| gmann | dansmith: perfect, thanks | 00:03 |
|---|---|---|
| gmann | dansmith: for cinder tempest plugin also you need to pass the validation resources https://review.opendev.org/c/openstack/cinder-tempest-plugin/+/881764/1/cinder_tempest_plugin/api/volume/test_volume_backup.py#107 | 00:09 |
| dansmith | gmann: I would but not after this: https://review.opendev.org/c/openstack/tempest/+/881926/2 | 00:10 |
| dansmith | I haven't updated the c-t-p patch to depend on that top one, but I figure that would be easier than adding it manually | 00:12 |
| gmann | dansmith: but this is for tempest tests only. cinder-tempest-plugin use separate create_server() method in their base class | 00:13 |
| gmann | https://github.com/openstack/cinder-tempest-plugin/blob/master/cinder_tempest_plugin/api/volume/base.py#L136 | 00:13 |
| dansmith | gmann: ah, I thought base.BaseVolumeTest there was the same | 00:14 |
| dansmith | okay | 00:14 |
| gmann | it is different as tempest change can break them many times as base.py is not stable interface. | 00:14 |
| dansmith | ack | 00:16 |
| dansmith | so I might need to change the network resource stuff too | 00:16 |
| dansmith | I'll wait to push that up until the current nova patch finishes running | 00:21 |
| gmann | k | 00:23 |
| opendevreview | Ghanshyam proposed openstack/tempest master: Add wait for server SSH-able in base attach_volume method https://review.opendev.org/c/openstack/tempest/+/842240 | 00:51 |
| opendevreview | Ghanshyam proposed openstack/tempest master: Add wait for server SSH-able in base attach_volume method https://review.opendev.org/c/openstack/tempest/+/842240 | 01:05 |
| opendevreview | Lukas Piwowarski proposed openstack/tempest master: [WIP] Catching broken tests in tempest-full-test-account-* jobs https://review.opendev.org/c/openstack/tempest/+/880630 | 10:03 |
| opendevreview | Luigi Dino Tamagnone proposed openstack/tempest master: Avoiding the clean of not tempest created resources https://review.opendev.org/c/openstack/tempest/+/879923 | 14:58 |
| kopecmartin | #startmeeting qa | 15:00 |
| opendevmeet | Meeting started Tue May 2 15:00:14 2023 UTC and is due to finish in 60 minutes. The chair is kopecmartin. Information about MeetBot at http://wiki.debian.org/MeetBot. | 15:00 |
| opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 15:00 |
| opendevmeet | The meeting name has been set to 'qa' | 15:00 |
| kopecmartin | #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_next_Office_hours | 15:00 |
| kopecmartin | agenda ^^ | 15:00 |
| lpiwowar | o/ | 15:01 |
| kopecmartin | o/ | 15:03 |
| kopecmartin | let's go through the usual | 15:03 |
| kopecmartin | #topic Announcement and Action Item (Optional) | 15:03 |
| kopecmartin | none this week | 15:03 |
| kopecmartin | #topic Bobcat Priority Items progress | 15:03 |
| gmann | o/ but in another meeting in parallel | 15:03 |
| kopecmartin | #link https://etherpad.opendev.org/p/qa-bobcat-priority | 15:03 |
| kopecmartin | hi o/ np, me too | 15:03 |
| * kopecmartin checking the priority items | 15:04 | |
| gmann | patrole retirement is completed | 15:04 |
| kopecmartin | yeah, there is only one review left | 15:06 |
| kopecmartin | #link https://review.opendev.org/c/openstack/tripleo-quickstart/+/880224 | 15:06 |
| gmann | yeah, that can be merged in its own time and seeing TripleO deprecated this is less priority too | 15:06 |
| gmann | but I will mark it completed as most of other work done | 15:06 |
| kopecmartin | i agree | 15:07 |
| kopecmartin | i made some progress with 'New cirros image update' | 15:07 |
| kopecmartin | #link https://review.opendev.org/c/openstack/devstack/+/881504 | 15:07 |
| kopecmartin | if that makes sense ^^ we can proceed with this too: | 15:08 |
| kopecmartin | #link https://review.opendev.org/c/openstack/devstack/+/881437 | 15:08 |
| kopecmartin | when you have a moment, please check ^ | 15:08 |
| kopecmartin | i'm not aware of any other updates on priority items | 15:09 |
| gmann | sure | 15:09 |
| kopecmartin | #topic Gate Status Checks | 15:09 |
| 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) | 15:10 |
| kopecmartin | 3 reviews | 15:10 |
| kopecmartin | i'll check as soon as i can (hopefully later today) | 15:10 |
| gmann | yeah, those are SSHABLE fixes by dansmith, will be good to check them | 15:11 |
| kopecmartin | #topic Bare rechecks | 15:12 |
| kopecmartin | #link https://etherpad.opendev.org/p/recheck-weekly-summary | 15:12 |
| kopecmartin | no updates there | 15:12 |
| gmann | on related one, I am trying to make server SSH-able by default in API tests #link https://review.opendev.org/c/openstack/tempest/+/842240 | 15:12 |
| gmann | but this is not ready, just info | 15:12 |
| kopecmartin | thanks, added to my list | 15:12 |
| kopecmartin | #topic Periodic jobs Status Checks | 15:12 |
| kopecmartin | periodic stable full | 15:12 |
| kopecmartin | #link https://zuul.openstack.org/builds?pipeline=periodic-stable&job_name=tempest-full-yoga&job_name=tempest-full-xena&job_name=tempest-full-zed&job_name=tempest-full-2023-1 | 15:12 |
| kopecmartin | periodic stable slow | 15:12 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=tempest-slow-2023-1&job_name=tempest-slow-zed&job_name=tempest-slow-yoga&job_name=tempest-slow-xena | 15:12 |
| gmann | I think it will make sense to do in scenario tests too but as that include backward incompatible change in create_server() interface of scenario manager, we need to be careful to not break the plugins | 15:12 |
| kopecmartin | periodic extra tests | 15:12 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=tempest-full-2023-1-extra-tests&job_name=tempest-full-zed-extra-tests&job_name=tempest-full-yoga-extra-tests&job_name=tempest-full-xena-extra-tests | 15:13 |
| kopecmartin | periodic master | 15:13 |
| kopecmartin | #link https://zuul.openstack.org/builds?project=openstack%2Ftempest&project=openstack%2Fdevstack&pipeline=periodic | 15:13 |
| kopecmartin | +1 | 15:13 |
| kopecmartin | oh, there seems to be some kind of a problem with tempest-centos9-stream-fips | 15:14 |
| kopecmartin | hitting retry and retry limit | 15:14 |
| kopecmartin | started just today | 15:14 |
| kopecmartin | hopefully it's something temporary, better to monitor it for a while | 15:16 |
| kopecmartin | it's most likely related to the mirror issues , there was an email about that on the ML | 15:17 |
| kopecmartin | moving on | 15:17 |
| kopecmartin | #topic Distros check | 15:17 |
| kopecmartin | cs-9 | 15:17 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=tempest-full-centos-9-stream&job_name=devstack-platform-centos-9-stream&skip=0 | 15:17 |
| kopecmartin | fedora | 15:17 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=devstack-platform-fedora-latest&skip=0 | 15:17 |
| kopecmartin | debian | 15:17 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=devstack-platform-debian-bullseye&skip=0 | 15:17 |
| kopecmartin | focal | 15:17 |
| kopecmartin | #link https://zuul.opendev.org/t/openstack/builds?job_name=devstack-platform-ubuntu-focal&skip=0 | 15:17 |
| kopecmartin | rocky | 15:17 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=devstack-platform-rocky-blue-onyx | 15:17 |
| kopecmartin | openEuler | 15:17 |
| kopecmartin | #link https://zuul.openstack.org/builds?job_name=devstack-platform-openEuler-22.03-ovn-source&job_name=devstack-platform-openEuler-22.03-ovs&skip=0 | 15:17 |
| kopecmartin | centos job suffers from the mirror issue | 15:18 |
| kopecmartin | however the openeuler job got improved, that's nice | 15:18 |
| kopecmartin | the rest looks within standard | 15:18 |
| kopecmartin | #topic Sub Teams highlights | 15:19 |
| kopecmartin | Changes with Review-Priority == +1 | 15:19 |
| 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) | 15:19 |
| kopecmartin | no reviews | 15:19 |
| kopecmartin | #topic Open Discussion | 15:19 |
| kopecmartin | anything for the open discussion? | 15:19 |
| lpiwowar | Nothing on my side. I'm still working on fixing the cleaning up of the resources for the object storage tests. | 15:20 |
| kopecmartin | great, thanks lpiwowar | 15:20 |
| kopecmartin | #topic Bug Triage | 15:22 |
| kopecmartin | #link https://etherpad.openstack.org/p/qa-bug-triage-bobcat | 15:22 |
| kopecmartin | numbers recorded ^^ | 15:22 |
| kopecmartin | no change since last week | 15:23 |
| kopecmartin | this is all from my side | 15:23 |
| lpiwowar | thanks kopecmartin | 15:23 |
| kopecmartin | see you online everyone | 15:24 |
| kopecmartin | #endmeeting | 15:24 |
| opendevmeet | Meeting ended Tue May 2 15:24:14 2023 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 15:24 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/qa/2023/qa.2023-05-02-15.00.html | 15:24 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/qa/2023/qa.2023-05-02-15.00.txt | 15:24 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/qa/2023/qa.2023-05-02-15.00.log.html | 15:24 |
| gmann | thanks kopecmartin | 15:24 |
| dansmith | gmann: so, for that top patch that enforces that validation stuff is provided.. I need to change the scenario manager to always provide those things, even if conf is disabled | 15:26 |
| dansmith | but I think that's okay because get_test_resources is a no-op if config is disabled | 15:26 |
| dansmith | is that okay/ | 15:26 |
| dansmith | since the runtime check is really to prevent new tests from being added that don't provide those, I think it's better to make that check not depend on config, as people could have validation disabled locally or in a plugin repo and not realize they broke an SSHABLE check until after they merge | 15:27 |
| gmann | dansmith: you are right, config is more for *run* validation but keeping all validation resources ready is good. and at least it will allow to ping server if ssh is disabled in guest | 15:31 |
| dansmith | gmann: ack cool, I noticed it in the cinder tempest run, which has a job with validation disabled (the ceph job!) | 15:31 |
| opendevreview | Dan Smith proposed openstack/tempest master: Allow wait_until=SSHABLE in scenario manager https://review.opendev.org/c/openstack/tempest/+/881675 | 15:33 |
| opendevreview | Dan Smith proposed openstack/tempest master: Fail create if validation flags do not agree https://review.opendev.org/c/openstack/tempest/+/881926 | 15:33 |
| dansmith | gmann: is there an easy way to get cinder-tempest-plugin installed locally into my tempest? pip installing it into the tox venv doesn't seem to do it | 17:26 |
| dansmith | oh maybe I need to use the tempest venv | 17:28 |
| dansmith | yup, okay | 17:28 |
| opendevreview | Dan Smith proposed openstack/tempest master: Fail create if validation flags do not agree https://review.opendev.org/c/openstack/tempest/+/881926 | 17:40 |
| gmann | dansmith: yeah, tempest venv | 17:40 |
| dansmith | yep, all good now | 17:41 |
| dansmith | gmann: should this be get_test_validation_resources()? https://github.com/openstack/tempest/blob/master/tempest/api/compute/admin/test_volumes_negative.py#L59 | 19:09 |
| dansmith | i.e. not the class variant | 19:09 |
| dansmith | the class variant returns None if conf has validation disabled | 19:09 |
| dansmith | so it behaves differently | 19:09 |
| dansmith | or could I make get_class_* return an empty fixture like the get_test_* one does? | 19:10 |
| gmann | create_test_server is class level server creation so we need to use class level validation resouce | 19:11 |
| gmann | oh that is bug if get_class_* and get_test_* return different way based on config | 19:11 |
| gmann | yeah we need to fix get_class_* one | 19:11 |
| dansmith | okay I don't understand why you would use one and not the other from a non-classmethod test | 19:12 |
| dansmith | but yeah, lemme make the class one behave the same and we'll see | 19:12 |
| dansmith | gmann: well, one returns a fixture and the other does not, | 19:13 |
| dansmith | and I guess the class one probably doesn't/shouldn't useFixture it right? | 19:13 |
| dansmith | oh no, nevermind | 19:14 |
| dansmith | it doesn't return the fixture, it returns fixture.resources | 19:14 |
| gmann | dansmith: which is just bool after matching from conf | 19:14 |
| gmann | oh no, return what is configured | 19:15 |
| dansmith | get_test returns the fixture.resources regardless of config, which is an empty dict if validation is disabled | 19:15 |
| dansmith | so I think just making the get_class one reutrn {} instead of None if conf has it disabled will be okay | 19:15 |
| gmann | ah got it. | 19:16 |
| dansmith | and that seems to work locally | 19:17 |
| gmann | that is more confusing, I do not think we check validation_resource['<resouece>'] is None or not instead we do check validation_resouces is None or not | 19:17 |
| gmann | I think returning None make sense from both | 19:17 |
| opendevreview | Dan Smith proposed openstack/tempest master: Fail create if validation flags do not agree https://review.opendev.org/c/openstack/tempest/+/881926 | 19:18 |
| dansmith | gmann: no I want to always return a dict, not always return None | 19:18 |
| gmann | but now thinking to your morning question, I think we do check run_validation in creating the validation resources otherwise it is just a empty things | 19:18 |
| dansmith | gmann: if we use None, we can't tell the difference between "disabled" and "not passed to create_test_server" | 19:18 |
| dansmith | see ^ | 19:18 |
| gmann | you mean when rn_validation is true and to check if validation_resources are passed from test or not? | 19:19 |
| gmann | got it, I think dict is good to differentiate both case | 19:20 |
| dansmith | gmann: I want to *always* check that validation_resources was passed if you asked for SSHABLE, regardless of whether run_validation is true, | 19:20 |
| dansmith | because it's a coding error if you ask for SSHABLE but don't set up to make it actually work, if it is enabled | 19:20 |
| gmann | but if run_validation is False, validation resources are not going to be created and ssh cannot be done right | 19:21 |
| dansmith | right, but you should still have passed me validation_resources={} in that case | 19:21 |
| dansmith | because otherwise it will "work" locally with validation disabled, but silently not validate in CI if run_validation=True | 19:21 |
| gmann | yeah i agree on {} as None if create_server default arg also | 19:22 |
| gmann | dansmith: in that case, you want to check {} too here https://review.opendev.org/c/openstack/tempest/+/881926/5/tempest/common/compute.py#303 | 19:24 |
| gmann | as test can skip passing it if run_validation is false. I mean they still ask for SSHABLE but do not call for validation resource creation itself when run_validation is false itself | 19:24 |
| gmann | and I think run_validation also should be checked ? | 19:25 |
| gmann | I mean check and fail if validation resource not passed when test ask for SSHABLE and run_validation true | 19:26 |
| gmann | because all three things are going to be needed for actual doing ssh | 19:26 |
| gmann | if any test asking for SSHABLE and not passing validation resource based on run_validation that still be ok and just skip the SSH instead of failing | 19:27 |
| dansmith | gmann: no, this is what I'm trying to say... | 19:45 |
| dansmith | gmann: if you allow tests to ask for SSHABLE but when they do not provide validation_resources, | 19:45 |
| dansmith | then the test writer thinks that we will actually wait for SSHABLE, but we will not | 19:46 |
| dansmith | so I'm trying to say I think we should be asserting that you cannot ask for SSHABLE and not provide the validation resources, because that yields what we have today which is tests that look like they wait for SSHABLE but do not | 19:46 |
| dansmith | get_test_resources() and get_class_resources() already check the CONF and avoid the setup if they don't need to, but we still need to pass the resources (empty dict if disabled) to the create_test_server() | 19:47 |
| dansmith | so that it knows we *would* provide actual resources if CONF was enabled | 19:47 |
| dansmith | gmann: if I'm not making sense, perhaps we should gmeet :) | 19:51 |
| gmann | dansmith: you have point i think. it will avoid coding error. we need to make sure if any test creating resources based on run_validation should always do | 19:52 |
| gmann | and you patch check that | 19:52 |
| dansmith | yeah | 19:52 |
| dansmith | gmann: oh pingable requires validation resources too because of the IP I guess? | 19:53 |
| gmann | yeah at least ip | 19:53 |
| dansmith | gotcha, I was thinking it didn't because it doesn't need the ssh key, but makes sense :) | 19:53 |
| gmann | yeah atleast in env where only ping allowed test should create/pass validation resource and based on config validation resource will be created | 19:55 |
| dansmith | yep, cool, I'll add | 19:55 |
| gmann | thanks | 19:56 |
| opendevreview | Dan Smith proposed openstack/tempest master: Fail create if validation flags do not agree https://review.opendev.org/c/openstack/tempest/+/881926 | 21:57 |
| opendevreview | Dan Smith proposed openstack/tempest master: WIP: Also require resources for PINGABLE https://review.opendev.org/c/openstack/tempest/+/882102 | 21:57 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!