15:00:40 <gouthamr> #startmeeting manila
15:00:41 <openstack> Meeting started Thu May 21 15:00:40 2020 UTC and is due to finish in 60 minutes.  The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:42 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:44 <openstack> The meeting name has been set to 'manila'
15:00:49 <carloss> o/
15:00:51 <dviroel> o/
15:01:08 <andrebeltrami> o/
15:01:09 <vhari> hi
15:01:15 <gouthamr> courtesy ping: xyang toabctl ganso vkmc amito lseki tbarron danielarthurt
15:01:21 <lseki> o/
15:01:21 <tbarron> hi
15:01:26 <amito> o/ hey
15:01:32 <gouthamr> hello o/
15:01:50 <gouthamr> Here's our agenda for the day:
15:01:56 <gouthamr> #link https://wiki.openstack.org/wiki/Manila/Meetings#Next_meeting
15:02:10 <gouthamr> lets get started
15:02:12 <gouthamr> #topic Announcements
15:03:00 <gouthamr> not much going on this week, in terms of announcements - folks are probably looking forward to a long weekend in some countries
15:03:04 <vkmc> o/
15:03:20 <gouthamr> to do fun things, at home :}
15:03:31 <vhari> :)
15:03:44 <gouthamr> if you haven't registered for the upcoming virtual PTG, please do so:
15:03:55 <gouthamr> #link  https://www.eventbrite.com/e/virtual-project-teams-gathering-june-2020-tickets-103456996662
15:04:35 <gouthamr> also add your name and tz info here:
15:04:37 <gouthamr> #link https://etherpad.opendev.org/p/vancouver-ptg-manila-planning
15:04:53 <gouthamr> any other announcements?
15:05:12 <gouthamr> #topic Requesting a new manila tempest plugin tag (gouthamr)
15:05:21 * gouthamr pages me
15:06:08 <gouthamr> okay, so we have some downstream requests piling up to tag a version of tempest with bugfixes that we've landed since the ussuri release
15:06:44 <gouthamr> we have a handful of these fixes, but, i was hoping to collect a few more
15:07:28 <gouthamr> so, if you have a tempest fix, point me to it, so i can review, wait for it to merge before requesting a release
15:07:56 <gouthamr> i'm trying to get this tag early next week at the latest
15:08:12 <gouthamr> any objections/concerns?
15:08:18 <dviroel> gouthamr: is there any upstream job that can test andrebeltrami fix?
15:08:34 <dviroel> this one: https://review.opendev.org/#/c/728604/
15:08:57 <gouthamr> #link https://review.opendev.org/#/c/728604/ (Add security service to share networks)
15:09:01 <gouthamr> ^ for the highlight
15:09:05 <dviroel> ack
15:09:25 <gouthamr> dviroel: sadly no, none of our first party DHSS=True drivers support security services
15:09:50 <dviroel> Seems to work with NetApp driver, but andrebeltrami is fixing some CI issues right now
15:09:57 <gouthamr> dviroel: your best bet is the NetApp/EMC CI systems
15:10:17 <dviroel> we will post the results as soon we achieve all tests passing on CI
15:10:42 <andrebeltrami> ++
15:10:49 <dviroel> gouthamr: yeah, we may send an email to EMC guys also, so they can test too
15:10:50 <gouthamr> dviroel: great... lets get this one, its been a pita for a bit
15:11:00 <gouthamr> dviroel: sure thing! thanks for considering it
15:11:23 <dviroel> gouthamr: yeah, if you have any other concern with the code, please ping us
15:11:26 <gouthamr> lkuchlan's not around, but he had a change https://review.opendev.org/#/c/725487/ that he's tested locally
15:12:19 <gouthamr> it would be a "nice-to-have" for us, given we've never run scenario tests against the cephfs native protocol in manila
15:12:44 <gouthamr> we'll need a custom image and some job changes to test this change
15:12:52 <gouthamr> those are still pending..
15:13:25 <gouthamr> i'll ping folks once they're ready...
15:13:47 <gouthamr> that's the two changes i'm tracking, speak up now if there are others
15:13:52 <gouthamr> or let me know soon :)
15:14:06 <vkmc> lkuchlan++
15:14:38 <gouthamr> cool, lets move on..
15:14:41 <gouthamr> #topic Reviews needing attention
15:14:44 <gouthamr> #link https://etherpad.openstack.org/p/manila-victoria-review-focus
15:15:26 <gouthamr> looks like haixin's issues aren't fixed on https://review.opendev.org/#/c/700393/
15:15:40 <gouthamr> i haven't had cycles to look at it again and help..
15:16:15 <gouthamr> that's still on my backlog however, and others are welcome to take a look - its certainly a positive change for us to have, and a sort of tech debt
15:16:40 <gouthamr> carloss: gave you plenty to work with wrt https://review.opendev.org/#/q/status:open+project:openstack/manila+branch:master+topic:bug/1843953 :)
15:17:00 <gouthamr> no rush though!
15:17:14 <carloss> yep :p ... it's in my todo list
15:17:19 <carloss> intend to work on it soon
15:17:30 <gouthamr> ack
15:17:43 <gouthamr> lets go down that list
15:18:00 <gouthamr> carloss vkmc: i noticed updates on https://review.opendev.org/#/c/703025/
15:18:33 <gouthamr> have you had a chance to re-review yet?
15:18:37 <carloss> yes, the author seems to have fixed the issue I pointed out, I'll try it in my env today
15:18:50 <gouthamr> good stuff, thanks carloss
15:18:59 <carloss> the code changes I've proposed were answered
15:19:09 <carloss> yw gouthamr
15:19:36 <gouthamr> https://review.opendev.org/#/c/722418/ - this has a +1 from vkmc; will need more time? eyes?
15:20:10 <gouthamr> both? :)
15:20:25 <carloss> I think more eyes would be nice... thanks for reviewing it, vkmc :)
15:20:33 <gouthamr> tbarron, dviroel can you help here?
15:20:51 <tbarron> sure
15:21:02 <dviroel> will take a look again, saw it on downstream some time ago
15:21:02 <gouthamr> carloss: also add carthaca to the review
15:21:19 <gouthamr> oh i see him there already
15:21:26 <carloss> yep :)
15:21:36 <gouthamr> good stuff
15:22:06 <gouthamr> huge props to carloss dviroel and tbarron for their war-footing merges of the zuulv3 backports
15:22:20 <gouthamr> that's just the tip of the iceberg, and there are a lot more coming :)
15:22:30 <vkmc> ++
15:22:42 <gouthamr> but, i've updated the review etherpad so you can track what's left in the current list
15:23:09 <gouthamr> some of the backports haven't been uploaded yet to prevent merge conflicts and unnecessary rebases
15:24:15 <gouthamr> alright, any one have any other review to bring up
15:24:16 <gouthamr> ?
15:24:34 <gouthamr> s/have/has
15:25:04 <dviroel> i'll add another NetApp bug fix to the etherpad
15:25:24 <gouthamr> dviroel: sure..
15:25:31 <dviroel> #link https://review.opendev.org/#/c/718233/ ([NetApp] Fix falsely report migration cancelation success)
15:25:33 <dviroel> ^ this one
15:26:06 <gouthamr> cool, please add it and reach out to folks that can help review it..
15:26:12 <dviroel> ack
15:26:25 <gouthamr> lets move on to our next item..
15:26:28 <gouthamr> #topic Bugs (vhari)
15:26:45 <vhari> o/
15:26:47 <gouthamr> o/ vhari - the floor is yours
15:26:57 <vhari> ty gouthamr
15:27:01 <vhari> #link https://bugs.launchpad.net/manila/+bug/1879907
15:27:01 <openstack> Launchpad bug 1879907 in OpenStack Shared File Systems Service (Manila) "restapi filter can't work correctly if response has 'null' value of the lookup key." [Undecided,New]
15:27:32 <vhari> this is a new one -
15:27:42 <vhari> needs minor triage for now
15:28:08 <gouthamr> weird
15:28:21 <gouthamr> "The test case query all shares with snapshot_id='b17eb2a9-c7a5-4367-ba0f-e43eeb76ee9f'. However manila returns 3 shares, 2 of them have snapshot_id=null and only 1 of them has the requested snapshot_id."
15:29:27 <gouthamr> the log actually confirms this, but i haven't seen this on the CI
15:31:02 <gouthamr> its a really old test too, so i wonder how this is broken now - and only in third party CI
15:31:21 <vhari> ack
15:31:27 <gouthamr> oh waiy
15:31:29 <gouthamr> wait*
15:31:30 <gouthamr> https://review.opendev.org/#/c/703025/
15:32:00 <gouthamr> this "breakage" is on a bugfix we just discussed
15:32:12 <gouthamr> so it looks like a regression introduced by the patch
15:32:56 <gouthamr> yeah, ctrl+F on the bug report for 703025, all the logs are from that change
15:33:19 <dviroel> exactly
15:34:00 <gouthamr> vhari: not a bug - the test is doing what its supposed to do, the code change is introducing a regression, that that has to be fixed before it can be merged
15:34:45 <gouthamr> vhari: so lets mark it invalid - wdyt?
15:35:15 <vhari> gouthamr, sounds ok, retest later and open bug if persist
15:35:48 <dviroel> need to tell this to the reporter, it's not the first time iirc
15:36:14 <gouthamr> dviroel: ack
15:36:21 <vhari> yes,  adding ^^  comment
15:36:31 <dviroel> thanks vhari
15:36:40 <vhari> sure, moving on
15:36:43 <vhari> #link https://bugs.launchpad.net/manila/+bug/1879884
15:36:43 <openstack> Launchpad bug 1879884 in OpenStack Shared File Systems Service (Manila) "lack of "cephfs_volume_mode " for other backend file systes" [Undecided,New]
15:36:59 <vhari> ditto for this one .. looking for minor triage
15:37:32 <tbarron> this is a curious one
15:38:25 <gouthamr> hmmm, yeah we don't really dictate what the unix directory permissions must be
15:38:51 <gouthamr> i wonder if we should; is this a common use case?
15:39:15 <vkmc> hmm, seems like something the operator should config for their use case
15:39:27 <vkmc> I'd make a default config for permissions to 777
15:40:30 <gouthamr> this is the change they're referring to when they say cephfs supports it: https://github.com/openstack/manila/blob/721bb70a81b80bf065705e8790bd117eae806e7d/releasenotes/notes/cephfs-set-mode-b7fb3ec51300c220.yaml#L5
15:40:42 <tbarron> so each back end would have to pick up on the option and act on it, right?
15:40:57 <gouthamr> yes
15:41:19 <tbarron> that's why we did  it just for cephfs, that's where we had a customer requesting it
15:41:20 <gouthamr> and that probably matters only in case of NFS
15:42:57 <gouthamr> hmmm, this one is a tough one - we'll probably have to think about it a bit more
15:44:03 <gouthamr> for one, we should ask the reporter what's the ZFS backend they're using - we can change stuff easily if its zfsonlinux, but if they're referring to zfssa, that's an unmaintained backend
15:44:50 <gouthamr> does anyone have any opinions on this bug, or is interested in tackling this use case?
15:45:12 <tbarron> i'd like to understand the use case better
15:45:34 <tbarron> e.g. if netapp or emc are used for nfs back end
15:45:41 <tbarron> is there a need?
15:46:02 <tbarron> e.g. when k8s runs on openstack?
15:46:12 <tbarron> what is is his zfs use case?
15:46:13 <gouthamr> its uniform, iirc, everyone sets the default to 755
15:46:52 <tbarron> gouthamr: right, so if there is a need for a difft mode, what is the use case?
15:47:17 <tbarron> if we implement this, do we have to have driver support all at once?
15:48:02 <tbarron> gouthamr: i can ask some questions in the bug, or you can, but to prioritize need more info unless the reporter wants to start working on it
15:48:38 <gouthamr> tbarron: ack, please do.. we can talk about it outside of this meeting, i don't understand *why* they can't deal with 755 to begin with ..
15:49:19 <gouthamr> tbarron: i haven't dabbled too much into openshift, but, on native k8s with manila-csi, we are able to use the lvm backend fine, without having to toggle unix permissions
15:50:51 <gouthamr> on netapp, you can control this with "unixPermissions" during share creation - i vaguely recall that the container plugins could do this
15:51:17 <gouthamr> cool, vhari - tbarron will engage, and we'll follow up
15:51:44 <vhari> got it .. so that's a wrap for bugs almost out of time :)
15:51:50 <gouthamr> great thank you vhari
15:51:58 <vhari> thanks all for input
15:52:00 <gouthamr> #topic Open Discussion
15:52:32 <gouthamr> wow time's flying, its already week 3 of Victoria Release: https://releases.openstack.org/victoria/schedule.html
15:52:50 <gouthamr> next week* :)
15:53:07 <gouthamr> anyway, i'll take some time to add time slots to our ptg discussions
15:53:17 <gouthamr> please let me know if you have a preferred slot
15:53:28 <gouthamr> i'll share deets on the ML asap
15:53:54 <gouthamr> noone
15:54:06 <gouthamr> has claimed any of the slots yet, so its first come first served
15:54:44 <dviroel> +1
15:55:08 <gouthamr> alrighty folks, anything else for today?
15:56:02 <gouthamr> cool, i want to end today's meeting with this:
15:56:51 <gouthamr> if you haven't already seen this, please take a moment to read up about maaritamm's latest adventures
15:56:53 <gouthamr> #link https://twitter.com/xahteiwi/status/1262675749149999105
15:57:25 <tbarron> gouthamr:  i *think* it was a limitiation in openshift 3 that may have been resolved
15:57:30 <gouthamr> we're grateful for her work here, and are proud of being her initial buddies :)
15:57:46 <gouthamr> tbarron: ah!
15:57:54 <maaritamm> lol, thanks gouthamr
15:58:06 <maaritamm> I'm still sneaking around in here too :D
15:58:13 <vhari> maaritamm++
15:58:21 <carloss> maaritamm++
15:58:38 <dviroel> maaritamm++
15:58:45 <amito> maaritamm++
15:58:46 <gouthamr> oh dont go away, maaritamm :)
15:58:49 <tbarron> i asked for info and marked it incomplete
15:59:01 <lseki> maaritamm++
15:59:07 <gouthamr> cool ty tbarron
15:59:07 <vkmc> maaritamm++
15:59:08 <tbarron> will become bug or wishlist and prioritized depending on response
15:59:24 <gouthamr> awesome, thanks all for joining today
15:59:34 <gouthamr> hope you stay safe and healthy!
15:59:37 <gouthamr> #endmeeting