*** ricolin_ is now known as ricolin | 05:05 | |
*** gthiemon1e is now known as gthiemonge | 06:46 | |
*** rpittau|afk is now known as rpittau | 07:06 | |
*** jpena|off is now known as jpena | 07:37 | |
slaweq | yoctozepto hi | 08:54 |
---|---|---|
slaweq | may I have a question? | 08:54 |
slaweq | I wanted to test how devstack will work with enforce new roles and system scopes in Neutron | 08:55 |
slaweq | and it seems that it's not working, e.g. due to the fact that in https://github.com/openstack/devstack/blob/master/lib/neutron_plugins/services/l3 there is everywhere "devstack-admin" hardcoded | 08:55 |
slaweq | and if new policies requires system_scope admin, it's failing | 08:55 |
slaweq | do You know if there is any work on going in devstack to make it working with new defaults? Or should I just propose something for neutron? | 08:56 |
yoctozepto | slaweq: hi; we need help with these bits; TheJulia has actually proposed some coverage for the parts she needed for Ironic | 09:07 |
slaweq | yoctozepto ok, so I will try to do something for Neutron part | 09:11 |
slaweq | I just didn't want to double work which maybe already was done somewhere :) | 09:11 |
yoctozepto | slaweq: yeah, good that you asked | 09:23 |
*** dviroel|out is now known as dviroel | 11:20 | |
*** jpena is now known as jpena|lunch | 11:36 | |
opendevreview | Martin Kopec proposed openstack/tempest master: DNM: Experimenting with test_fix_argument_yes test https://review.opendev.org/c/openstack/tempest/+/797436 | 11:53 |
*** osmanlicilegi is now known as Guest187 | 12:25 | |
*** ChanServ changes topic to "Discussion of OpenStack Quality Assurance Projects (tempest, devstack, grenade, and friends) | docs https://wiki.openstack.org/wiki/QA | specs https://review.openstack.org/#/q/project:openstack/qa-specs | channel logs http://eavesdrop.openstack.org/irclogs/%23openstack-qa/" | 12:30 | |
*** jpena|lunch is now known as jpena | 12:31 | |
TheJulia | Out of curiosity, has anyone seen jobs failing with https://zuul.opendev.org/t/openstack/build/d3159791bcca4ac89aceddbec6c43611 ? | 13:20 |
TheJulia | yoctozepto: slaweq: specifically I just wanted to fire up ironic, at least, first in enforcing mode. My second patch seems to like to fail jobs randomly in devstack :( | 13:24 |
TheJulia | well, to be fair, I do manage to get ironic up in enforcing mode with my patches, so it gives me a place to start with the tempest plugin | 13:24 |
opendevreview | Slawek Kaplonski proposed openstack/devstack master: WIP Deploy Neutron with enforced new RBAC rules https://review.opendev.org/c/openstack/devstack/+/797450 | 13:36 |
slaweq | yoctozepto TheJulia I have some problems with that, I made ^^ for now but it's still failing for me on subnetpool creation. And I don't understand really why because "--os-cloud devstack-system-admin" is used there so it should be system scoped token. When I copy same command into terminal and run it, it is running fine but when it's run during stack.sh, it's failing for me every time :/ | 13:40 |
TheJulia | I guess if devstack-system-admin is not working then it is a bug that needs to be fixed for the service | 13:45 |
TheJulia | slaweq: fwiw, I don't care what we set it to, as long as we move the effort forward, I was only setting stuff to devstack-admin because everything was using the environment variables before hand and the environment variables are toxic to the entire effort | 13:47 |
TheJulia | Which makes me think... could the env vars be in the mix? | 13:47 |
slaweq | TheJulia problem for us is that when new defaults and scopes are enforced, creation of shared subnetpool is available only for system_admin, so system_scope='all' has to be set in the context. Without that neutron returns 403. So I wanted to use devstack-system-admin as this one has system_scope=all configured in the clouds.yaml and it works for me when devstack is deployed, but for some reason not during stack.sh | 13:49 |
slaweq | I will need to dig more into it | 13:50 |
TheJulia | seriously?! | 13:50 |
TheJulia | that sounds like a defect in the policy | 13:51 |
TheJulia | stack.sh *does* have the environment variables loaded | 13:51 |
TheJulia | and any reference to project ID will get transmitted | 13:51 |
TheJulia | which will change the effective scope the policy enforcement views it as | 13:52 |
slaweq | TheJulia if You would have few minutes, can You try to deploy devstack with https://review.opendev.org/c/openstack/devstack/+/797450 ? | 13:53 |
slaweq | Maybe I made some obvious mistake there :) | 13:53 |
TheJulia | I really don't have the spoons or time to smash and rebuild my devstack right now | 13:54 |
TheJulia | but, the environment variables are still there. We ran into that with some ironic commands where we actually needed to unset project id | 13:55 |
slaweq | TheJulia ahh, so it can be some env variable which wasn't cleaned, good idea to check, thx | 13:58 |
kopecmartin | #startmeeting qa | 14:00 |
opendevmeet | Meeting started Tue Jun 22 14:00:34 2021 UTC and is due to finish in 60 minutes. The chair is kopecmartin. Information about MeetBot at http://wiki.debian.org/MeetBot. | 14:00 |
opendevmeet | Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. | 14:00 |
opendevmeet | The meeting name has been set to 'qa' | 14:00 |
kopecmartin | #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_next_Office_hours | 14:00 |
kopecmartin | agenda ^^^ | 14:00 |
kopecmartin | o/ | 14:00 |
TheJulia | slaweq: env vars are global in the stack.sh run | 14:00 |
soniya | hello | 14:02 |
kopecmartin | hi soniya | 14:02 |
yoctozepto | o/ | 14:04 |
jparoly | o/ | 14:04 |
kopecmartin | let's start, let's go through usual agenda first | 14:04 |
kopecmartin | #topic Announcement and Action Item | 14:04 |
kopecmartin | no announcements from my side | 14:04 |
kopecmartin | #topic Xena Priority Items progress | 14:05 |
kopecmartin | #link https://etherpad.opendev.org/p/qa-xena-priority | 14:05 |
kopecmartin | any updates regarding the priority items? | 14:05 |
kopecmartin | none from my side this week unfortunately | 14:05 |
kopecmartin | gmann: i was reviewing your "Use older run-tempest" change .. i have a few questions just to make sure | 14:06 |
kopecmartin | #link https://review.opendev.org/c/openstack/tempest/+/787455 | 14:06 |
kopecmartin | please check whenever you have time | 14:06 |
kopecmartin | if nothing else, let's move on | 14:08 |
kopecmartin | #topic OpenStack Events Updates and Planning | 14:08 |
kopecmartin | nothing here, we can skip | 14:08 |
kopecmartin | #topic Gate Status Checks | 14:08 |
kopecmartin | anything regarding the gates to bring up? | 14:08 |
yoctozepto | nope | 14:09 |
kopecmartin | have you seen something like this? | 14:09 |
kopecmartin | #link https://zuul.opendev.org/t/openstack/build/0231be392bb94c4a8fee4ada8df79ee4 | 14:09 |
kopecmartin | "refusing to convert from file to symlink for /usr/local/bin/tox" | 14:09 |
kopecmartin | ensure-tox role is failing in a new job I'm working on | 14:10 |
yoctozepto | first time I see this message | 14:11 |
kopecmartin | interesting, I didn't change anything close to it .. just copy paste of a standard code we have everywhere | 14:11 |
kopecmartin | yeah | 14:11 |
kopecmartin | #topic Periodic jobs Status Checks | 14:11 |
kopecmartin | #link https://zuul.openstack.org/builds?job_name=tempest-full-victoria-py3&job_name=tempest-full-ussuri-py3&job_name=tempest-full-train-py3&pipeline=periodic-stable | 14:11 |
kopecmartin | #link https://zuul.openstack.org/builds?job_name=tempest-all&job_name=tempest-full-oslo-master&pipeline=periodic | 14:12 |
kopecmartin | seems all good | 14:12 |
kopecmartin | #topic Sub Teams highlights | 14:12 |
kopecmartin | #link https://review.openstack.org/#/q/project:openstack/tempest+status:open | 14:12 |
kopecmartin | #link https://review.openstack.org/#/q/project:openstack/patrole+status:open | 14:12 |
kopecmartin | #link https://review.openstack.org/#/q/project:openstack/devstack+status:open | 14:12 |
kopecmartin | #link https://review.openstack.org/#/q/project:openstack/grenade+status:open | 14:12 |
kopecmartin | #link https://review.opendev.org/#/q/project:openstack/hacking+status:open | 14:12 |
kopecmartin | anything from any subteam? | 14:12 |
kopecmartin | patrole gates still failing, i'm having problems to find time to look into it | 14:13 |
kopecmartin | let's see Changes with Review-Priority == +1 | 14:13 |
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) | 14:14 |
kopecmartin | just one from patrole | 14:14 |
kopecmartin | blocked on the gates | 14:14 |
kopecmartin | #topic Gate Blocker Fix / Urgent Change | 14:14 |
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) | 14:15 |
kopecmartin | No reviews | 14:15 |
kopecmartin | #topic Open Discussion | 14:15 |
kopecmartin | no topics there | 14:15 |
kopecmartin | anyone anything to discuss? | 14:16 |
yoctozepto | too hot to think | 14:16 |
kopecmartin | agree on that :) | 14:16 |
kopecmartin | #topic Bug Triage | 14:18 |
kopecmartin | #link https://etherpad.opendev.org/p/qa-bug-triage-xena | 14:18 |
kopecmartin | i don't have anything specific, numbers are reported there as always ^^ | 14:18 |
kopecmartin | gmann: would you check this one when you have a moment,thanks https://review.opendev.org/c/openstack/stackviz/+/796488 | 14:19 |
kopecmartin | #link https://review.opendev.org/c/openstack/stackviz/+/796488 | 14:19 |
kopecmartin | If there is nothing else from anyone, we can close the office hour | 14:20 |
kopecmartin | and go for an ice cream :D | 14:20 |
yoctozepto | or just hide under a rock | 14:20 |
kopecmartin | no wifi connection there | 14:20 |
kopecmartin | thanks | 14:21 |
kopecmartin | #endmeeting | 14:21 |
opendevmeet | Meeting ended Tue Jun 22 14:21:15 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:21 |
opendevmeet | Minutes: https://meetings.opendev.org/meetings/qa/2021/qa.2021-06-22-14.00.html | 14:21 |
opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/qa/2021/qa.2021-06-22-14.00.txt | 14:21 |
opendevmeet | Log: https://meetings.opendev.org/meetings/qa/2021/qa.2021-06-22-14.00.log.html | 14:21 |
opendevreview | Lukas Piwowarski proposed openstack/tempest master: WIP: Fix test_network_quota_exceeding test https://review.opendev.org/c/openstack/tempest/+/797466 | 14:46 |
opendevreview | Julia Kreger proposed openstack/devstack master: Use specific credentials for tempest plugin setup https://review.opendev.org/c/openstack/devstack/+/780315 | 14:49 |
*** gthiemon1e is now known as gthiemonge | 15:38 | |
opendevreview | Rodolfo Alonso proposed openstack/tempest master: Use "resolvectl dns" to retrieve the DNS nameservers https://review.opendev.org/c/openstack/tempest/+/794776 | 15:57 |
gmann | slaweq: yeah, system scope in devstack need more work. I started for few services but those are in progress | 16:29 |
*** gfidente is now known as gfidente|afk | 16:36 | |
*** rpittau is now known as rpittau|afk | 17:11 | |
*** jpena is now known as jpena|off | 17:32 | |
*** iurygregory_ is now known as iurygregory | 18:39 | |
opendevreview | Merged openstack/devstack master: os_vif: Add support for configuring os_vif_ovs plugin https://review.opendev.org/c/openstack/devstack/+/796826 | 19:03 |
frickler | slaweq: I think something like https://review.opendev.org/c/openstack/devstack/+/780417 is needed to avoid interference between cloud.yaml config and env vars | 19:25 |
opendevreview | Ade Lee proposed openstack/devstack master: WIP/DNM: Add centos-multinode nodeset https://review.opendev.org/c/openstack/devstack/+/797535 | 19:55 |
*** dviroel is now known as dviroel|out | 20:59 | |
TheJulia | gmann: so question comes to mind, how in the world are services with plugins supposed to be able to use the pre-created os_system_* stuffs ? It almost seems like we can't and that we're just going to have to spawn our client attached to it | 21:46 |
TheJulia | looks like keystone had the clients in tempest already so they didn't have to do that, but structurally without being able to dynamically grab the credenitals since cls.os_system_admin and friends are instances of manager classes, individual attachment seems to be the only way :\ | 22:01 |
TheJulia | hmm, that may work actually | 22:06 |
gmann | TheJulia: as plugins service client are also using init via tempest base class then you can access os_system* etc same way you are doing os_admin or so | 22:41 |
gmann | or even better option for plugins is to register the service clients in tempest registry via this interface https://github.com/openstack/tempest/blob/7e96c8e854386f43604ad098a6ec7606ee676145/tempest/test_discover/plugins.py#L121 | 22:41 |
gmann | so that any plugins can access those service client in same way like they are accessing Tempest in-tree service cleint | 22:42 |
TheJulia | gmann: are there actually examples of os_admin, I didn't see any on codesearch | 23:33 |
TheJulia | gmann: to be more precise, in an actually tempest plugin which is not keystone's | 23:33 |
TheJulia | looks like tempest is the only thing leveraging register_service_client | 23:35 |
TheJulia | it looks like maybe service_clients mixed with get_service_clients in plugin.py | 23:42 |
TheJulia | I guess something to look at tomorrow | 23:43 |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!