14:01:24 <rosmaita> #startmeeting cinder 14:01:26 <openstack> Meeting started Wed Sep 23 14:01:24 2020 UTC and is due to finish in 60 minutes. The chair is rosmaita. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:27 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:01:30 <openstack> The meeting name has been set to 'cinder' 14:01:33 <whoami-rajat__> Hi 14:01:33 <rosmaita> #topic roll call 14:01:33 <eharney> hi 14:01:34 <smcginnis> o/ 14:01:37 <kaisers> o/ 14:01:41 <enriquetaso> i/ 14:02:09 <lseki> o/ 14:02:25 <geguileo> hi! o/ 14:02:30 <e0ne> hi 14:02:34 <walshh_> hi 14:02:34 <LiangFang> o/ 14:02:59 <rajinir> o/ 14:03:20 <rosmaita> hello everyone 14:03:33 <rosmaita> #link https://etherpad.opendev.org/p/cinder-victoria-meetings 14:03:34 <jungleboyj> Happy Wednesday. 14:03:48 <rosmaita> #topic updates 14:04:02 <rosmaita> i guess that's "announcements" on the agenda 14:04:21 <rosmaita> we got both our proposed forum sessions accepted 14:04:29 <rosmaita> here's the info: 14:04:38 <rosmaita> #link https://www.openstack.org/summit/2020/summit-schedule/events/24748/cinder-user-messages 14:04:47 <rosmaita> #link https://www.openstack.org/summit/2020/summit-schedule/events/24749/cinder-third-party-ci-whats-in-it-for-you 14:05:17 <rosmaita> these sessions are supposed to be an opportunity for operators and users to interact with us 14:05:19 <enriquetaso> \o/ 14:05:44 <rosmaita> so, it would be good for as many cinder community members (that is, you) to attend 14:05:52 <jungleboyj> ++ 14:05:59 <rosmaita> so please mark them on your calendar 14:06:58 <rosmaita> ok, next item is that we have a virtual ptg coming up 14:07:09 <rosmaita> #link https://etherpad.opendev.org/p/wallaby-ptg-cinder-planning 14:07:17 <rosmaita> please add topics to the above 14:07:32 <rosmaita> or at the very least, make sure you register for the PTG (it's free) 14:07:47 <rosmaita> finally, election season has begun 14:07:57 <rosmaita> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-September/017493.html 14:08:23 <rosmaita> self-nominations are due by 29 September 14:08:27 <rosmaita> details in the email 14:08:49 <rosmaita> would be nice to keep some cinder representation on the TC 14:09:02 <rosmaita> talk to jungleboyj if you want an idea of what the time commitment is like 14:09:17 <jungleboyj> :-) 14:09:46 <rosmaita> that's all from me -- anyone have an announcement? 14:10:12 <jungleboyj> There may still be Cinder representation on the TC. :-) 14:10:41 <rosmaita> is smcginnis on the TC these days? 14:10:55 <jungleboyj> rosmaita: No. Not at the moment. 14:11:16 <rosmaita> jungleboyj: i misread the subject of your emal 14:11:29 <jungleboyj> rosmaita: Which one? 14:11:32 <rosmaita> you are not running for oslo PTL, looks like 14:11:47 <jungleboyj> :-) Right. Was thanking Ben. 14:11:58 <rosmaita> i though it was about the TC .. my mistake 14:12:18 <rosmaita> well, it would be good to have *more* cinder representation on the TC! 14:12:32 <rosmaita> my fault for not looking at the ML in threaded view 14:12:35 <rosmaita> :) 14:12:54 <rosmaita> #topic RC-1 14:13:06 <rosmaita> we will be releasing tomorrow 14:13:23 <rosmaita> i will put up an overall release notes patch after the meeting 14:13:35 <rosmaita> i think we just need to mention the 2 new microversions 14:13:38 * e0ne wants to propose own patch https://review.opendev.org/#/c/752497/ 14:14:42 <rosmaita> and maybe OSSN-0086, i think that happened after the ussuri release, though that whole thing is a blur 14:14:54 <rosmaita> e0ne: let's add it to the etherpad 14:15:11 <rosmaita> so we need to make sure release-critical patches make it into the release 14:15:18 <rosmaita> which is why they are called release-critical 14:15:34 <geguileo> rosmaita: The RBD keyring for cinderlib patch is needed, or it will break cinderlib... 14:15:34 <rosmaita> #link https://etherpad.opendev.org/p/cinder-victoria-release-critical-patches 14:15:46 <rosmaita> geguileo: number 1 on the list! 14:15:50 <geguileo> oh, I see it there 14:15:52 <geguileo> XD 14:16:03 <geguileo> thanks 14:16:29 <rosmaita> e0ne: that patch you posted looks appropriate 14:16:46 <rosmaita> and small and easy to review 14:16:53 <e0ne> rosmaita: at lease, it makes users more happy :) 14:17:05 <rosmaita> and probably operators too 14:17:10 <e0ne> rosmaita: +1 14:17:47 <whoami-rajat__> i think this patch is also needed to make the system scope thing work 14:17:48 <whoami-rajat__> https://review.opendev.org/#/c/752396/ 14:18:09 <e0ne> geguileo, rosmaita: I understand why we need this for cinderlib, but it looks like a one more dirty hack to get cinderlib working :( 14:18:29 <geguileo> e0ne: it's basically a revert of what we removed 14:18:45 <geguileo> but making sure it cannot be used in Cinder 14:18:55 <geguileo> the initial patches didn't have to add it, just made a small change 14:19:15 <geguileo> to remove it from cinder usage 14:19:19 <rosmaita> yeah, the way i see it is that cinderlib has some requirements that are slightly different from cinder 14:19:33 <geguileo> because in Cinder this is a security concern 14:19:39 <geguileo> but it isn't for cinderlib 14:19:54 <rosmaita> right, so even if the option is included in a cinder.conf, it will be ignored for cinder 14:20:58 <e0ne> this behaviour should be documented somewhere 14:21:33 <rosmaita> well, there are comments in the code for developers 14:21:34 <geguileo> it's documented in OS-Brick 14:22:02 <e0ne> geguileo: can you please drop me a link? 14:22:24 <whoami-rajat__> rosmaita: incase it got missed, should i add the above patch to the etherpad as well? 14:22:27 <geguileo> e0ne: https://github.com/openstack/os-brick/blob/18fc838ea3400f81820d826732aef324410e90df/os_brick/initiator/connectors/rbd.py#L123-L125 14:22:51 <e0ne> geguileo: thanks 14:22:54 <geguileo> np 14:23:13 <rosmaita> whoami-rajat__: was there any followup patches from the project level default volume-types? 14:23:14 <whoami-rajat__> i see it added now, thanks rosmaita 14:23:38 <whoami-rajat__> rosmaita: yes, there was one for NITs but it is merged 14:23:47 <rosmaita> ok, great 14:24:07 <geguileo> whoami-rajat__: I think we should add a release note, and maybe a unit test... 14:24:27 <rosmaita> ok, so the way things work now is that when RC-1 is released, the stable/victoria branch is cut 14:24:40 <rosmaita> all bugfixes need to be proposed for master 14:24:48 <whoami-rajat__> geguileo: sure, i will add those 14:24:53 <geguileo> whoami-rajat__: thanks 14:25:03 <rosmaita> and backports will have to be proposed to stable/victoria to get into RC-2 14:25:15 <rosmaita> so, to keep this process manageable 14:25:35 <rosmaita> if you have a release-critical bug, please add it to the etherpad so the stable team is aware of it 14:25:38 <rosmaita> and can evaluate it 14:26:00 <rosmaita> keep in mind that if a bug misses the victoria release, it can still be backported to victoria later for a stable release 14:26:21 <rosmaita> and TBH, i don' t know that people rush out and install victoria on release day anyway 14:26:38 <rosmaita> any questions? 14:27:25 <rosmaita> #topic stable releases 14:27:44 <rosmaita> i want to release from stable/ussuri and train soon 14:27:52 <rosmaita> ussuri is ready to go, i think 14:28:07 <rosmaita> train could still use this backport: https://review.opendev.org/752916 14:28:36 <rosmaita> and there are some other open patches for stable/train that could go if people are interested in making revisions and reviewing 14:28:52 * whoami-rajat__ was going to mention the same 14:29:01 <rosmaita> but once the above backport to train lands, i'll propose the releases 14:29:18 <jungleboyj> rosmaita: Done. 14:29:38 <rosmaita> this will give us up to date ussuri and train with os-brick in the requirements that addresses all the OSSN-0086 followups 14:30:03 <rosmaita> #topic migrate-to-focal community goal status 14:30:30 <rosmaita> ok, this goal is continuing during RC time 14:30:55 <rosmaita> the most immediate effect on us is this problem with barbican being run in focal: 14:31:05 <rosmaita> #link https://storyboard.openstack.org/#!/story/2007732 14:31:33 <rosmaita> so gmann posted patches so that our jobs that use barbican will run in bionic where it's not a problem 14:31:49 <rosmaita> os-brick: https://review.opendev.org/#/c/743083/ 14:31:58 <rosmaita> cinder-tempest-plugin: https://review.opendev.org/#/c/743079/ 14:32:14 <rosmaita> i just want everyone to be aware of this in case something else breaks 14:32:34 <rosmaita> the above reviews, i think one has been approved and the other still needs approval 14:32:40 <rosmaita> so please take a look 14:32:54 <rosmaita> because if they don't get merged before the big focal switchover, our gates will break 14:33:07 <rosmaita> and that switchover is imminent 14:33:42 <rosmaita> hopefully there will be no bad surprises 14:34:18 <rosmaita> ok, on to the other community goal, which is to get rid of all legacy zuul jobs 14:34:31 <rosmaita> #topic update on last legacy job 14:34:41 <rosmaita> tosky: you weren't sure you'd be able to make it 14:34:53 <tosky> tl;dr I'm late, sorry 14:35:02 <rosmaita> no problem, you are right on time actually 14:35:28 <rosmaita> i think that's the same patch you mentioned last week? 14:35:37 <rosmaita> #link https://review.opendev.org/751579 14:36:04 <tosky> technically a) the community goals should be done at RC1 time (or even before) b) this is cinder, I should have invested more time before c) well, we will need an extra backport, at least it's an experimental job 14:36:36 <rosmaita> i don't think you have any reason to apologize 14:36:36 <tosky> yes, it is that same patch; no updates, but I think now I know how to solve the reporting part 14:36:43 <jungleboyj> rosmaita: ++ 14:37:00 <rosmaita> ok, great, that gives me an excuse to wait for the next patch to review :) 14:37:48 <rosmaita> thanks for the update, and thank you for working on this ... you have pretty much done it alone 14:38:12 <rosmaita> #topic update on NFS online extend bugfix 14:38:18 <rosmaita> lseki: that's you 14:38:21 <lseki> o/ 14:38:45 <lseki> so I haven't worked on this since last time we discussed about it 14:38:59 <lseki> and won't be able to address it in the next few months 14:39:28 <lseki> so if anyone else has interest in adopting it 14:39:39 <lseki> I'll be happy to help with any questions 14:40:13 <lseki> the summary of what I found so far is in the etherpad 14:40:13 <lseki> #link https://etherpad.opendev.org/p/fix-nfs-online-extend 14:40:34 <lseki> and there's a patch from kaisers that is under progress 14:40:44 <kaisers> Note on that: I can fix the reno, np. I'm still working on our new CI setup as much as i can which is why i'm not doing much else here currently 14:40:53 <rosmaita> one question ... is the current status that if you try to do an online extend, we now reject it instead of going into error? 14:40:54 <lseki> which removes support from generic NFS and quobyte drivers 14:40:58 <lseki> #link https://review.opendev.org/#/c/725805/ 14:41:53 <rosmaita> or is that what ^^ does? 14:42:12 <lseki> rosmaita: no, it goes to error, and ^ makes it reject 14:42:15 <lseki> yes 14:42:35 <rosmaita> ok, i knew we had discussed it, couldn't remember if it had merged or not 14:43:13 <lseki> kaisers thanks 14:43:18 <rosmaita> i think this might be a good candidate patch for RC-2 14:43:40 <lseki> ++ 14:44:38 <rosmaita> ok, thanks for the update 14:45:06 <rosmaita> #topic open discussion 14:48:44 <jungleboyj> *crickets* 14:48:46 * lseki zzz 14:49:04 <eharney> i have questions after looking at https://bugs.launchpad.net/cinder/+bug/1896764 if nothing else :) 14:49:14 <openstack> eharney: Error: Could not gather data from Launchpad for bug #1896764 (https://launchpad.net/bugs/1896764). The error has been logged 14:49:28 <eharney> err 14:49:29 <smcginnis> eharney broke the bot 14:49:55 <e0ne> is launchpad down? 14:49:55 <eharney> anyway, this looks like a small problem from the volume local cache work 14:50:21 <eharney> but in fixing it, i'm not sure if "storage_protocol" is supposed to always been in volume_stats, or volume_stats['pools'] like this driver seems to do 14:51:05 <eharney> which makes me wonder if the local cache works w/ volumes using pools currently 14:51:09 <eharney> *with drivers using pools 14:51:44 <lseki> e0ne: LP is pretty slow today 14:52:02 <rosmaita> bug page just loaded for me 14:53:15 <LiangFang> eharney: sorry if I caused the bug 14:53:24 <eharney> i can submit the simple fix and we can look there to see if there are bigger problems, if that works 14:54:16 <eharney> LiangFang: nah, this is likely more about us not having sufficient documentation on driver API details 14:55:26 <rosmaita> eharney: this looks release-critical to me 14:55:59 <LiangFang> I'm happy to be included as the reviewer for such issue 14:56:20 <LiangFang> I will take a look of the bug tomorrow 14:56:26 <eharney> i'll post the patch in a moment with a follow-up question there 14:56:47 <rosmaita> sounds good, thanks eharney 14:57:24 <rosmaita> just a few minutes left 14:59:10 <rosmaita> i guess that's all for today then ... let's clear out and make way for horizon 14:59:17 <whoami-rajat__> thanks! 14:59:18 <jungleboyj> Thanks! 14:59:19 <e0ne> rosmaita: thanks! 14:59:32 <tosky> thanks! 14:59:34 <lseki> thanks 14:59:37 <LiangFang> thanks 14:59:47 <rosmaita> #endmeeting