15:05:07 <enriquetaso> #startmeeting cinder_bs
15:05:07 <opendevmeet> Meeting started Wed Feb  1 15:05:07 2023 UTC and is due to finish in 60 minutes.  The chair is enriquetaso. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:05:07 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:05:07 <opendevmeet> The meeting name has been set to 'cinder_bs'
15:05:26 <whoami-rajat> hi
15:05:27 <enriquetaso> Today we have 4 bugs
15:05:34 <enriquetaso> hello
15:05:40 <enriquetaso> #topic Quotas rollback error when retype failed
15:05:44 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/2004239
15:05:51 <enriquetaso> If admin tenants do retype to a volume owned by other tenants, and it fails for some reason, quota rollback will fail. Because it will try to find reservations with the admin tenant's project_id but not the volume owner's.
15:06:22 <enriquetaso> i forget to ask the cinder version the reporter is using
15:06:31 <enriquetaso> but looks like something cinder should fix soon
15:08:33 <whoami-rajat> maybe it's addressed in geguileo quota work but that might take some time to merge
15:09:40 <enriquetaso> oh, true, i'll check and add a comment for that then
15:09:43 <enriquetaso> moving on
15:09:58 <enriquetaso> #topic  Hitachi: Cinder scheduler is down if backend is down
15:10:02 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/2004140
15:10:15 <enriquetaso> When the backend is down, the Cinder scheduler is also down with a key error,
15:10:15 <enriquetaso> because Hitachi driver doesn't return all required keys in update_volume_stats().
15:10:23 <enriquetaso> Fix proposed to master:
15:10:28 <enriquetaso> #link https://review.opendev.org/c/openstack/cinder/+/871269
15:10:46 <enriquetaso> i left a comment on the patch, needs update
15:10:53 <eharney> if the bug report is correct, we need a fix for the scheduler, not just the hitachi driver, which is what is proposed
15:10:53 <enriquetaso> moving on
15:11:10 <eharney> that fix misses the actual problem
15:11:22 <enriquetaso> good point
15:12:28 <enriquetaso> im not sure how other drivers deal with the same problem
15:12:52 <enriquetaso> if confirmed, we would need a new bug report for the scheduler along
15:13:43 <enriquetaso> i'll left a comment and try to reproduce if nobody volunteers to do it
15:17:25 <enriquetaso> okay.. last two bugs
15:17:40 <enriquetaso> #topic  [doc] Update 'Introduction to the Block Storage service in cinder'
15:17:44 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/2004419
15:17:51 <enriquetaso> The documentation states that "With the Block Storage service, you can attach a device to only one instance." However, Cinder multiattach was introduced in Queens, as described here https://docs.openstack.org/cinder/latest/admin/volume-multiattach.html. Please update the documentation to reflect that multiattach is possible in some scenarios.
15:18:03 <enriquetaso> This could be a potential intern first contribution for Outreachy potential intern
15:18:17 <rosmaita> good idea!
15:18:54 <whoami-rajat> +1
15:19:05 <enriquetaso> :D cool
15:19:07 <enriquetaso> last one
15:19:16 <enriquetaso> #topic Replace plain Mock() calls with autospec=True or Mock(spec)
15:19:21 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/2004174
15:19:43 <enriquetaso> this is intended for potential Outreachy interns too
15:19:48 <enriquetaso> Cinder has some tests with assertion typo's or plain calls to Mock() or MagicMock() resulting in those tests not testing code correctly.
15:19:55 <enriquetaso> "The refactored code would replace plain Mock() calls with Mock(spec) calls
15:19:55 <enriquetaso> or you can either pass autospec=True to patch() / patch.object() or use the create_autospec() function to create a mock with a spec. We want stronger mocks by using specs, do the right thing where needed to get this."
15:21:14 <enriquetaso> okay!
15:21:15 <eharney> i have one other bug to mention
15:21:18 <opendevreview> Merged openstack/cinder stable/victoria: Check VMDK subformat against an allowed list  https://review.opendev.org/c/openstack/cinder/+/871628
15:21:23 <enriquetaso> sure eharney
15:21:35 <opendevreview> Merged openstack/cinder master: [Pure Storage] Add support for 3-site, trisync, replication.  https://review.opendev.org/c/openstack/cinder/+/862365
15:21:40 <eharney> i just wrote https://bugs.launchpad.net/cinder/+bug/2004484 - rosmaita might be interested in this one
15:21:53 <eharney> tox4 isn't showing useful output that tox3 did, we should fix it somehow
15:21:53 <enriquetaso> #topic  list dependencies with tox4
15:22:00 <eharney> because this is annoying
15:22:00 <enriquetaso> #link https://bugs.launchpad.net/cinder/+bug/2004484
15:22:34 <enriquetaso> sounds fine
15:22:41 <eharney> previously we logged all packages used in the env, now it doesn't unless you pass in an env variable
15:22:46 <rosmaita> we do have pip freeze output on the ci jobs, though
15:22:54 <eharney> oh, well that's good i guess
15:23:14 <rosmaita> i'm just being a jerk because i didn't think of it
15:23:24 <enriquetaso> like tox -e pipfreeze?
15:23:44 <rosmaita> no, more like .tox/envname/bin/pip freeze
15:23:52 <rosmaita> use the pip from the testenv
15:24:04 <eharney> i didn't see an obvious way to set something in tox.ini to just make it do this, but it would be good for local developers to have it IMO
15:24:19 <rosmaita> eharney: are you talking about the list of stuff that's all in one line just before the tests run?
15:24:28 <eharney> rosmaita: yeah
15:24:47 <rosmaita> ok, i always ignore it as hard to read and grep the pip freeze output
15:25:10 <rosmaita> yeah, if you find the output useful, i don't object to adding the CI=1
15:25:27 <eharney> right, but i don't know how to add that for the local developer case :)
15:26:03 <rosmaita> i think just in the setenv= in the testenv definition?
15:26:25 <rosmaita> i can try it out later if you don't have time
15:26:25 <eharney> i think that sets it too late for it to do anything
15:26:30 <eharney> i tested it and didn't work, anyway
15:26:51 <rosmaita> ok, well i am all out of ideas then
15:26:52 <eharney> anyway, i wrote a bug because i wanted to do something else today, no need to figure it out now
15:27:12 <rosmaita> ok, cool
15:27:16 <rosmaita> i have one
15:27:27 <rosmaita> https://review.opendev.org/c/openstack/cinder/+/871976
15:27:40 <rosmaita> enriquetaso: i asked on the patch to file a new bug
15:27:54 <rosmaita> it can be a public security bug, it's a "hardening opportunity"
15:27:54 <enriquetaso> oh okay
15:28:07 <rosmaita> just want to give you a heads-up so you don't freak out when you see it
15:28:15 <eharney> looks like a nice hardening patch
15:28:17 <enriquetaso> thanks rosmaita
15:28:40 <enriquetaso> sounds good then
15:28:47 <rosmaita> yeah, i think it's worth doing, just object to it being snuck in as related to the latest CVE bug
15:29:09 <enriquetaso> okay, do we have more bugs?
15:29:26 <enriquetaso> let's close for today's
15:29:33 <enriquetaso> thank you for joining!!
15:29:36 <rosmaita> thanks enriquetaso
15:29:39 <enriquetaso> #endmeeting