14:00:04 <rosmaita> #startmeeting cinder 14:00:05 <openstack> Meeting started Wed Jan 20 14:00:04 2021 UTC and is due to finish in 60 minutes. The chair is rosmaita. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:06 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:08 <openstack> The meeting name has been set to 'cinder' 14:00:13 <rosmaita> #topic roll call 14:00:19 <jungleboyj> o/ 14:00:42 <michael-mcaleer> hi 14:00:42 <whoami-rajat___> Hi 14:00:44 <sfernand> hi 14:01:18 <enriquetaso> hi 14:01:26 <eharney> hi 14:01:30 <tosky> hi 14:02:09 <e0ne> hi 14:03:19 <rosmaita> hello everyone, let's get started 14:03:26 <rosmaita> #topic announcements 14:03:38 <rosmaita> just for awareness, the Xena python runtimes have been proposed 14:03:48 <rosmaita> #link https://review.opendev.org/c/openstack/governance/+/770860 14:04:00 <rosmaita> python 3.6 and 3.8 14:04:17 <rosmaita> if you have any opinions or comments, leave them on the review ^^ 14:04:34 <rosmaita> also, if you are interested in requirements-type stuff 14:04:44 <rosmaita> the lower-constraints discussion is continuing 14:04:54 <rosmaita> #link http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019672.html 14:05:02 <rosmaita> that's the email thread (a lot of messages) 14:05:21 <rosmaita> there will also be some discussion at the TC meeting tomorrow 14:05:24 <rosmaita> #link https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee#Next_Meeting 14:05:48 <rosmaita> so far, it's looking like we may keep lower-constraints but only for direct dependencies 14:06:03 <rosmaita> and we will be updating requirements (and hence lower constraints) more frequently 14:06:33 <rosmaita> an issue on the ML is whether there is an easy way to determine that the requirements list actually contains all direct dependencies 14:06:48 <rosmaita> because it's easy to pick them up via another dependency 14:06:58 <rosmaita> so if you have ideas, opinions, please chime in 14:07:32 <rosmaita> #topic announcements - upcoming deadlines 14:07:47 <rosmaita> R-12 - this week - wallaby milestone 2 14:08:15 <rosmaita> key issue is the new driver merge deadline is M-2 14:08:22 <rosmaita> we will discuss that in detail later 14:08:31 <rosmaita> R-9 - Feb 8 - virtual midcycle part 2 14:08:46 <rosmaita> R-6 - March 1 - wallaby os-brick release 14:08:56 <rosmaita> R-5 - March 8 - wallaby cinderclient release, Milestone 3, 3rd Party CI Compliance Check 14:09:11 <rosmaita> i should mention that those dates are "week of" dates 14:09:21 <rosmaita> actual deadline is usually thursday of that week 14:09:56 <rosmaita> something i want to ask right now is whether the week of Feb 8 still looks good for a midcycle 14:10:15 <rosmaita> is anyone aware of any major conflicts? 14:10:39 <rosmaita> you can check your calendars and bring it up at open discussion 14:11:00 <rosmaita> if there aren't any conflicts, i'll circulate a poll about the day/time for the meeting 14:11:19 <rosmaita> next announcement is some sad news 14:11:31 <rosmaita> Last week of michael-mcaleer as bug deputy 14:11:43 <rosmaita> and even worse, moving teams and no longer working on cinder 14:11:47 <enriquetaso> :( 14:11:49 <michael-mcaleer> yeah found out last week im moving to another platform 14:12:02 <rosmaita> on behalf of the team, i'd like to thank michael-mcaleer for his work as bug deputy 14:12:06 <michael-mcaleer> I dont think ill be working with openstack but if I was it would likely be manila 14:12:07 <rosmaita> it's been very helpful 14:12:24 <rosmaita> well, i'm sure the manila team will be glad to have you 14:12:25 <eharney> yes, thanks for the work for sure 14:12:32 <jungleboyj> :-( 14:12:34 <michael-mcaleer> thanks rosmaita :) short but enjoyable as bug deputy, and a very very fulfilling 5 years working as a contributor 14:12:53 <rosmaita> yes, you have left your mark on the project! 14:12:55 <michael-mcaleer> ive come a long way since a green behind the ears sw dev 14:12:55 <jungleboyj> michael-mcaleer: Thank you for all your contributions. 14:13:09 <michael-mcaleer> thanks everyone :) 14:13:32 <michael-mcaleer> I might be back in the future, I see myself staying active in the open-source space 14:13:48 <rosmaita> best wishes 14:14:03 <rosmaita> on the positive side, there is now an opening for bug deputy 14:14:25 <rosmaita> i'm sure michael-mcaleer will be happy to talk with you about how he's been doing it if you are interested in picking that up 14:14:34 <rosmaita> ok, last item 14:14:37 <michael-mcaleer> no problem at all, feel free to reach out 14:14:56 <rosmaita> next week is the final meeting of the month, so will be held simultaneously on irc and video 14:15:24 <rosmaita> connection info will most likely be the usual, but i will send out a confirming email early next week 14:15:44 <rosmaita> ok, that's it for announcements 14:15:55 <rosmaita> #topic new drivers status 14:16:30 <rosmaita> ok, we've got 4 volume drivers and 1 backup driver proposed, and tomorrow is the merge deadline 14:16:38 <rosmaita> so, let's see what's up with them 14:16:50 <rosmaita> Dell EMC PowerVault ME Series 14:17:02 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/758684/ 14:17:21 <rosmaita> \o/ merged!!! 14:17:46 <rosmaita> s3 cinder backup driver 14:17:46 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/746561 14:18:18 <rosmaita> the code on this one looks basically OK to me 14:18:30 <e0ne> +1. but I didn't test it 14:18:46 <rosmaita> there's a proposed CI job, though it doesn't seem to be passing 14:19:00 <rosmaita> does anyone have time to test this one out? 14:19:23 <rosmaita> or, to help with the CI job, which would be a decent alternative 14:19:33 <rosmaita> to manual testing, i mean 14:20:20 <rosmaita> do we traditionally require CI on backup drivers? 14:20:59 <e0ne> rosmaita: I didn't see job proposal before. I'll check it 14:20:59 <jungleboyj> I can at least look through the code on that one. I haven't looked at that driver yet. 14:21:15 <e0ne> rosmaita: unfortunately, we didn't :( 14:21:30 <jungleboyj> rosmaita: I think we talked about the fact that we 'wished' we had those CIed, but we don't. 14:21:40 <jungleboyj> So, don't think it would be fair to block them on that. 14:21:45 <rosmaita> ok, just thought i'd ask. the CI job would be nice, then 14:21:48 <eharney> what does the s3 backup CI job back up to? s3 itself? something that looks like s3? 14:22:09 <rosmaita> i think it's backing up to swift 14:22:21 <eharney> ah ok 14:22:27 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/770210 14:23:03 <e0ne> jungleboyj: fair enough but somebody should confirm that it works with S3 14:23:14 <jungleboyj> Agreed. 14:23:31 <rosmaita> tosky: if you have a few minutes to look at that CI job, i'm not sure they've picked the best base job to use 14:24:50 <rosmaita> i guess an alternative would be for the driver proposer to deploy the driver in devstack and run tempest and post the logs? 14:26:39 <rosmaita> what i propose for this driver is let's review the code (i haven't looked at the latest patch set) but hold off on merging until we have some kind of testing verification 14:26:58 <e0ne> rosmaita: sound good 14:27:43 <rosmaita> ok, to be specific, let's aim for two +2s by the merge deadline tomorrow (though not merging), and then an extension of 1 week (to next thursday) to get testing verification 14:28:09 <rosmaita> if the above is met, it is eligible for wallaby 14:28:18 <e0ne> +1 14:28:23 <rosmaita> driver for TOYOU ACS5000 14:28:23 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/767290/ 14:28:44 <jungleboyj> ++ for the backup driver plan. 14:28:51 <rosmaita> ty 14:29:34 <rosmaita> last i looked at this one, it was really close. the CI is running, though not reporting reliably 14:29:59 <rosmaita> i told them as long as they get it responding properly by M-3, that is OK 14:30:27 <rosmaita> so this is mainly an issue of code, which i think jungleboyj and smcginnis had given them comments on 14:31:01 <rosmaita> so at this point it's a matter of verifying the revisions 14:31:07 <jungleboyj> Yeah. They had six.text_type in there. 14:31:22 <jungleboyj> Looks like they have removed that. I just need to do a quick scan over it. 14:31:41 <tosky> rosmaita: sorry, I will look at that 14:31:44 <jungleboyj> I am ok with your plan on the CI. So, we should be able to merge this. 14:31:47 <rosmaita> tosky: ty 14:32:04 <jungleboyj> Once I look through it and if smcginnis is good. 14:32:11 <rosmaita> ok, i know smcginnis is in a time crunch these days, so i will review and make sure his comments were addressed 14:32:33 <jungleboyj> rosmaita: Sounds like a good plan. 14:32:41 <rosmaita> cool, lets us both review today and wait to merge until smcginnis gives a final OK 14:32:50 <jungleboyj> ++ 14:33:01 <rosmaita> all right, so unless we discover something horrible, i think this one is good for the merge deadline 14:33:14 <rosmaita> ceph iscsi volume driver 14:33:21 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/662829 14:33:47 <rosmaita> i think this is looking good, but the CI hasn't been passing 14:33:54 <rosmaita> (or at least wasn't the last time i looked) 14:34:37 <rosmaita> this is the CI patch: 14:34:46 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/667108 14:35:53 <rosmaita> ok, so that's in the gate now 14:35:58 <jungleboyj> I would like to see this get in. Could we give it the M-3 deadline to have that working? 14:36:25 <rosmaita> i was going to ask, has anyone tried this one out? 14:36:35 <rosmaita> i know a lot of people are interested in it 14:38:07 <rosmaita> m-3 seems kind of far away ... how about 2 weeks? 14:38:53 <rosmaita> that would be feb 5 14:39:02 <jungleboyj> hemna: ^^ 14:39:31 <jungleboyj> I think it is up to hemna as to if he can get things done. 14:39:31 <rosmaita> i think the code is in good shape, so it would just be small changes if there's a bug preventing the CI from working 14:40:06 <rosmaita> ok, let's say feb 5 and we can have a status checkpoint next week 14:40:21 <jungleboyj> Ok. Sounds fair. 14:40:50 <rosmaita> ok, i think that leaves the Kumoscale driver and associated brick changes 14:41:03 <rosmaita> cinder volume driver 14:41:22 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/768574 14:41:42 <rosmaita> brick nvmeof connector mdraid support 14:41:51 <rosmaita> #link https://review.opendev.org/c/openstack/os-brick/+/768575 14:41:57 <rosmaita> brick "healing agent" 14:42:11 <rosmaita> #link https://review.opendev.org/c/openstack/os-brick/+/768576 14:42:46 <rosmaita> so, these are in various states. i understand from the team that they do have CI running but i don't think it's reporting yet 14:43:23 <rosmaita> i know there are some people interested in the updated nvmeof connector 14:44:31 <rosmaita> my opinion on this is that our (cinder team) review bandwidth has been low this cycle, so i would like to give them a little more time 14:45:16 <rosmaita> and, it looks like the other proposed drivers are all reviewed, it's a matter of CI for the others at this point 14:45:40 <rosmaita> so we could concentrate on this final driver 14:46:08 <jungleboyj> I have only quickly looked at the KumoScale driver. Big driver. 14:46:19 <jungleboyj> Whatever you think is best here. 14:47:15 <rosmaita> well, it looks like we have some current FC drivers responding on the os-brick patches 14:47:34 <rosmaita> so we'll hopefully be able to detect any bad regressions 14:47:58 <rosmaita> but, this is a heads-up to FC driver maintainers to look at those brick patches! 14:49:38 <rosmaita> unless there's an objection, i'm willing to extend the merge deadline out to Feb 8 for this 14:50:47 <rosmaita> ok, i guess there are no objections 14:50:55 <jungleboyj> Ok. 14:51:02 <rosmaita> OK, i will put all the above in an email to the ML today 14:51:40 <rosmaita> so, a bunch of new drivers in wallaby ... that's good news 14:51:56 <rosmaita> #topic "ospurge" via openstackclient via openstacksdk 14:52:10 <rosmaita> this is really just an awareness thing at this point 14:52:21 <rosmaita> it came up on the ML: 14:52:33 <rosmaita> #link http://lists.openstack.org/pipermail/openstack-discuss/2021-January/019874.html 14:53:00 <rosmaita> short story is, there is support in the openstackclient for a "purge" operation that is supposed to clean up when a project is deleted 14:53:10 <rosmaita> is uses the openstacksdk: 14:53:21 <rosmaita> #link https://opendev.org/openstack/openstacksdk/src/branch/master/openstack/block_storage/v3/_proxy.py#L547 14:53:34 <rosmaita> that's already merged code 14:53:39 <rosmaita> the osc patch is here: 14:53:50 <rosmaita> https://review.opendev.org/c/openstack/python-openstackclient/+/734485 14:54:40 <rosmaita> i think it would be a good midcycle topic to look at what's actually being purged and see whether stuff could be added 14:55:06 <eharney> well i guess it doesn't purge things like cinder quotas configured for a project 14:55:29 <rosmaita> or private volume-type where only that project has access 14:56:28 <rosmaita> i guess on that note, i should mention that with the R-9 midcycle coming up, please add topics: 14:56:44 <rosmaita> #link https://etherpad.opendev.org/p/cinder-wallaby-mid-cycles 14:56:56 <rosmaita> #topic stable releases 14:57:02 <rosmaita> whoami-rajat___: that's you 14:57:08 <whoami-rajat___> thanks 14:57:52 <whoami-rajat___> so just wanted to spread awareness that we've planned for a stable release on R-7 which is in 2 weeks gap with the mid-cycle on R-9 14:58:23 <rosmaita> i agree with that timing 14:58:40 <rosmaita> we can take a look at potential backports at the midcycle 14:59:02 <whoami-rajat___> i will add a topic in the mid-cycle if anyone would like to include any important backport 14:59:09 <whoami-rajat___> yes, as rosmaita said 14:59:55 <whoami-rajat___> I will create a new tracker doc for the current open patches and provide updates on it from next meeting 15:00:04 <rosmaita> ok, thanks ... we are out of time 15:00:10 <rosmaita> let's clear out for horizon 15:00:11 <whoami-rajat___> thanks! 15:00:17 <jungleboyj> Thanks! 15:00:25 <rosmaita> sorry no open discussion, there's always the cinder channel 15:00:30 <rosmaita> #endmeeting