15:00:29 <gouthamr> #startmeeting manila
15:00:29 <openstack> Meeting started Thu Mar 26 15:00:29 2020 UTC and is due to finish in 60 minutes.  The chair is gouthamr. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:33 <openstack> The meeting name has been set to 'manila'
15:00:37 <dviroel> o/
15:00:58 <tbarron> hey
15:01:02 <maaritamm> o/
15:01:17 <gouthamr> courtesy ping: xyang toabctl ganso vkmc amito lseki carloss
15:01:29 <lseki> o/
15:01:30 <carloss> hi :)
15:01:38 <gouthamr> we'll be missing vhari likely today, i'll drive the bug triage
15:01:39 <vhari> o/
15:01:44 <andrebeltrami> hi
15:02:05 <ganso> hello
15:02:15 <vhari> gouthamr, ack may need to step out for a bit, ty for driving the bugs on the list today
15:02:24 <gouthamr> o/
15:02:28 <gouthamr> vhari: sure thing
15:02:39 <gouthamr> agenda for this meeting: https://wiki.openstack.org/wiki/Manila/Meetings#Next_meeting
15:02:50 <gouthamr> Thank you all for joining
15:03:40 <gouthamr> Hope you're all doing well in these virulent times
15:03:49 <gouthamr> #topic Announcements
15:04:03 <gouthamr> First up, a release time line check
15:04:13 <gouthamr> Ussuri-3/Feature Freeze is in 2 weeks
15:04:32 <gouthamr> that deadline also includes Client Library Releases for Ussuri - python-manilaclient and manila-ui
15:04:36 <gouthamr> #link: https://releases.openstack.org/ussuri/schedule.html#u-ff
15:05:25 <gouthamr> So, please give features that touch CLI review preference and we'll discuss them in a bit
15:05:55 <gouthamr> Does anyone anticipate anything to slip at this stage? Or any gotchas that we must be aware of?
15:06:20 <gouthamr> hoping not :)
15:06:35 <gouthamr> Please note two new stable branch updates for manila:
15:06:40 <gouthamr> #link https://releases.openstack.org/train/index.html#train-manila (manila 9.1.1)
15:06:43 <gouthamr> #link https://releases.openstack.org/stein/index.html#stein-manila (manila 8.1.1)
15:07:17 <gouthamr> These were released this morning and include important bug fixes, so users are recommended to update to these versions at the earliest
15:08:02 <gouthamr> continuing with release themes, i added manila project deadlines to the Victoria Release schedule
15:08:06 <gouthamr> #link https://releases.openstack.org/victoria/schedule.html (Victoria cycle schedule + manila deadlines)
15:08:14 <vkmc> o/
15:08:18 * vkmc sneaks in late
15:08:38 <gouthamr> just as i said her name
15:08:50 <lseki> haha
15:08:51 <vkmc> hehe
15:09:00 <dviroel> lol
15:09:16 <gouthamr> please take a look at the deadlines added, and lets discuss them here - these are not set in stone
15:09:43 <gouthamr> although i allowed the patch to merge before bringing your attention to it, my bad
15:10:05 <gouthamr> so, before we move on to our next topic, i'd like to note some peculiarities with the upcoming release
15:10:31 <gouthamr> If you missed this email on the ML:
15:10:37 <gouthamr> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-February/012735.html ([all] Proposed release schedule for Victoria)
15:11:14 * gouthamr Victoria will be a slightly shorter release cycle
15:11:25 <gouthamr> "Part of the reason for the slightly shorter schedule was to wrap up prior to the Summit"
15:12:11 <gouthamr> It's still not as short a cycle as Ocata was
15:12:29 <gouthamr> but, i needed to adjust some deadlines
15:13:02 <gouthamr> i'm proposing that Manila's spec freeze be three weeks before M-2
15:13:22 <gouthamr> the new driver deadline remains M-2
15:13:42 <gouthamr> feature proposal freeze is two weeks prior to feature freeze
15:14:01 <gouthamr> gah, three weeks*
15:14:19 <gouthamr> and you've four weeks between PTG and spec freeze
15:14:37 <gouthamr> does anyone disagree with this proposal?
15:15:39 <gouthamr> you don't have to - right now, please let me know if you think this can't work for any reason and we'll bring it up in an upcoming meeting and adjust timelines
15:16:40 <gouthamr> My last announcement today is for the upcoming TC/PTL elections:
15:16:46 <gouthamr> #link http://lists.openstack.org/pipermail/openstack-discuss/2020-March/013597.html ([all][elections][ptl][tc] Combined PTL/TC Nominations Kickoff)
15:17:26 <gouthamr> dunno if you're shy or respectful, but, no-one reached out to me to attempt to be PTL :P
15:18:07 <gouthamr> so i've a plan brewing to groom folks to step up
15:18:53 <gouthamr> if no-one self-nominates to lead us, i'll seek self-nominations for the Wallaby release, and i'll mentor you through the Victoria cycle
15:19:12 <gouthamr> *crickets*
15:19:57 <gouthamr> anywho, it gives me time to document all things that one does being a project team lead and discuss how we can make things better
15:20:05 <vkmc> we are not yet done with gouthamr's ptlness
15:20:10 <vkmc> seems so
15:20:12 <vkmc> xD
15:20:12 <carloss> vkmc ++
15:20:14 <carloss> :D
15:20:20 <dviroel> :)
15:20:34 <gouthamr> <3
15:20:44 <gouthamr> Any other announcements for today?
15:20:50 <dviroel> yes
15:20:59 <gouthamr> go for it, dviroel
15:21:04 <dviroel> we have new manila contributor from our side
15:21:13 <dviroel> felipe_rodrigues, welcome!
15:21:17 <lseki> 🎉
15:21:20 <carloss> welcome, felipe_rodrigues :D
15:21:31 <felipe_rodrigues> Thank you!  :)
15:21:36 <andrebeltrami> Welcome! :p
15:21:45 <vkmc> wee, welcome!
15:21:51 <gouthamr> felipe_rodrigues: welcome, thank you for joining us!
15:21:58 * gouthamr reaches for the zorilla stamp
15:22:46 <dviroel> hope to see lots of patches from felipe_rodrigues soon :)
15:22:54 <gouthamr> felipe_rodrigues: are you sure your last name isn't viroel, because we just heard of another felipe around here and i'm always suspicious of dviroel
15:23:20 <andrebeltrami> hahaha!
15:23:30 <carloss> hehe
15:23:31 <lseki> haha
15:23:34 <gouthamr> never too young to be a zorilla
15:23:47 * dviroel dviroel already reserved fviroel id for the future
15:24:01 <gouthamr> dviroel: hah! smart dad
15:24:05 <felipe_rodrigues> ahahaha
15:24:28 <gouthamr> cool, thank you for making the introduction dviroel
15:24:36 <gouthamr> anything else before we move on?
15:24:51 <gouthamr> #topic Tracking our work
15:25:03 <gouthamr> #link https://etherpad.openstack.org/p/manila-ussuri-review-focus (Ussuri cycle review focus etherpad)
15:25:19 <gouthamr> How're things looking here? lets do a rollcall
15:25:31 <gouthamr> Support query user message by timestamp
15:25:39 <gouthamr> #link https://review.opendev.org/#/c/706438/ (manila)
15:25:40 <gouthamr> #link https://review.opendev.org/#/c/708807/ (python-manilaclient)
15:26:00 <gouthamr> this has gotten some review attention - any reason to think we can't merge soon'
15:26:49 <gouthamr> i'll take a look at both these today, i don't see any blockers
15:27:00 <gouthamr> [Unity] Manage/unmanage share server/share/snap
15:27:07 <gouthamr> #link https://review.opendev.org/#/c/688227/
15:27:54 <gouthamr> a number of us have taken a look at this code, couple of issues we've tried conveying to the author/s
15:28:06 <gouthamr> their CI is currently broken, we'll discuss why in a second
15:28:17 <dviroel> ^ not sure why they don't want to enable manage/unmanage tests on their ci
15:29:15 <dviroel> should be possible just to enable them only for this change
15:29:16 <carloss> yep, dviroel... i've asked that as well in my most recent comment
15:29:26 <carloss> but didn't get any response yet
15:29:38 <gouthamr> carloss: can you -1 with your comment, might get their attention
15:29:59 <carloss> gouthamr ok
15:30:19 <gouthamr> I don't see any reviewers added for the next one
15:30:20 <gouthamr> Graduate share groups feature
15:30:24 <gouthamr> #link https://review.opendev.org/#/q/topic:bp/graduate-share-groups-feature+(status:open+OR+status:merged)
15:31:29 <gouthamr> ah ty dviroel
15:31:33 <carloss> i've got a review from vkmc
15:31:39 <carloss> on the manila change
15:31:45 * vkmc hides
15:31:45 <gouthamr> cool, so it's moving fine
15:31:53 <vkmc> oh it's was a good thing
15:31:53 <vkmc> ok
15:31:54 <gouthamr> Tenant based quotas for share replicas and replica sizes
15:31:54 <vkmc> :)
15:32:02 <gouthamr> lol
15:32:04 <gouthamr> #link https://review.opendev.org/#/q/topic:bp/graduate-share-groups-feature+(status:open+OR+status:merged)
15:32:16 <gouthamr> gah
15:32:17 <gouthamr> #link https://review.opendev.org/#/q/topic:bp/limit-share-replicas-per-share+(status:open+OR+status:merged)
15:32:33 <gouthamr> this has been getting good reviewer attention too
15:32:43 <gouthamr> and looks good for the most part, anything to add carloss?
15:32:47 <carloss> ty for reviewing, gouthamr and ty for the review/QA, dviroel
15:32:57 <dviroel> didn't  added my comments on that yet, but I'm working on QA'ing
15:33:03 <gouthamr> awesome
15:33:04 <gouthamr> Create share from snapshot in another pool or backend
15:33:11 <gouthamr> #link https://review.opendev.org/#/q/topic:bp/create-share-from-snapshot-in-another-pool-or-backend+(status:open+OR+status:merged)
15:33:25 <gouthamr> this is the most complicated set of changes in my view for our attention
15:33:34 <gouthamr> i'd like some more eyes on it, if possible
15:33:53 <gouthamr> including the driver patch from NetApp:
15:33:55 <gouthamr> #link https://review.opendev.org/#/c/712642/
15:34:33 <gouthamr> carthaca: your review would be valuable too ^
15:34:37 * carthaca bookmarks the 3 links above to do some review
15:34:55 * gouthamr notes how people are reading his mind today
15:35:06 <gouthamr> thank you carthaca
15:35:17 <dviroel> andrebeltrami is working on QA NetApp driver also
15:35:32 <dviroel> so please andrebeltrami, post your comments there too
15:35:42 <andrebeltrami> Sure!
15:35:44 <gouthamr> alright, the rest of the reviews on that etherpad aren't really features, and we have some time to get to them
15:36:28 <gouthamr> please correct me if that's not the case
15:36:56 <dviroel> agree
15:37:08 <gouthamr> cool, anything else about these reviews before we move on?
15:37:27 <gouthamr> #topic NetApp Share Server NFS modify implementation (carthaca)
15:37:33 <gouthamr> #link: https://etherpad.openstack.org/p/manila-netapp-share-server-nfs-modify (NetApp Share Server NFS modify proposal etherpad)
15:37:38 <gouthamr> carthaca: you're up
15:37:45 <carthaca> Hi, thanks
15:37:51 <carthaca> I hope it is okay I just put this very netapp specific topic on the agenda.
15:37:57 <carthaca> But I wanted to collect feedback before coding something that I would never be able to contribute back and had to maintain in my fork forever ^^
15:38:05 <carthaca> Please have a look at my etherpad and comment, ask, discuss ;)
15:38:38 <vkmc> josecastroleon, o/
15:38:40 <josecastroleon> hi
15:39:40 <gouthamr> carthaca: is this entirely netapp specific? or did you intend on generalizing this solution?
15:39:41 <gouthamr> i've a couple of questions
15:40:54 <gouthamr> "give my customers the possibiility to select certain nfs vserver settings"
15:41:04 <carthaca> well not all of that options are netapp specific, but I'm not that familiar with the other back ends
15:41:29 <gouthamr> carthaca: that isn't entirely congruent with the way share servers are designed
15:41:40 <gouthamr> carthaca: users have no control over them
15:42:06 <carthaca> currently only via security services
15:42:34 <gouthamr> carthaca: yes, those and share networks are the only aspects they can view/manipulate about their share servers
15:43:34 <gouthamr> carthaca: and these server side settings may not make sense to share networks, security services or individual shares (as you noted)
15:43:56 <gouthamr> carthaca: but, the netapp driver already handles something like this today
15:44:37 <gouthamr> i forget how, but, nfs versions can be configured
15:44:56 <carthaca> yes, you can set this in the config
15:45:25 <gouthamr> ah, so its not exactly defined per tenant then
15:45:36 <carthaca> but that would mean I would have to run multiple manila-share services for the same backend or something
15:45:48 <gouthamr> true
15:46:28 <gouthamr> hmmm, at the outset, it feels like you could define scoped share type extra-specs and set these options when the first share is created
15:46:58 <dviroel> this looks like qos, but for share-servers
15:47:11 <dviroel> could work in the same way, i guess
15:47:18 <gouthamr> but, when the share type extra-specs are manipulated and the next share is created, the driver can choose to either manipulate the existing share server or create a new one
15:48:31 <gouthamr> dviroel: true, or any other scoped extra-spec that we'd plumb directly to the driver
15:49:13 <gouthamr> carthaca: i'm not against generalizing this solution, but we'll probably need to discuss the possibilities/merits in a longer meeting
15:49:31 <dviroel> might need a common capability also, to identify backends that support it and have it enabled
15:50:04 <dviroel> the rest is up to the driver
15:51:09 <gouthamr> i recall a similar request to determine root squash and all squash nfs options
15:52:00 <gouthamr> i can dig up that discussion from a past PTG to see what ideas were considered, a proposal was not made
15:53:55 <carthaca> sounds related, I think this option is called ignore-nt-acl-for-root at netapp nfs modify
15:54:10 <gouthamr> ack
15:54:18 <gouthamr> cool, i think we've some AIs on this carthaca - we'll see what was discussed earlier wrt share server side customizations, please explore if scoped extra-specs can be considered
15:54:37 <gouthamr> if you're planning on a generalized implementation, a specification would be required
15:55:04 <carthaca> cool, thanks for the feedback
15:55:06 <gouthamr> if it just pertains to netapp, a spec is not necessary, just file a bp for project tracking
15:55:27 <gouthamr> cool, we only have 5 minutes
15:55:41 * gouthamr looks into the bug etherpad
15:56:03 <gouthamr> dviroel: we might have to look at https://bugs.launchpad.net/manila/+bug/1868586 outside of this meeting
15:56:05 <openstack> Launchpad bug 1868586 in Manila "tempest complains 'Share network xxx could not be found.' " [Undecided,New]
15:56:28 <gouthamr> i want to skip bug triage today and give ~3 mins to open discussion
15:56:32 <gouthamr> taht okay?
15:56:38 <vkmc> +1
15:56:40 <gouthamr> #topic Open Discussion
15:56:47 <dviroel> gouthamr: ok
15:56:59 <priteau> gouthamr: you can take a bit more time if you need
15:57:26 <gouthamr> ah, thank you priteau - you're kind :) we'll do the triage on our channel
15:57:49 <gouthamr> anyone has anything to discuss?
15:57:52 <josecastroleon> yes
15:57:59 <gouthamr> josecastroleon: o/
15:58:47 <josecastroleon> we have just had a nasty behavior on our manila setup
15:59:03 <josecastroleon> we have 4 different cephfs backends behind
15:59:11 <josecastroleon> each of them mapped to a share type
15:59:21 <josecastroleon> so scheduling should be easy right?
15:59:41 <josecastroleon> we were seeing a continuous increase of scheduling time
16:00:16 <gouthamr> josecastroleon: oh, weird - do you mind us moving over to #openstack-manila
16:00:21 <josecastroleon> sure
16:00:41 <gouthamr> josecastroleon: thank you, lets continue there
16:01:04 <gouthamr> thanks all for attending, we'll discuss this scheduling issue on #openstack-manila, please join us there if you can
16:01:08 <dviroel> thanks!
16:01:10 <gouthamr> #endmeeting