15:00:20 <carloss> #startmeeting manila
15:00:20 <opendevmeet> Meeting started Thu Sep  8 15:00:20 2022 UTC and is due to finish in 60 minutes.  The chair is carloss. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:20 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:20 <opendevmeet> The meeting name has been set to 'manila'
15:00:27 <felipe_rodrigues> o/
15:00:32 <carloss> courtesy ping: vkmc dviroel vhari gouthamr
15:00:33 <haixin> o/
15:00:36 <vkmc> o/
15:00:37 <lucasmoliveira059> \o
15:00:44 <gouthamr> o/
15:00:46 <HelenaDantas[m]> o/
15:00:47 <dviroel> o/
15:01:11 <francie_> o/
15:01:40 <vhari> hi
15:02:35 <nahimsouza[m]> hi
15:02:59 <caiquemello[m]> \o
15:03:20 <carloss> hello everyone! good to have you here today
15:03:27 <luizsantos[m]> o/
15:04:07 <carloss> today's meeting agenda:
15:04:11 <carloss> #link https://wiki.openstack.org/wiki/Manila/Meetings#Next_meeting
15:04:14 <MatheusAndrade[m]> o/
15:04:38 <raffaelacunha[m]> o/
15:05:08 <carloss> let's start with our announcements for today
15:05:15 <carloss> #topic Announcements
15:06:30 <carloss> I would like to share with you today that francie_'s outreachy internship ended two weeks ago
15:06:45 <carloss> and I would like to publicly thank her for all the amazing work she did for manila in such short time
15:06:57 <carloss> we can still see her around working on changes (which is awesome)
15:07:31 <carloss> thank you for all the work francie_!
15:07:53 <carloss> https://review.opendev.org/q/owner:franciedigital%2540gmail.com - I'm glad to see all of these changes merged
15:08:19 <carloss> and also glad to see how much you enjoyed working on every single change on this list
15:08:20 <gouthamr> indeed; great job francie_!
15:09:04 <carloss> I would also like to appreciate the work of maaritamm during this Outreachy internship cycle :D
15:09:17 <francie_> Thanks all
15:11:18 <francie_> It was a great pleasure working with you guys
15:11:37 <carloss> maaritamm and I were co-mentors and she was impressive with organization and her OSC skills
15:11:44 <carloss> it was great!
15:11:58 <carloss> remember that you'll ever be welcome in this community francie_ :)
15:13:01 <carloss> second thing on announcements:
15:13:02 <carloss> Schedule and Deadlines:
15:13:14 <carloss> #link https://releases.openstack.org/zed/schedule.html
15:13:29 <carloss> we're 1 week past the official feature freeze
15:13:31 <carloss> but
15:13:45 <carloss> I bought some of our candidate changes one extra week to be merged
15:14:22 <carloss> #link https://lists.openstack.org/pipermail/openstack-discuss/2022-September/030281.html
15:15:05 <carloss> we can go over the mentioned changes in a few, they are not merged yet but they are getting close
15:15:24 <carloss> and the last announcement I have for today:
15:15:43 <carloss> As you might have seen in the mailing list: Bugsquash postponed for next week! (Sep 13th to 15th)
15:16:15 <carloss> we had a bugsquash event scheduled for this week but due to the feature freeze exceptions we needed to postpone it for a week
15:16:29 <carloss> so we would keep the reviewers and change owners attention on the FFE changes
15:17:42 <carloss> that's all I had for $topic
15:17:55 <carloss> do you have an announcement you'd like to share with us today?
15:20:16 * carloss taking silence as a no
15:20:26 <carloss> #topic Review Focus
15:20:34 <carloss> #link https://etherpad.opendev.org/p/manila-zorilla-review-focus (Review Focus Etherpad)
15:22:09 <carloss> let's go over the three changes that are our focus now
15:22:15 <carloss> starting with
15:22:51 <carloss> #link https://review.opendev.org/c/openstack/manila/+/825008 (Metadata for Share Snapshots Resource)
15:23:20 <carloss> this change got more reviewers' traction since last week and issues found were solved and comments addressed
15:23:43 <carloss> I'm testing this issue again now
15:24:12 <carloss> I see the rebase on top of share transfer and the conflict resolution (thanks ashrodri for anticipating that)
15:25:37 <carloss> s/testing this issue/testing this change
15:26:13 <carloss> I intend to post my tests result soon, so there is enough time to act on them
15:27:06 <carloss> something else to add to this change?
15:28:00 <carloss> thank you to the reviewers and ashrodri for solving the comments :)
15:28:19 <carloss> second change in the list is:
15:29:00 <carloss> #link https://review.opendev.org/c/openstack/manila/+/843832 Update micversion to 2.73, support share transfer between project |
15:29:18 <carloss> this has also received more reviews since last week - thanks haixin for addressing all the comments
15:29:27 <carloss> this is the next to test on my quere
15:29:31 <carloss> queue*
15:29:44 <haixin> thanks for review:)
15:30:13 <carloss> you're welcome haixin
15:30:35 <gouthamr> i'm looking at this right now; if there are minor/non-breaking issues, i'll note them so we can get the patch merged and push up a change later for minor fixes
15:30:40 <felipe_rodrigues> thanks Haixin, there is still an open point.. it's a corner case, If it's too complicated to be solved now, we could merge this way and work on it as a follow-up.
15:31:28 <felipe_rodrigues> overall looks good, nice work haixin :)
15:32:07 <carloss> gouthamr++ felipe_rodrigues++ - thank you for the reviews on this change
15:32:21 <carloss> and thank you for the work haixin
15:33:09 <carloss> third change on the list:
15:33:39 <carloss> #link https://review.opendev.org/c/openstack/manila/+/848987 Refactor the Ceph NFS driver to use Cephadm NFS
15:34:04 <carloss> I feel this is getting quite close too - thank you for addressing the comments related to latest review vkmc
15:34:45 <gouthamr> waiting on results here: https://review.opendev.org/c/openstack/manila-tempest-plugin/+/834220
15:36:23 <gouthamr> if the ceph-nfs job does better here than on the manila patch, we can merge this - we've run into resource exhaustion :/
15:36:54 <gouthamr> manila-tempest-plugin-cephfs-nfs: https://zuul.opendev.org/t/openstack/build/7dd2735257ed4c658416f4b42c525b66
15:37:46 <gouthamr> resource exhaustion makes manila-share, nova-api, nova-cpu etc lose connection to the db and fail tests
15:39:14 <gouthamr> and its apparent this only happens when scenario tests are run - so we may have to think of some better/long term strategies here
15:41:01 <carloss> hmm - I see concurrency on the new job set to two... would if help if we bump down? that's one approach we took with the generic driver job when we switched to a more recent image that needed more resources
15:41:58 <carloss> I think there's a chance to hit timeout in the job
15:42:23 <carloss> but I'm unsure what would happen more often: the resource exhaustion or timeout in the job
15:44:24 <gouthamr> yeah, hard to tell; also this could be a test node on a slow hypervisor
15:44:55 <carloss> ^ that's a possibility as well :)
15:45:44 <dviroel> what about create a DNM change, with a new job which runs only scenario test, with concurrency set to 1
15:46:41 <gouthamr> some other strategies: run api tests with max concurrency, then run scenario tests with concurrency 1 --- it'll need a new custom job definition
15:47:01 <carloss> yep
15:47:39 <gouthamr> or do multi-node testing like we were planning and isolate ceph on a node by itself -- we can still hopefully run with a good test concurrency with this
15:48:23 <gouthamr> i like dviroel_'s idea for now though - we can do a DNM patch if the manila-tempest-plugin change suffers the same fate
15:49:05 <carloss> ++ - that would help to validate at this point
15:50:23 <dviroel> multi-node approach is a good one, but may take some time to get everything working.
15:51:54 <carloss> ++
15:53:43 * gouthamr net-split and a disconnect?
15:53:45 <gouthamr> https://review.opendev.org/c/openstack/manila-tempest-plugin/+/856540
15:53:47 <gouthamr> ^ testing with concurrency=1 in parallel because i'm impatient :D
15:54:40 <carloss> xD
15:54:41 <carloss> thanks gouthamr
15:57:34 <carloss> okay, so we have a direction - thanks for checking the tests gouthamr
15:57:42 <carloss> we can follow up on the test approach
15:57:53 <carloss> we have ~3 minutes on this meeting yet
15:58:27 <carloss> sorry we didn't have time to go over the bugs list again today vhari :(
15:58:55 <carloss> okay, that's all I had for $topic
15:59:09 <carloss> and I think we can get back two minutes to our days :D
15:59:39 <carloss> thank you for joining, let's continue the reviews to get these changes moving
15:59:55 <carloss> #endmeeting