13:00:03 <mnasiadka> #startmeeting kolla
13:00:03 <opendevmeet> Meeting started Wed Aug 27 13:00:03 2025 UTC and is due to finish in 60 minutes.  The chair is mnasiadka. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:00:03 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
13:00:03 <opendevmeet> The meeting name has been set to 'kolla'
13:00:06 <mnasiadka> #topic rollcall
13:00:13 <mmalchuk> \o
13:00:17 <frickler> o/
13:00:20 <mattcrees> o/
13:00:24 <amir58118> \o
13:00:30 <mnasiadka> o/
13:00:56 <bbezak> \o
13:01:48 <mnasiadka> #topic agenda
13:01:48 <mnasiadka> * CI status
13:01:48 <mnasiadka> * Release tasks
13:01:48 <mnasiadka> * Current cycle planning
13:01:48 <mnasiadka> * Additional agenda (from whiteboard)
13:01:49 <mnasiadka> * Open discussion
13:01:52 <mnasiadka> #topic CI status
13:02:11 <mnasiadka> Seems green overall, unless somebody corrects my mistake ;-)
13:03:39 <bbezak> looks ok
13:06:55 <mnasiadka> #topic Release tasks
13:07:25 <mnasiadka> So, UCA is sort of done, right bbezak ?
13:07:36 <bbezak> done
13:07:48 <mnasiadka> So, we need to decide what to do with RDO
13:07:56 <mnasiadka> It seems there will not be RDO Flamingo
13:08:26 <mnasiadka> And I guess the options are - use RDO Epoxy, - use RDO trunk, get rid of RDO usage
13:08:48 <bbezak> to be fair we were not using it for most of epoxy cycle and it was fine. IIRC
13:08:54 <frickler> is this also needed for rocky or only for centos?
13:09:26 <mnasiadka> also for rocky
13:09:31 <mnasiadka> it's not probably needed anymore
13:09:46 <mnasiadka> majority of things can come from epel
13:10:16 <mnasiadka> and we can use nfv-openvswitch and ceph repos as we do today, we don't need rdo-release package pulling these in
13:11:09 <mnasiadka> I did have some approach in https://review.opendev.org/c/openstack/kolla/+/948520
13:11:13 <mnasiadka> #link https://review.opendev.org/c/openstack/kolla/+/948520
13:13:45 <mnasiadka> So, I'd say let's try to drop the reliance on RDO, given they haven't still produced EL10 Epoxy
13:13:48 <bbezak> one less dependency to think about
13:14:10 <mnasiadka> If we don't manage to do it - we can just use trunk for now, or pin it to Epoxy, whatever works
13:14:25 <mnasiadka> Ok, another thing is - cycle highlights
13:14:43 <mnasiadka> We've been rather pushing them closer to trailing projects release, instead of now
13:14:52 <mnasiadka> because it's sort of a mid cycle for us now ;-)
13:15:40 <mnasiadka> Do we want to push something this week and then update - or rather do it later?
13:16:16 <opendevreview> Michal Nasiadka proposed openstack/kolla master: DNM: Test removing Delorean  https://review.opendev.org/c/openstack/kolla/+/948520
13:16:20 <opendevreview> Michal Nasiadka proposed openstack/kolla master: DNM: Test removing Delorean  https://review.opendev.org/c/openstack/kolla/+/948520
13:17:11 * frickler doesn't care, it's only marketing after all
13:17:34 <opendevreview> Michal Nasiadka proposed openstack/kolla master: DNM: Test removing Delorean  https://review.opendev.org/c/openstack/kolla/+/948520
13:17:52 <mnasiadka> Yeah, let's maybe defer that
13:18:06 <mnasiadka> #topic Current cycle planning
13:18:39 <mnasiadka> I've been trying to wrap my head around glance uwsgi failing in centos10 and not in debian/ubuntu due to chunked encoding - I'll continue doing that
13:18:58 <mnasiadka> Also we need to delete ironic-inspector - so been putting some cycles to that today
13:19:05 <mnasiadka> bbezak: how's the mariadb upgrade?
13:19:58 <bbezak> Mr seunghunlee is looking into that
13:20:00 <bbezak> looks promising
13:21:08 <mnasiadka> ok then
13:22:24 <mnasiadka> #topic Additional agenda
13:22:39 <mnasiadka> r-krcek [2025-08-20]
13:22:39 <mnasiadka> please review https://review.opendev.org/c/openstack/kolla-ansible/+/920714
13:22:39 <mnasiadka> please review together with depends-on https://review.opendev.org/c/openstack/kolla-ansible/+/948526
13:23:04 <amir58118> Please note that all Swift-related tests have passed and are green.
13:23:44 <jaydg> Could somebody please look into https://review.opendev.org/c/openstack/kolla-ansible/+/924575 ? Would it be possible to get it into Flamingo?
13:23:52 <mnasiadka> getting destroy working with limit could be a nice thing
13:25:01 <amir58118> +1
13:25:25 <mnasiadka> ok, next one from the whiteboard is:
13:25:28 <mnasiadka> amir58118 [2028-08-27]
13:25:28 <mnasiadka> please review this patch together with its depends-on https://review.opendev.org/c/openstack/kolla-ansible/+/953576
13:25:33 <mnasiadka> I assume it's the swift one
13:25:48 <amir58118> here we go again
13:26:05 <mnasiadka> At least I'm currently busy with something else, and unfortunately swift is at the distant end of my priority list
13:26:25 <mnasiadka> But if some other core reviewers are happy to have a look - then you might be lucky
13:27:03 <amir58118> Understood.
13:27:07 <mnasiadka> Just to note - release deadline for cycle trailing projects is Dec 1st-5th
13:27:16 <mnasiadka> in theory a lot of time, practically not that much
13:27:31 <amir58118> Got it.
13:27:48 <mnasiadka> jaydg: the same situation as with swift, we might get to it some day
13:28:06 <mnasiadka> ok then, let's go forward
13:28:10 <mnasiadka> #topic Open discussion
13:28:13 <mnasiadka> Anybody anything?
13:28:18 <opendevreview> Bartosz Bezak proposed openstack/kolla-ansible stable/2025.1: [letsencrypt] Avoid rendering empty --key-type argument  https://review.opendev.org/c/openstack/kolla-ansible/+/958653
13:28:46 <Vii> I have a few commits pending, mostly simple errors, like a documentation error and ansible output: 952818: Standardize naming for MariaDB recovery and backup files | https://review.opendev.org/c/openstack/kolla-ansible/+/952818
13:28:54 <Vii> Direct log sending: 953458: Fluentd: Switch to direct log shipping to OpenSearch nodes | https://review.opendev.org/c/openstack/kolla-ansible/+/953458
13:28:58 <mnasiadka> ah, I have something
13:29:03 <Vii> Too much backend (monitoring alarm :)) 940810: Letsencrypt disabled redundant entries in configs | https://review.opendev.org/c/openstack/kolla-ansible/+/940810
13:29:08 <Vii> Barbican, tests passed: 935704: Added vault support to barbican as back-end secret | https://review.opendev.org/c/openstack/kolla-ansible/+/935704
13:29:11 <mnasiadka> bbezak, frickler: What are we doing with that damn common_options?
13:30:16 <frickler> mnasiadka: good question, it is really tough to review that change, for questionable gain, too
13:30:44 <mnasiadka> right
13:30:46 <bbezak> I added my review
13:31:14 <bbezak> imho it is not backportable
13:31:42 <mnasiadka> I was thinking it's maybe easier to make it smaller by creating service-* roles for some repetitive actions and breaking it into respective changes
13:31:50 <mnasiadka> but yes, nothing will be backportable
13:33:05 <mnasiadka> I'll try to come up with something, but from another perspective we're leaking more passwords in other places, so maybe we need to start with a check what needs to be fixed.
13:34:14 <frickler> or maybe we don't need to, just better document that the deploy host needs to be hardened appropriately?
13:34:40 <mnasiadka> well, we could do that docs change for backportable fix
13:35:39 <mnasiadka> I'll try to document that in some spare cycles
13:35:55 <mnasiadka> Ok then - I think that's it for today
13:36:03 <mnasiadka> Thanks for coming - see you next week!
13:36:04 <mnasiadka> #endmeeting