| *** dviroel is now known as dviroel|out | 00:15 | |
| opendevreview | Lee Yarwood proposed openstack/devstack master: WIP nova: Ensure each compute uses a unique iSCSI initiator https://review.opendev.org/c/openstack/devstack/+/812391 | 07:17 |
|---|---|---|
| *** jpena|off is now known as jpena | 07:33 | |
| opendevreview | Christian Schwede proposed openstack/tempest master: Improve cleanup after Swift testing https://review.opendev.org/c/openstack/tempest/+/812455 | 07:54 |
| opendevreview | Christian Schwede proposed openstack/tempest master: Improve cleanup after Swift testing https://review.opendev.org/c/openstack/tempest/+/812455 | 08:26 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Add microversion schemas for os-instance-actions https://review.opendev.org/c/openstack/tempest/+/812471 | 09:39 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Test Network QoS minimum packet rate scenarios https://review.opendev.org/c/openstack/tempest/+/806257 | 09:39 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Remove race from wait_for_interface_detach waiter https://review.opendev.org/c/openstack/tempest/+/812472 | 09:39 |
| opendevreview | Lee Yarwood proposed openstack/devstack master: nova: Ensure each compute uses a unique iSCSI initiator https://review.opendev.org/c/openstack/devstack/+/812391 | 09:46 |
| opendevreview | Lee Yarwood proposed openstack/devstack master: nova: Ensure each compute uses a unique iSCSI initiator https://review.opendev.org/c/openstack/devstack/+/812391 | 10:36 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Remove race from wait_for_interface_detach waiter https://review.opendev.org/c/openstack/tempest/+/812472 | 10:40 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Test Network QoS minimum packet rate scenarios https://review.opendev.org/c/openstack/tempest/+/806257 | 10:40 |
| opendevreview | Christian Schwede proposed openstack/tempest master: Improve cleanup after Swift testing https://review.opendev.org/c/openstack/tempest/+/812455 | 11:02 |
| *** jpena is now known as jpena|lunch | 11:24 | |
| opendevreview | Lukas Piwowarski proposed openstack/tempest master: Limit jsonschema package to versions older than 4.2.0 https://review.opendev.org/c/openstack/tempest/+/812487 | 11:52 |
| *** dviroel|out is now known as dviroel | 12:07 | |
| *** jpena|lunch is now known as jpena | 12:22 | |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Remove race from wait_for_interface_detach waiter https://review.opendev.org/c/openstack/tempest/+/812472 | 12:50 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Test Network QoS minimum packet rate scenarios https://review.opendev.org/c/openstack/tempest/+/806257 | 12:50 |
| *** artom_ is now known as artom | 13:18 | |
| opendevreview | Slawek Kaplonski proposed openstack/devstack master: Install OVS from source when it was configured like that https://review.opendev.org/c/openstack/devstack/+/799621 | 13:36 |
| *** artom_ is now known as artom | 13:55 | |
| opendevreview | Michal proposed openstack/devstack master: tools: Fix use of continue https://review.opendev.org/c/openstack/devstack/+/812507 | 13:57 |
| kopecmartin | #startmeeting qa | 14:00 |
| opendevmeet | Meeting started Tue Oct 5 14:00:57 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 |
| gmann | o/ | 14:01 |
| kopecmartin | #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Agenda_for_next_Office_hours | 14:01 |
| soniya29 | o/ | 14:01 |
| kopecmartin | agenda ^^ | 14:01 |
| kopecmartin | hi o/ | 14:01 |
| kopecmartin | let's start | 14:02 |
| kopecmartin | #topic Announcement and Action Item (Optional) | 14:02 |
| kopecmartin | we are done with the releases | 14:02 |
| kopecmartin | i think that's it regarding the announcements | 14:03 |
| yoctozepto | o/ | 14:03 |
| kopecmartin | #topic Xena Priority Items progress | 14:03 |
| kopecmartin | any progress on this front? | 14:04 |
| kopecmartin | #link https://etherpad.opendev.org/p/qa-xena-priority | 14:04 |
| gmann | nothing from me this week. | 14:04 |
| soniya29 | i am very much stuck over certain patrole issues | 14:04 |
| kopecmartin | nothing from my side neither | 14:05 |
| kopecmartin | #topic OpenStack Events Updates and Planning | 14:05 |
| kopecmartin | #link https://etherpad.opendev.org/p/qa-yoga-ptg | 14:05 |
| kopecmartin | PTG is in 2 weeks | 14:05 |
| kopecmartin | feel free to add your proposals to the etherpad above , if you haven't already | 14:05 |
| kopecmartin | we have booked the following slots: 13-15 UTC October 18th and 15-17 UTC October 19th | 14:06 |
| kopecmartin | for PTG sessions | 14:06 |
| kopecmartin | #topic Gate Status Checks | 14:06 |
| 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) | 14:07 |
| kopecmartin | any issues to bring up? | 14:07 |
| kopecmartin | I don't see any urgent patches | 14:07 |
| kopecmartin | i have noticed refstack projects started failing on jsonschema - the new version released lately - 4.0.1 | 14:08 |
| kopecmartin | it is installed by tempest (all refstack projects inherit from it) | 14:08 |
| gmann | yeah, nothing urgent on gate afaik | 14:08 |
| kopecmartin | but we're currently investigating options | 14:08 |
| kopecmartin | apart from that i haven't noticed anything else | 14:08 |
| kopecmartin | good | 14:08 |
| kopecmartin | #topic Periodic jobs Status Checks | 14:09 |
| 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 | 14:09 |
| kopecmartin | #link https://zuul.openstack.org/builds?project=openstack%2Ftempest&project=openstack%2Fdevstack&pipeline=periodic | 14:09 |
| kopecmartin | all green | 14:09 |
| kopecmartin | #topic Sub Teams highlights | 14:10 |
| kopecmartin | any updates, highlights we need to discuss today ? | 14:10 |
| kopecmartin | Changes with Review-Priority == +1 | 14:10 |
| 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) | 14:10 |
| kopecmartin | there are none | 14:10 |
| yoctozepto | there is open discussion point by me afair | 14:11 |
| kopecmartin | yup, let's get to it | 14:11 |
| kopecmartin | #topic Open Discussion | 14:11 |
| soniya29 | kopecmartin, gmann, we have certain api listed here - https://docs.openstack.org/api-ref/network/v2/index.html? | 14:11 |
| kopecmartin | soniya29: what about them? | 14:12 |
| gmann | which one we are discussing ? | 14:13 |
| soniya29 | where some apis lack neutron policy implementation, so my question, in such case, do we need to implement the policy first? | 14:13 |
| soniya29 | gmann, kopecmartin, https://review.opendev.org/c/openstack/patrole/+/807131 | 14:13 |
| gmann | soniya29: ohk, in that case we need to disucss with neutron team if those are missing or not needed | 14:13 |
| kopecmartin | yup, let's ask the neutron team | 14:15 |
| soniya29 | gmann, okay, if they require extension i can help them out there otherwise we can remove task from storyboard | 14:15 |
| gmann | soniya29: kopecmartin replied, this is extension https://github.com/openstack/neutron-lib/blob/master/neutron_lib/api/definitions/logging_resource.py | 14:15 |
| gmann | we are just not using the correct name. | 14:15 |
| soniya29 | gmann, i would correct it | 14:16 |
| gmann | and more importantly we need to tests 1. if test running 2. test skipping on enabling the extension with list not with 'All' | 14:16 |
| gmann | because with 'All' which is default will run tests even you have wring extensions name. | 14:16 |
| yoctozepto | (let me know when I can go ahead with "test with MariaDB rather than MySQL (yoctozepto, frickler)") | 14:17 |
| soniya29 | gmann, okay | 14:18 |
| kopecmartin | thanks gmann for the info and comments | 14:18 |
| kopecmartin | yoctozepto: you may go ahead :) | 14:18 |
| yoctozepto | ok | 14:19 |
| yoctozepto | #subtopic test with MariaDB rather than MySQL (yoctozepto, frickler) | 14:19 |
| yoctozepto | so, as the topic says - could we switch the default database to MariaDB? | 14:19 |
| yoctozepto | it seems it's the most common option chosen by deployment projects; yet devstack defaults to MySQL | 14:20 |
| yoctozepto | so far I have only identified openstack charms to use mysql | 14:20 |
| yoctozepto | all others use mariadb | 14:20 |
| kopecmartin | we're talking only about changing the default option, right? | 14:20 |
| yoctozepto | right | 14:21 |
| frickler | yoctozepto: do you use special repos for mariadb? we might want to mirror those in infra | 14:21 |
| frickler | with "you" = kolla | 14:22 |
| yoctozepto | frickler: in kolla we do; we are talking devstack here which could at least use mariadb from distro (like we are doing for debian now) | 14:22 |
| yoctozepto | yeah, I figured :-) | 14:22 |
| yoctozepto | it's a good point to improve kolla's use of mirrors ;-) | 14:22 |
| yoctozepto | but let's focus on devstack in this very discussion | 14:22 |
| frickler | yes, I was thinking about where devstack would install mariadb from | 14:23 |
| frickler | if you think distro versions are good enough, I'm fine with that, at least for the initial step | 14:23 |
| yoctozepto | I think we can rely on distros for this one; but I'm open to going upstream | 14:23 |
| yoctozepto | ok | 14:23 |
| yoctozepto | I think OSA also uses upstream | 14:24 |
| yoctozepto | but would have to check | 14:24 |
| yoctozepto | yup, they seem to | 14:25 |
| yoctozepto | https://opendev.org/openstack/openstack-ansible-galera_server/src/branch/master/vars for curious here | 14:26 |
| yoctozepto | so yeah, two official deployments projects on upstream mariadb | 14:26 |
| yoctozepto | tripleo uses rdo-provided one | 14:26 |
| yoctozepto | so kinda-distro-but-not-quite | 14:26 |
| kopecmartin | cool, i'm ok with aligning with the other projects on this one | 14:28 |
| kopecmartin | are there any additional changes required? or is it simple as just changing the default option? | 14:28 |
| yoctozepto | Helm also uses upstream it seems, but container image of course | 14:28 |
| yoctozepto | kopecmartin: changing it is simple; there is one quirk | 14:28 |
| yoctozepto | in that neutron tests with special mariadb job | 14:29 |
| yoctozepto | which will now not be different | 14:29 |
| yoctozepto | wonder if they want to test mysql per se | 14:29 |
| yoctozepto | probably not :D | 14:29 |
| yoctozepto | so can just drop that job | 14:29 |
| opendevreview | Christian Schwede proposed openstack/tempest master: Improve cleanup after Swift testing https://review.opendev.org/c/openstack/tempest/+/812455 | 14:31 |
| kopecmartin | good, anything else to add to this? | 14:31 |
| kopecmartin | anything else for the open discussion? | 14:31 |
| yoctozepto | no, I'm good | 14:31 |
| yoctozepto | we agreed to switch, right? :P | 14:31 |
| kopecmartin | yes | 14:31 |
| yoctozepto | gmann? | 14:31 |
| yoctozepto | ok | 14:32 |
| gmann | sorry, i was in another discussion. | 14:32 |
| gmann | do we have any job testing that currently ? | 14:32 |
| gmann | or we are proposing changing default directly | 14:32 |
| yoctozepto | I guess we merge after the xena release and hope for the best (-: | 14:33 |
| yoctozepto | neutron tests with mariadb | 14:33 |
| gmann | I thinkwe have once for postgresql | 14:33 |
| slaweq | yoctozepto: are You talking about our periodic job with mariadb? | 14:33 |
| yoctozepto | slaweq: yeah, we will want to drop it after we change the default | 14:34 |
| gmann | https://github.com/openstack/tempest/blob/a7bcabc8976e6e646d5e4379e3289b43586261c1/zuul.d/integrated-gate.yaml#L284 | 14:34 |
| slaweq | yoctozepto: it would be fine for me | 14:34 |
| yoctozepto | slaweq: ack, thanks | 14:34 |
| slaweq | we added that periodic job actually because default was still mysql | 14:34 |
| gmann | yoctozepto: yeah, after Xena release of course. | 14:34 |
| slaweq | and once we had some issue with mariadb and some db migrations (I don't remember exactly now what it was) | 14:35 |
| gmann | or may be we can add a job and seeing all good and working then change default. yoctozepto frickler ? | 14:35 |
| yoctozepto | slaweq: yeah, those were reported by me :-) | 14:35 |
| slaweq | so to avoid such issues we added one periodic job to test explicitly with mariadb | 14:35 |
| slaweq | ahh, ok :) | 14:35 |
| slaweq | too many bugs to remember for me :) | 14:35 |
| yoctozepto | gmann: we know it works on ubuntu, just can't tell if for all the projects | 14:35 |
| gmann | yeah, for all projects/configuration | 14:35 |
| yoctozepto | I say we change and await feedback, most effective :-) | 14:36 |
| gmann | and meanwhile we can start the ML about it and how other feel though it is configurable to any DB | 14:36 |
| yoctozepto | yeah, but sending ML notice is nice | 14:36 |
| gmann | and in PTG which is not so far we can see if anyone object on that | 14:37 |
| yoctozepto | yeah, that seems most reasonable, thanks gmann for your input | 14:37 |
| gmann | it should not be issue though we we do not test DB itself just any workable we use in devstack. | 14:37 |
| gmann | yoctozepto: thanks for initiating on this. +1 | 14:38 |
| yoctozepto | :-) | 14:38 |
| frickler | actually the interesting bugs will likely only be seen in a multinode galera setup | 14:40 |
| * yoctozepto wonders if us switching to MariaDB will invalidate the "the most popular RDBMS" slogan for MySQL ;p | 14:40 | |
| yoctozepto | frickler: well, that neutron one was regardless of galera | 14:40 |
| yoctozepto | anyhow, galera is not a concern for devstack | 14:41 |
| yoctozepto | we already use mariadb on centos jobs and will on debian | 14:41 |
| yoctozepto | and now will switch to mariadb on ubuntu | 14:42 |
| yoctozepto | makes sense to me | 14:42 |
| gmann | +1 | 14:42 |
| yoctozepto | shameless plug: cores, please review adding debian testing https://review.opendev.org/c/openstack/devstack/+/789083 | 14:43 |
| yoctozepto | :-) | 14:43 |
| kopecmartin | i'm sorry, i had to jump on a different meeting | 14:46 |
| kopecmartin | anything else for the discussion? | 14:46 |
| kopecmartin | #topic Bug Triage | 14:47 |
| kopecmartin | #link https://etherpad.opendev.org/p/qa-bug-triage-xena | 14:47 |
| kopecmartin | i recorded the numbers as always ^^ | 14:47 |
| kopecmartin | that's all from my side , so if there isn\t anything else | 14:48 |
| kopecmartin | i think we can close the office hour | 14:48 |
| kopecmartin | thank you all for the great discussion today | 14:48 |
| kopecmartin | #endmeeting | 14:48 |
| gmann | kopecmartin: seems like end meeting did not work? | 14:49 |
| kopecmartin | interesting | 14:50 |
| kopecmartin | lol | 14:50 |
| kopecmartin | #endmeeting | 14:50 |
| opendevmeet | Meeting ended Tue Oct 5 14:50:13 2021 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot . (v 0.1.4) | 14:50 |
| opendevmeet | Minutes: https://meetings.opendev.org/meetings/qa/2021/qa.2021-10-05-14.00.html | 14:50 |
| opendevmeet | Minutes (text): https://meetings.opendev.org/meetings/qa/2021/qa.2021-10-05-14.00.txt | 14:50 |
| opendevmeet | Log: https://meetings.opendev.org/meetings/qa/2021/qa.2021-10-05-14.00.log.html | 14:50 |
| gmann | wow :) | 14:50 |
| kopecmartin | ah a space before the "#" | 14:50 |
| kopecmartin | maybe | 14:50 |
| gmann | ahh | 14:50 |
| yoctozepto | xD | 14:52 |
| yoctozepto | thanks kopecmartin | 14:52 |
| frickler | yoctozepto: btw., not sure if you saw my comments about debian memory usage yesterday in #-infra. | 15:03 |
| frickler | seems that qemu has roughly 2x the rss footprint in debian compared to ubuntu | 15:04 |
| yoctozepto | frickler: I did not see them; that's very... interesting and infuriating I guess? | 15:05 |
| yoctozepto | what does it need it for | 15:05 |
| frickler | sean suggested checking the machine type, but it is pc-i440fx-4.2 vs. pc-i440fx-5.2 , so I don't expect such a large delta from that | 15:05 |
| yoctozepto | yeah, the delta sounds absurd | 15:06 |
| frickler | one idea would be the memory-backend=pc.ram option, I guess that is added by libvirt, not nova | 15:06 |
| frickler | let me put those into a paste and possibly move to #-nova actually | 15:06 |
| clarkb | is same page merging enabled on ubuntu but not debian? | 15:06 |
| clarkb | that might explain it too? | 15:06 |
| yoctozepto | also, centos has 5.1, debian has 5.2 so no kidding | 15:07 |
| frickler | https://paste.opendev.org/show/809797/ | 15:08 |
| yoctozepto | what does memory-backend=pc.ram do? | 15:08 |
| yoctozepto | is not it somehow a default? | 15:08 |
| frickler | I tried to find that out, but didn't find a proper doc | 15:08 |
| yoctozepto | eh, qemu and their docs | 15:09 |
| frickler | there is some docs about a memory-file backend. which seems to be kind of file backed memory | 15:09 |
| yoctozepto | https://wiki.qemu.org/ChangeLog/5.0 | 15:10 |
| frickler | so possibly this is memory-backed memory | 15:10 |
| yoctozepto | yeah, this sounds like an explicit of what was implicit previously | 15:10 |
| yoctozepto | but mayhaps it might be reserving mem twice? | 15:10 |
| yoctozepto | due to some misinterpretation | 15:10 |
| frickler | I'll ask in nova now | 15:12 |
| yoctozepto | ok | 15:12 |
| yoctozepto | I'll continue there too | 15:12 |
| *** whoami-rajat__ is now known as whoami-rajat | 15:13 | |
| opendevreview | James Parker proposed openstack/whitebox-tempest-plugin master: Add AMD Sev instance creation test https://review.opendev.org/c/openstack/whitebox-tempest-plugin/+/809441 | 15:21 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Remove race from wait_for_interface_detach waiter https://review.opendev.org/c/openstack/tempest/+/812472 | 16:17 |
| opendevreview | Balazs Gibizer proposed openstack/tempest master: Test Network QoS minimum packet rate scenarios https://review.opendev.org/c/openstack/tempest/+/806257 | 16:17 |
| *** jpena is now known as jpena|off | 16:32 | |
| opendevreview | Slawek Kaplonski proposed openstack/devstack master: Install OVS from source when it was configured like that https://review.opendev.org/c/openstack/devstack/+/799621 | 18:46 |
| opendevreview | Slawek Kaplonski proposed openstack/devstack master: Don't fail if there is no nf_conntrack_proto_gre module available https://review.opendev.org/c/openstack/devstack/+/812597 | 18:46 |
| kopecmartin | gmann: do we need jsonschema here? maybe we can just check whether the instance is an instance of str? https://opendev.org/openstack/tempest/src/commit/a7bcabc8976e6e646d5e4379e3289b43586261c1/tempest/lib/common/jsonschema_validator.py#L34 | 19:15 |
| kopecmartin | more context here: https://review.opendev.org/c/openstack/tempest/+/812487 | 19:15 |
| kopecmartin | in my comment there | 19:15 |
| gmann | kopecmartin: checking | 19:30 |
| gmann | kopecmartin: but we need it here for sure https://opendev.org/openstack/tempest/src/commit/367e148d439042ab43ea3fb610d7f6d456b89b47/tempest/lib/common/rest_client.py#L927 | 19:34 |
| gmann | kopecmartin: commented, I think for newer tempest we can update the tempest code as per new jsonschwma | 19:35 |
| kopecmartin | gmann: thanks .. yeah , we need that, i'm questioning this particular call - https://opendev.org/openstack/tempest/src/branch/master/tempest/lib/common/jsonschema_validator.py#L34 | 19:54 |
| kopecmartin | was its purpose just to make sure instance is a string? | 19:54 |
| gmann | kopecmartin: not sure, its very old code so not remember why I am checking str_type and not validate ISO 8601 format. but we can remove it and let oslo return value error if not string. | 19:59 |
| gmann | we can remove that while moving to latest jsonschema | 20:00 |
| lyarwood | https://review.opendev.org/c/openstack/devstack/+/812391 - Looking for devstack core reviews on this bugfix to unblock live migration testing in Nova and likely multinode attached volume testing elsewhere if anyone has time. | 20:16 |
| *** dviroel is now known as dviroel|out | 20:55 | |
Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!