14:00:24 <whoami-rajat> #startmeeting cinder 14:00:24 <opendevmeet> Meeting started Wed Dec 7 14:00:24 2022 UTC and is due to finish in 60 minutes. The chair is whoami-rajat. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:24 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:24 <opendevmeet> The meeting name has been set to 'cinder' 14:00:28 <whoami-rajat> #topic roll call 14:00:34 <hemna> mep 14:02:41 <jungleboyj> o/ 14:02:50 <rosmaita> o/ 14:03:34 <whoami-rajat> did people get my ping in cinder channel? or maybe there is some sort of holiday ... 14:03:53 <jungleboyj> I saw your ping. :-) 14:04:16 <whoami-rajat> oh ok, good to know 14:04:25 <rosmaita> could be ... tuesday and thursday are holidays in some locations, so people may be taking today off too 14:04:26 <whoami-rajat> #link https://etherpad.opendev.org/p/cinder-antelope-meetings 14:04:49 <whoami-rajat> rosmaita, ok, sounds good, just wanted to confirm if people weren't notified 14:04:55 <whoami-rajat> so we can get started 14:05:18 <whoami-rajat> #topic announcements 14:05:31 <whoami-rajat> first, Midcycle summary 14:05:47 <whoami-rajat> we had a very productive midcycle last week where we discussed about the current specs and drivers proposed 14:05:51 <whoami-rajat> also other topics 14:05:59 <whoami-rajat> the summary is available here 14:06:03 <whoami-rajat> #link https://wiki.openstack.org/wiki/CinderAntelopeMidCycleSummary 14:06:06 <rosmaita> thanks for writing up the summary, it looks good 14:06:18 <jungleboyj> ++ 14:06:32 <whoami-rajat> I would recommend to go through the etherpad/recording for detailed version of topics since the summary is minimal 14:06:38 <whoami-rajat> thanks rosmaita and jungleboyj for taking a look 14:07:07 <whoami-rajat> and jungleboyj uploaded the recording so you can find it on youtube! 14:07:12 <whoami-rajat> the link is in the summary 14:07:28 <jungleboyj> Also in the ehterpad. 14:07:34 <whoami-rajat> yes! 14:08:18 <whoami-rajat> thanks everyone for joining the midcycle and making it useful 14:08:35 <whoami-rajat> next announcement, cinderclient tests failing with python 3.11 14:08:43 <whoami-rajat> #link https://lists.openstack.org/pipermail/openstack-discuss/2022-December/031392.html 14:08:49 <whoami-rajat> #link https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1025028 14:09:02 <whoami-rajat> so some of the cinderclient tests are failing with py311 14:09:10 <whoami-rajat> the report is in the debian bug reported 14:09:39 <rosmaita> kind of concerning that taskflow has a 3.11 bug 14:09:47 <whoami-rajat> I will take a look at them later this week but feel free to take a look 14:10:13 <whoami-rajat> I'm not very familiar with taskflow so can't comment 14:10:21 <whoami-rajat> current runtimes of Antelope are 3.8 and 3.10 so we are good 14:10:28 <whoami-rajat> but good to get them fixed early 14:11:00 <rosmaita> yeah, i was going to say, you can back-burner this 14:11:47 <whoami-rajat> yes, doesn't seem to be an issue for this cycle at least 14:12:07 <whoami-rajat> next, Cinderlib release 14:12:16 <whoami-rajat> #link https://lists.openstack.org/pipermail/openstack-discuss/2022-December/031386.html 14:12:28 <whoami-rajat> so the time has come to release cinderlib which is next week 14:13:04 <whoami-rajat> I'm not sure if any changes were made but i think we are still on the gate failure situation? 14:13:08 <rosmaita> 211 unit test failures at this point 14:13:16 <whoami-rajat> :( 14:13:22 <whoami-rajat> that's pretty bad 14:13:35 <rosmaita> eyeballing it, it's mostly in the persistence module 14:13:51 <whoami-rajat> ok 14:14:10 <whoami-rajat> maybe it's time we can drag geguileo in since we only have a week left for the release 14:14:10 <rosmaita> that said, i haven't looked closely, but will do so today 14:14:29 <whoami-rajat> great, thanks for keeping track of it 14:14:48 <rosmaita> he will be on holiday tomorrow 14:15:34 <rosmaita> my goal is to at least narrow it down to what exactly the failures are 14:15:43 <whoami-rajat> yes, what i meant was he can take a look when he's available 14:15:52 <whoami-rajat> but if you think things are under control 14:15:58 <whoami-rajat> then that's good as well 14:16:15 <rosmaita> yeah, most of the errors look like this: 14:16:18 <rosmaita> ValueError: Backend named fake_backend already exists with a different configuration 14:16:40 <rosmaita> so I'm hoping there's just one change to make and we'll be golden 14:17:07 <rosmaita> and also a bunch of these: 14:17:10 <rosmaita> oslo_db.exception.NoEngineContextEstablished: No TransactionContext is established for this <cinder.context.RequestContext object at 0x7f95228e28e0> object within the current thread; the 'session' attribute is unavailable. 14:17:14 <whoami-rajat> hmm, that's a strange error, looks like multiple backends with same name 'fake_backend' are defined 14:17:26 <rosmaita> which looks like stephen's change 14:17:30 <whoami-rajat> yeah the session thing was removed 14:17:47 <whoami-rajat> do you have a link for the failed job(s)? 14:17:52 <whoami-rajat> maybe i can take a look as well 14:18:10 <rosmaita> no, ran them locally, i don't think the logs are available on the "open for zed" patch anymore 14:18:43 <rosmaita> btw, if you want to look into this, you need to apply the "open for zed" patch first, otherwise everything passes 14:18:48 <whoami-rajat> ok, i will recheck that then to get it since I'm out of local envs 14:18:56 <whoami-rajat> oh ok 14:19:01 <whoami-rajat> thanks for the heads up 14:19:06 <rosmaita> np 14:19:31 <rosmaita> https://review.opendev.org/c/openstack/cinderlib/+/848846 14:19:37 <rosmaita> (just so you have the patch) 14:20:03 <whoami-rajat> thanks again, rechecked it 14:20:10 <rosmaita> cool 14:20:56 <whoami-rajat> great 14:21:10 <rosmaita> it's getting toward the end of the day for you, i will email an update at the end of my day saying how far i got looking into this 14:21:10 <enriquetaso> hi 14:21:23 <rosmaita> Sofia! 14:21:33 <whoami-rajat> hey enriquetaso 14:21:37 <enriquetaso> sorry, Im late 14:22:15 <whoami-rajat> rosmaita, that would be great, although i will probably check after few hours 14:22:25 <whoami-rajat> enriquetaso, no problem, it's been a light agenda 14:22:44 <whoami-rajat> so we've no other announcement or topic for the day 14:22:54 <whoami-rajat> but one thing i remembered that i missed 14:23:41 <ganso> o/ whoami-rajat I actually just added one item to the list, sorry the delay 14:24:46 <rosmaita> ganso: i have your patch open with some comments, but haven't published them yet 14:25:02 <whoami-rajat> ganso, oh the review, let's discuss it after my last announcement (which i just added) 14:25:17 <whoami-rajat> final announcement, Spec freeze next week 14:25:29 <whoami-rajat> so we have 3 specs in discussion for this cycle 14:25:41 <whoami-rajat> 1) Encrypted backups 14:25:47 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder-specs/+/862601 14:26:12 <whoami-rajat> I've reviewed it, others can review it as well so geguileo can update it in one go 14:26:19 <whoami-rajat> 2) extend volume support in FS drivers 14:26:25 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder-specs/+/866718 14:26:46 <whoami-rajat> we had the original proposal of using a sync API but that was recently rejected by nova 14:26:57 <whoami-rajat> so kgube_ had to write another spec for an alternative method 14:27:05 <whoami-rajat> this again needs reviews 14:27:31 <whoami-rajat> 3) backup specific field in volume attributes 14:27:54 <whoami-rajat> this was discussed during midcycle and is solution to the problem when we want to live migrate an instance with attached volume that is backing up 14:28:20 <whoami-rajat> Christian is working on it but i don't see any new spec 14:28:36 <whoami-rajat> so we can probably focus on the first two for the time being 14:29:16 <whoami-rajat> that's all the announcements for the day 14:29:20 <whoami-rajat> moving on to topics 14:29:25 <whoami-rajat> #topic Requesting review of https://review.opendev.org/c/openstack/cinder/+/812685 14:29:38 <whoami-rajat> ganso, I need to check again where we left the reviews last time but will take a look 14:30:10 <ganso> whoami-rajat: Rafael has updated the patch to only remove the properties when uploading the volume back to glance 14:30:26 <ganso> so now the metadata properties are being saved in cinder 14:30:46 <whoami-rajat> that sounds in agreement to what was proposed last time 14:30:50 <whoami-rajat> and also in sync with nova 14:32:10 <rosmaita> actually, it looks like the filtering is only being done in the backup driver? 14:32:10 <whoami-rajat> anything else on this ganso ? 14:32:21 <ganso> that's all I had to mention, just a reminder for re-review 14:33:25 <rosmaita> ganso: i will publish my half-baked comments, please tell Rafael not to take it personally if they are stupid because i missed something 14:34:10 <ganso> rosmaita: I do not have direct contact with Rafael, but maybe we can discuss it in the comments 14:34:17 <rosmaita> ok 14:34:46 <whoami-rajat> rosmaita, yeah that looks to be the case but maybe I'm missing something with a quick look 14:37:10 <whoami-rajat> that's all we had for topics 14:37:11 <ganso> rosmaita, whoami-rajat: that seems to be the case for me too. The other modifications are just refactoring 14:37:27 <ganso> are there other ways to upload the volume back to glance, such as not using the backup driver ? 14:37:42 <whoami-rajat> ganso, yes, we've upload volume to image operation 14:38:09 <whoami-rajat> ganso, refer to upload_volume method in image_utils 14:38:44 <whoami-rajat> also in volume_utils, not sure where exactly the metadata is passed 14:39:26 <ganso> whoami-rajat: ok, so maybe it is missing that. I compared all methods before and after the latest update and it didn't seem like something was missing, but perhaps it was a side-effect fix when we were removing all properties. Now that we no longer are, it is missing filtering in that scenario 14:41:07 <whoami-rajat> ack, I'm thinking do we even need to filter things out when restoring backup to a volume since it doesn't involve glance 14:41:26 <whoami-rajat> anyways i will leave comments on the patch 14:41:51 <ganso> whoami-rajat: thanks! 14:42:15 <whoami-rajat> thanks for working on that 14:42:54 <whoami-rajat> since we don't have any more topics, let's move to open discussion 14:42:58 <whoami-rajat> #topic open discussion 14:43:17 <ganso> whoami-rajat: at the moment I'm just "shepherding", Rafael is doing the actual work 14:45:13 <whoami-rajat> even following up and discussing is an important part :) 14:46:47 <whoami-rajat> looks like there is nothing more to discuss, let's end the meeting early 14:46:53 <whoami-rajat> thanks everyone for joining 14:47:00 <whoami-rajat> #endmeeting