15:00:15 <kopecmartin> #startmeeting qa 15:00:15 <opendevmeet> Meeting started Tue Mar 29 15:00:15 2022 UTC and is due to finish in 60 minutes. The chair is kopecmartin. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:15 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:15 <opendevmeet> The meeting name has been set to 'qa' 15:00:21 <kopecmartin> #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Weekly_QA_Team_meeting 15:00:22 <kopecmartin> agenda ^^ 15:00:38 <gmann> o/ 15:01:21 <kopecmartin> o/ 15:01:51 <frickler> \o 15:02:14 <kopecmartin> o/ 15:02:17 <kopecmartin> let's start 15:02:18 <kopecmartin> #topic Announcement and Action Item (Optional) 15:02:31 <kopecmartin> we should be done with releases 15:02:44 <kopecmartin> couple of patches are hanging there though 15:02:54 <gmann> few grenade setup are pending, which is in gate 15:02:58 <gmann> but we can discuss that later 15:03:05 <gmann> yeah 15:03:29 <kopecmartin> ok 15:03:44 <kopecmartin> no other updates from my side 15:03:56 <kopecmartin> #topic Yoga Priority Items progress 15:04:01 <kopecmartin> any updates on this front? 15:04:24 <kopecmartin> i'm going through and am checking the status of the efforts 15:05:04 <gmann> nothing much from me. just reviewed the rbac patch on devstack side. 15:05:12 <kopecmartin> there is a long list of patches for the manager cleanup i'm failing to get to 15:05:38 <gmann> yeah it is all in plugins sides, I did not see revision after my first review 15:05:47 <gmann> but keeping eyes on those. 15:06:07 <gmann> may be it will be good to use the single topic for those, replace-tempest-sm 15:06:44 <kopecmartin> the patches should be under one topic 15:06:46 <kopecmartin> #link https://review.opendev.org/q/topic:tempest-scenario-manager-cleanup 15:07:02 <gmann> cool 15:07:26 <gmann> I think there are many new one, I will review those 15:07:44 <kopecmartin> cool thanks 15:08:02 <kopecmartin> #topic OpenStack Events Updates and Planning 15:08:11 <kopecmartin> PTG is gonna be next week 15:08:25 <kopecmartin> which reminds, we're cancelling the office hour for the next week, right? 15:08:34 <gmann> yeah 15:08:55 <kopecmartin> ack, i'll send an email 15:09:02 <kopecmartin> #link https://etherpad.opendev.org/p/qa-zed-ptg 15:09:05 <kopecmartin> topics ^^ 15:09:09 <kopecmartin> i'm about to add 2 more 15:09:21 <kopecmartin> regarding the current status and next plans for s-rbac and fips 15:10:03 <kopecmartin> i'll finalize the order of the topics by the end of the week and will send another email to the ML announcing it 15:10:14 <gmann> kopecmartin: we are also going to discuss both in TC PTG and on FIPs to be a goal or not 15:10:45 <gmann> for rbac, nova is ready with new rbac and that i will target for zed cycle to migrate tempest to new policy 15:10:50 <kopecmartin> good, right during the monday's session? 15:11:00 <gmann> kopecmartin: no, in thursday or friday 15:11:05 <kopecmartin> okey 15:11:48 <gmann> but we can discuss/plan the QA part as per the current progress on rbac #link http://lists.openstack.org/pipermail/openstack-discuss/2022-March/027837.html 15:12:05 <gmann> which is nova and keystone for now 15:12:21 <gmann> neutron in progress so that also a candidate for QA work 15:13:05 <kopecmartin> sounds good 15:13:47 <kopecmartin> #topic Gate Status Checks 15:13:55 <kopecmartin> #link https://review.opendev.org/q/label:Review-Priority%253D%252B2+status:open+(project:openstack/tempest+OR+project:openstack/patrole+OR+project:openstack/devstack+OR+project:openstack/grenade+OR+project:openstack/hacking) 15:14:05 <kopecmartin> no high priority patches 15:14:13 <frickler> I noticed hacking doesn't have RP 15:14:21 <slaweq> Hi. If you will be talking about neutron RBAC QA I will be interested to join 🙂 15:14:23 <frickler> otherwise I'd tagged my py310 fix 15:14:38 <gmann> slaweq: great 15:14:47 <frickler> #link https://review.opendev.org/c/openstack/hacking/+/835413 15:15:09 <gmann> frickler: I think there are very less traffic there so we can consider all review there as quick to do 15:15:44 <frickler> yeah, makes sense, but kopecmartin can drop it from the url above then as it will never match 15:15:46 <gmann> +2, lgtm 15:16:08 <gmann> ah, right. good point 15:16:33 <kopecmartin> slaweq: good, thanks, i'll keep that in mind 15:16:39 <kopecmartin> frickler: hacking patch merged 15:16:44 <frickler> thx 15:16:45 <kopecmartin> sure, i'll drop it 15:17:17 <kopecmartin> #topic Periodic jobs Status Checks 15:17:29 <kopecmartin> periodic stable 15:17:32 <kopecmartin> #link https://zuul.openstack.org/builds?job_name=tempest-full-xena&job_name=tempest-full-wallaby-py3&job_name=tempest-full-victoria-py3&job_name=tempest-full-ussuri-py3&pipeline=periodic-stable 15:17:40 <kopecmartin> periodic master 15:17:41 <kopecmartin> #link https://zuul.openstack.org/builds?project=openstack%2Ftempest&project=openstack%2Fdevstack&pipeline=periodic 15:18:16 <kopecmartin> seems all good 15:18:20 <frickler> pretty green 15:18:47 <kopecmartin> oh, that reminds me, is it time to add tempest yoga jobs? 15:18:50 <frickler> remember for next week to add periodic-weekly 15:19:26 <gmann> we did right no? 15:19:36 <frickler> ah, no, we didn't move fedora there 15:19:43 <frickler> then disregard 15:20:02 <gmann> #link https://review.opendev.org/c/openstack/tempest/+/835229/1/zuul.d/project.yaml 15:20:20 <gmann> kopecmartin: you mean in url? 15:20:47 <kopecmartin> gmann: oh, thanks, i forgot i pushed that review .. yeah, in that case i just need to update the url in the ageda 15:20:50 <kopecmartin> agenda 15:20:54 <gmann> kopecmartin: :) 15:20:58 <kopecmartin> frickler: periodic-weekly? 15:21:21 <gmann> we do not have peridioc-weekly jobs I think 15:21:31 <frickler> I thought we wanted to move the fedora jobs there, but then we didn't 15:21:42 <gmann> ah, yeah. 15:21:42 <frickler> or was that normal periodic? 15:21:52 <frickler> anyway, not happening for now 15:21:52 <gmann> it was normal periodic but we reverted that 15:22:54 <kopecmartin> ah, i see, okey then 15:23:08 <kopecmartin> #topic Distros check 15:23:12 <kopecmartin> our new topic :) 15:23:18 <kopecmartin> centos 8/9 15:23:21 <kopecmartin> #link https://zuul.openstack.org/builds?job_name=tempest-full-centos-9-stream&job_name=tempest-full-py3-centos-8-stream&job_name=devstack-platform-centos-8-stream&job_name=devstack-platform-centos-9-stream&skip=0 15:23:28 <kopecmartin> fedora 15:23:29 <kopecmartin> #link https://zuul.openstack.org/builds?job_name=devstack-platform-fedora-latest&skip=0 15:23:38 <kopecmartin> openEuler 15:23:40 <kopecmartin> #link https://zuul.openstack.org/builds?job_name=devstack-platform-openEuler-20.03-SP2+&skip=0 15:23:46 <kopecmartin> and debian 15:23:47 <kopecmartin> #link https://zuul.openstack.org/builds?job_name=devstack-platform-debian-bullseye&skip=0 15:23:58 <gmann> we can drop centos8 stream now? in zed TC defined testing runtime is centos9 stream only 15:24:11 <kopecmartin> gmann: ack, will do 15:24:44 <gmann> but can we do? i mean many jobs are still centos8-stream so we cannot drop it from devstackat least at this stage ? 15:24:45 <kopecmartin> centos 9 looks quite green \o/ 15:25:21 <gmann> frickler: usually we do keep support for one more cycle than they are not defined/targetted for testing? 15:25:29 <kopecmartin> oh, i meant we can delete that from the link in the agenda, for the rest i really need to check , yeah, it may be used on many places 15:25:58 <gmann> ah I mean testing and devstack support for centos8-stream not just list to monitor 15:26:35 <kopecmartin> yup, i get it now 15:26:48 <frickler> I think we would get pushback from various projects when dropping c8s already 15:27:05 <frickler> and as long as we need to keep support, we should also keep testing 15:27:06 <kopecmartin> need to do a research for that, we can discuss on ptg 15:27:19 <kopecmartin> frickler: correct 15:27:20 <gmann> yeah, let's wait. may be we can do in AA 15:27:36 <gmann> and see how projects are moving to centos9-stream in zed 15:27:49 <gmann> kopecmartin: +1 on check in PTG 15:28:05 <frickler> it sure would be good to incentivise (sp?) projects to migrate 15:28:49 <frickler> but even adding a 10 foot deprecation warning would mostly go unnoticed I fear 15:29:34 <kopecmartin> :D likely 15:29:42 <gmann> :) yeah. that does not help actually 15:30:15 <gmann> frickler: may be drop, break and revert. that is good warning signal :) 15:30:33 <opendevreview> Merged openstack/hacking master: Fix test errors with python3.10 https://review.opendev.org/c/openstack/hacking/+/835413 15:30:41 <frickler> gmann: ack, I was thinking the same but hesitated to propose it ;) 15:30:54 <gmann> and as it is just start of cycle, we can try 15:30:58 <kopecmartin> :D 15:31:33 <gmann> kopecmartin: what you say? drop, send ML, if we get pushback then revert saying move your jobs soon 15:32:19 <kopecmartin> in case we will want to drop that job, yeah, it sounds like an effective way how to do it 15:32:27 <kopecmartin> if we drop 15:32:44 <kopecmartin> let's see how it goes on ptg 15:32:48 <gmann> yeah distro drop+testing goes together 15:33:51 <kopecmartin> moving on 15:33:52 <opendevreview> Brian Haley proposed openstack/devstack master: Add support for IPv6 tunnel endpoints https://review.opendev.org/c/openstack/devstack/+/710519 15:33:58 <kopecmartin> Changes with Review-Priority == +1 15:34:04 <kopecmartin> #link https://review.opendev.org/q/label:Review-Priority%253D%252B1+status:open+(project:openstack/tempest+OR+project:openstack/patrole+OR+project:openstack/devstack+OR+project:openstack/grenade+OR+project:openstack/hacking) 15:34:28 <opendevreview> Brian Haley proposed openstack/tempest master: Add an IPv6 tunnel endpoint job definition https://review.opendev.org/c/openstack/tempest/+/828431 15:34:34 <kopecmartin> #link https://review.opendev.org/c/openstack/devstack/+/835227 15:34:40 <kopecmartin> #link https://review.opendev.org/c/openstack/devstack/+/835228/2 15:34:41 <frickler> someone proposed "fixing" setup.py for devstack 15:34:44 <haleyb> oops, sorry for interrupting 15:34:54 <frickler> so I though why not drop it instead? 15:35:22 <frickler> haleyb: np, submitting patches shouldn't stop just because we are meeting 15:36:32 <kopecmartin> frickler: sounds good to me, there isn't any info which wouldn't be somewhere else in the doc too, such as ML or the home page 15:36:44 <gmann> I think it make sense to drop. it create more confusion than thinking of in doc building 15:36:45 <kopecmartin> so there isn't any reason why to keep the files then i guess 15:36:56 <gmann> yeah 15:37:06 <frickler> 835228 needs a rebase I think 15:37:38 <gmann> +2 no setup one 15:38:12 <opendevreview> Martin Kopec proposed openstack/devstack stable/yoga: Stop installing Tempest at system wide for stable branch https://review.opendev.org/c/openstack/devstack/+/835228 15:38:12 <gmann> yeah, I think extensions are good now. it can be rebased or just rebase on master 15:39:05 <kopecmartin> #topic Open Discussion 15:39:10 <kopecmartin> anything for the open discussion? 15:40:14 <frickler> https://review.opendev.org/c/openstack/devstack/+/832873 may be interesting for you to look at 15:41:21 <frickler> after some discussion, I'd go the simple way of simply adding everything we need to PATH hardcoded, because it doesn't cost much 15:43:47 <haleyb> since i wound-up here could i ask a question about the two IPv6 patches i just rebased? one devstack, one tempest 15:43:56 <haleyb> i hadn't added to agenda 15:44:33 <kopecmartin> frickler: it makes sense, i can't think of reasons why not to do so 15:44:47 <kopecmartin> haleyb: sure, can you send the links? 15:44:55 <kopecmartin> let's have a ook 15:44:56 <kopecmartin> look 15:45:02 <haleyb> https://review.opendev.org/c/openstack/devstack/+/710519 and https://review.opendev.org/c/openstack/tempest/+/828431 15:45:43 <haleyb> my question was mostly about the ipv6 jobs - i added a new one "ipv6-all" so as to not break the 'ipv6-only' users 15:46:07 <haleyb> but i didn't know if having yet another one was good, or should it be folded into the other 15:46:40 <haleyb> the devstack change has been on my list for a while, and i think gets things to almost 100% ipv6 15:47:21 <haleyb> i've so far only tested with a neutron patch depending on them 15:48:22 <frickler> not answering your question, but how about making a job that also only uses v6 for tenant networks? 15:48:55 <frickler> I just happened to get that to work in kolla https://review.opendev.org/c/openstack/kolla-ansible/+/712768 15:49:38 <haleyb> frickler: that would be the next logical step, since those would support IPv6 endpoints, the only thing left would be the multi-node job definitions, which i haven't had time to get working yet 15:49:38 <frickler> but I think we should keep those two jobs at least for a while. 15:50:09 <frickler> we could deprecate and later remove the ipv6-only job 15:52:16 <haleyb> frickler: yes, that would be an option once the new one is working everywhere. i haven't actually looked to see what repos inherit from those jobs to then put test patches out 15:52:53 <kopecmartin> good, so it sounds like it's a good idea to have it as a separate job for now 15:53:07 <kopecmartin> i'll do the review a bit later 15:53:10 <gmann> I think most of the project have the IPv6-only jobs 15:53:44 <gmann> can we just modify the original job itself with TUNNEL_IP_VERSION check ? 15:53:45 <haleyb> yes, searching in codesearch for devstack-tempest-ipv6 has a lot 15:54:03 <gmann> yeah, it was a community wide goal and I added in most of the projects 15:54:27 <gmann> and projects can enable it in their ipv6-only jobs if they wat 15:54:54 <gmann> or we can do enable it by default in ipv6-only and if anyone is breaking then disable/fix there ? 15:55:21 <frickler> +1 for the latter, seems reasonable 15:55:37 <haleyb> gmann: i guess that was my question, and i didn't want to break anything... 15:55:49 <kopecmartin> gmann: yup, that's a good idea (the last one), +1 15:56:10 <frickler> breaking things but providing a fix along with it seems fine 15:56:20 <kopecmartin> well, sometimes breaking is good 15:56:22 <kopecmartin> frickler: yeah 15:57:09 <kopecmartin> anything else? 15:57:12 <frickler> thinking about it, likely few projects would even care how tunneling is set up 15:57:18 <gmann> yeah 15:57:23 <frickler> maybe octavia 15:57:24 <haleyb> gmann: so just so i understand correctly, modify the devstack change to enable the tunnel version as well, then the tempest change is almost not needed 15:57:29 <gmann> in start of cycle we can try these things 15:58:07 <gmann> haleyb: yeah, and it get enable in every *ipv6-only* jobs on projects side automatically. 15:58:22 <gmann> and after that we can send it on ML if anyone is surprised with that if broken/ 15:59:02 <haleyb> gmann: they could be. as i said the only thing i didn't get working yet is multi-node w/ipv6, but there aren't jobs for that yet anyways 15:59:26 <kopecmartin> let's skip bug triage today 15:59:31 <gmann> haleyb: yeah there are very few I think 15:59:40 <haleyb> thanks for the feedback everyone 15:59:42 <kopecmartin> gmann: when you have a moment, can you check these 2 (relation chain): https://review.opendev.org/c/openstack/tempest/+/835271 15:59:51 <kopecmartin> haleyb: yw 16:00:00 <gmann> haleyb: but you can disable in multinode job until you make it working 16:00:17 <gmann> kopecmartin: ack, will check after nova meeting 16:00:29 <kopecmartin> thanks 16:00:44 <kopecmartin> that's it, time for the office hour is up 16:00:51 <kopecmartin> see you everyone on PTG 16:00:59 <frickler> thx all 16:01:05 <kopecmartin> #endmeeting