| opendevreview | melanie witt proposed openstack/tempest master: WIP Add configuration for compute unified limits feature https://review.opendev.org/c/openstack/tempest/+/790186 | 01:10 |
|---|---|---|
| opendevreview | melanie witt proposed openstack/tempest master: WIP tests for nova unified quotas https://review.opendev.org/c/openstack/tempest/+/804311 | 01:10 |
| opendevreview | yatin proposed openstack/devstack master: [WIP] Support CentOS Stream 9 https://review.opendev.org/c/openstack/devstack/+/800903 | 05:11 |
| frickler | yoctozepto: gmann: ianw's pip cleanup stack starting at https://review.opendev.org/c/openstack/devstack/+/802642/8 seems ready to me (assuming the weird failure goes away with my recheck) | 05:30 |
| *** akekane_ is now known as abhishekk | 06:03 | |
| *** rpittau|afk is now known as rpittau | 06:22 | |
| *** jpena|off is now known as jpena | 07:33 | |
| opendevreview | Slawek Kaplonski proposed openstack/devstack master: Deploy Neutron with enforced new RBAC rules https://review.opendev.org/c/openstack/devstack/+/797450 | 07:40 |
| *** elodilles_pto is now known as elodilles | 08:12 | |
| *** akekane_ is now known as abhishekk | 09:08 | |
| opendevreview | Martin Kopec proposed openstack/tempest master: Updating NON_ACTIVE_LIST https://review.opendev.org/c/openstack/tempest/+/796698 | 09:56 |
| opendevreview | Martin Kopec proposed openstack/tempest master: WIP: updating NON_ACTIVE_LIST https://review.opendev.org/c/openstack/tempest/+/784899 | 09:56 |
| opendevreview | Martin Kopec proposed openstack/tempest master: WIP: updating NON_ACTIVE_LIST https://review.opendev.org/c/openstack/tempest/+/784899 | 09:59 |
| *** dviroel|ruck|out is now known as dviroel|ruck | 11:24 | |
| *** jpena is now known as jpena|lunch | 11:33 | |
| opendevreview | Martin Kopec proposed openstack/tempest master: WIP: updating NON_ACTIVE_LIST https://review.opendev.org/c/openstack/tempest/+/784899 | 11:43 |
| *** jpena|lunch is now known as jpena | 12:24 | |
| *** jpena is now known as jpena|off | 12:33 | |
| opendevreview | Martin Kopec proposed openstack/tempest master: WIP: updating NON_ACTIVE_LIST https://review.opendev.org/c/openstack/tempest/+/784899 | 14:20 |
| opendevreview | Merged openstack/devstack master: Revert "Workaround for new pip 20.3 behavior" https://review.opendev.org/c/openstack/devstack/+/802642 | 14:27 |
| opendevreview | Merged openstack/devstack master: install_pip: don't fail when not installed https://review.opendev.org/c/openstack/devstack/+/804037 | 14:29 |
| opendevreview | Merged openstack/devstack master: install_pip: Use packaged pip on Fedora https://review.opendev.org/c/openstack/devstack/+/804038 | 14:29 |
| opendevreview | Merged openstack/devstack master: Add Fedora 34 support https://review.opendev.org/c/openstack/devstack/+/802989 | 14:30 |
| tosky | can I also request another vote on this wallaby backport of a devstack feature (s3 support for cinder-backup) needed to test the service feature itself? https://review.opendev.org/c/openstack/devstack/+/785449 | 14:32 |
| gmann | tosky: checking | 14:42 |
| *** jpena|off is now known as jpena | 14:43 | |
| gmann | +A | 14:44 |
| gmann | kopecmartin: yoctozepto frickler ianw making few of the cleanup and job voting in devstack gate, please review what you think? https://review.opendev.org/q/topic:%22devstack-gate-jobs%22+(status:open%20OR%20status:merged) | 14:47 |
| opendevreview | Merged openstack/devstack master: Add devstack-enforce-scope job to enable the rbac scope checks https://review.opendev.org/c/openstack/devstack/+/804263 | 14:52 |
| gmann | lbragstad: ^^ now we have enforce_scope job to detect the what all things needed when we add/enable enforce_scope for any services. | 14:53 |
| frickler | gmann: oh, that's "gate jobs for devstack", not "devstack-gate jobs" ;) I'm not sure about adding that much additional load to the gate, though. we still are facing enough instability with the existing jobs, too. | 14:54 |
| lbragstad | gmann nice - should we add keystone to that list? | 14:54 |
| gmann | lbragstad: +1, and that will help to move the seeting to devstack side | 14:54 |
| lbragstad | gmann yeah - we need to start flushing out all the places in devstack where it tries to do system API stuff with project-scoped tokens | 14:55 |
| lbragstad | (all the failures we hit the other day) | 14:55 |
| lbragstad | and start fixing those | 14:55 |
| gmann | frickler: :), i rethought when i added topic. on extra load, actually it is just making n-v jobs to voting which are stable enough on project side gate | 14:55 |
| gmann | lbragstad: +1. I am going to start with that nova as first candidate but i think keystone is needed before that? | 14:56 |
| yoctozepto | gmann: I am with frickler on this; especially the tempest-slow does not sound good to go directly to gating | 14:56 |
| gmann | lbragstad: but let's see how the flow goes | 14:56 |
| lbragstad | gmann yeah - keystone's probably going to fail first since it's the first service devstack setups\ | 14:56 |
| lbragstad | sets up* | 14:56 |
| gmann | yoctozepto: tempest-slow is voting on tempest and in many project gate | 14:56 |
| gmann | lbragstad: ack. | 14:56 |
| gmann | yoctozepto: frickler i mean none of them are non voting on project side. so if they fail project gate fail first so they fix on priority | 14:57 |
| yoctozepto | hmm | 14:58 |
| gmann | because many times, we have seen devstack change failing project gate. like ironic for glance quota things | 14:58 |
| gmann | and always forget to check the n-v jobs result | 14:58 |
| yoctozepto | well, it opens a possibility that we have ugly random issues making it hard to merge fixes | 14:59 |
| yoctozepto | I agree we forget | 14:59 |
| yoctozepto | let's merge at least some of these | 14:59 |
| frickler | is tempest-multinode-full-py3 a subset of tempest-slow-p3? so we could only use the latter instead of running both at least? | 15:01 |
| yoctozepto | frickler: I think they are actually complementary | 15:04 |
| yoctozepto | (last time I tried to understand it) | 15:04 |
| gmann | frickler: yoctozepto different, slow run only slow tests which does not need to be all multinode . | 15:10 |
| gmann | frickler: yoctozepto but let me check the overlap between those. | 15:11 |
| frickler | I think I'd also be fine with enabling all those gate jobs as proposed, watching gate stability for while, and let's say reevaluate at the PTG? | 15:15 |
| yoctozepto | works for me; under the condition we agree to disable them if they block other vital patches | 15:18 |
| frickler | yoctozepto: I'd assume we'd agree on that anyway, anytime | 15:21 |
| tosky | elodilles: I guess this change can be abandoned now? https://review.opendev.org/c/openstack/devstack/+/742953 | 15:22 |
| frickler | .oO(do we agree on the definition of "vital", though? ;) | 15:22 |
| yoctozepto | frickler: well, we try not to, to maintain good posture and QA higiene | 15:22 |
| yoctozepto | frickler: hahaha, good point | 15:22 |
| frickler | I think the difficulty most of the time lies in judging whether the error we see is actually related to the patch in question or not | 15:24 |
| frickler | with gmann's patches, we will at least have to look at things in more details. that's likely better than just not noticing an error triggered in a n-v job | 15:25 |
| yoctozepto | frickler: yeah, I hate it that we have this instability in gates that we can't just run everything voting | 15:29 |
| *** rpittau is now known as rpittau|afk | 15:31 | |
| elodilles | tosky: i think so. however stable/ocata still exists for devstack. anyway, if someone would want to fix it it's not just cinder that should be pulled in via ocata-eol tag | 15:32 |
| ykarel | gmann, when you get chance please also check https://review.opendev.org/c/openstack/devstack/+/803706 | 15:33 |
| *** jpena is now known as jpena|off | 15:34 | |
| tosky | elodilles: uhm, shouldn't be there an automatic fallback then if the proper branch is not found? | 15:34 |
| elodilles | tosky: i'll push a patch to EOL devstack's ocata-eol as it is broken for quite some time now | 15:35 |
| elodilles | tosky: hmmm, maybe that would be the best | 15:35 |
| elodilles | tosky: it doesn't select the $series-eol if stable/$series doesn't exist now, so I guess that would be a nice new feature to add | 15:36 |
| yoctozepto | gmann, frickler, kopecmartin: I suggest we discuss adding the extra voting jobs during the next QA meeting; I am tentatively OK to merge them | 15:45 |
| opendevreview | yatin proposed openstack/devstack master: Set swap size to 4G for c8 jobs https://review.opendev.org/c/openstack/devstack/+/803706 | 16:01 |
| gmann | frickler: yoctozepto tempest-multinode-py3 run all the test excluding slow test. basically slow tests are only run in tempest-slow-py3 | 16:13 |
| gmann | but we can make tempest-multinode-py3 not to run all tests(excluding slow) instead run only multinode test + slow that way we can remove tempest-slow-py3 itself | 16:14 |
| gmann | let me try that. and I will add it in next QA meeting | 16:14 |
| yoctozepto | thanks | 16:15 |
| gmann | yoctozepto: replied for irrelevant-file comment https://review.opendev.org/c/openstack/devstack/+/804282/1 | 16:16 |
| gmann | those are needed for uulv3 native jobs too, not related to legacy things | 16:16 |
| gmann | ykarel: ack. I might be able to check next week though | 16:17 |
| ykarel | gmann, ack | 16:17 |
| *** ykarel is now known as ykarel|away | 16:31 | |
| opendevreview | Merged openstack/devstack stable/wallaby: Configure Cinder backup driver https://review.opendev.org/c/openstack/devstack/+/785449 | 17:18 |
| opendevreview | melanie witt proposed openstack/tempest master: WIP tests for nova unified quotas https://review.opendev.org/c/openstack/tempest/+/804311 | 17:43 |
| melwitt | anyone know if this is a known failure in the openstack-tox-py36 job? https://zuul.opendev.org/t/openstack/build/d6cebb39e7d24972b73fb46ed88e2e3e/log/job-output.txt#3314 ModuleNotFoundError: No module named 'check-uuid-yesgjl2k6p0' | 18:09 |
| yoctozepto | I have not seen it yet | 18:11 |
| gmann | melwitt: i have seen it, I think I filed bug too or workaround let me check | 18:12 |
| melwitt | ok cool | 18:13 |
| gmann | https://bugs.launchpad.net/tempest/+bug/1918316 | 18:14 |
| gmann | but there was not known issue and i could not reproduce it so skipped test | 18:14 |
| gmann | and then unskiped when we did not see this issue | 18:15 |
| gmann | I will reopen that bug | 18:15 |
| melwitt | ah ok, so it's a intermittent fail. they all seem to be 😝 | 18:15 |
| gmann | yeah, difficult to find root cause too, anyways you can recheck that and I have re-open that bug | 18:17 |
| melwitt | ack thanks | 18:17 |
| opendevreview | Merged openstack/devstack master: Support optimized cinder backend for glance https://review.opendev.org/c/openstack/devstack/+/757934 | 18:54 |
| eliadcohen | @Team, is anyone unable to reach: http://opendev.org/openstack/requirements/raw/branch/master/upper-constraints.txt ? | 18:56 |
| melwitt | works for me ^ | 19:05 |
| opendevreview | Merged openstack/devstack master: Simplify pip install https://review.opendev.org/c/openstack/devstack/+/804197 | 19:54 |
| *** dviroel|ruck is now known as dviroel|ruck|out | 21:51 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!