14:00:35 <rosmaita> #startmeeting cinder
14:00:35 <opendevmeet> Meeting started Wed Aug 18 14:00:35 2021 UTC and is due to finish in 60 minutes.  The chair is rosmaita. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:35 <opendevmeet> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:35 <opendevmeet> The meeting name has been set to 'cinder'
14:00:42 <e0ne> hi
14:00:44 <rosmaita> #topic roll call
14:01:00 <sfernand> hi
14:01:08 <eharney> hi
14:01:17 <tosky> hi
14:01:21 <enriquetaso> thi
14:01:23 <simondodsley> hi
14:01:33 <walshh_> hi
14:02:05 <fabiooliveira> hi
14:02:08 <whoami-rajat> Hi
14:02:23 <rosmaita> looks like a good turnout, let's get started
14:02:28 <rosmaita> #topic announcements
14:02:44 <rosmaita> actually, i forgot to post the agenda link
14:02:52 <rosmaita> #link https://etherpad.opendev.org/p/cinder-xena-meetings
14:02:59 <rosmaita> ok, on to announcements
14:03:11 <rosmaita> Festival of Reviews on Friday, 1400-1600 UTC
14:03:27 <rosmaita> it's our monthly fun time to review small patches together
14:03:43 <rosmaita> #link https://etherpad.opendev.org/p/cinder-festival-of-reviews
14:04:07 <rosmaita> next up: the openstack PTL nomination period has opened
14:04:09 <enriquetaso> ++
14:04:18 <rosmaita> #link http://lists.openstack.org/pipermail/openstack-discuss/2021-August/024191.html
14:04:31 <rosmaita> self-nominations have to happen before 24 august
14:05:26 <rosmaita> i am planning to nominate myself for one last time
14:05:43 <rosmaita> but that doesnt' mean someone else can't nominate themselves also
14:06:18 <rosmaita> but if i am elected, i will be very much interested in working with anyone interested in becoming PTL for future cycles
14:06:41 <rosmaita> so contact me if you are interested, or want info on what being PTL entails
14:06:53 <rosmaita> next, some upcoming deadlines
14:07:04 <rosmaita> xena os-brick release is tomorrow (!!)
14:07:18 <rosmaita> next week's cinder meeting is in video + irc
14:07:44 <rosmaita> by the way, we did the monthly video meeting as an experiment ... does anyone have feedback on whether we should continue it?
14:08:32 <rosmaita> is this thing on?
14:08:40 <eharney> i think it's worth continuing
14:08:46 <walshh_> me too
14:08:48 <whoami-rajat> i feel it's very productive and we also get to interact so +1 from me
14:09:04 <rosmaita> ok, sounds good ... thanks for the feedback
14:09:15 <rosmaita> ok, coming up in 2 weeks we have:
14:09:25 <rosmaita> cinderclient and cinderclient extension release
14:09:43 <rosmaita> milestone-3 for cinder, which is the feature freeze
14:10:31 <rosmaita> just looking at cinderclient patches ...
14:10:40 <rosmaita> #link https://review.opendev.org/q/project:openstack/python-cinderclient+status:open+branch:master+is:mergeable
14:10:49 <rosmaita> doesn't look like anything major
14:11:01 <rosmaita> is anyone working on a cinder feature that will impact the cinderclient?
14:11:53 <rosmaita> sounds like no
14:12:32 <rosmaita> ok, as far as review priorities go over the next week, it's cinder features
14:12:46 <rosmaita> #link https://blueprints.launchpad.net/cinder/xena
14:13:06 <rosmaita> there are a lot of things there that need review, so please jump in
14:13:34 <rosmaita> that's all i have for announcements?  anyone have anything else?
14:14:40 <rosmaita> ok, moving on, then
14:14:50 <rosmaita> #topic xena os-brick release
14:15:07 <rosmaita> we have two feature patches, and some bugfixes
14:15:16 <rosmaita> #link https://review.opendev.org/q/project:openstack/os-brick+status:open+branch:master+is:mergeable
14:16:05 <rosmaita> there are a few extra-small and small patches that are in good shape
14:16:21 <rosmaita> would be good to get those reviewed (and hopefully merged) today
14:16:59 <rosmaita> the feature patches are:
14:17:08 <rosmaita> #link https://review.opendev.org/c/openstack/os-brick/+/800014
14:17:23 <rosmaita> "Add support for multiple volumes within subsystem to NVMe-OF connector"
14:17:48 <rosmaita> this was in pretty good shape last time i looked, an update to test data caused a pep8 problem yesterday, but looks like that's fixed now
14:17:56 <rosmaita> we also have
14:18:06 <rosmaita> #link https://review.opendev.org/c/openstack/os-brick/+/802691
14:18:18 <rosmaita> "NVMeOF Connection Agent"
14:18:55 <simondodsley> The whole NVMe-oF connector is not in a good way IMHO as it is very focused on Kumoscale - it doesn't support multipath either.
14:18:57 <rosmaita> i asked Zohar to re-architect it a bit, but i have not looked at the latest version yet
14:19:19 <simondodsley> The Pure NVMe driver doesn't work well with it
14:19:41 <rosmaita> simondodsley: yes, i think we should discuss this at the PTG
14:20:11 <rosmaita> but i do want to remind people that when you are interested in a technology, please review patches that impact it
14:21:04 <rosmaita> because as long as the CIs pass, it's going to look good to non-experts in that technology
14:21:11 <rosmaita> (speaking for myself here)
14:21:27 <rosmaita> the strength of the community is having lot of eyes on changes
14:21:44 <rosmaita> so, please exercise your vision, everyone!
14:22:55 <rosmaita> ok so i need people to look over the NVMeOF Connection Agent patch
14:23:34 <rosmaita> we need candid feedback about how close it is to being ready
14:24:38 <simondodsley> given the few drivers that use it I'd let this pass but we need to completely refactor the code I think for better support
14:25:38 <rosmaita> yeah, refactoring seems reasonable
14:27:18 <rosmaita> i believe that simondodsley has "volunteered" to lead a PTG session about next steps for NVMeOF connector
14:27:47 <rosmaita> so other people interested in that or who know a lot about the latest nvmeof protocols, please reach out to simon
14:28:23 <rosmaita> it would be good to have an organized community effort to get this working well
14:28:37 <rosmaita> because there seems to be a lot of interest in it now, industry-wise
14:29:00 <rosmaita> #topic general reviewing comments
14:29:07 <simondodsley> there are a number of vendors that support nvme so getting us all to come up with a lowest common denominator connector would be best for everyone
14:29:17 <rosmaita> i agree entirely
14:29:40 <e0ne> simondodsley: +1
14:30:49 <rosmaita> well, my general reviewing comments are prompted by a lot of review requests in the cinder channel who, when i look,  haven't been doing much reviewing themselves
14:31:28 <rosmaita> this is a community project, and part of the community is posting patches and an even more important part is reviewing patches
14:31:48 <rosmaita> it only take 1 person to write a patch, but it takes 2 people to approve a patch
14:32:05 <rosmaita> so, everyone who posts a patch should be reviewing at least 2 other patches
14:32:26 <rosmaita> we have helpful info in the cinder contributor docs if you aren't sure about how to review
14:32:44 <rosmaita> and a +1 with no comments is not a helpful review
14:33:40 <rosmaita> the other thing i would like people to think about is what we can do to structure this better
14:34:06 <rosmaita> what i mean is, Zohar for instance has been good about responding to reviews on his agent patch
14:34:27 <rosmaita> but we are running out of time and i'm afraid it's going to miss xena
14:34:41 <rosmaita> so it would have been better if more reviews were happening earlier
14:35:09 <rosmaita> i believe i have been mentioning the deadlines at meetings fairly regularly
14:35:35 <rosmaita> but it looks like we need something other than "please remember to review this patch"
14:35:48 <rosmaita> and i am not sure what that something else is
14:36:06 <enriquetaso> good question, i'm not sure what we can do, but we need to do something
14:36:20 <rosmaita> yeah, everyone please think about this
14:36:41 <rosmaita> maybe some other open source community has a good idea we can use
14:37:24 <rosmaita> ok, so in the mean time:
14:37:34 <rosmaita> today: os-brick reviews (see link above)
14:37:49 <rosmaita> immediately after tomorrow: xena feature reviews
14:38:05 <rosmaita> and on friday: come to the festival and help do XS reviews
14:38:57 <rosmaita> ok, next topic
14:39:06 <rosmaita> #topic gate failure fix
14:39:14 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/805026
14:39:20 <rosmaita> that obviously is high priority
14:39:27 <rosmaita> eharney: anything you want to add?
14:39:36 <eharney> no, just caught this this morning
14:40:08 <rosmaita> what is it with the lvxxx commands?
14:40:40 <rosmaita> that is a rhetorical question
14:40:49 <eharney> heheh
14:40:57 <rosmaita> anyway, we do have a pattern now for how to deal with these things, so it should be quick review
14:41:04 <eharney> if this looks familiar, it's because it's a bug in lvm commands in general, but we've only applied this so far in places where we've seen issues
14:41:29 <rosmaita> ok, i will commit to reviewing it after the meeting
14:41:43 <rosmaita> it's eric's patch, so he can;t
14:41:58 <rosmaita> need one more core to commit to reviewing it today
14:43:37 <whoami-rajat> i can take a look
14:43:43 <rosmaita> thank you!
14:43:43 <enriquetaso> is this affecting os-brick as well?
14:44:06 <eharney> not sure when we call lvextend via os-brick
14:44:27 <rosmaita> maybe not yet ... if you have time, enriquetaso, would be good to take a look
14:44:40 <eharney> the code is duplicated in os-brick, so we'll have to get it sorted out eventually
14:45:19 <rosmaita> #topic cinder-mypy non-voting job failing on all patches
14:45:31 <rosmaita> eric has a patch up that addresses this
14:45:38 <eharney> just a heads up that i know this is failing everywhere currently, there's a fix for it
14:45:41 <rosmaita> #link https://review.opendev.org/c/openstack/cinder/+/784453
14:45:58 <rosmaita> i will re-review the patch today
14:46:19 <rosmaita> would be good to get it fixed, i don't want people to start completely ignoring the mypy job results
14:47:05 <eharney> the fix is the third patch in a series -- the second has been approved, the first needs another +2
14:47:43 <enriquetaso> #link https://review.opendev.org/c/openstack/cinder/+/780933
14:48:17 <rosmaita> thanks enriquetaso
14:48:58 <rosmaita> i;m already the +2 on that one, so will need someone else to take a look
14:50:05 <rosmaita> i hate to do this again, but i would like someone to commit to looking at https://review.opendev.org/c/openstack/cinder/+/780933 today
14:50:33 <rosmaita> it would be good to have the mypy job fixed today
14:50:51 <enriquetaso> can take a look
14:50:59 <rosmaita> \o/
14:51:02 <rosmaita> thanks enriquetaso
14:51:17 <rosmaita> ok, that's everything from the agenda
14:51:39 <rosmaita> #topic open discussion
14:51:49 <rosmaita> just want to mention this shout in the cinder channel:
14:51:51 <rosmaita> shoffmann: Like discussed last week I proposed an overview about nas_secure options. I guess, we won't have time at todays meeting but maybe you can have a look later and give some hints, what is missing and how we can proceed. https://etherpad.opendev.org/p/gSotXYAZ3JfJE8FEpMpS
14:52:14 <eharney> i'll remind that https://etherpad.opendev.org/p/cinder-xena-ci-tracking still exists
14:52:15 <rosmaita> i think shoffmann also sent something to the ML about this
14:52:41 <shoffmann> Yes, I raised this also in the ML.
14:52:52 <rosmaita> shoffmann: glad you are here
14:52:59 <eharney> i will read over shoffmann's pad, looks useful, thanks for assembling that
14:53:04 <rosmaita> thanks for working on this
14:53:21 <rosmaita> also, i am assuming that anyone here from netapp will be interested
14:53:31 <shoffmann> Not sure, if something is missing there or we can find better examples/use cases
14:53:57 <rosmaita> to follow up simondodsley's point from earlier, now is the time to give some feedback to make sure we don't wind up with a narrow solution
14:54:28 <sfernand> sure, I will take a look
14:54:30 <rosmaita> shoffmann has listed some options and then an action proposal
14:54:46 <rosmaita> thanks sfernand
14:55:10 <rosmaita> so, anyone interested in NFS, please take a look at the etherpad and leave some feedback
14:56:08 <rosmaita> I want to second eharney's mention of https://etherpad.opendev.org/p/cinder-xena-ci-tracking
14:56:33 <rosmaita> if you hit a failure, please take a few minutes to add it to the etherpad
14:59:58 <rosmaita> ok, looks like that's all for today ... please review!
15:00:05 <rosmaita> don't forget about the festival on friday
15:00:16 <whoami-rajat> Review request: Would like to request for review on my backup user message patch, it adds a base framework that helps with loosen chain of patches for other operations https://review.opendev.org/c/openstack/cinder/+/786627
15:00:33 <rosmaita> thanks, everyone
15:00:38 <whoami-rajat> thanks!
15:00:38 <enriquetaso> thanks brian
15:00:38 <rosmaita> #endmeeting