15:00:54 <gouthamr> #startmeeting manila
15:00:55 <openstack> Meeting started Thu May 14 15:00:54 2020 UTC and is due to finish in 60 minutes.  The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:56 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:58 <openstack> The meeting name has been set to 'manila'
15:01:07 <tbarron> hi
15:01:16 <dviroel> o/
15:01:29 <andrebeltrami> o/
15:01:43 <carloss> hi
15:01:56 <vkmc> o/
15:01:59 <danielarthurt> o/
15:02:08 <lseki> o/
15:02:15 <carthaca> hi
15:02:42 <gouthamr> courtesy ping: xyang toabctl ganso amito
15:02:44 <gouthamr> o/ hello everyone, thank you for joining..
15:02:54 <gouthamr> here's our agenda for this meeting
15:03:00 <gouthamr> #link https://wiki.openstack.org/wiki/Manila/Meetings#Next_meeting
15:03:08 <amito> Hey o/
15:03:21 <gouthamr> #topic Announcements
15:03:55 <gouthamr> It's no surprise, we've had yet another smooth on-time release!
15:04:46 <gouthamr> Thank you all for your efforts in getting Ussuri out of the gate; now it's officially the last supported release from the community
15:05:06 <tbarron> gouthamr++ you make  it go smoothly
15:05:18 <vkmc> gouthamr++
15:05:35 <gouthamr> :)
15:05:40 <gouthamr> To that effect, there's a virtual celebration gathering that's planned
15:05:51 <gouthamr> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-May/014781.html (Ussuri Virtual Celebration)
15:06:16 <gouthamr> please do join and represent this community :)
15:07:15 <gouthamr> We had a couple of community meetings yesterday
15:07:33 <gouthamr> where a bunch of contributors provided the highlights of the release
15:07:39 <gouthamr> you can watch them here
15:07:43 <gouthamr> #link https://www.youtube.com/playlist?list=PLKqaoAnDyfgpYADSiOfIVwgKb5zbL0GJE
15:08:59 <gouthamr> cool... and next, a reminder that the Victoria PTG is a couple of weeks away
15:09:09 <gouthamr> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-April/014567.html (ML Post with slots)
15:09:09 <gouthamr> #link https://etherpad.opendev.org/p/vancouver-ptg-manila-planning (Victoria PTG Planning Etherpad)
15:09:09 <gouthamr> #link https://ethercalc.openstack.org/126u8ek25noy (PTG Slot Signup sheet)
15:09:35 <gouthamr> I see new topics every time i look at that etherpad, i'll give it another week before discussing time slots :)
15:10:01 <gouthamr> that's all i had in terms of announcements today, anyone else got any?
15:10:35 <gouthamr> #link https://www.openstack.org/ptg/
15:10:57 <gouthamr> ^ this website's been updated with the schedule and registration details
15:11:29 <gouthamr> moving on...
15:11:33 <gouthamr> #topic Generic driver test consolidation
15:12:02 <gouthamr> so, as you're aware, we're working on converting all our test jobs to zuulv3
15:12:30 <gouthamr> native zuulv3 format*
15:12:44 <gouthamr> and we discussed some test job consolidation last week
15:12:56 <gouthamr> #link https://review.opendev.org/726291/ (Generic Driver test job consolidation)
15:13:50 <gouthamr> i wrote a ballad for a commit message, i apologize, but, i wanted to explain all the permutations of testing we had, and what we'll lose wrt this consolidation
15:14:22 <gouthamr> i'd be happy to have feedback to improve that change..
15:15:35 <gouthamr> the tl;dr remains that we're testing the generic driver in more combinations than we need; and we've been neglecting the health of these jobs since they fail so often due to reasons not entirely in the manila driver
15:15:55 <vkmc> gouthamr, not sure if there is something already, but I was thinking about submitting some docs on that
15:16:00 <vkmc> if you are not up to it
15:16:10 <vkmc> it would be nice to have this in the contributors guide or something like that
15:16:16 <gouthamr> vkmc: what kind of docs are you thinking?
15:16:22 <vkmc> I think we have something about ci... but for third party ci
15:16:54 <gouthamr> oh, a doc about adding/modifying a CI job?
15:16:56 <vkmc> just leaving some record on the decisions we take and why for the ci jobs so contributors know what we are testing and where
15:17:39 <vkmc> adding/modifying ci jobs, I think we have something on that
15:17:44 <vkmc> at least for third party ci
15:18:13 <gouthamr> ack
15:18:31 <vkmc> I can push a draft for review and we can discuss there, sounds good?
15:18:37 <gouthamr> vkmc++
15:18:45 <vkmc> cool :)
15:19:32 <gouthamr> thank you, vkmc.. so, a bit more about this consolidated job... it runs more tests than any of the individual jobs did
15:19:59 <gouthamr> and catches an old bug that we had an open change for...
15:20:18 <gouthamr> #link Closes-Bug: 1703581
15:20:18 <openstack> bug 1703581 in OpenStack Shared File Systems Service (Manila) "[Generic driver] CIFS extension failing because of volume in use thus unable to unmount" [Undecided,In progress] https://launchpad.net/bugs/1703581 - Assigned to Goutham Pacha Ravi (gouthamr)
15:20:18 <gouthamr> gah
15:20:22 <gouthamr> oh
15:20:27 <gouthamr> good bot
15:21:07 <gouthamr> ^ currently assigned to me, but was a patch from Jan Vondra...
15:21:51 <gouthamr> if we revive that change, i think we'll fix that issue; i'll follow up: https://review.opendev.org/#/c/498445/
15:22:07 <gouthamr> any concerns/questions?
15:22:56 <gouthamr> taking that for a no.. :)
15:23:46 <gouthamr> please review that change when possible...
15:23:52 <gouthamr> #topic Reviews needing attention
15:24:00 <gouthamr> #link https://etherpad.openstack.org/p/manila-victoria-review-focus
15:24:27 <gouthamr> I don't see any new notes, i presume changes there are ack'ed..
15:24:36 <gouthamr> is there anything that we need to talk about?
15:26:08 <carthaca> I would like to have a second reviewer for https://review.opendev.org/#/c/725791/
15:26:09 <dviroel> We are trying to contact the contributor on this patch https://review.opendev.org/#/c/703025/
15:27:31 <gouthamr> carthaca: ack.. ty for bringing that up..
15:27:47 <dviroel> added carthaca change to the etherpad
15:28:06 <carthaca> dviroel: thanks
15:28:15 <gouthamr> dviroel: great... if the owner of 703025 doen't respond we might have to push a patch with the changes ourselves..
15:29:15 <gouthamr> dviroel: it's possible they're not working at the moment, or things are getting lost in translation.. thanks for being patient :)
15:29:46 <gouthamr> any other changes to discuss?
15:30:33 <gouthamr> else let's move on to the next section
15:30:38 <gouthamr> #topic Bugs (vhari)
15:31:08 <gouthamr> vhari has a conflict today
15:31:23 <gouthamr> lets look at the etherpad that she's updated
15:31:29 <gouthamr> #link https://etherpad.openstack.org/p/manila-bug-triage-pad-new (Bug Triage etherpad)
15:31:43 <gouthamr> https://bugs.launchpad.net/manila/+bug/1878477
15:31:43 <openstack> Launchpad bug 1878477 in OpenStack Shared File Systems Service (Manila) "Devstack plugin sets service instance parameters for all drivers" [Low,Confirmed] - Assigned to Victoria Martinez de la Cruz (vkmc)
15:32:04 <gouthamr> this one is triaged, needs a milestone
15:32:07 <vkmc> oh hi, I just triaged that one
15:32:19 <gouthamr> thanks vkmc - i'll assign it to victoria-1
15:32:24 <vkmc> ++
15:32:29 <tbarron> haha
15:32:55 * gouthamr it's gonna be this, all release
15:32:58 <gouthamr> #link https://bugs.launchpad.net/manila/+bug/1878244
15:32:58 <openstack> Launchpad bug 1878244 in OpenStack Shared File Systems Service (Manila) "[Doc] Create share from snapshot in another pool or backend" [Undecided,New] - Assigned to Douglas Viroel (dviroel)
15:33:08 <vkmc> gouthamr, count on that xD
15:33:35 <gouthamr> haha! thanks for opening a doc bug just like the bot would do dviroel
15:33:53 <dviroel> what went wrong?
15:33:56 <gouthamr> wonder what happened to the bot
15:34:13 <dviroel> DOCImpact vs DocImpact?
15:34:24 <carloss> ^ may be this
15:35:38 <gouthamr> perhaps, we can check the system-config repo to confirm
15:35:51 <gouthamr> this one is a medium impact, can i target to Victoria-1?
15:36:10 <dviroel> Will upload a patch soon, didn't see too much to add since API Ref was already upload together with the main change
15:36:12 <dviroel> gouthamr: yes
15:36:20 <gouthamr> thanks
15:36:29 <gouthamr> #link https://bugs.launchpad.net/manila/+bug/1877613
15:36:30 <openstack> Launchpad bug 1877613 in OpenStack Shared File Systems Service (Manila) "Share servers cannot be deleted with the generic driver" [Undecided,New]
15:37:12 <gouthamr> hmmm, "If a share server is deleted in nova"
15:37:26 <vkmc> queens
15:37:54 <dviroel> alpha23 didn't like your template
15:38:10 <gouthamr> lol, yes..
15:38:17 <gouthamr> but i'm thinking there are a few issues here
15:38:53 <gouthamr> one, they're expecting the driver to realize that the share server is gone, and provision a new one for new shares
15:38:58 <gouthamr> that's a reasonable ask...
15:39:32 <gouthamr> but, i think for their current situation, they can force-delete the share and share-server from manila
15:40:15 <gouthamr> i don't know if we have a force delete API for share servers
15:40:54 <gouthamr> we certainly can delete the shares.. although, we could be leaving behind the underlying cinder volumes
15:41:52 <gouthamr> this situation isn't normal - i.e., regular users can't remove their share servers from nova - they'd have to assume the manila service role, or be admin
15:43:18 <gouthamr> i'll update this bug with these thoughts... we'll also remind them that this driver is used for testing only, and we could use help to get it working
15:43:51 <gouthamr> #link https://bugs.launchpad.net/manila/+bug/1877410
15:43:51 <openstack> Launchpad bug 1877410 in OpenStack Shared File Systems Service (Manila) "[RFE] Container driver uses deprecated brctl commands " [Low,New] - Assigned to Goutham Pacha Ravi (gouthamr)
15:44:01 <gouthamr> oh, what's missing here
15:44:11 <gouthamr> nothing, it's triaged..
15:44:20 <gouthamr> #link https://bugs.launchpad.net/manila/+bug/1877063
15:44:21 <openstack> Launchpad bug 1877063 in OpenStack Shared File Systems Service (Manila) "netapp driver scheduler-stats missing max_over_subscription_ratio" [Undecided,In progress] - Assigned to Maurice Escher (maurice-escher)
15:44:54 <gouthamr> couple of labels missing, what's your impact assessment dviroel/carthaca?
15:46:03 <dviroel> gouthamr: seems to be low, don't know what carthaca thinks about it
15:46:32 <gouthamr> perhaps, low? it doesn't break the scheduler's oversubscription calculations..
15:46:33 <carthaca> I'm fine with low
15:46:45 <gouthamr> thanks
15:46:47 <gouthamr> #link https://bugs.launchpad.net/manila/+bug/1876827
15:46:47 <openstack> Launchpad bug 1876827 in OpenStack Shared File Systems Service (Manila) "Generic driver no-share-server setup fails on the CI" [Undecided,New]
15:47:04 <gouthamr> ^ i'm tearing down this job
15:47:29 <gouthamr> i didn't get an opportunity to debug what's going on though, it's not got anything to do with the generic driver
15:48:39 <gouthamr> the devstack plugin creates two nova VMs, and both of them remain unreachable for 5 minutes...
15:49:23 <gouthamr> it's unusual that it should happen, perhaps there's something broken that we ought to investigate and fix or let nova/neutron folks know...
15:49:35 <gouthamr> for now though, i'd propose wishlisting this bug
15:50:01 <tbarron> we can't chase every openstack integration issue at tip of master
15:50:07 <gouthamr> +1
15:50:20 <gouthamr> if someone intends to maintain testing the generic driver in a DHSS=False fashion, they can revive this job easily and fix it up..
15:50:20 <tbarron> manila isn't the osp qe integration team
15:51:03 <gouthamr> any concerns wishlisting this one?
15:51:17 <gouthamr> nope?
15:51:22 <gouthamr> okay last one !
15:51:24 <gouthamr> #link https://bugs.launchpad.net/manila/+bug/1839175
15:51:24 <openstack> Launchpad bug 1839175 in OpenStack Shared File Systems Service (Manila) "ganesha: the export IDs are not reused" [Medium,New]
15:52:08 <tbarron> legit issue, though not reported by customers
15:52:41 <gouthamr> hmmm, so a cloud with 64535 shares will hit this
15:53:22 <gouthamr> that's an aweful lot of shares, but, i suppose it's not unviable..
15:53:34 * tbarron looks forward to having to deal with the problem
15:54:00 <tbarron> i'm being a bit flip, we should try to schedule work on this
15:54:12 <gouthamr> hold on one minute
15:54:24 <gouthamr> doesn't the cephfs driver create one export per share per client?
15:54:43 <gouthamr> so this could occur with far lesser shares, tbarron?
15:54:45 <tbarron> yeah so we could hit this sooner
15:54:59 <tbarron> but still not a problem from the field yet
15:55:04 <gouthamr> ack
15:55:13 <tbarron> still lookiing forward to that scale
15:55:39 <gouthamr> okay, it's been triaged as a Medium, do we have a volunteer?
15:55:47 <tbarron> we should make a test wher the cross product of clients and shares gets us to there
15:56:30 <tbarron> gouthamr: we should likely have a RH person ( can be me if you want) but that doesn't mean it's top priority in backlog yet
15:56:46 <gouthamr> okay..
15:56:51 <tbarron> non-RH people are of course welcome to work it!
15:56:56 <gouthamr> lets discuss it more outside of this meeting then
15:57:01 <gouthamr> that's a wrap on the bugs
15:57:08 <gouthamr> vhari would be proud, we got 'em all
15:57:21 <tbarron> vhari++ gouthamr++
15:57:27 <gouthamr> only fixes are left, but small celebrations :)
15:57:37 <gouthamr> #topic Open Discussion
15:58:18 <gouthamr> how does everyone feel about the time slots we picked for the PTG?
15:58:27 <vkmc> pokemons?
15:58:31 <vkmc> oh bugs
15:58:32 <vkmc> sorry
15:58:37 <tbarron> heh
15:58:42 <carloss> haha
15:58:46 <vkmc> ++ for slots
15:58:56 <carloss> ++ for slots :)
15:59:13 <tbarron> gouthamr: thanks for doing the work on that
15:59:43 <dviroel> i'm ok with the slots also
16:00:05 <gouthamr> cool, thank you for the feedback
16:00:09 <gouthamr> we're at the hour
16:00:28 <gouthamr> thanks all for joining, hope you continue to remain safe and healthy!
16:00:33 <gouthamr> #endmeeting