13:00:33 <bbezak> #startmeeting kolla 13:00:33 <opendevmeet> Meeting started Wed Aug 6 13:00:33 2025 UTC and is due to finish in 60 minutes. The chair is bbezak. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:00:33 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 13:00:33 <opendevmeet> The meeting name has been set to 'kolla' 13:00:38 <bbezak> #topic rollcall 13:00:47 <jovial> o/ 13:00:49 <amir58118> o/ 13:02:18 <bbezak> #topic agenda 13:02:29 <bbezak_2> * Roll-call * Agenda * Announcements * Review action items from the last meeting * CI status * Release tasks * Regular stable releases (first meeting in a month) * Current cycle planning * Additional agenda (from whiteboard) * Open discussion 13:02:47 <bbezak> #topic CI status 13:03:13 <bbezak> Kolla and kolla-ansible with depends on to kolla are red 13:03:18 <bbezak> waiting for https://review.opendev.org/c/openstack/kolla/+/956456 13:03:36 <mmalchuk> o/ 13:04:05 <bbezak> hopefully we will merge that today 13:04:26 <bbezak> #topic Regular stable releases (first meeting in a month) 13:04:43 <bbezak> I'll raise new point releases when we will finish this CI drama 13:05:27 <bbezak> thx for everybody who helped with merging/fixing CI issues in recent weeks 13:05:43 <amir58118> +1 13:05:48 <bbezak> #topic Current cycle planning 13:06:21 <bbezak> I probably will look into bumping mariadb soon 13:06:25 <bbezak> as it is long overdue 13:07:09 <bbezak> looking at Valkey support - I will chase gkoper 13:07:51 <bbezak> mnasiadka: surely continues his uwsgi crusade shortly 13:09:17 <bbezak> recent docker/kolla issues is showing that we should pursue other options - I saw some patches already. and it will ease doing a multiarch containers as well 13:09:24 <bbezak> but that is different story 13:10:06 <bbezak> #topic Open discussion 13:10:33 <amir58118> The patches to restore Swift support in Kolla and Kolla-ansible are ready for review after about 100 hours of work. The core implementation is complete and available for feedback and structural review. https://review.opendev.org/q/topic:%22dev-swift-modernization%22 13:11:26 <mmalchuk> bbezak shouldn't we pin docker version for the deploy like we pin docker-py in https://opendev.org/openstack/requirements/src/branch/master/upper-constraints.txt#L537 ? 13:11:43 <seunghunlee> It would be nice we can review the process of https://bugs.launchpad.net/kolla-ansible/+bug/2080402 13:12:25 <frickler> late o/ 13:13:16 <mmalchuk> amir58118 did we run integration tests during deploy in your change in k-a ? 13:13:38 <bbezak> excuse me a bit as frickler just joined - we would need review for https://review.opendev.org/c/openstack/kolla/+/956456 13:14:07 <bbezak> and interesting point mmalchuk - I was testing that and that indeed works - https://review.opendev.org/c/openstack/kolla-ansible/+/956364 13:14:28 <bbezak> however that will not solve other people deployments 13:14:44 <bbezak> as not all people are using ansible-collection-kolla bits to deploy hosts 13:14:55 <bbezak> at least I don't think so 13:15:00 <frickler> yes, we have our own docker pins downstream, so I'd rather try to avoid that 13:15:03 <mmalchuk> bbezak sure! we pin python libs, but use latest os packages - that's not good) 13:15:15 <frickler> I'll look at the other fix today 13:15:35 <bbezak> well, kolla-ansible is not about host OS 13:15:40 <bbezak> kayobe surely 13:16:24 <mmalchuk> yep. but we can mention version pin recomendations in docs and use them on CI 13:16:34 <amir58118> mmalchuk: I'm not exactly sure, I need to check. 13:16:59 <bbezak> that is another bit to maintain 13:17:15 <mmalchuk> amir58118 ok. we can continue to discuss in the review 13:17:24 <bbezak> docker-py should fix it, but I didn't see traction there 13:17:48 <bbezak> so we're fixing it in kolla for the time being 13:18:29 <mmalchuk> ok. but it'll work until the next version with a new issue))) 13:19:02 <bbezak> that is also true :) 13:19:09 <mnasiadka> sorry for being late 13:19:24 <fungi> oh, thought you were still on vacation 13:19:26 <bbezak> kolla assumes stable docker engine libraries. I think that is a good assumption 13:19:40 <bbezak> (podman too) 13:19:50 <mnasiadka> fungi: got back yesterday :) 13:19:59 <fungi> i considered asking whether there's been any more thoughts about the contributor/maintainer survey results and metrics discussion from last week, but wanted to give you time to catch up on the meeting log 13:20:03 <mmalchuk> sometime ago mnasiadka said something like 'We're putting out fires' on my comment like this) 13:20:46 <mnasiadka> yeah, that's the life of a deployment project mainly ;-) 13:20:58 <mmalchuk> sad 13:22:06 <bbezak> seunghunlee: could you please elaborate your point on https://bugs.launchpad.net/kolla-ansible/+bug/2080402 ? 13:24:00 <seunghunlee> I don't think we should introduce new feature like this. The bug report is not explaining much and the review of the feature was not don thoroughly in my opinion. So,the feature ended up introducing a new bug https://bugs.launchpad.net/kolla-ansible/+bug/2119344 13:24:01 <frickler> iiuc https://review.opendev.org/c/openstack/kolla-ansible/+/956409 needs reviews 13:25:59 <seunghunlee> The feature doesn't provide default value or disable this new option by default. It was merged with the author's own +2. As a user, this looks rushed. 13:27:02 <bbezak> I agree. I think this is general problem of lack of meaningful reviews 13:27:18 <bbezak> as there are not many active reviewers 13:27:18 <mnasiadka> I think this can be said to hundreds of other patches 13:27:21 <mmalchuk> seunghunlee it was merged after +2 from mnasiadka 13:27:23 <mnasiadka> But it is what it is 13:28:03 <bbezak> there is last comment mentioning that it was a wrong way of doing things 13:28:13 <frickler> we can't test everything in CI, so I'd say we need to live with occasional regressions 13:28:48 <mnasiadka> we don't have an OIDC scenario, so yes - we need to live with that 13:28:55 <frickler> the question is whether a second reviewer would have had a better chance of spotting the regression, which may not be likely, too 13:29:11 <bbezak> true. but I also understand frustration about such cases 13:29:13 <mnasiadka> But I'm a bit worried about people writing what needs to be fixed in bug report, but not submitting the patch in Gerrit 13:29:59 <frickler> iiuc the fix is submitted now, so we can just review it and move on? 13:30:25 <bbezak> yeap 13:30:28 <seunghunlee> I agree not everything can be tested through CI and also we're lacking reviews but this particular case looked bad, so I wanted to raise 13:30:29 <mmalchuk> there is now good or wrong way to fix things. we should fix. lets merge one new fix) 13:30:37 <seunghunlee> Yeah I proposed a fix, so let's move on 13:30:41 <mmalchuk> there is no* 13:30:53 <mmalchuk> seunghunlee good! thanks 13:31:11 <bbezak> fungi: it looks like we need more time to analyze your points 13:31:46 <mnasiadka> seunghunlee: basically people historically have been putting a list of patches they need reviews on in the Kolla Whiteboard 13:31:53 <mnasiadka> And I liked that approach 13:32:48 <frickler> yes, but even that has been difficult to get to recentish 13:33:03 <mnasiadka> seunghunlee: https://etherpad.opendev.org/p/KollaWhiteBoard#L66 - here 13:33:20 <seunghunlee> mnasiadka: Thanks 13:33:24 <bbezak> and as usual if anybody have some change to review, and it feels stale. whiteboard is a good place, but IRC as well 13:33:51 <mnasiadka> frickler: well, I'd say it's mainly me and bbezak doing reviews, you probably have $other priorities and kevko seems to not be that active as in the past 13:34:46 <frickler> mnasiadka: yes, I try to look at the important things, but more often than not do not get to it in time 13:35:16 <mnasiadka> We don't really have candidates for more core reviewers, so it is what it is. 13:35:49 <opendevreview> Seunghun Lee proposed openstack/kolla-ansible master: Re-add missing permission attribute https://review.opendev.org/c/openstack/kolla-ansible/+/956667 13:36:23 <seunghunlee> That's unfortunate 13:36:34 <bbezak> ok, good discussion today 13:37:36 <bbezak> any other points to discuss? 13:38:34 <opendevreview> Michal Nasiadka proposed openstack/kolla-ansible master: WIP: Remove ironic-inspector deployment https://review.opendev.org/c/openstack/kolla-ansible/+/951864 13:38:42 <opendevreview> Michal Nasiadka proposed openstack/kolla-ansible master: WIP: Remove ironic-inspector deployment https://review.opendev.org/c/openstack/kolla-ansible/+/951864 13:39:27 <opendevreview> Seunghun Lee proposed openstack/kolla-ansible stable/2025.1: Re-add missing permission attribute https://review.opendev.org/c/openstack/kolla-ansible/+/956668 13:39:32 <bbezak> thank you all for lively chat today! 13:39:40 <bbezak> #endmeeting