkg names might be non trivial 18:14:40 <fungi> note that pypi project names don't 1:1 match git repository names for a few reasons (primarily normalization, but also some are renamed in their setup.cfg because the original name was already taken) 18:14:44 <gouthamr> ^ agree 18:14:47 <gouthamr> https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup#L413 18:14:48 <gmann> or we can do try-cleanup in current list in share basis. this tasks has been open since long and I think we should start cleaning up 18:15:11 <gmann> that seems faster than all other possible steps. 18:15:14 <gouthamr> #link https://etherpad.opendev.org/p/openstack-pypi-maintainers-cleanup#L413 (repos left to review for pypi packages because of metadata issues) 18:15:34 <gouthamr> gmann++ 18:16:35 <fungi> frickler: i didn't mean to imply it's trivial, just that the release team probably knows what is needed to do it and can hopefully provide guidance to whomever builds the list 18:16:57 <gouthamr> okay we've spent some time on this..i think we have raised enough awareness among the tc about the ongoing work.. 18:17:00 <gouthamr> lets chat on this channel outside of this meeting.. 18:17:07 <gouthamr> and follow up on this AI.. 18:17:13 <fungi> (e.g. scraping the project.name from the setup.cfg file of each deliverable repo in our projects list) 18:18:01 <gouthamr> gtema: do you have any update on the other AI we had? "Marking inactive projects prominently" (gtema) 18:18:20 <gtema> not yet, sorry 18:18:21 <gouthamr> i was going to suggest moving this one to https://etherpad.opendev.org/p/tc-2024.2-tracker 18:18:59 <gouthamr> nope don't be :) it isn't an urgent issue.. we can track it like we do our usual trackers 18:19:35 <gouthamr> alright; any other thoughts on $topic? 18:20:07 <gmann> I updated status for a few of the one assigned to me 18:20:30 <gouthamr> ty gmann.. /me didn't get around to checking this week 18:20:42 <gouthamr> #topic 2024.2 TC Tracker 18:20:46 <gouthamr> #link https://etherpad.opendev.org/p/tc-2024.2-tracker (Technical Committee activity tracker) 18:21:36 <gouthamr> lots of progress on the "Leaderless projects and inactivity" topic 18:21:57 <gouthamr> ty gmann and all those participating in the reviews and discussions on the ML 18:22:51 <frickler> how long do we want to wait for the charms ptl assignment review to be updated before considering the project inactive? 18:22:58 <gouthamr> :( 18:23:08 <gmann> what we do on this, its been open for a month and no response form volunteer leader? https://review.opendev.org/c/openstack/governance/+/914254 18:23:18 <gmann> frickler: yeah, that one 18:23:35 <gmann> same with Trove, no response on ML on another reminder 18:23:45 <gouthamr> frickler: its a typo in his name that was flagged, seems like an awfully low hanging fruit to knock out .. sigh.. 18:23:52 <gmann> I think we should move both projects to inactive state? 18:24:15 <gmann> gouthamr: it is not about typo, it is about how active the volunteer is for leading the project. 18:24:47 <frickler> yes, not reacting on their own changes is a bad indicator on that 18:24:50 <gmann> in past we have seen many cases where we struggled to get activities on projects we have been assigning the volunteer PTL 18:25:11 <gouthamr> i do see activity in the charms repos: https://review.opendev.org/q/charm 18:25:54 <fungi> i think canonical may still be relying on those for some commercial product of theirs, i wonder if jamespage knows what the situation is with its seeming abandonment 18:26:13 <slaweq> maybe we can ask jamespage about charms / 18:26:15 <slaweq> ? 18:26:29 <fungi> maybe they're willing to take further support fully downstream like red hat did with tripleo 18:26:34 <gmann> yeah, it seems Felipe was active on charm 4 days ago https://review.opendev.org/c/openstack/charm-ceph-osd/+/919794 18:26:35 <slaweq> sorry, fungi was faster :) 18:26:54 <gmann> it seems they might missed the governance change 18:27:21 <slaweq> fungi aren't charms also used in their new product which is sunbeam IIRC? 18:27:59 <fungi> yes, so may be a similar situation to tripleo where they're fine moving support for their legacy stuff downstream until they eol the product 18:28:12 <frickler> there's also some recent updates on the release team tracker, I think we'll need to look at all these teams well before milestone-2 https://etherpad.opendev.org/p/dalmatian-relmgt-tracking#L331 18:28:31 <fungi> i think sunbeam has its own separate charms, but i don't know the full story 18:29:05 <spotz[m]> I know when we approved Sunbeam James said there was enough difference for it to not be the same project 18:29:32 <gouthamr> #link https://review.opendev.org/c/openstack/governance/+/903490 (Retire all single charm repositories) 18:29:46 <gouthamr> ^ this is WIP 18:31:21 <gouthamr> maybe we can move this discussion to the ML and seek clarification.. 18:31:50 <spotz[m]> + 18:32:05 <gouthamr> #action gouthamr will start a mail thread on the status of the charms project and the PTL volunteer 18:34:34 <gouthamr> there are some open changes on openstack/project-team-guide that can use some review attention: 18:34:45 <gouthamr> #link https://review.opendev.org/q/project:openstack/project-team-guide+status:open (open changes to project-team-guide) 18:35:33 <frickler> + openstack-manuals 18:36:12 <gouthamr> ah good point 18:36:24 <gouthamr> #link https://review.opendev.org/q/project:openstack/openstack-manuals+status:open (open changes on openstack-manuals) 18:37:42 <dansmith> gouthamr: two of those p-t-g changes are waiting on fixes from the submitter 18:38:05 <spotz[m]> I'll check those out 18:38:18 <gouthamr> ty dansmith spotz[m] 18:40:06 <gouthamr> lets move on.. 18:40:11 <gouthamr> #topic Open Discussion 18:40:38 <gouthamr> ty slaweq for volunteering to be an election official 18:40:56 <opendevreview> Merged openstack/project-team-guide master: Make the project removal from infra as step#6 https://review.opendev.org/c/openstack/project-team-guide/+/919976 18:40:59 <gouthamr> ianychoi just published the first draft on the dates/process for the next election 18:41:02 <gmann> ++ thanks slaweq 18:41:17 <gouthamr> #link https://review.opendev.org/q/topic:%222025.1-elections%22 (kicking off 2025.1 elections) 18:41:48 <gmann> will check today 18:42:20 <gouthamr> i'm looking to achieve a couple of things with this: 1) real early notification so we don't have the few issues we've had where PTL candidates said they were away or didn't look at the ML 18:42:53 <gouthamr> 2) i wanted to do some more election awareness emails than usual so that we can have a better participation 18:43:17 <gouthamr> we did note that having ~50 people vote was pretty bad for the last TC election.. 18:47:01 <opendevreview> Merged openstack/openstack-manuals master: Re-add project data for 2023.1 Antelope https://review.opendev.org/c/openstack/openstack-manuals/+/916823 18:48:09 <gouthamr> not to pile on to the meeting fatigue :) lets get back to work 18:48:39 <gouthamr> last call for any other discussion to go on record here 18:48:56 <fungi> remember to submit to the summit cfp! 18:48:57 <spotz[m]> Someone needs to AC me until my repos count:) 18:49:03 <fungi> deadline is the end of the month 18:49:28 <JayF> spotz[m]: TC is technically final call on that. I'd suggest you submit a governance change to AC yourself. 18:49:47 <gouthamr> fungi ++ 18:50:14 <gouthamr> #link https://2024.openinfraasia.org/ (OpenInfra Summit Asia '24 Call for Papers) 18:51:56 <gouthamr> spotz[m]: like gmann mentioned in the last meeting, we can merge your governance change, and update the tooling and we'll not require an AC change for you.. 18:52:39 <gouthamr> but, this can take a bit more time.. can i work with you on that? 18:52:52 <spotz[m]> Sure 18:52:53 * gouthamr will probably ask questions and whip up votes at best 18:54:46 <gouthamr> awesome; if there's nothing else.. i'd like to give you back 5 mins 18:54:53 <gouthamr> thank you all for attending! 18:55:00 <gouthamr> #endmeeting