15:00:17 <carloss> #startmeeting manila 15:00:17 <opendevmeet> Meeting started Thu Nov 21 15:00:17 2024 UTC and is due to finish in 60 minutes. The chair is carloss. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:17 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:17 <opendevmeet> The meeting name has been set to 'manila' 15:00:35 <carloss> courtesy ping: dviroel vhari gouthamr carthaca msaravan pulluri ashrodri 15:00:43 <carthaca> hi 15:00:45 <dviroel> o/ 15:00:50 <vhari> hi 15:00:51 <msaravan> hi 15:01:37 <haixin> o/ 15:02:46 <gireesh> hi 15:03:43 <ashrodri> o/ 15:05:00 <carloss> o/ hello everyone 15:05:06 <carloss> good quorum, let's get started 15:05:35 <carloss> today's meeting agenda: 15:05:44 <carloss> #link https://wiki.openstack.org/wiki/Manila/Meetings#Next_meeting (Meeting agenda) 15:05:58 <carloss> #topic Announcements 15:06:32 <carloss> Schedule and deadlines 15:06:46 <carloss> we are 1 week past the epoxy-1 milestone 15:07:17 <carloss> I'll be retargeting the bugs taht did't make it 15:07:42 <carloss> and I finally proposed the change that adds all of the manila specific deadlines to the official schedule 15:08:05 <carloss> #link https://review.opendev.org/c/openstack/releases/+/935891 (Add manila deadlines to the epoxy schedule) 15:08:05 <carloss> please take a look at this change and provide feedback 15:09:12 <carloss> most of these dates shouldn't sound like news 15:09:17 <carloss> because we've discussed them at the ptg 15:09:42 <carloss> so again: feedback is welcome 15:09:58 <carloss> following the proposed deadlines, spec freeze is not that far away 15:10:20 <carloss> and we will need to focus on the reviews for the next couple of weeks 15:10:36 <carloss> that's all I had for announcements though. Is there something else you'd like to share with us? 15:12:14 <carloss> #topic CI documentation 15:12:31 <carloss> so we've had a lot of discussions in the past around $topic 15:13:14 <carloss> and it's always been something I to focus really bad 15:13:32 <carloss> s/I/I wanted 15:14:07 <carloss> there was a discussion in a TC meeting two weeks ago where people mentioned some issues with setting up CI systems 15:14:34 <carloss> and one thing that was mentioned was the lack of documentation that we have 15:14:56 <carloss> and I'm saying we but I'm referring to other projects that also require third party systems to be set up and reporting 15:14:59 <kpdev> o/ 15:15:21 * carloss waves at kpdev 15:15:53 <carloss> so it would be nice if we could get something started and I know there are people that have been through that process here 15:17:22 <carloss> our cinder folks are currently looking at consolidating what they have and enhancing that too 15:17:54 <carloss> that would be a good thing for us too 15:18:17 <carloss> I can get this started but I will definitely need some help from people that have been working with third party CI systems lately 15:19:01 <carloss> for example: msaravan and gireesh mentioned they would start looking at issues and the deployment of their CI soon 15:19:09 <carloss> I believe we can benefit from their experience 15:19:47 <carloss> I will start an etherpad and draft something as a starting point, but would be ideal if you get involved as well 15:20:26 <carloss> the CI setup for vastdata has been completed not that long ago too, so that's other maintainer that I can reach out to help as well 15:21:10 <carloss> we should do this collaboratively because I believe the experience from everyone will benefit any new CI setup from now on 15:21:20 <carloss> how does that sound? 15:22:01 <vhari> ++ 15:24:22 <gireesh> Yes, I have deploy the zuul from scratch, I have experience around this. there is node pool issue which msaravan is working on 15:25:21 <msaravan> ++ 15:25:47 <carloss> ack, thank you very much. Your knowledge will be very much appreciated 15:26:01 <carloss> any other thoughts on $topic? 15:28:32 <carloss> taking silence as no 15:28:47 <carloss> #topic Review Focus 15:28:55 <carloss> #link https://etherpad.opendev.org/p/manila-epoxy-review-focus (Epoxy review focus etherpad) 15:29:18 <carloss> two specs to review on this cycle so far 15:29:25 <carloss> #link https://review.opendev.org/c/openstack/manila-specs/+/933558 (Add spec for share group affinity policy) 15:29:44 <carloss> kpdev carthaca: thank you for sharing your thoughts on this spec 15:30:20 <carloss> I intend to review it soon and I am also hoping I can get some reviewers' attention to this too 15:30:42 <haixin> i will review that spec. 15:30:50 <carloss> haixin: thank you very much 15:31:28 <carloss> gireesh gouthamr ashrodri: can you please take a look too? 15:32:00 <carloss> the other spec that is being proposed is: 15:32:05 <carloss> #link https://review.opendev.org/c/openstack/manila-specs/+/932637 (Add spec for subnet force deletion) 15:32:42 <carloss> another feature we've discussed at the PTG 15:32:48 <carloss> also needs some reviews 15:33:30 <carloss> carthaca gireesh kpdev ashrodri haixin: can you please take a look at it too? 15:33:43 <haixin> sure 15:33:46 <carloss> tyvm 15:33:48 <carthaca> ack 15:33:49 <gireesh> sure 15:34:08 <kpdev> ack, we better add them are reviewers 15:34:39 <kpdev> them *as 15:35:12 <carloss> yep 15:37:12 <carloss> #link https://review.opendev.org/c/openstack/manila/+/898306 (Improve DB queries by join-loading Share model ) 15:37:20 <carloss> this change has been reviewed in the past 15:37:30 <carloss> so I'd encourage the reviewers to revisit it 15:37:38 <carloss> doesn't seem far to be merged 15:38:27 <carloss> #link https://review.opendev.org/c/openstack/manila/+/931050 (Allow to update access rule for type 'ip') 15:38:52 <carloss> one of the features being proposed 15:38:56 <carloss> seems to be in a good shape 15:39:14 <carloss> I'll take a look and also ask gouthamr when he's back :) 15:40:09 * ashrodri stepped away for a bit. yes i will take a look at both specs 15:40:15 <carloss> #link https://review.opendev.org/c/openstack/manila/+/929091 (Pass share network subnet metadata updates to backend drivers) 15:40:37 <carloss> another feature we've discussed and it should sound familiar to the recent share metadata updates we've introduced 15:41:30 <carloss> there are some reviewers assigned, I hope they (myself included) can take a look at this and provide feedback until next week 15:42:52 <carloss> and there are several other fixes waiting for reviews in the etherpad 15:43:54 <carloss> my ask is: please proactively take a look in the fixes in the etherpad 15:44:49 <carloss> we've a lot of bugs to go through today 15:45:06 <carloss> so I'd like to leave at least a bit of time for that in this meeting 15:45:49 <carloss> kpdev: thanks for sending your priority list to #openstack-manila 15:46:18 <carloss> I hope that helps guiding reviewers 15:46:26 <carloss> #topic Bug Triage 15:46:35 <carloss> #link https://etherpad.openstack.org/p/manila-bug-triage-pad-new (Bug Triage etherpad) 15:46:39 <vhari> ty carloss .. have a couple of bugs to start with 15:46:45 <vhari> #link https://bugs.launchpad.net/manila/+bug/2088269 15:48:25 <vhari> there is a fix proposed https://review.opendev.org/c/openstack/manila/+/935348 15:48:40 <carloss> yep, fix looks good, I've asked for a release note now 15:49:00 <carloss> we can target it to epoxy-2 and priority can be low on this 15:49:03 <carloss> but we are very close 15:49:13 <vhari> I will move the bug to in progress 15:49:18 <carloss> ack, ty 15:49:19 <vhari> ack gr8 15:49:32 <vhari> next #link https://bugs.launchpad.net/manila/+bug/2089062 15:50:05 <carloss> so our nova folks also dedicated some time to test share backups alongside the virtiofs patch series 15:50:12 <carloss> virtiofs is quite close to being merge 15:50:18 <carloss> merged* 15:50:33 <carloss> however, there are two big backup issues, and one of them we've triaged last week 15:50:41 <carloss> the other I filed this week and we can triage it 15:51:02 <carloss> during the data copy process in the generic backup approach, we remove all of the access rules of the share 15:51:18 <carloss> and that will make the VM lose connectivity to the share 15:51:35 <carloss> we should avoid that 15:52:28 <carloss> so the initial idea is to prevent the backup to start in case the share and its access rules are locked, because that means someone is using the share and it will be harmful if we just drop them 15:52:48 <carloss> maybe this also impacts the generic share migration approach 15:53:14 <carloss> so what we can do? in the backups API, if the share or its access rules have locks, we should prevent the backup creation 15:54:04 <carloss> ideally we'd do it only when we know we'll use the generic backup approach, but that can leave the API request hanging 15:54:16 <carloss> and we'd fall into the same issue that we've discussed last week 15:54:27 <carloss> we need an assignee for this 15:54:43 <carloss> as the response might change in some cases, we need to evaluate if we'll need a new API microversion 15:55:02 <carloss> anyone familiar with backups willing to fix this issue? :) 15:56:56 <vhari> while being considered ^^ :) that's all the time we have for bugs today 15:57:05 <vhari> and back to you carloss 15:57:40 <carloss> ack, tyvm vhari 15:58:20 <carloss> we can set this to medium priority and target to epoxy-3 15:58:29 <carloss> just to get some closure to the triaging part :) 15:58:32 <vhari> sure, will do 15:58:33 <carloss> let's wrap up 15:58:50 <carloss> thanks for participating 15:59:04 <carloss> let's get back to #openstack-manila 15:59:08 <carloss> have a great day! :D 16:00:19 <carloss> #endmeeting