| opendevreview | Merged openstack/tempest master: Deprecate the old API microversion fixture https://review.opendev.org/c/openstack/tempest/+/820976 | 04:40 |
|---|---|---|
| opendevreview | Ghanshyam proposed openstack/tempest master: Update aggregate tests for bug#1907775 https://review.opendev.org/c/openstack/tempest/+/821647 | 04:41 |
| *** amoralej|off is now known as amoralej | 08:15 | |
| opendevreview | wangxiyuan proposed openstack/devstack master: openEuler 20.03 LTS SP2 support https://review.opendev.org/c/openstack/devstack/+/760790 | 08:32 |
| opendevreview | wangxiyuan proposed openstack/devstack master: openEuler 20.03 LTS SP2 support https://review.opendev.org/c/openstack/devstack/+/760790 | 08:45 |
| opendevreview | wangxiyuan proposed openstack/devstack master: openEuler 20.03 LTS SP2 support https://review.opendev.org/c/openstack/devstack/+/760790 | 09:04 |
| opendevreview | wangxiyuan proposed openstack/devstack master: Add openEuler nodeset https://review.opendev.org/c/openstack/devstack/+/821677 | 09:07 |
| opendevreview | wangxiyuan proposed openstack/devstack master: Add openEuler nodeset https://review.opendev.org/c/openstack/devstack/+/821677 | 09:08 |
| opendevreview | wangxiyuan proposed openstack/devstack master: openEuler 20.03 LTS SP2 support https://review.opendev.org/c/openstack/devstack/+/760790 | 09:12 |
| opendevreview | Lajos Katona proposed opendev/elastic-recheck master: Add query for bug 1954663 https://review.opendev.org/c/opendev/elastic-recheck/+/821684 | 09:56 |
| opendevreview | Lajos Katona proposed opendev/elastic-recheck master: Add query for bug 1799790 https://review.opendev.org/c/opendev/elastic-recheck/+/821684 | 10:22 |
| *** jpena|off is now known as jpena | 11:42 | |
| *** soniya29 is now known as soniya29|afk | 12:04 | |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Update aggregate tests for bug#1907775 https://review.opendev.org/c/openstack/tempest/+/821647 | 12:45 |
| gibi | gmann: fixed an issue in ^^ | 12:45 |
| *** amoralej is now known as amoralej|lunch | 13:16 | |
| *** soniya29|afk is now known as soniya29 | 13:58 | |
| *** amoralej|lunch is now known as amoralej | 14:04 | |
| opendevreview | Andre Aranha proposed openstack/tempest master: WIP - Refactor ssh.Client to allow other clients https://review.opendev.org/c/openstack/tempest/+/820860 | 14:42 |
| kopecmartin | #startmeeting qa | 15:01 |
| opendevmeet | Meeting started Tue Dec 14 15:01:46 2021 UTC and is due to finish in 60 minutes. The chair is kopecmartin. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:01 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:01 |
| opendevmeet | The meeting name has been set to 'qa' | 15:01 |
| kopecmartin | #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Weekly_QA_Team_meeting | 15:01 |
| kopecmartin | agenda ^^ | 15:01 |
| frickler | o/ | 15:02 |
| kopecmartin | hey o/ | 15:03 |
| lpiwowar | hi o/ | 15:03 |
| kopecmartin | let's start by going through the usual | 15:04 |
| kopecmartin | #topic Announcement and Action Item (Optional) | 15:04 |
| kopecmartin | nothing specific from me here | 15:04 |
| gmann | o/ | 15:04 |
| kopecmartin | #topic Yoga Priority Items progress | 15:04 |
| kopecmartin | #link https://etherpad.opendev.org/p/qa-yoga-priority | 15:04 |
| kopecmartin | any updates on the priority items? | 15:05 |
| soniya29 | hello | 15:05 |
| kopecmartin | hey lpiwowar, gmann, soniya29 o/ | 15:05 |
| jparoly | o/ | 15:06 |
| kopecmartin | no updates from my side w.r.t priority items | 15:06 |
| soniya29 | same with me | 15:06 |
| kopecmartin | we just should proceed with the fips patches, but let's talk about that in a minute | 15:07 |
| kopecmartin | gmann: anything for "Migration of Devstack and Tempest tests to new secure RBAC"? anything i can help with? | 15:08 |
| gmann | kopecmartin: not yet, few things under work but not up as such | 15:08 |
| kopecmartin | ok, thanks | 15:08 |
| kopecmartin | soniya29: what about "Cleanup of duplicated scenario.manager"? any blockers i could help with? | 15:09 |
| soniya29 | kopecmartin, i haven't started with it yet :) | 15:10 |
| kopecmartin | soniya29: np, just let me know if i can help you in any way :) | 15:10 |
| soniya29 | kopecmartin, yeah, sure | 15:10 |
| kopecmartin | sure, moving on | 15:11 |
| kopecmartin | #topic OpenStack Events Updates and Planning | 15:11 |
| kopecmartin | nothing here, let's skip | 15:11 |
| kopecmartin | #topic Gate Status Checks | 15:11 |
| 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) | 15:11 |
| kopecmartin | no changes with priority == +2 | 15:11 |
| kopecmartin | #topic Periodic jobs Status Checks | 15:12 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=tempest-full-xena-py3&job_name=tempest-full-wallaby-py3&job_name=tempest-full-victoria-py3&job_name=tempest-full-ussuri-py3&job_name=tempest-full-train-py3&pipeline=periodic-stable | 15:12 |
| kopecmartin | #link https://zuul.openstack.org/builds?project=openstack%2Ftempest&project=openstack%2Fdevstack&pipeline=periodic | 15:12 |
| kopecmartin | all green \o/ | 15:12 |
| kopecmartin | any gate issues to bring up? | 15:12 |
| gmann | nothing I have seen | 15:13 |
| frickler | the was a pypi hickup earlier today | 15:13 |
| frickler | should be fixed by now | 15:13 |
| kopecmartin | oh, perfect | 15:13 |
| kopecmartin | #topic Sub Teams highlights | 15:14 |
| kopecmartin | Changes with Review-Priority == +1 | 15:14 |
| 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) | 15:14 |
| kopecmartin | no changes there | 15:14 |
| kopecmartin | any change to bring up? | 15:14 |
| kopecmartin | i have this one regarding ecdsa keys | 15:15 |
| kopecmartin | #link https://review.opendev.org/c/openstack/tempest/+/807465 | 15:15 |
| kopecmartin | from what i remember it doesn't change anything, just adds support for different keys, I need to take a look again | 15:16 |
| kopecmartin | but there shouldn't be any harm to proceed, we need to start somewhere with the fips thingy | 15:16 |
| gmann | I will also check that, I think we have job setup in nova to run with that key and see if that is working? | 15:16 |
| kopecmartin | yeah, i think it was | 15:17 |
| gmann | do you have that nova job patch? | 15:17 |
| kopecmartin | let me check | 15:17 |
| kopecmartin | #link https://review.opendev.org/c/openstack/nova/+/790519/15/.zuul.yaml#521 | 15:17 |
| kopecmartin | #link https://review.opendev.org/c/openstack/nova/+/790519 | 15:17 |
| gmann | but that does not depends-on on tempest | 15:18 |
| gmann | ah it is | 15:18 |
| gmann | we need that nova job to pass first so that we can verify the tempest change | 15:19 |
| kopecmartin | it does, on a third review after the one with the ecdsa keys support | 15:19 |
| gmann | yeah | 15:19 |
| kopecmartin | ok, recheck then? or you mean to have that job gating in tempest? | 15:19 |
| kopecmartin | at some point i'd like to see those jobs (at least one with fips) gating in tempest, but i'm kinda lost with all the prerequisites, like which patch needs to be merged before we can have that | 15:20 |
| gmann | yeah, adding in tempest will be good also to verify the change | 15:20 |
| kopecmartin | i need to take a deeper look | 15:20 |
| gmann | added comment in that patch to add tempest job | 15:21 |
| gmann | without that it is difficult to check if new changes/testrs are working on service side or not | 15:21 |
| kopecmartin | yeah, agree, i'll try to propose that after the meeting as a non-voting of course and then we can recheck the patch with the ecdsa keys support | 15:22 |
| gmann | +1, thanks | 15:23 |
| gmann | we can do in same patch so that we sync on merging the stack | 15:23 |
| kopecmartin | yeah, good, let's move on | 15:24 |
| kopecmartin | #topic Open Discussion | 15:24 |
| kopecmartin | (gmann) Turning off the openstack-health service and so does retirement of it? | 15:24 |
| kopecmartin | there will be a call tomorrow with the tripleo ci team where we will discuss next steps on how we can collaborate together as they are building their version of openstakc-health | 15:25 |
| kopecmartin | the day after tomorrow, not tomorrow, sorry | 15:25 |
| gmann | kopecmartin: thanks for the call schedule. | 15:25 |
| kopecmartin | let me find the emails | 15:25 |
| gmann | did we send on ML | 15:25 |
| kopecmartin | #link http://lists.openstack.org/pipermail/openstack-discuss/2021-December/026250.html | 15:25 |
| gmann | yeah Thursday December 16th 1630 UTC | 15:26 |
| gmann | +1 | 15:26 |
| kopecmartin | so in 2 days and one hour | 15:26 |
| kopecmartin | the next topic is | 15:26 |
| kopecmartin | (gmann) This came up during pain point discussion in TC. With fewer contributors issues in QA (especially devstack where we have more incoming requests/changes), how about expanding the devstack team with potential candidates? | 15:26 |
| kopecmartin | #link https://etherpad.opendev.org/p/pain-point-elimination | 15:27 |
| gmann | yeah, it is just a thought in case we can extend | 15:27 |
| gmann | frickler: yoctozepto ianw ^^ what you think | 15:27 |
| frickler | are there any candidates? also it might make sense to clean up inactive cores | 15:27 |
| gmann | and before that we need to find the candidates | 15:27 |
| gmann | frickler: yeah that is good point | 15:27 |
| gmann | I think we can discuss on the possible candidates like we do for adding core process among us first to avoid any offending | 15:28 |
| frickler | I would be very glad to mentor potential candidates, but I haven't seen any | 15:28 |
| gmann | and then see if we can solve the issue brought up in pain point discussion | 15:28 |
| gmann | frickler: yeah. there were few in my thought and before i discuss about those they prioirty change | 15:29 |
| gmann | anyways, I just added the topic as it came up in pain point discussion but we can follow our normal process of finding the candidates first and then proceed | 15:30 |
| frickler | in particular there still isn't anyone with a focus on centos/fedora, would be good if someone within redhat could bring that up internally | 15:30 |
| gmann | I will start few on email | 15:30 |
| kopecmartin | eliadcohen: ^ | 15:30 |
| gmann | +1, yeah good point, we have seen lot of interest from centos side on testing, py3.6 etc I think it make sense to bring this to them | 15:31 |
| gmann | and if they can help to maintain devstack part of that, it will be helpful | 15:31 |
| kopecmartin | do we wanna write an email to ML? and let's see who'll reply | 15:31 |
| frickler | e.g. we need to urgently move to fedora35 I think | 15:31 |
| frickler | before f34 gets removed from infra | 15:31 |
| gmann | kopecmartin: for redhat, yes please. or you can send on rdo community ML also | 15:32 |
| gmann | kopecmartin: frickler for current possible candidate, let me start email among us first | 15:32 |
| kopecmartin | gmann: ack | 15:32 |
| frickler | gmann: make sure to use my current email then | 15:33 |
| gmann | frickler: ack | 15:33 |
| kopecmartin | anything else for the open discussion? | 15:34 |
| gmann | that's all from me | 15:34 |
| kopecmartin | #topic Bug Triage | 15:36 |
| kopecmartin | #link https://etherpad.opendev.org/p/qa-bug-triage-yoga | 15:36 |
| kopecmartin | bug numbers are recorded at ^ | 15:36 |
| kopecmartin | however, i don't have any specific bug to bring up | 15:37 |
| kopecmartin | so if there isn't anything else to discuss, i think that's it then | 15:37 |
| kopecmartin | thank you all | 15:38 |
| kopecmartin | #endmeeting | 15:38 |
| opendevmeet | Meeting ended Tue Dec 14 15:38:20 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:38 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/qa/2021/qa.2021-12-14-15.01.html | 15:38 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/qa/2021/qa.2021-12-14-15.01.txt | 15:38 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/qa/2021/qa.2021-12-14-15.01.log.html | 15:38 |
| gmann | kopecmartin: thanks | 15:38 |
| gmann | gibi: ack and thanks for update | 15:48 |
| gibi | gmann: I haven't checked yet if it is helped or not :/ | 15:48 |
| gmann | gibi: it failed on some requirement conflict not sure why. anyways I will check that in my afternoon | 15:50 |
| gibi | ahh, I see now. OK | 15:50 |
| gibi | thanks | 15:50 |
| opendevreview | Andre Aranha proposed openstack/tempest master: WIP - Refactor ssh.Client to allow other clients https://review.opendev.org/c/openstack/tempest/+/820860 | 16:11 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Introduce @serial test execution decorator https://review.opendev.org/c/openstack/tempest/+/821732 | 16:26 |
| gibi | gmann, sean-k-mooney: ^^ this is my first trial to introduce a @serial decorator with an inter process read write lock | 16:27 |
| gibi | lets see how badly it fails :) | 16:27 |
| gmann | gibi: ack, thanks | 16:28 |
| sean-k-mooney | oh cool ill take a look. i assume you implemetned it yourself | 16:29 |
| sean-k-mooney | e.g. the lock | 16:29 |
| gibi | sean-k-mooney: yeah I used oslo's inter process lock to implement an rw | 16:35 |
| gibi | sean-k-mooney: It is not yet writer preferring lock but I can hack on that if needed | 16:36 |
| sean-k-mooney | ack | 16:36 |
| sean-k-mooney | as long as it does not allow readers and writers to run conncureently it shoudl be enough | 16:36 |
| gibi | yeah that is the idea :D | 16:36 |
| gibi | and it allows infinite amount of readers or one writer | 16:37 |
| sean-k-mooney | well some RW locks dont wait for all readers to complete before the writer starts | 16:37 |
| sean-k-mooney | thats the important thing to capture if we dont do werter perfering | 16:37 |
| sean-k-mooney | /werter/writer/ | 16:38 |
| sean-k-mooney | gibi: i suspect that the most likely failure mode will be test timeouts | 16:38 |
| sean-k-mooney | if we have writer stavation | 16:38 |
| gibi | yeah we can have that | 16:39 |
| gibi | but oslo has some fair kwargs for the lock | 16:39 |
| gibi | I haven't looked yet but it feels like a solution for starvation | 16:39 |
| sean-k-mooney | perhaps yes that will fix any stravation issue between writers | 16:40 |
| sean-k-mooney | but not nessisarly between readers and writers | 16:40 |
| sean-k-mooney | we can see if its needed | 16:40 |
| sean-k-mooney | well fair shoudl be the default | 16:40 |
| sean-k-mooney | but what i mean is we can see of medeation between readers and writers is needed | 16:40 |
| sean-k-mooney | idealy the lock time would not apply to the test timeout time | 16:41 |
| sean-k-mooney | but not sure we can really do that | 16:41 |
| gibi | ahh I see your point, the impl uses two locks so between the two locks we have still starvation issue even if each lock is fair | 16:41 |
| gibi | yeah I also don't think we can much hack on the timer | 16:41 |
| sean-k-mooney | ya although we have relivly few aggeat tests what will need thw witer lock so hopefully that wont be a problem in practice | 16:42 |
| gibi | one drawback of the current setup is that I need the lock on class level as the class setup already creates resources like instances | 16:42 |
| sean-k-mooney | you can still use grouping with regix of by class or other method to mitgate it | 16:42 |
| sean-k-mooney | ah i see how the lock is working thats smart without being clever. | 16:46 |
| gibi | that is wikipedia not me :D | 16:46 |
| sean-k-mooney | it will be interesting to see 1 if it works and 2 if it has any meaningful impact of execution time | 16:48 |
| gibi | yeah | 16:48 |
| gmann | impact should not be much once it work. as it is very few aggregate tests we have currently | 16:49 |
| sean-k-mooney | gibi: by the way rather then explict unlocking in teardownClass i might consider using the self.addCleanup approch instead | 16:49 |
| gmann | and i think many of them can be covered/already covered in nova fcuntioanl too so we can clean that up if timing is issue | 16:50 |
| sean-k-mooney | that said this is also fine | 16:50 |
| gibi | sean-k-mooney: yeah, if there is class level cleanup then I will move there. | 16:50 |
| gibi | sean-k-mooney: good point | 16:50 |
| sean-k-mooney | i belvie there is at least in testtool | 16:50 |
| gibi | sean-k-mooney: what we have to make sure that the unlock is the last cleanup in the list though | 16:51 |
| gmann | we want to release at test level? it should be released at class level right | 16:51 |
| gibi | gmann: it is on class level | 16:51 |
| gmann | we cna do at the end of tearDownclass | 16:51 |
| sean-k-mooney | ya for now doing it explcitly is proably for the best | 16:51 |
| gibi | gmann: due to resources are set up at class level too | 16:51 |
| sean-k-mooney | i dont actuly see the calse version in the docs so perhaps it does not exist | 16:51 |
| gmann | gibi: yeah, but self. addCleanup is test level so currrent one is ok | 16:51 |
| sean-k-mooney | https://testtools.readthedocs.io/en/latest/api.html#testtools.TestCase.addCleanup i only see | 16:52 |
| gibi | gmann: yes, hence my conditional that if there is class level equivalent for addCleanup then I can move there | 16:52 |
| gmann | it exist but it is same things, called in tearDownclass only so releasng in tearDownclass is fine | 16:52 |
| sean-k-mooney | gibi: the reason i was considering addCleanup was for excption safty | 16:52 |
| sean-k-mooney | so if we dont use it maybe a try finaly would be needed | 16:52 |
| sean-k-mooney | in tearDownClass | 16:53 |
| sean-k-mooney | to ensure its alwasy done | 16:53 |
| gmann | this is internal in tempest but that is called in tearDownclass onlty addClassResourceCleanup | 16:53 |
| gmann | doing in addCleanup will not work as we have to acquire lick in setUp then | 16:53 |
| gibi | sean-k-mooney: good point about a finally block | 16:53 |
| gmann | yeah doing at the end in tearDownclass is best place | 16:54 |
| sean-k-mooney | well it sound liek addClassResourceCleanup would have the right semantics if im parsing what gmann said above correctly | 16:54 |
| gmann | sean-k-mooney: we do not need to do there as it is nothing but called in tearDownclass only | 16:54 |
| sean-k-mooney | but lets see how the job run goes | 16:54 |
| gmann | addClassResourceCleanup is just a internal sync of doing cleanup in tearDownclass | 16:55 |
| gmann | it is not like called after tearDownclass like addCleanup for tearDown() | 16:55 |
| sean-k-mooney | yep im jsut concerend that if there is an excption between https://review.opendev.org/c/openstack/tempest/+/821732/1/tempest/test.py#263 and 275 it wont be called currently | 16:55 |
| sean-k-mooney | with that said i think its very unlikely | 16:57 |
| sean-k-mooney | that we will have an issue between 263 and 275 | 16:57 |
| sean-k-mooney | so it likely will work as writen | 16:57 |
| *** amoralej is now known as amoralej|off | 17:26 | |
| *** jpena is now known as jpena|off | 17:37 | |
| opendevreview | Ghanshyam proposed openstack/tempest master: Update aggregate tests for bug#1907775 https://review.opendev.org/c/openstack/tempest/+/821647 | 17:48 |
| opendevreview | Andre Aranha proposed openstack/tempest master: WIP - Refactor ssh.Client to allow other clients https://review.opendev.org/c/openstack/tempest/+/820860 | 19:57 |
| opendevreview | Alex Yefimov proposed openstack/tempest master: This is a fix for intermittent tempest unittest failure of "test_fix_argument_yes". The expectation is that the test will fail less often, but it is acknowledged that this is not a complete fix. There should be a failure rate decrease of ~50% ... based on my testing https://review.opendev.org/c/openstack/tempest/+/821322 | 21:05 |
| opendevreview | Alex Yefimov proposed openstack/tempest master: This is a fix for intermittent tempest unittest failure of "test_fix_argument_yes". The expectation is that the test will fail less often, but it is acknowledged that this is not a complete fix. There should be a failure rate decrease of ~50% ... based on my testing https://review.opendev.org/c/openstack/tempest/+/821322 | 21:08 |
| gmann | gibi: looks like lock is causing the timeout in jobs. | 23:35 |
| opendevreview | Ghanshyam proposed openstack/tempest master: Update aggregate tests for bug#1907775 https://review.opendev.org/c/openstack/tempest/+/821647 | 23:40 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!