15:00:27 <mgoddard> #startmeeting kolla 15:00:32 <openstack> Meeting started Wed Sep 23 15:00:27 2020 UTC and is due to finish in 60 minutes. The chair is mgoddard. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:33 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:36 <openstack> The meeting name has been set to 'kolla' 15:00:38 <mgoddard> #topic rollcall 15:00:47 <mgoddard> put your hands in the air like you just don't care 15:00:49 <priteau> o/ 15:00:50 <mgoddard> \o/ 15:01:18 <hrw> [o] 15:01:21 <Fl1nt> \o 15:01:24 <wuchunyang> o 15:01:38 <yoctozepto> o/ 15:01:41 <mgoddard> that's a new one hrw 15:01:47 <hrw> ha! 15:01:54 <yoctozepto> mgoddard: but we care :-( 15:02:07 <mnasiadka> o/ 15:02:10 <Fl1nt> he's grumpy ^^ 15:02:21 <hrw> Fl1nt: who is not? 15:02:34 <yoctozepto> hrw: santa claus 15:02:37 <Fl1nt> I'm not ^^ 15:03:00 <mgoddard> #topic agenda 15:03:00 <yoctozepto> <sherlock-mode>Fl1nt is Santa Claus</sherlock-mode> 15:03:10 <mgoddard> * Roll-call 15:03:13 <mgoddard> * Announcements 15:03:14 <mgoddard> ** Kolla Klub meeting returns tomorrow @ 15:00 UTC! https://docs.google.com/document/d/1EwQs2GXF-EvJZamEx9vQAOSDB5tCjsDCJyHQN5_4_Sw/ 15:03:16 <mgoddard> ** Kolla feature freeze fast approaching: Sep 28 - Oct 02 15:03:19 <mgoddard> ** Submit Virtual PTG topic proposals: https://etherpad.opendev.org/p/kolla-wallaby-ptg 15:03:20 <mgoddard> ** TC & PTL elections coming up: http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017306.html 15:03:23 <mgoddard> ** Ubuntu qemu bug http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017390.html 15:03:25 <mgoddard> * Review action items from the last meeting 15:03:27 <mgoddard> * CI status 15:03:29 <mgoddard> * Victoria release planning 15:03:31 <mgoddard> ** Bump minimum Ansible version to 2.9 but blacklist 2.10 on CentOS 8? https://github.com/ansible/ansible/issues/71786 15:03:33 <mgoddard> #topic announcements 15:03:52 <mgoddard> #info Kolla Klub meeting returns tomorrow @ 15:00 UTC! 15:04:15 <mgoddard> #link https://docs.google.com/document/d/1EwQs2GXF-EvJZamEx9vQAOSDB5tCjsDCJyHQN5_4_Sw/ 15:04:36 <mgoddard> agenda: 15:04:39 <mgoddard> Quick upstream update 15:04:41 <mgoddard> Improving engagement with the community in non EU/US timezones 15:04:43 <mgoddard> Operator feedback 15:04:53 <yoctozepto> ++ 15:05:54 <mgoddard> #info Kolla feature freeze fast approaching: Sep 28 - Oct 02 15:06:03 <mgoddard> Weekly reminder 15:06:24 <mgoddard> #info Submit Virtual PTG topic proposals 15:06:31 <mgoddard> #link https://etherpad.opendev.org/p/kolla-wallaby-ptg 15:06:54 <mgoddard> No topics yet :) 15:07:03 * yoctozepto proposes to copy-paste unfinished victoria ones 15:07:03 <Fl1nt> I'll had some ^^ 15:07:07 <mgoddard> Great 15:07:35 <mgoddard> yoctozepto: into the Incomplete Victoria Priorities section please 15:07:48 <mgoddard> although we're not finished yet :) 15:07:52 <yoctozepto> mgoddard: sure, just closer to victoria release 15:07:55 <yoctozepto> indeed! 15:08:07 <yoctozepto> I need to get you out of my head 15:08:23 <mgoddard> cue Kylie 15:08:27 <mgoddard> #info TC & PTL elections coming up 15:08:36 <mgoddard> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017306.html 15:09:09 <mgoddard> Again, anyone is free to nominate themselves to join the TC, or become PTL 15:09:27 <mgoddard> Anyone thinking of running for kolla? (you don't have to say now) 15:09:56 <yoctozepto> mgoddard: I hope you are just teasing and you are going to run for it :-) 15:10:17 <mgoddard> yoctozepto: all good things must come to an end... 15:10:20 <hrw> I do not plan to 15:10:27 <yoctozepto> mgoddard: :O :-(((((( 15:10:37 <mgoddard> I do intend to run this time however 15:11:10 <yoctozepto> mgoddard: we can also use the ptl-less model if it makes you feel better 15:11:13 <hrw> mgoddard: done! you are PTL for Wcycle ;D 15:11:33 <wuchunyang> +1 15:11:54 <mgoddard> #info Ubuntu qemu bug 15:11:54 <yoctozepto> mgoddard: but you definitely have my vote whenever, wherever 15:12:16 <mgoddard> yoctozepto: are you deliberately throwing out pop lyrics? 15:12:32 <mgoddard> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017390.html 15:12:35 <yoctozepto> mgoddard: singer's soul 15:13:06 <mgoddard> ^ potentially nasty bug for ubuntu users on victoria/focal 15:13:18 <yoctozepto> indeed, hence why I added it to agenda 15:13:23 <yoctozepto> not sure we can do anything about it 15:13:38 <yoctozepto> based on further discussion it's not 100% reproducible 15:14:08 <yoctozepto> so it belongs to the most feared category of bufs 15:14:10 <yoctozepto> bugs* 15:14:28 <mgoddard> #topic Review action items from the last meeting 15:14:32 <mgoddard> yoctozepto to try bumping ansible to 2.9-2.10 15:14:38 <mgoddard> he tried, he failed 15:14:48 <mgoddard> not his fault 15:14:55 <yoctozepto> it works on Ubuntu 15:14:57 <yoctozepto> and Debian 15:15:02 <mgoddard> looks like we will be sticking with 2.9 15:15:03 <yoctozepto> just not CentOS 8 15:15:04 <openstackgerrit> Michal Nasiadka proposed openstack/kolla-ansible master: Use Docker healthchecks for core services https://review.opendev.org/676389 15:15:08 <mgoddard> py3.6 issue then 15:15:08 <yoctozepto> so I managed to bump 15:15:12 <yoctozepto> but it failed to run it 15:15:15 <yoctozepto> seems o 15:15:18 <yoctozepto> so* 15:15:30 <mgoddard> great 15:15:30 <yoctozepto> I did not get much help from the ansible community 15:15:46 <yoctozepto> and unable to reproduce minimal example 15:15:49 <mgoddard> why would they want ansible to work on RHEL distros? 15:15:50 <priteau> We won't consider bumping to 2.9 only? 15:16:03 <mgoddard> +1, that was the backup plan 15:16:10 <yoctozepto> it's the final agenda point 15:16:16 <yoctozepto> but we can discuss it now 15:16:23 <mgoddard> let's get it over with 15:16:34 <yoctozepto> anyone disagrees? 15:16:35 <mgoddard> bump minimum ansible to 2.9? 15:16:43 <yoctozepto> #vote yes 15:16:55 <mgoddard> #vote yes 15:17:03 <mnasiadka> #vote yes 15:17:19 <wuchunyang> #vote yes 15:17:21 <hrw> + 15:17:53 <yoctozepto> regarding maximum - we can bump it for deb to keep it tested but deliberately note it is not supported on centos 8 15:18:10 <mgoddard> sounds complicated 15:18:26 <yoctozepto> well, we kinda test it like this already anyway 15:18:27 <yoctozepto> ;p 15:18:34 <yoctozepto> lowest on centos 15:18:37 <yoctozepto> highest on ubuntu 15:18:43 <mgoddard> true. lucky you caught it 15:19:06 <yoctozepto> so it's a matter of documenting the failure fact 15:19:19 <yoctozepto> and *maybe* we get it fixed for centos anyway 15:19:35 <yoctozepto> testing 2.10 has the benefit of catching it bugs early 15:19:37 <yoctozepto> "benefit" 15:20:10 <yoctozepto> bump maximum ansible to 2.10? 15:20:13 <yoctozepto> #vote yes 15:20:21 <Fl1nt> Goes for the 2.9 on my own 15:20:29 <Fl1nt> sorry, I had a work call :( 15:21:02 <mgoddard> hmm, unsure 15:21:04 <Fl1nt> #vote no 15:21:41 <priteau> 2.10 was just released today if I understood correctly, and we just started testing with it 15:21:52 <Fl1nt> because I feel like it will be yet another complication to get to maintain two different syntax/requirements on different platform and it doesn't really add a lot of benefit yet. 15:21:57 <priteau> Sounds like something we should be bumping in Wallaby? 15:22:13 <priteau> (and hoping that CentOS supports it by then) 15:22:14 <Fl1nt> +1 15:22:34 <mnasiadka> well, can we bump testing on debian/ubuntu but not advertise 2.10 support in docs? 15:22:47 <Fl1nt> for testing purpose sure ^^ 15:22:48 <mnasiadka> once centos 8 issues are solved, we'll fix the support note? 15:23:07 <mgoddard> it's only a loose dependency, more of a constraint. We could potentially bump to 2.10 in a minor release? 15:23:11 <Fl1nt> but be careful of any downward compatibility issue tho 15:23:32 <Fl1nt> especially on keywords etc. 15:23:40 <mgoddard> of course, it would need to be tested 15:23:40 <mnasiadka> mgoddard: sure 15:23:45 <mnasiadka> but let's get rid of 2.8 ;) 15:24:00 <Fl1nt> I already had a bad surprise with those nasty changes with keywords ^^ 15:24:19 <mnasiadka> once 2.11 will be out - 2.8 will be eol anyways 15:24:23 <Fl1nt> for sure let's abbandon 2.8 :D 15:24:28 <mgoddard> seems like the general feeling is no 2.10 in V, reassess later 15:24:40 <mgoddard> anyone want to bump min to 2.9? 15:24:41 <yoctozepto> no promise of 2.10 compat 15:24:41 <openstackgerrit> Michal Nasiadka proposed openstack/kolla-ansible master: Use Docker healthchecks for core services https://review.opendev.org/676389 15:25:02 <yoctozepto> easy stuff, folks, get your hands dirty :-) 15:25:45 <priteau> I can give it a try, I am not too familiar with this part in K-A 15:26:18 <hrw> +1 for 2.9 15:26:22 <yoctozepto> priteau: perfect, the ideal candidate 15:26:24 <mnasiadka> well, yoctozepto can guide you in how many places that's done :D 15:26:32 <mgoddard> priteau: we should do the same in kayobe 15:26:35 <yoctozepto> or my change really 15:26:41 <mgoddard> #action priteau to bump minimum Ansible to 2.9 15:26:46 <priteau> mgoddard: of course, I know where to change it in Kayobe :-) 15:27:09 <yoctozepto> https://review.opendev.org/#/c/752420/2 15:27:09 <patchbot> patch 752420 - kolla-ansible - [DNM] Test Ansible 2.10.0rc1 - 5 patch sets 15:27:15 <mgoddard> #topic CI status 15:28:56 <mgoddard> how are things looking? 15:29:07 <mgoddard> whiteboard looks ok 15:29:19 <mgoddard> ubuntu binary still unhappy 15:29:20 <priteau> Kayobe is RED :( 15:29:56 <mgoddard> priteau: any idea what the problem is? I saw you mentioned ansible 2.10 15:30:00 <priteau> I'll recheck just to be sure, but I am able to reproduce the issue reliably & locally 15:30:16 <priteau> Well I thought it was related, but it fails in my tox venv which uses 2.9.13 15:30:36 <yoctozepto> it seems k&k-a stabilised a bit 15:30:49 <priteau> so far I've only found that if we pip3 uninstall ansible, remove the site-package ansible folder and reinstall it, it works 15:30:50 <yoctozepto> sad ubuntu binary still fails but it's not really our fault... 15:31:07 <yoctozepto> priteau: :O 15:31:15 <yoctozepto> MAGIC 15:31:37 <priteau> and except for the ansible 2.10 release, there were no changes to the other packages in the last days 15:31:42 <priteau> CI was still green yesterday 15:31:59 <mgoddard> priteau: are you able to continue looking at it? 15:32:09 <priteau> yeah I'll try some more 15:32:35 <mgoddard> #action priteau to investigate kayobe kayobe-tox-molecule failures 15:32:45 <mgoddard> #topic Victoria release planning 15:32:59 <mgoddard> Victoria approaches 15:33:13 <mnasiadka> yoctozepto: I'll look into ubuntu-binary case, I think it can be fixed 15:33:28 <mgoddard> #action mnasiadka to investigate ubuntu binary CI failures 15:33:31 <yoctozepto> mnasiadka: superb 15:33:47 <yoctozepto> will it be a victory for kolla 15:34:04 <mgoddard> no doubt 15:34:41 <mgoddard> hrw: what's the current status of infra images? Do you think you'll be able to sort out the CI in time? 15:35:45 <hrw> hm. good question 15:36:18 <hrw> will look at making change to have some sort of tree 15:36:34 <hrw> https://marcin.juszkiewicz.com.pl/2020/09/23/8-years-of-my-work-on-aarch64/ 15:36:37 <hrw> ops 15:36:42 <mgoddard> lol 15:36:51 <mgoddard> this is not an advertising platform :p 15:37:03 <mnasiadka> no? 15:37:24 <hrw> mgoddard: if I would like to advertise I would add #link ;D 15:37:34 <mgoddard> I think we need to decide now whether to pursue for V or not 15:37:44 <mgoddard> we have basically 1 week left 15:37:55 <yoctozepto> +/- FFE period 15:38:10 <hrw> move to W I would say 15:38:17 <yoctozepto> then W it is 15:38:17 <mgoddard> it's a big enough change that FFE seems unfriendly 15:38:27 <mgoddard> e.g. kayobe will need to react to it 15:38:30 <hrw> get kolla_infra_install_type for k-a into V 15:38:33 <mgoddard> ok 15:38:56 <yoctozepto> we can have an experimental job for infra as well 15:39:02 <yoctozepto> to just test that it exists 15:39:07 <mgoddard> the work can continue, but we can drop the review priority 15:39:08 <hrw> so if you want to be brave then you can build infra+binary/source and install it using k-a in V 15:39:10 <yoctozepto> much easier than the CI warfare 15:39:42 <openstackgerrit> Pierre Riteau proposed openstack/kolla-ansible master: Bump minimum Ansible version to 2.9 https://review.opendev.org/753776 15:39:45 <yoctozepto> I felt sad we didn't deliver any support for letsencrypt in this cycle 15:39:52 <hrw> yoctozepto: https://review.opendev.org/753260 + https://review.opendev.org/706982 and you have base for such CI job 15:39:52 <patchbot> patch 753260 - kolla - [WIP] add centos8 images for source+infra CI job - 1 patch set 15:39:54 <patchbot> patch 706982 - kolla-ansible - Introduce 'kolla_infra_install_type' variable - 13 patch sets 15:40:02 <yoctozepto> so I started upstreaming bite-sized portions of semi-automatic approach now 15:41:22 <mgoddard> yoctozepto: will it be compatible with the full approach? 15:41:42 <yoctozepto> mgoddard: yes, that's the goal 15:41:46 <mgoddard> ok 15:41:49 <yoctozepto> it just requires user intervention and docs 15:41:57 <mgoddard> better be quick then :) 15:41:58 <yoctozepto> I also left a comment where you asked about cycle 15:42:05 <yoctozepto> sure, sire 15:43:05 <mgoddard> priteau: any progress on pull/retag/push? 15:43:29 <priteau> yes. I have a very rough proof of concept, I need a bit more time to clean it up 15:43:40 <mgoddard> nice 15:43:46 <priteau> Honestly not sure we can get it ready for V now 15:45:30 <mgoddard> ok 15:45:58 <mgoddard> I would very much like to get wuchunyang's work on octavia merged 15:46:07 <mgoddard> #link https://review.opendev.org/#/q/topic:bp/implement-automatic-deploy-of-octavia+(status:open+OR+status:merged) 15:46:51 <mgoddard> reviews would be appreciated 15:46:55 <wuchunyang> need some review now.. 15:47:40 <mgoddard> I will do another pass through 15:47:45 <johnsom> I think I have reviewed this stack at least once. I will try to give another pass later today. It was in good shape last time I looked. 15:48:08 <openstackgerrit> Pierre Riteau proposed openstack/kolla-ansible master: Bump minimum Ansible version to 2.9 https://review.opendev.org/753776 15:48:20 <mgoddard> thanks johnsom 15:48:20 <wuchunyang> mgoddard johnsom thanks 15:48:22 <mnasiadka> yoctozepto: seems ubuntu-binary will get solved after all - https://bugs.launchpad.net/cloud-archive/+bug/1893935 15:48:23 <openstack> Launchpad bug 1893935 in Ubuntu Cloud Archive "Missing dependency on vitrageclient and zunclient in master" [Undecided,Confirmed] 15:48:40 <mgoddard> good investigative work mnasiadka 15:48:57 <yoctozepto> mnasiadka: because it hit charms 15:49:00 <mgoddard> would also be nice to land keystone federation https://review.opendev.org/#/c/695432/34 15:49:00 <patchbot> patch 695432 - kolla-ansible - Add support to OpenID Connect Authentication flow - 34 patch sets 15:49:08 <mnasiadka> yoctozepto: yeah :) 15:49:12 <mgoddard> although that needs an update 15:49:29 <mgoddard> health checks have had some activity 15:49:36 <mgoddard> https://review.opendev.org/743958 15:49:36 <patchbot> patch 743958 - kolla-ansible - Add healthchecks option to kolla_docker - 14 patch sets 15:49:38 <mgoddard> https://review.opendev.org/676389 15:49:38 <patchbot> patch 676389 - kolla-ansible - Use Docker healthchecks for core services - 83 patch sets 15:49:47 <mgoddard> more reviews welcome 15:50:31 <mgoddard> https://review.opendev.org/#/c/728448/ has a +2... 15:50:32 <patchbot> patch 728448 - kolla-ansible - Support TLS encryption of RabbitMQ client-server t... - 12 patch sets 15:51:02 <mgoddard> priteau: do you plan to work on kayobe multiple envs before V? 15:53:00 <priteau> I would like to but it's getting quite late now 15:53:33 <priteau> (in the cycle) 15:54:18 <mgoddard> indeed 15:54:49 <mgoddard> I could see if I can spare some cycles, if that helps? 15:55:26 <priteau> It would help for sure 15:55:58 <mgoddard> let's see how the next week goes 15:56:30 <mgoddard> any other features for V? 15:56:48 <wuchunyang> designate ? 15:57:10 <wuchunyang> the designate default_pool_id bug has fixed. 15:57:12 <yoctozepto> that did not happen, unfortunately 15:57:49 <wuchunyang> i can work on it .. tomorrow. 15:58:15 <mgoddard> #topic open discussion 15:58:50 <yoctozepto> wuchunyang: that would be awesome 15:58:58 <mgoddard> kplant mentioned ozzzo is a new colleague, who will be spending some time upstream 15:59:02 <mgoddard> welcome ozzzo :) 15:59:24 <yoctozepto> welcome ozzzo indeed 15:59:27 <kplant> :-) 15:59:35 <kplant> welcome ozzzo 15:59:36 <yoctozepto> time to wrap up (for me at least) 15:59:37 <mgoddard> feel free to drop questions into IRC as you climb the learning curve 15:59:40 * yoctozepto afk 15:59:56 <mgoddard> indeed, time for us all to wrap up 15:59:58 <mgoddard> thanks all 16:00:12 <mgoddard> #endmeeting