14:00:20 <whoami-rajat> #startmeeting cinder 14:00:20 <opendevmeet> Meeting started Wed Nov 8 14:00:20 2023 UTC and is due to finish in 60 minutes. The chair is whoami-rajat. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:20 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:20 <opendevmeet> The meeting name has been set to 'cinder' 14:00:23 <whoami-rajat> #topic roll call 14:00:29 <simondodsley> o/ 14:00:33 <eharney> hi 14:00:41 <jayaanand> o/ 14:00:43 <msaravan> Hi 14:00:48 <akawai> o/ 14:00:50 <jungleboyj> o/ 14:00:50 <Saikumar> o/ 14:01:55 <keerthivasansuresh> o/ 14:03:10 <whoami-rajat> #link https://etherpad.opendev.org/p/cinder-caracal-meetings 14:03:14 <rosmaita> o/ 14:03:17 <whoami-rajat> hello everyone 14:03:20 <rosmaita> (sorry i'm late) 14:03:28 <whoami-rajat> no worries, we are just getting started 14:03:40 <whoami-rajat> let's start with announcements 14:03:43 <whoami-rajat> #topic nnouncements 14:03:47 <whoami-rajat> #topic announcements 14:03:56 <whoami-rajat> first, 2024.1 Caracal Cinder specific release schedule 14:04:01 <whoami-rajat> #link https://review.opendev.org/c/openstack/releases/+/899847 14:04:14 <whoami-rajat> the html render is more readable 14:04:15 <whoami-rajat> #link https://23a76c7dd4af37bd2885-b0e51bf7ce4fe92c12b00e3eba8de233.ssl.cf5.rackcdn.com/899847/3/check/openstack-tox-docs/35626f6/docs/caracal/schedule.html 14:04:31 <whoami-rajat> we discussed this last week but a lot of team members weren't around 14:04:37 <whoami-rajat> so bringing it up again today 14:04:55 <whoami-rajat> there are 2 dates that I've selected for midcycles 14:04:56 <whoami-rajat> Midcycle 1 : Wednesday 6th December 2023 (1400-1600 UTC) 14:04:56 <whoami-rajat> Midcycle 2 : Wednesday 14th February 2024 (1400-1600 UTC) 14:05:13 <whoami-rajat> let me know if there is any conflict with any major event on the following date and time 14:06:09 <whoami-rajat> and also you can check other dates 14:06:16 <whoami-rajat> I've tried to avoid dates closer to end of the year 14:06:22 <rosmaita> Dec 6 is a holiday in Finland and Spain 14:06:23 <whoami-rajat> since a lot of team members aren't around that time 14:07:02 <whoami-rajat> oh i remember that you mentioned it ... 14:07:40 <whoami-rajat> i can change it to 13 December, which is 1 week before spec freeze 14:07:53 <whoami-rajat> i don't see any deadline on R-16 week 14:08:38 <simondodsley> Showing some love for OpenStack for Midcycle 2 14:08:57 <rosmaita> 13 december looks clear, but 6 would be better ... we need to check with Gorka about whether he is taking the 6th off or not 14:09:16 <rosmaita> simondodsley: everybody loves cinder! 14:10:34 <whoami-rajat> i was planning Midcycles 2 weeks prior to some deadlines, didn't realize it fell on a perfect date! 14:10:56 <whoami-rajat> rosmaita, i will take a note to ask him when he's back 14:12:30 <whoami-rajat> okay moving on 14:12:46 <whoami-rajat> next, pep 517 14:12:59 <whoami-rajat> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/HVFN5RBSHRTM3B2UUKPAWKH6H6AT6CYR/ 14:13:24 <whoami-rajat> (the links were much shorter in the previous archive list) 14:14:09 <whoami-rajat> i don't know a lot about the install process but from what i understand, usage of setup.py has been deprecated and will removed in pip 23.1 14:14:24 <whoami-rajat> in favor of pyproject.toml files 14:14:52 <whoami-rajat> Sean Mooney has proposed a patch for cinder and other projects to have a minimal pyproject.toml file 14:14:56 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder/+/899952 14:15:28 <whoami-rajat> i don't know if this is discussed and approved by TC to be an openstack wide thing 14:15:37 <whoami-rajat> also the cinder patch is failing gate currently 14:15:49 <simondodsley> it's not that difficult of a change, especially as the `setup.py` we have in cinder is so minimal 14:15:51 <whoami-rajat> so I'm not sure what we should do as of now 14:17:04 <whoami-rajat> yeah, it only has a constraint of pbr and the version looks pretty old 14:17:21 <whoami-rajat> I just want to make sure if this is the approach we are going with openstack wide 14:17:22 <simondodsley> i think the suggested patch could have some additional `[project]` info in there 14:18:31 <simondodsley> i also wonder if `pbr` should be `6.0.0` 14:19:19 <rosmaita> that functional-py39 job that jon bernard flagged -- the error looks like what i thought this patch was supposed to address 14:19:29 <whoami-rajat> pbr 6.0.0 released today so that might be the reason the patch has old version 14:20:01 <simondodsley> actually our `requirements.txt` states `>=5.8.0` 14:20:34 <rosmaita> well, in the gates, we'll be most likely using whatever upper-constraints allows 14:21:11 <simondodsley> is `pbr` even mentioned in the upper-constraints? 14:21:30 <simondodsley> so this all become moot anyway 14:21:53 <rosmaita> pbr===6.0.0 14:22:11 <simondodsley> well that answers that 14:22:45 <rosmaita> i can follow up on this if no one else is interested 14:22:46 <simondodsley> we should get the all the pyproject patches to reflect 6.0.0 IMHO 14:23:48 <whoami-rajat> rosmaita, do you mean for cinder or openstack wide? 14:24:02 <rosmaita> hmmm ... i guess both 14:24:10 <rosmaita> i was originally thinking cinder-only 14:24:31 <whoami-rajat> ok, i can help out with the cinder part though i don't understand the pep517 completely 14:24:44 <whoami-rajat> yeah first we can address the cinder issue 14:24:51 <simondodsley> Just get Sean to update all his patches, but we should also fix our requirements 14:25:07 <whoami-rajat> but again, my idea was to only merge it in cinder if it's going to be an openstack wide thing 14:25:25 <simondodsley> well it has to be done 14:26:07 <rosmaita> i haven't actually read through that email thread yet, will do that after the meeting 14:26:29 <rosmaita> but i think simondodsley is right, it needs to be done soon-ish 14:26:50 <whoami-rajat> simondodsley, there are different solutions proposed by Sean in his original mail and i still haven't seen any acceptance on a particular one 14:27:38 <simondodsley> i'll read the thread again - it was very long... 14:31:12 <whoami-rajat> we can keep an eye on the thread for updates, i just wanted to raise some awareness that this is something that might affect our project as well 14:31:37 <whoami-rajat> with that, let's move on to next announcement 14:31:40 <whoami-rajat> next, Call for Writers or OpenInfra Live Participants 14:31:46 <whoami-rajat> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/QEE7O3A3DWIX2P57G75OCB3GMUPLQPOY/ 14:32:17 <whoami-rajat> TBH, i have never participated in this but if you want details, refer to the mail or Contact @kbarrientos:matrix.org 14:33:03 <whoami-rajat> the general idea here is if you have any topic for the openinfra live, please write to Kristin 14:33:40 <whoami-rajat> next, [docs] Encourage search engines to show newest version of OpenStack docs first 14:33:45 <whoami-rajat> #link https://lists.openstack.org/archives/list/openstack-discuss@lists.openstack.org/thread/Q3AJ5C54IZSGBJ76JTVXTWVTFNXDRD7N/ 14:34:07 <whoami-rajat> I've faced this many times and finally someone wrote it on the ML 14:34:31 <whoami-rajat> whenever I try to search for any doc, its usually the older release versions that show up, pike, queens etc and rarely the latest one 14:34:59 <whoami-rajat> so i end up replacing <release_name> -> latest which works but it's little inconvenient 14:35:18 <whoami-rajat> there were 2 solutions proposed 14:35:22 <whoami-rajat> 1. Prevent Google from indexing the older versions by using `noindex`: https://developers.google.com/search/docs/crawling-indexing/block-indexing 14:35:46 <whoami-rajat> i don't know if we can do something about this since it needs to be included in the html meta tag itself 14:35:57 <whoami-rajat> 2. Removing old documentation so it is flushed from Google 14:36:11 <whoami-rajat> this one i didn't like because a lot of deployments are running on older releases 14:36:20 <whoami-rajat> and they might want to refer to info specific for their release 14:36:29 <whoami-rajat> the latest docs info might not be relevant for them 14:37:09 <whoami-rajat> anyway, i wasn't looking for any discussion around this, just wanted to mention this since i felt others might have bumped into the same issue 14:37:28 <whoami-rajat> that's all the announcements from my side 14:37:34 <whoami-rajat> anyone has anything else to announce? 14:37:48 <simondodsley> i agree that the old docs are an issue. 14:38:02 <simondodsley> noindex is probbaly the best of the terrible solutions available 14:38:32 <whoami-rajat> +1 14:39:31 <whoami-rajat> looks like that was all for announcements 14:39:33 <whoami-rajat> let's move to topics 14:39:50 <whoami-rajat> #topic Increase size of volume image metadata values 14:39:56 <whoami-rajat> drencrom, that's you 14:40:01 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder/+/868485 14:40:48 <whoami-rajat> this is the discussion summary from the PTG regarding this 14:40:50 <drencrom> Hi, I just wanted to know it this topic was discussed during the PTG 14:40:51 <whoami-rajat> #link https://wiki.openstack.org/wiki/CinderCaracalPTGSummary#Image_Metadata_in_Cinder.2C_Nova.2C_and_Glance 14:41:02 <rosmaita> yes, it was 14:41:17 <rosmaita> and that reminds me that i had the action item to send something to the ML about this 14:42:18 <drencrom> Ok, so I'll follow in the ML then 14:42:47 <rosmaita> the tl;dr is that we do not want to allow metadata values that long 14:43:15 <drencrom> Ok, that might be a problem for the people that is currently using large metadata vaues 14:43:19 <rosmaita> and like sean pointed out on the review (i think), nova just disregards them anyway 14:44:02 <rosmaita> drencrom: you are correct, we don't want to unnecesarily break people 14:44:17 <whoami-rajat> yeah we need to be consistent about this else there might be loss of metadata values when these properties are send/received between different projects 14:45:45 <drencrom> I think the easy path is to ignore them in nova and document that, but the IO issue is harder to solve without breaking things 14:48:48 <whoami-rajat> so we can follow up on this on the ML 14:49:27 <whoami-rajat> and raise concerns there to see what other teams (glance, nova) thinks 14:49:43 <rosmaita> i'll try to get the email out in the next day or so 14:50:01 <drencrom> thanks rosmaita 14:50:06 <whoami-rajat> great, thanks rosmaita 14:50:12 <whoami-rajat> next topic 14:50:19 <whoami-rajat> #topic Review priorities for M-1 (November 16, 2023) 14:50:35 <whoami-rajat> next week is M-1 and we targeted 2 things for it that we should prioritize 14:50:49 <rosmaita> drencrom: please bug me in #openstack-cinder if you don't see something by 1500 UTC on friday 14:51:08 <whoami-rajat> 1. SqlAlchemy 2.0 support 14:51:15 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder/+/887032 14:51:49 <whoami-rajat> it's not a single patch change but a series if you look on the right side in Relation Chain 14:52:28 <whoami-rajat> ideally we should get all patches in but merging till "Add job to test with SQLAlchemy master (2.x)" should address the 2.0 issues 14:52:41 <whoami-rajat> which in total are 4 patches 14:53:00 <whoami-rajat> so request everyone to please take a look and review on priority 14:54:14 <whoami-rajat> 2. NFS extend in-use volume support spec 14:54:20 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder-specs/+/896014 14:54:38 <whoami-rajat> I reviewed the change again and it still makes sense to do 14:54:56 <whoami-rajat> remember it's not only fixing the extend in-use problem for generic nfs but all vendors inheriting from it 14:55:13 <whoami-rajat> the use case was mainly for netapp nfs but dell nfs should also benefit from it 14:55:27 <whoami-rajat> it's a reproposal so should be pretty quick to review it 14:55:39 <whoami-rajat> there are additions to the testing part which is good 14:56:09 <whoami-rajat> that's all for the topics 14:56:17 <whoami-rajat> let's quickly go through review requests 14:56:25 <whoami-rajat> #topic review requests 14:56:52 <whoami-rajat> in the PTG, we discussed about this and since the last meeting we came up with a structure that should be followed for review requests 14:56:57 <whoami-rajat> Structure: 14:56:57 <whoami-rajat> One author should only add one patch for review 14:56:57 <whoami-rajat> The patch should be based on the approaching deadling (M1, M2, M3 etc) 14:56:57 <whoami-rajat> There should be a comment explaining the patch being added 14:57:38 <whoami-rajat> hope that helps reviewers focus on only the important patches 14:57:43 <whoami-rajat> we have 2 review requests this week 14:57:47 <whoami-rajat> 1. HPE 3par: Unable to create clone of replicated vol 14:57:52 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder/+/884594 14:58:21 <whoami-rajat> i think this was also a RC2 request which didn't make it 14:58:41 <whoami-rajat> so good to review 14:58:56 <whoami-rajat> 2. [Hitachi] Fix exception when deleted volume is busy 14:58:58 <whoami-rajat> #link https://review.opendev.org/c/openstack/cinder/+/885986 14:59:18 <whoami-rajat> this was also part of the RC2 targets last cycle 14:59:23 <whoami-rajat> already has a +2 from me 14:59:40 <whoami-rajat> so other reviewers can take a look 15:00:02 <whoami-rajat> we are out of time 15:00:08 <whoami-rajat> thanks everyone for attending 15:00:16 <jungleboyj> Thank you! 15:00:22 <simondodsley> cheers 15:00:22 <whoami-rajat> #endmeeting