14:00:53 <whoami-rajat> #startmeeting cinder
14:00:53 <opendevmeet> Meeting started Wed Dec 20 14:00:53 2023 UTC and is due to finish in 60 minutes.  The chair is whoami-rajat. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:53 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:53 <opendevmeet> The meeting name has been set to 'cinder'
14:00:57 <whoami-rajat> #topic roll call
14:01:31 <simondodsley> o/
14:02:05 <rosmaita> o/
14:02:41 <akawai> o/
14:02:45 <whoami-rajat> #link https://etherpad.opendev.org/p/cinder-caracal-meetings
14:03:56 <whoami-rajat> very few people around today
14:04:06 <whoami-rajat> which gives a good seg-way to our first announcement
14:04:11 <whoami-rajat> #topic announcement
14:04:20 <whoami-rajat> first, Canceling next week's meeting (27 December)
14:04:35 <whoami-rajat> since it's Year end break for many team members, we will be canceling next week's meeting
14:04:47 <whoami-rajat> i will send a mail after the meeting to ML
14:04:58 <whoami-rajat> next, Release stable branches
14:05:04 <whoami-rajat> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/EK2OQUO26ITYPYFYIVFSYM5CR2TVCZJL/
14:05:11 <tkajinam> o/
14:05:33 <whoami-rajat> it's been some time we entered 2024.1 development cycle
14:05:48 <whoami-rajat> i think it's a good time to EM yoga at this point and release other stable branches
14:05:55 <whoami-rajat> i will talk to Jon regarding this
14:06:06 <whoami-rajat> next, Python 3.12 issues and classes of bugs
14:06:14 <whoami-rajat> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/6QRQTJQF3KCJDYEYN6BVTUMFGXU7DTHW/
14:06:47 <whoami-rajat> so someone ran UTs with py3.12 in debian and found a bunch of failures all around openstack
14:06:49 <whoami-rajat> There are two bugs related to Cinder
14:06:49 <whoami-rajat> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1031057
14:06:49 <whoami-rajat> https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=993311
14:07:05 <whoami-rajat> the full list is here
14:07:08 <whoami-rajat> #link https://bugs.debian.org/cgi-bin/pkgreport.cgi?which=maint&data=team%2Bopenstack%40tracker.debian.org&archive=no&raw=yes&bug-rev=yes&pend-exc=fixed&pend-exc=done
14:07:35 <whoami-rajat> i guess we have time so we will deal with it after the break when the team is back
14:07:43 <whoami-rajat> next, Upcoming Deadlines & Dates
14:07:48 <whoami-rajat> Cinder spec freeze: December 22nd, 2023
14:07:49 <whoami-rajat> Caracal-2 Milestone: January 11th, 2024
14:07:57 * jungleboyj is lurking
14:08:44 <whoami-rajat> there is one spec that was discussed as a potential one for this cycle
14:08:45 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder-specs/+/862601
14:09:09 <whoami-rajat> sadly in the recent update, no review comments were addressed so I've left a -1
14:09:41 <whoami-rajat> there is a lot to address actually, 24 comments, I'm not sure if it will make it in time (in 2 days)
14:09:45 <whoami-rajat> but let's see
14:10:03 <whoami-rajat> okay, final announcement for today
14:10:05 <whoami-rajat> cinderlib status
14:10:11 <whoami-rajat> and i think it's added by rosmaita
14:10:28 <rosmaita> yes, sorry i have not filled in the info
14:10:58 <rosmaita> first thing is that the governance patch has merged, so we are OK stopping cinderlib development
14:11:08 <rosmaita> next thing is to merge some patches that clean things up
14:11:21 <rosmaita> https://review.opendev.org/q/topic:%22deprecate-cinderlib%22
14:11:40 <rosmaita> i will try to explain what's happening with those, because you can see they are all over the place
14:12:04 <rosmaita> i think the most important ones right now are these:
14:12:21 <rosmaita> patch: https://review.opendev.org/c/openstack/cinder/+/903263
14:12:21 <rosmaita> what: removes cinderlib testing from some jobs defined in cinder repo (but not the ones defined in cinder-tempest-plugin)
14:12:21 <rosmaita> who approves: cinder team
14:12:25 <rosmaita> and
14:12:35 <rosmaita> patch: https://review.opendev.org/c/openstack/cinder-tempest-plugin/+/903262
14:12:35 <rosmaita> what: stops testing cinderlib with master
14:12:35 <rosmaita> who approves: cinder team
14:13:00 <rosmaita> i don't know what will happen to our gate if we delete cinderlib master before merging ^^
14:13:15 <rosmaita> technically, we won't delete the master branch, just most of the stuff in it
14:13:33 <rosmaita> that's https://review.opendev.org/c/openstack/cinderlib/+/903264 if you are interested
14:13:59 <rosmaita> 903264 can't be merged until the infra team merges https://review.opendev.org/c/openstack/project-config/+/903260
14:14:51 <rosmaita> 903260 turns off zuul testing of cinderlib (just cinderlib) completely
14:15:20 <rosmaita> so i think what we need to do is merge https://review.opendev.org/c/openstack/cinder/+/903263 and https://review.opendev.org/c/openstack/cinder-tempest-plugin/+/903262 as soon as we can
14:15:59 <rosmaita> there isn't a dependency between the 2
14:16:07 <rosmaita> but we need them both
14:16:49 <rosmaita> i just did a recheck on 903262, zuul was complaining that the governance patch had not merged yet
14:16:57 <rosmaita> so hopefully it will be green
14:18:54 <whoami-rajat> thanks for working on this, so from cinder side we can focus on 3 patches
14:18:55 <rosmaita> ok, thanks to whoever is adding the notes in the agenda etherpad
14:19:08 <whoami-rajat> 1 in cinder, 1 in cinderlib and 1 in cinder-tempest-plugin
14:19:15 <tkajinam> feel free to add anything missing.
14:19:31 <tkajinam> (I saw someone already did
14:19:48 <rosmaita> the only one missing is an infra patch that handles the documentation redirection, but i don't think we need to worry about that
14:20:06 <rosmaita> that's https://review.opendev.org/c/openstack/openstack-manuals/+/903261
14:20:23 <rosmaita> i need to work with the infra team to determine exactly what's needed on that one
14:20:45 <rosmaita> (it has a -2 right now)
14:21:04 <rosmaita> but it doesn't impact any of the important stuff, which is getting our cinder gates in order
14:22:29 <rosmaita> ok, that's all from me, i guess what would be good would be cinder-core volunteers to watch the 2 key patches
14:22:49 <rosmaita> to be clear, those are:
14:22:58 <rosmaita> https://review.opendev.org/c/openstack/cinder/+/903263
14:22:58 <rosmaita> https://review.opendev.org/c/openstack/cinder-tempest-plugin/+/903262
14:23:46 <rosmaita> one problem is that since i'm the author, i probably shouldn't +2 my own work
14:24:24 <rosmaita> but i will be around all day today, if anyone has questions about what's going on in those patches
14:24:36 <rosmaita> the cinder one is pretty straightforward
14:24:51 <whoami-rajat> I've reviewed the cinder patch but i think Gorka should also take a look (and be the second reviewer)
14:24:56 <rosmaita> and it only affects master, so cinder in the stable branches will still continue to test with cinderlib in the stable branches
14:26:11 <rosmaita> the cinder-tempest-plugin one is a little more complicated, but conceptually, it builds on stuff already in the cinder-tempest-plugin .zuul.yaml that creates a base lvm-barbican job and then creates child jobs for different branches
14:26:24 <whoami-rajat> quite optimistic to mention that we will reach 9999.2 cycle
14:26:32 <rosmaita> :D
14:26:36 <rosmaita> i couldn't resist
14:26:42 <whoami-rajat> :D
14:26:57 <tkajinam> lol
14:27:05 <rosmaita> also, we can make that change once no cinderlib stable branches are supported/unmaintained
14:27:14 <rosmaita> so around 2030 or so
14:27:47 <happystacker> lol
14:27:49 <rosmaita> ok, thanks everyone, that's all from me
14:28:17 <whoami-rajat> thanks again rosmaita !
14:28:40 <rosmaita> btw, does anyone want a link to test RE2 expressions?
14:28:53 <rosmaita> (to review the cinder-tempest-plugin patch)
14:29:23 <whoami-rajat> please go ahead
14:29:40 <rosmaita> this one seems pretty good:
14:29:43 <rosmaita> #link https://www.regexplanet.com/advanced/golang/index.html
14:30:31 <rosmaita> i'm pretty sure go uses RE2
14:30:37 <rosmaita> (at least i hope so!)
14:31:15 <whoami-rajat> yeah i was confused as it said golang but looks like a good tool
14:32:11 <whoami-rajat> okay moving on to topics
14:32:16 <whoami-rajat> #topic Migration from Eventlet to AsyncIO
14:32:26 <whoami-rajat> i don't see Jon around
14:32:43 <whoami-rajat> i think he was planning to work on the eventlet issue
14:32:53 <whoami-rajat> we already have a community goal for that in place
14:32:58 <whoami-rajat> though I haven't gone through it
14:33:13 <whoami-rajat> let's see if we have project specific work items
14:33:24 <whoami-rajat> #link https://review.opendev.org/c/openstack/governance/+/902585
14:33:36 <rosmaita> (it's still under review ^^)
14:35:15 <whoami-rajat> ok, so Jon can take a look at that
14:35:25 <whoami-rajat> i think that's all for his topic?
14:36:39 <zaitcev> I see a big problem with that review
14:37:08 <zaitcev> It approaches the problem by trying to highlight the issue and set a general direction.
14:37:18 <zaitcev> That's not going to yield a result
14:37:40 <zaitcev> What we need instead is a guy to sit on the problem and deep-think about it, then implement a solution.
14:38:36 <zaitcev> It's going to happen anyway, no matter what that review says and regardless of Herve's efforts. Only we aren't helping it along, except in a most vague way by being supportive.
14:39:25 <zaitcev> But I'm not volunteering, so I continue playing in that charade by reviewing and discussing.
14:39:53 <whoami-rajat> if the spec doesn't lay out a high level plan on how to tackle the issue, i see that as a problem for the person going to implement it, though i would need to go through the spec first to comment on it
14:40:10 <whoami-rajat> but thanks for looking into it zaitcev
14:41:01 <whoami-rajat> i think that's all for this, moving on to next topic
14:41:04 <whoami-rajat> #topic Storpool driver discussion
14:41:40 <whoami-rajat> so during the midcycle, we discussed about the design of storpool reporting multiple pools under a backend
14:41:47 <whoami-rajat> which seems to be an acceptable behavior
14:42:01 <sp-bmilanov> hi all, I think Peter won't be able to make it to discuss this, but he wanted me to tell you that he says he has no objections to Rajat's suggestion of sending a second query without the host match, but for the same backend, but he is not sure exactly which cinder.db method to use for that, and he has not had the chance to try to experiment with such a second query and see whether it will help the StorPool "create a
14:42:01 <sp-bmilanov> from a Glance image" case
14:42:45 <whoami-rajat> sp-bmilanov, no worries, i added the topic to share an update
14:43:09 <whoami-rajat> i discussed the scenario with Gorka and he agrees that reporting the cross pool cloning as a capability is a viable option
14:43:37 <whoami-rajat> regarding simondodsley's concern about advertising it in the support matrix, we can remove that if storpool team is fine with it
14:44:09 <sp-bmilanov> I will check with Peter
14:44:14 <whoami-rajat> so we will have a capability reported by storpool driver that will allow it to perform the desired operation
14:44:24 <whoami-rajat> at the same time we won't be adding it as a feature in the support matrix
14:44:33 <whoami-rajat> sp-bmilanov, sure
14:45:15 <whoami-rajat> that's all from my side on this
14:45:15 <sp-bmilanov> whoami-rajat: thanks for making time for this
14:46:08 <whoami-rajat> np, thanks for working on this
14:46:13 <whoami-rajat> okay, next topic
14:46:15 <whoami-rajat> #topic NFS online volume extend
14:46:23 <whoami-rajat> it was again added by me
14:47:11 <whoami-rajat> so the author is working on adding a new API in cinder that nova will call for the nfs drivers to report success failure of extend operation from nova side
14:47:40 <whoami-rajat> i think we can extend the effort for nova to call the new API (extend volume completion) for all drivers
14:47:54 <whoami-rajat> as this will allow us to report a user message if the extend fails on the nova side
14:48:01 <whoami-rajat> i will mention that in the nova review
14:48:34 <whoami-rajat> but just wanted to highlight here that it will improve our error reporting for other drivers as well
14:49:07 <whoami-rajat> that's all from my side on this one
14:49:46 <whoami-rajat> i guess we can move on to the next topic
14:49:50 <whoami-rajat> #topic Implementation of the dedicated backup status field for cinder volumes
14:50:04 <whoami-rajat> that's added by Christian
14:50:04 <whoami-rajat> but he doesn't seem to be around
14:50:37 <whoami-rajat> maybe we can skip some topics and move on to Takashi's topic since he is around (and we've less time)
14:50:47 <whoami-rajat> #topic Deprecation of VMWare drivers
14:50:50 <whoami-rajat> tkajinam, that's you
14:51:14 <tkajinam> I wasnted to give you some updates about the global discussion.
14:51:52 <tkajinam> so I started the mailing thread to ask if we can deprecate all the vmware support in multiple projects. there are some discussions going on in that thread and in short some people are interested in maintaining vmware virt driver in nova because they are using it
14:52:36 <tkajinam> however nova team is careful about the situation and they will watch the status of the whole effort to restore CI coverage and maintain the driver for a while, even if they decide not to remove the vmware virt driver during this cycle.
14:53:36 <tkajinam> I think the reasonable approach here is to anyway deprecate the driver, to warn more "hidden" users. if the situation with virt driver is recovered then we can undeprecate the volume driver as well.
14:54:18 <tkajinam> note that VMWare CI in cinder has been broken for a while. if we keep the vmware volume driver then we may need to ask these people to restore that CI as well.
14:55:54 <whoami-rajat> i think if the CI is broken, it makes sense to deprecate the driver irrespective of the nova support
14:56:25 <rosmaita> we should also put up a patch adding SUPPORTED = False for the driver
14:56:45 <whoami-rajat> rosmaita, there is a patch already
14:56:46 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder/+/903737
14:56:50 <tkajinam> I've already pushed a patch to deprecate these vmware drivers and mark these unsupported
14:56:56 <tkajinam> yes, that's the one
14:57:04 <rosmaita> great!
14:57:39 <tkajinam> CI failed for unrelated reasons but I'll try to get it passed. it'd be nice if we can merge this during this cycle.
14:57:45 <whoami-rajat> ok, thanks for working on this tkajinam , can you let me know once the CI passes and i can take a look
14:57:52 <tkajinam> thanks
14:58:01 <whoami-rajat> sorry for the recent gate failures, we are also struggling with it ...
14:58:17 <tkajinam> and also take a look about windows deprecation patches I added to the agenda. I think we all agree with it since Winstackers has retired.
14:58:33 <tkajinam> yeah, no problem.
14:59:08 <tkajinam> I'll update the ml thread to hightlight again the situation of cinder CI to get some attention from people working on nova ci.
14:59:24 <whoami-rajat> sounds good
14:59:31 <tkajinam> that's all from my side
14:59:35 <whoami-rajat> thanks again
14:59:50 <whoami-rajat> and with that we are out of time
14:59:59 <whoami-rajat> i will move rest of the topics to next meeting agenda
15:00:09 <whoami-rajat> remember we are canceling next week's meeting on 27th December
15:00:17 <whoami-rajat> so next meeting will happen on 3rd Jan 2024
15:00:30 <whoami-rajat> merry christmas and a happy new year to the team!
15:00:33 <whoami-rajat> see you all next year
15:00:40 <whoami-rajat> #endmeeting