Wednesday, 2026-04-15

opendevreviewMaxim Sava proposed openstack/tempest master: Add scenario tests for Glance image multistore functionality  https://review.opendev.org/c/openstack/tempest/+/95836906:34
*** dmellado47174 is now known as dmellado471707:31
opendevreviewMaor Blaustein proposed openstack/tempest master: Migrate QoS with placement tests to NTP  https://review.opendev.org/c/openstack/tempest/+/98455908:42
*** dmellado47172 is now known as dmellado471709:39
*** ykarel_ is now known as ykarel09:57
*** efoley_ is now known as efoley13:24
opendevreviewMaor Blaustein proposed openstack/tempest master: Migrate QoS with placement tests to NTP  https://review.opendev.org/c/openstack/tempest/+/98455914:45
gmaanqa ofice hour  time18:00
gmaan#startmeeting qa18:00
opendevmeetMeeting started Wed Apr 15 18:00:58 2026 UTC and is due to finish in 60 minutes.  The chair is gmaan. Information about MeetBot at http://wiki.debian.org/MeetBot.18:00
opendevmeetUseful Commands: #action #agreed #help #info #idea #link #topic #startvote.18:00
opendevmeetThe meeting name has been set to 'qa'18:00
gmaanwho all here today18:01
abongaleo/18:01
gmaanlet's start, 18:02
gmaan#link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_next_Office_hours18:02
gmaantoday agnda ^^18:03
gmaan#topic Announcement and Action Item (Optional)18:03
gmaanprevious meeting action items18:03
gmaangmaan to test the stable branch job with at least 1 tempest plugin, try to adjust it in existing job instead of new one18:03
gmaanthis is done18:03
abongaleworking on devstack-platform-rocky-blue-onyx job18:04
gmaan#link https://review.opendev.org/c/openstack/tempest/+/97896018:04
gmaanperfect, thans abongale 18:04
abongaleraise bug https://bugs.launchpad.net/devstack/+bug/214741818:05
gmaangmaan to create the 2026.2 priority etherpad18:05
gmaandone18:05
gmaan#link https://etherpad.opendev.org/p/qa-2026-2-priority18:05
gmaangmaan to add stable/2026.1 jobs in the periodic jobs monitoring links18:05
gmaandone, i updated the link18:05
gmaanabongale: I am continuing your action item18:05
gmaan#action abongale to look into the failing stable/2026.1 devstack-platform-rocky-blue-onyx18:05
gmaan#topic Gate Status / Blocker Fix / Urgent Change18:06
gmaan#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)18:06
gmaannothing pending on review18:06
gmaanthe one which are actuall gate blocker or urgent18:07
sean-k-mooney o/ just lurking incase there is a topic of interest18:07
gmaan#topic Priority Review18:08
gmaan#link https://review.opendev.org/q/label:Review-Priority%3D%252B1+status:open+(project:openstack/tempest+OR+project:openstack/devstack+OR+project:openstack/grenade)18:08
gmaanthis one is updated, I will check that18:09
gmaan#link https://review.opendev.org/c/openstack/tempest/+/95761218:09
gmaanerst all are already reviewed or pending on author18:09
gmaan#topic Periodic jobs Status Checks18:09
gmaanPeriodic Master jobs:18:10
gmaan#link https://zuul.openstack.org/builds?project=openstack%2Ftempest&project=openstack%2Fdevstack&pipeline=periodic18:10
gmaanall ok18:10
gmaanPeriodic Stable jobs:18:10
gmaan#link https://zuul.openstack.org/builds?job_name=tempest-full-2024-2&job_name=tempest-full-2025-1&job_name=tempest-full-2025-2&job_name=tempest-full-2026-1&pipeline=periodic-stable&skip=018:10
gmaan#link https://zuul.openstack.org/builds?job_name=tempest-slow-2024-2&job_name=tempest-slow-2025-1&job_name=tempest-slow-2025-2&job_name=tempest-slow-2026-1&skip=018:11
gmaan#link https://zuul.openstack.org/builds?job_name=tempest-multinode-2025-1&job_name=tempest-multinode-2025-2&job_name=tempest-multinode-2024-2&job_name=tempest-multinode-2026-1&pipeline=periodic-stable&skip=018:11
gmaan#lin https://zuul.openstack.org/builds?job_name=tempest-full-2024-2-extra-tests&job_name=tempest-full-2025-1-extra-tests&job_name=tempest-full-2025-2-extra-tests&job_name=tempest-full-2026-1-extra-tests&pipeline=periodic-stable&skip=018:11
gmaanall green18:11
gmaan#topic Distros check:18:11
gmaanCentos18:11
gmaan#link https://zuul.openstack.org/builds?job_name=tempest-full-centos-9-stream&job_name=devstack-platform-centos-9-stream&job_name=devstack-platform-centos-10-stream&skip=018:12
gmaanall ok18:12
gmaanRocky Linux18:12
gmaan#link https://zuul.openstack.org/builds?job_name=devstack-platform-rocky-blue-onyx&job_name=tempest-full-rocky&skip=018:12
gmaansome timeout but latest run is passing18:12
gmaanUbuntu18:12
gmaan#link https://zuul.openstack.org/builds?job_name=devstack-platform-ubuntu-noble-ovn-source&job_name=devstack-platform-ubuntu-noble-ovs&job_name=devstack-platform-ubuntu-jammy&skip=018:12
gmaanall ok18:12
gmaan#topic Bugs:18:13
gmaannothing on agenda but any bug to discuss today18:13
gmaan#link https://etherpad.opendev.org/p/qa-bug-triage-and-fix18:13
sean-k-mooneyc9s was supported on 2025.1 but not 2025.2 so after this release we coudl trime that form the list18:13
sean-k-mooneyon master at least for devstack.18:13
gmaansure18:14
gmaanusuauly we can drop it after one cycle but yes after this release this can be done18:14
sean-k-mooneywell master doesnt need to supprot it any more18:14
sean-k-mooneyi guess what i ment is once 2025.1 goes to unmaintend18:15
sean-k-mooneywe rpoably dont need to monitor c9s18:15
gmaansure18:15
abongaleack18:16
gmaan#topic Open Discussion18:16
gmaannothing on agenda in open discussion18:16
gmaananything else for today otherwise we can close early18:17
abongaleso trunk repos required for 2025.2 & 2026.1 is not present https://trunk.rdoproject.org/ 18:17
sean-k-mooneywell tehre is master18:17
sean-k-mooneybut the last release rdo ddi was epoxy 2025.118:18
sean-k-mooneyso its expecte that there are no release for 2025.2 or 2026.118:18
sean-k-mooneyrdo anoched that epoxy woudl be the last release unless new maintainers steped up a year ago18:19
abongaleSo should we point wget repos  stable branch of 2025.2 & 2026.1 to masters for devstack-platform-rocky-blue-onyx job18:20
sean-k-mooneyidealy we woudl remove allusage of rdo form devstack and point to the centos sig repos or rocky equivlents18:20
sean-k-mooneyor epel/copr which ever makes the most sense in that context18:21
abongale:sean-k-mooney ack18:21
sean-k-mooneymaster for c9 and c10 has had fail to build form souce issues for a while18:22
gmaanyeah, I am not sure we can keep rdo things 18:22
sean-k-mooneyi guess one other topic18:23
sean-k-mooneyubuntu 26.04 iss coming out next week. i belive there is a patch to create an image already18:23
sean-k-mooneyi.e. on opendev zuul but without uploadign it to the ci providers18:23
sean-k-mooneydoes anyone know more on its current state18:24
gmaani remember the change but not sure if that is merged or not18:24
gmaanlet me find18:24
sean-k-mooneyi plan to test it out after the offcial release but have not really been folowing18:24
gmaani think this one18:26
gmaan#link https://review.opendev.org/c/opendev/zuul-providers/+/98218218:26
gmaanack18:26
sean-k-mooneyyep thats the one18:27
sean-k-mooneyi know there are some issues with the rust version of sudo18:27
sean-k-mooneyso we may need to revert back to the gnu sudo unless they fixed that18:28
sean-k-mooneyit didnt work with privsep properly18:28
sean-k-mooneybut that easy to do but im interested what other gotchas we will see18:28
gmaanyeah18:29
gmaanany other topic18:29
gmaanlet's close 18:31
gmaanthanks for joining18:31
gmaan#endmeeting18:31
opendevmeetMeeting ended Wed Apr 15 18:31:19 2026 UTC.  Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4)18:31
opendevmeetMinutes:        https://meetings.opendev.org/meetings/qa/2026/qa.2026-04-15-18.00.html18:31
opendevmeetMinutes (text): https://meetings.opendev.org/meetings/qa/2026/qa.2026-04-15-18.00.txt18:31
opendevmeetLog:            https://meetings.opendev.org/meetings/qa/2026/qa.2026-04-15-18.00.log.html18:31
sean-k-mooneygmaan: i could have brought this up in the meeting but its mroe an fyi. stephen and i have been talking about posibly supproting uv in devstack to replace the current use of pip and virtual env form disto python18:33
sean-k-mooneygmaan: i have a working poc of that locally which i might push to see what peopel think18:33
sean-k-mooneyas a totally optioanl mode of course18:34
sean-k-mooneythe intersing thing is i was able to disabel --site-pacakges in general with uv18:34
sean-k-mooneythre is one case where that might be required for nova file injection but we dont test that and its been deprecated for years18:35
sean-k-mooneyso indepently of uv im wonder if gating that behind a flag and not adding it by default woudl be a nice improvement18:35
gmaansure, we can check it once PoC is up18:36
gmaanfile injection test, yes it has been deprecated so i do not have strong opinion on its testing18:37
sean-k-mooneyi dont know if we even test it in nova at this point18:37
sean-k-mooneyi would have to check but i suspect we dont18:37
sean-k-mooneyand even so i think we are callign libgustfs as a cli rather then a lib18:37
sean-k-mooneyso im relaly not sure --site-packages is required.18:38
sean-k-mooneyits worth takeing a look at anyway because that cuases us a buch of other headaches18:38
sean-k-mooneyso if we can drop it by default or entirly it woudl be nice18:38
melwittgmaan: we have been seeing some intermittent fail of vtpm live migrate tests for NoValidHost and looking through the logs it seems it is likely just a timing thing when tests are running in parallel -- it is possible for compute resources to get tied up temporarily. what is the usual fix for this? run in serial? lower concurrency for just that job? or other?19:31
gmaanmelwitt: but that job does not run too many tests right? I am not sure if concurrency can solve it19:33
melwittgmaan: it doesn't run too many but if the "wrong" ones go at the same time, I could see how it could happen. with resize or live migration you automatically get one host that has to be ignored because that's the one you're on, and if some other test made a VM on the other host, maybe that takes up the resources. maybe i am wrong19:36
gmaanlooking at stackviz report, it seems it has only two test worker and both running vTPM test19:36
gmaanhttps://9ad4826b71b64bbb4cfa-962c398c9a043149e9c78cc12a8a1726.ssl.cf2.rackcdn.com/openstack/82999e425f924f5d9a80935c6a9f96e2/controller/logs/stackviz/index.html#/stdin/timeline19:36
melwittoh yeah I should look at stackviz19:37
gmaani think that is type of test running (seeing in stackviz report) in parralel as you desribed 19:37
gmaanI was thinking if tempest tests and we can make them in single test class that also make them run in seriel to each other but its tempest test running in parallel with whitebox-tempest-plugin tests19:39
melwittyeah, hm. I guess these really do need to be serial. I don't know why I thought parallel could work ok19:39
melwittright. so if I were to set tempest concurrency=1 would it work across both classes?19:40
melwittcurrently I had it set to 4 not realizing the problem https://github.com/openstack/nova/blob/master/.zuul.yaml#L45819:41
gmaanyeah we can set that 1. I was thinking to expliclty ask via --serial in job but we do not have that var for run-tempest role so  concurrency=1 seems ood option19:44
gmaangood19:44
melwittI'll propose a change and check its results. thanks19:45
gmaan++19:46
opendevreviewCarlos Eduardo proposed openstack/tempest master: WIP: Introduce share attachment client calls  https://review.opendev.org/c/openstack/tempest/+/98487120:41

Generated by irclog2html.py 4.1.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!