16:01:46 <jgriffith> #startmeeting cinder
16:01:46 <openstack> Meeting started Wed Aug 20 16:01:46 2014 UTC and is due to finish in 60 minutes.  The chair is jgriffith. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:48 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:50 <openstack> The meeting name has been set to 'cinder'
16:01:51 <jgriffith> hehe
16:01:53 <bswartz> .o/
16:01:55 <winston-d> o/
16:01:57 <smcginnis> o/
16:01:58 <avishay> hello
16:02:01 <e0ne> hi
16:02:05 <thingee> o/
16:02:15 <jgriffith> Nice!  Lots of peoples
16:02:19 <jgriffith> let's get started
16:02:23 <jgriffith> agenda https://wiki.openstack.org/wiki/CinderMeetings
16:02:27 <jgriffith> err...
16:02:30 <jgriffith> #agenda https://wiki.openstack.org/wiki/CinderMeetings
16:02:41 <jkremer> hi
16:02:48 <jgriffith> jkremer: hola
16:02:58 <jgriffith> #topic FPF
16:02:58 <DuncanT> Hey
16:03:06 <xyang1> hi
16:03:18 <jgriffith> So just a reminder, and maybe a surprise to a few folks
16:03:27 <jgriffith> tomorrow!  is the Feature Proposal Freeze for Juno
16:03:57 <jgriffith> if you look here: https://launchpad.net/cinder/+milestone/juno-3
16:04:08 <jgriffith> we've got a few issues :(
16:04:31 <jgriffith> I believe Vincent proposed a patch for his QoS work
16:04:44 <jgriffith> so I'll update that one manually.... however it didn't look promising
16:04:50 <bswartz> what issues?
16:05:15 <jgriffith> bswartz: not started ratio in targetted BP's
16:05:31 <bswartz> :-o
16:05:32 <jgriffith> bswartz: typically if FPF is 24 hours out it'd be good if everything was "proposed"
16:05:38 <jgriffith> at least typically that's the idea
16:05:51 <bswartz> yeah but doesn't this happen every release?
16:06:03 <jgriffith> bswartz: as I said in FTC I'd really like to change that
16:06:11 <jgriffith> bswartz: and we did do better this milestone
16:06:31 <jgriffith> bswartz: I'd like to not settle for status quo if that's *ok* with you
16:06:42 <bswartz> heh okay
16:07:16 <jgriffith> bswartz: do you know anything about navneets pool patch?
16:07:20 <jgriffith> or bp
16:07:26 <bswartz> what about it?
16:07:28 <jgriffith> https://blueprints.launchpad.net/cinder/+spec/windows-smbfs-volume-driver
16:07:36 <bswartz> We're going with winston's right?
16:07:43 <jgriffith> bswartz: correct
16:07:57 <bswartz> wrong link I think
16:08:00 <jgriffith> bswartz: were navneet and winston-d going ot share that BP?
16:08:05 <jgriffith> oops
16:08:11 <jungleboyj> o/
16:08:16 <jgriffith> https://blueprints.launchpad.net/cinder/+spec/multiple-service-pools
16:08:35 <bswartz> winston can use that BP if he doesn't already have one
16:08:46 <jgriffith> winston-d: ^^ ?
16:08:47 <bswartz> I can't wait for it to merge
16:08:52 <bswartz> winston's change that is
16:09:07 <jgriffith> I'll fix it up later if winston-d isn't around
16:09:08 <avishay> winston-d is using this one https://blueprints.launchpad.net/cinder/+spec/pool-aware-cinder-scheduler
16:09:14 <winston-d> i have one: https://blueprints.launchpad.net/cinder/+spec/pool-aware-cinder-scheduler
16:09:21 <winston-d> avishay: thx. :)
16:09:25 <avishay> winston-d: :)
16:09:28 <jgriffith> k... I'll mark the old one as obsolete
16:09:39 <jgriffith> winston-d: thanks.. that's the info I was looking for
16:09:42 <bswartz> jgriffith, winston-d: let's swap out the BPs and put winston's in j-3
16:09:58 <winston-d> bswartz: +1
16:10:03 <jgriffith> ronenkat: I need to get yours updated as well
16:10:05 <ronenkat> jgriffith: the replication blueprint is not on the list-  https://blueprints.launchpad.net/cinder/+spec/volume-replication
16:10:17 <jgriffith> ronenkat: it is, but wrong bp again
16:10:20 <jgriffith> ronenkat: same issue
16:10:42 <jgriffith> ronenkat: https://blueprints.launchpad.net/cinder/+spec/pool-aware-cinder-scheduler
16:10:45 <jgriffith> bahh
16:10:58 <jgriffith> ronenkat: look at the milestone paget
16:11:12 <ronenkat> jgriffith: that is the replication blueprint
16:11:30 <jgriffith> ronenkat: yeah
16:11:40 <jgriffith> I'm having copy pasteissues this am :(
16:11:42 <avishay> https://blueprints.launchpad.net/cinder/+spec/volume-replication
16:11:43 <alexpilotti> hi guys, show the reference to the SMB BP
16:11:52 <jgriffith> keep getting the wrong thing out of my clipboard
16:11:57 <jgriffith> avishay: thanks!
16:12:00 <jgriffith> avishay: that one!
16:12:01 <alexpilotti> s/show/saw/
16:12:11 <jgriffith> ronenkat: but look at the milestone page
16:12:19 <xyang1> jgriffith: I don't see CG there either: https://blueprints.launchpad.net/cinder/+spec/consistency-groups
16:12:20 <alexpilotti> jgriffith: please let me know if I can help in any way
16:12:24 <avishay> jgriffith: i have an advanced degree in ctrl-c/ctrl-v
16:12:30 <jkremer> jgriffith: Should driver bp be in the list too? I don't see https://blueprints.launchpad.net/cinder/+spec/oracle-zfssa-cinder-driver
16:12:40 <ronenkat> jgriffith: ok
16:13:01 <jgriffith> ronenkat: check my updates and make sure they're good
16:13:23 <xyang1> jgriffith: driver bp for VMAX: https://blueprints.launchpad.net/cinder/+spec/emc-vmax-driver-juno-update
16:13:23 <jgriffith> jkremer: yeah, I'll get yours in there
16:13:37 <jkremer> jgriffith: Thanks!
16:14:07 <jgriffith> xyang1: on it
16:14:09 <ronenkat> jgriffith: I would like to have the Storwize replication support patch there as well, pending merge of replication
16:14:17 <xyang1> jgriffith: xtremeIO driver: https://blueprints.launchpad.net/cinder/+spec/xtremio-cinder-volume-driver
16:14:20 <xyang1> jgriffith: thanks!
16:15:03 <jgriffith> xyang1: there's no work items for that one?
16:15:09 <jgriffith> xtremio
16:15:15 <jgriffith> ie nothing submitted
16:15:47 <jgriffith> xyang1: ?
16:15:54 <xyang1> jgriffith: https://review.openstack.org/#/c/104251/
16:16:06 <xyang1> jgriffith: we need to fix the bp link in commit msg
16:16:18 <jgriffith> xyang1: please do :)
16:16:26 <avishay> i started to review the extremio driver today
16:16:38 <xyang1> jgriffith: ok. thanks!  thanks Avishay!
16:16:48 <jgriffith> avishay: thoughts so far?
16:16:54 <jgriffith> avishay: going to be able to land this week?
16:16:55 <jgriffith> :)
16:17:41 <rushiagr> o/
16:17:47 <avishay> jgriffith: nothing too awful :)
16:17:47 <jgriffith> thingee: https://blueprints.launchpad.net/cinder/+spec/valid-states-api
16:17:59 <jgriffith> thingee: any chance of that making it?
16:18:45 <thingee> jgriffith: yes
16:18:49 <jgriffith> harlowja_away: thingee ^^
16:18:57 <jgriffith> k... what's the plan?
16:19:22 <jgriffith> thingee: harlowja_away the plan for that patch?
16:19:28 <thingee> my plan wasn't originally to work with harlowja_away
16:19:36 <thingee> it was just to stick the states in a common place.
16:19:42 <thingee> not something as extensive as his
16:19:54 <jgriffith> ok... so what are we going to propose by tomorrow?
16:20:22 <jgriffith> thingee: ok we can talk later
16:20:27 <jgriffith> you must be busy
16:20:29 <thingee> stick the states in a valid place, have both reset and list states pull from that
16:20:40 * thingee had someone coming up to his desk
16:20:49 <thingee> valid place = common place
16:20:53 <jgriffith> thingee: silly people, don't they know you're in a meeting :)
16:21:02 <jgriffith> thingee: Ok, let's talk after meeting
16:21:07 <thingee> certainly
16:21:50 <jgriffith> Ok... so the only one I really care about so to speak:
16:21:57 <jgriffith> https://blueprints.launchpad.net/cinder/+spec/secure-nfs
16:22:05 <jgriffith> anybody know Glenn?
16:22:10 <jgriffith> any idea of status here?
16:22:13 <bswartz> glenng: ping
16:22:20 <jgriffith> bswartz: :)
16:22:34 <jgriffith> bswartz: he's not in channel, already tried that :)
16:22:45 <bswartz> okay so it's already submitted
16:22:55 <bswartz> why is it failing pep8? >_<
16:23:20 <jgriffith> bswartz: I'll look at it
16:23:20 <bswartz> glenng sits down the aisle from him I'll go poke him
16:23:29 <bswartz> s/him/me/
16:23:41 <jgriffith> bswartz: none of these are making it into LP because it appears bad commit messages
16:24:08 <jgriffith> Ok, that's the list
16:24:20 <jgriffith> I don't anticipate adding anything after this meeting
16:24:32 <jgriffith> so speak now, or wait for Kilo
16:24:45 <jgriffith> although there's 3 items I wanted to add :(
16:24:50 <jungleboyj> jgriffith: We should also try to get this in https://review.openstack.org/#/c/112224/4/
16:25:03 <jungleboyj> The Reference implementation for replication.
16:25:04 <ronenkat> jgriffith: driver support for replication, how can we handle this?
16:25:17 <jgriffith> jungleboyj: so that's my next point :)
16:25:30 <jgriffith> jungleboyj: ronenkat we have two choices;
16:25:30 <bswartz> https://review.openstack.org/#/c/107693/
16:25:33 <jungleboyj> Also, I was late, do I have my ssh fix in the list?
16:25:35 <bswartz> that's the review for secure-nfs
16:25:39 <jgriffith> Tag on to the BP for the base work (I'm not a fan)
16:25:41 <bswartz> I don't see anything wrong w/ the commit msg
16:25:46 <jgriffith> Log as a bug
16:26:05 <jgriffith> and add to drivers as bug fixes
16:26:14 <jgriffith> not sure which is best here
16:26:18 <jgriffith> thoughts ^^
16:26:24 <jgriffith> we've tried it both ways in the past
16:26:39 <jgriffith> I hate adding on to the BP cuz they drag on forever when we do that
16:26:49 <jgriffith> and there's no good tracking of what's what
16:27:15 <jungleboyj> jgriffith: Sounds like bug is the way to then?  Though it is kind-of like marking a feature as a bug to get it in.  ;-)
16:27:31 <jgriffith> jungleboyj: that's exactly what it is
16:27:43 <avishay> as long as it doesn't get exploited
16:27:50 <jungleboyj> avishay: +2
16:28:03 <jgriffith> avishay: yeah, this late in the game I don't see much option
16:28:18 <jgriffith> avishay: jungleboyj but I would propose that people have 2 weeks after these features land
16:28:21 <jgriffith> and that's it
16:28:24 <jgriffith> then they wait for Kilo
16:28:36 <jungleboyj> ronenkat: Any concerns?
16:28:37 <jgriffith> and BTW if they don't land this week, they're out til Kilo
16:28:50 <jgriffith> this applies to pools, CG's and replication
16:29:03 <jgriffith> because there's a line of changes that follow them
16:29:09 <xyang1> jgriffith: The CG patch doesn't have a series goal.  maybe that is why it didn't show up in J-3 list.   https://blueprints.launchpad.net/cinder/+spec/consistency-groups
16:29:10 <bswartz> does land mean merge or submit?
16:29:19 <jgriffith> land == merge
16:29:24 <jgriffith> submit doesn't do any good
16:29:45 <bswartz> c'mon guys merge winston's pool-aware-scheduler!
16:29:49 <jgriffith> xyang1: fixing
16:29:53 <hemna> :(
16:29:56 <hemna> arg
16:29:58 <winston-d> bswartz: +2
16:29:59 <xyang1> jgriffith: thanks!
16:30:00 <hemna> the other face
16:30:13 <hemna> bswartz, we are working with winston-d on some issues w/ the pool scheduler
16:30:20 <hemna> wrt retyping
16:30:22 <bswartz> oh
16:30:29 <hemna> I think it's close though
16:30:34 <bswartz> cool
16:30:45 <jgriffith> so fyi, drop dead date is this week
16:30:52 <hemna> goosh
16:31:03 <jgriffith> so if it's not merged or ready to merge Monday it's out until Kilo for most cases
16:31:11 <jgriffith> hemna: sorry dude
16:31:38 <hemna> so it has to be merged by the 21st, not just submitted?
16:31:42 <jgriffith> hemna: but the fact is that it does no good to merge a new API feature at the last hour and NOT allow any of the drivers to submit against it
16:31:44 <hemna> I guess you just covered that
16:31:50 <jgriffith> hemna: no
16:32:02 <jgriffith> hemna: I said merged by Monday or ready to Merge
16:32:04 <jgriffith> Monday morning
16:32:05 <hemna> oh ok
16:32:19 <hemna> Monday.   sorry I was looking at the official schedule
16:32:21 <bswartz> warm up those +2 buttons
16:32:21 <kmartin> it so close maybe we should just open a bug for the retype issue, if it's even an issue?
16:32:34 <jgriffith> tomorrow is Proposal freeze
16:32:59 <jgriffith> kmartin: might want to consider that particularly if you're the only one impacted
16:33:54 <bswartz> if the feature works I say merge it so drivers can submit against it and file bugs so they can get fixed during j-4
16:33:54 <jungleboyj> kmartin, Sounds like an ok plan to me if you are ok with it.
16:34:11 <jgriffith> Ok... so is everyone clear on dates and deadlines here?
16:34:28 <kmartin> jgriffith: markstr is working in out with winston-d  as I type in the cinder channel :(
16:34:31 <hemna> yuh
16:34:38 <bswartz> yep
16:34:40 <jgriffith> no surprises?  No emails tonight at Midnight saying "oh crap, my bp wasn't targetted"
16:34:55 <jungleboyj> Where is the list jgriffith ?
16:35:00 <jungleboyj> Sorry.  Was late.
16:35:04 <jgriffith> jungleboyj: dude... really?
16:35:13 <jgriffith> jungleboyj: https://launchpad.net/cinder/+milestone/juno-3
16:35:27 <jgriffith> jungleboyj: so every milestone has a status page in LP
16:35:35 <jgriffith> jungleboyj: opens up with each milestone
16:35:42 <jgriffith> jungleboyj: should be a daily check :)
16:35:51 <jungleboyj> jgriffith: Sorry sir.
16:36:11 <jgriffith> jungleboyj: oh, now that's a bit uncalled for :)
16:36:24 <glenng> *wonders if anyone knows how to make the whip crack sound on IRC*
16:36:32 <jgriffith> jungleboyj: sir makes me feel sooo old, remember I got carded
16:36:34 <thingee> certainly
16:36:40 <thingee> whoops
16:36:43 <jungleboyj> jgriffith: https://blueprints.launchpad.net/cinder/+spec/configurable-ssh-host-key-policy
16:36:44 <thingee> http://status.openstack.org/reviews/#cinder
16:36:45 <jgriffith> alright... next topic... I'm going to change order a bit
16:36:57 <jungleboyj> jgriffith: Dude ^^^
16:37:12 <jungleboyj> jgriffith: That was the one I was asking about.  Can we still get that in?
16:37:41 <jgriffith> jungleboyj: that's alredy merged hasn't it?
16:38:06 <jungleboyj> jgriffith: No, that was what we were looking at at the meetup.
16:38:13 <jungleboyj> I need to apply the comments yet.
16:38:16 <markstur> jgriffith, was chatting on cinder.  I have a bp but it was just a retype/pool fix that was split off.  Probably could/should be bug?  S
16:38:47 <jgriffith> jungleboyj: well get on it man
16:38:49 <jgriffith> :)
16:38:53 <jgriffith> jungleboyj: you have it WIP
16:39:03 <jungleboyj> jgriffith: Right.
16:39:09 <jgriffith> ok, this has taken way longer than I had hoped
16:39:25 <jgriffith> let's get to some other topics and we can discuss in openstack-cinder after meeting
16:39:37 <jgriffith> I just wanted to make sure everyone was clear on dates and priorities
16:39:38 <jungleboyj> jgriffith: Didn't see it in the j3 list and wanted to make sure it wouldn't be rejected when I push up the code.
16:39:49 <jgriffith> jungleboyj: if you push it up by tomorrow it won't
16:39:56 <jgriffith> jungleboyj: if it's after tomorrow it will be rejected
16:40:03 <jungleboyj> jgriffith: Yikes.  Ok.
16:40:15 <jgriffith> jungleboyj: that's what Proposal Freeze means
16:40:23 <jgriffith> maybe I should explain that better...
16:40:34 <kmartin> jgriffith: what are the dates again? just kidding
16:40:39 <jgriffith> does everybody get what Feature Proposal Freeze is about?
16:40:42 <jgriffith> kmartin: LOL
16:41:07 <jgriffith> it means you can't propose any new Feature Work without an exception after said cutoff
16:41:16 <jgriffith> items that are in the gerrit queue count as making the cut
16:41:23 <avishay> jgriffith: what was the next topic? :)
16:41:25 <jgriffith> it's "new" patches/submissions
16:41:38 <jgriffith> avishay: thank you
16:42:02 <jgriffith> let's let ronenkat have at it quickly
16:42:11 <jgriffith> #topic replication patches
16:42:19 <jgriffith> I think the jest here is, reviews
16:42:28 <jgriffith> just raising awareness... right ronenkat ?
16:42:34 <ronenkat> Take a look at the Cinder meeting text, I put up a reference implementation for replication
16:42:44 <ronenkat> for any driver owner to look at
16:42:56 <ronenkat> jgriffith: thats it
16:43:07 <jgriffith> cool :)
16:43:29 <jgriffith> #topic mainters file
16:43:31 <jgriffith> DuncanT:
16:43:54 <jgriffith> Everybody that owns a driver, fill out the etherpad
16:43:58 <DuncanT> Ok, so I'm trying to collect up this info into a MAINTAINERS file in the cinder sources for now
16:44:04 <bswartz> link?
16:44:07 <DuncanT> Eventually I want it in the code
16:44:12 <DuncanT> https://etherpad.openstack.org/p/cinder-driver-maintainers
16:44:13 <jgriffith> bswartz: meeting agenda page
16:44:16 <bswartz> ty
16:44:22 * bswartz is lazy :-(
16:44:27 <jgriffith> bswartz: haha
16:44:33 <kmartin> DuncanT: did I not give you a link that had all the maintainers
16:44:42 <DuncanT> While building this, I noticed it is hard to tell classes that are not really drivers... the kilo abstract base class planned work should shake that out better
16:44:54 <jgriffith> DuncanT: +1
16:45:01 <DuncanT> kmartin: No, that was missing a bunch and a bunch were out-of-date
16:45:07 <kmartin> ok
16:45:26 <jgriffith> DuncanT: anything else?
16:45:27 <DuncanT> I will fill in any I know, but if people here could fill in the etherpad then it save me digging through my email
16:45:35 <DuncanT> Also CI usernames please
16:45:37 <DuncanT> That's it
16:45:39 <smcginnis> DuncanT: For maintainers are you looking for just an email contact?
16:45:43 <jgriffith> DuncanT: cool... thanks!
16:45:45 <DuncanT> Name and email
16:45:52 <jgriffith> DuncanT: don't go anywhere.. you get the next one too :)
16:45:55 <DuncanT> irc nick if you want
16:46:00 <ronenkat> Does it make sense to have the info in the openstack market place, the drivers part?
16:46:20 <jgriffith> ronenkat: that's come up before and currently my take is no
16:46:28 <jgriffith> ronenkat: that's a different "marketing" thing
16:46:28 <DuncanT> ronenkat: If somebody wants to update that from the maintainers file, they're very welcome
16:46:30 <ronenkat> :-)
16:46:44 <ronenkat> just looking to reduce redundent info
16:46:52 <jgriffith> ronenkat: good point
16:46:56 <DuncanT> I intend the git tree to be the single source of truth from the who to shout at PoV
16:47:06 <jgriffith> DuncanT: I like that
16:47:31 <jgriffith> DuncanT: a single source of truth list of drivers and maintainers would be welcome
16:47:42 <jgriffith> Ok...
16:47:48 <jgriffith> #topic simple CI
16:47:55 <DuncanT> I want to imbed it in the code at some point, but I need to think on that more
16:47:59 <DuncanT> Ok, simple CI
16:48:11 <DuncanT> Code will be at https://github.com/Funcan/kiss-ci today some time
16:48:20 <DuncanT> Hopefully within the hour
16:48:46 <smcginnis> DuncanT: Thanks for this!
16:48:46 <DuncanT> Probably won't quite work for anybody who isn't me initially but have a read, it is tiny
16:48:54 <e0ne> DuncanT: will it be based on openstack-infra/config or semethin others from infra?
16:49:04 <DuncanT> e0ne: Totally scratch written
16:49:16 <jgriffith> DuncanT: pull requests welcomed?
16:49:22 <DuncanT> e0ne: Much more lightweight, no jenkins, no nodepool etc
16:49:27 <DuncanT> Pull requests very welcome
16:49:28 <e0ne> looks interesting
16:49:41 <jgriffith> DuncanT: awesome
16:49:43 <DuncanT> I'll stackforge it once it is a bit more useful
16:49:52 <DuncanT> For now github pull requests are fine
16:50:02 <jgriffith> Everybody keep in mind we have no authoritative requirement for how this is done
16:50:09 <jgriffith> Offering up alternatives
16:50:11 <e0ne> DuncanT: hm.. i'm voting on leave jenkis w/o zuul, nodepool, etc:)
16:50:15 <DuncanT> I warn you now the code is terrible, I won't be upset by fixes of any type
16:50:21 <jgriffith> let people sort through and see what works best for them
16:50:45 <DuncanT> Ok, I'm done, I'll shout in the channel when it is up
16:50:51 <hemna> I had problems getting thingee's packer stuff working
16:51:04 <hemna> I have yet to get a single packer to work
16:51:16 <DuncanT> I'm using a hand built image at the moment... image creation is TBD
16:51:25 <jungleboyj> hemna: I was able to get one to work with virtualbox.
16:51:33 <e0ne> jgriffith: but almost all is looking to infra as a reference architecture, imp
16:51:37 <smcginnis> hemna: VirtualBox or KVM? I think there were kvm issues.
16:51:38 <e0ne> s/imp/imo
16:51:41 <jungleboyj> Had to use the i686 image for some reason.
16:51:45 <hemna> I'm using qemu
16:51:47 <jgriffith> e0ne: true statement
16:51:48 <hemna> or trying to at least
16:52:00 <jgriffith> e0ne: and I think if you go that route that's awesome
16:52:12 <e0ne> i'm trying...
16:52:20 * DuncanT hates jenkins enough to look into other routes
16:52:20 <jgriffith> e0ne: but I do think it's pretty heavy lift for some of us to just get this req fulfilled
16:52:38 <jgriffith> e0ne: and if there's a KISS approach that works well I'm all for it
16:52:44 <jgriffith> anywho....
16:52:52 <DuncanT> Then realised that there's an army of people who will hopefully fix up his crappy code ;-)
16:53:02 <jungleboyj> jgriffith: Are there expectations for 3rd Party CI reliability by the end of Juno?
16:53:05 <jgriffith> DuncanT: SocialCoding rocks!
16:53:22 <jgriffith> jungleboyj: there were expectations for reliability before they were turned on :)
16:53:38 <jgriffith> jungleboyj: I'm rather dissappointed in the current status
16:53:47 <jgriffith> jungleboyj: that's not a knock on anyone
16:53:49 <e0ne> jgriffith, DuncanT jenkins is good like a plugable-cron-launcher-and-publisher. all other staff - scripts with return code
16:53:52 <jgriffith> jungleboyj: just a statement of fact
16:53:53 * jungleboyj feels like I am asking all the good questions today.
16:54:02 <jgriffith> jungleboyj: :(
16:54:17 <jgriffith> Honestly... we've been pretty laxed here
16:54:35 <jgriffith> but IMO after J3 if your system isn't stable it shoudld be disabled until it is
16:54:42 <jgriffith> relegated to sandbox or whatever
16:55:00 <jgriffith> and it would be better for the maintainers to make that call than for myself or someone from infra
16:55:04 <jungleboyj> jgriffith: Agreed, what does that mean for the driver in the community though?
16:55:08 * avishay is waiting for the two most interesting topics on the agenda (6 minutes left) :(
16:55:19 <jgriffith> jungleboyj: I don't know what you're asking
16:55:44 <jgriffith> avishay: ok... let's cut over to it
16:55:49 <jgriffith> jungleboyj: and I can talk later
16:55:54 <jungleboyj> jgriffith: Sure.
16:55:54 <jgriffith> jungleboyj: ok with you?
16:55:57 <jgriffith> kk
16:56:09 <jgriffith> #topic idea of a maintenance rlease
16:56:11 <e0ne> jgriffith: may be it should run not all test(full tempest run)
16:56:25 <jgriffith> avishay: I'm assuming you wanted this one and not "summit" :)
16:56:32 <avishay> jgriffith: both
16:56:37 <jgriffith> e0ne: not an option
16:56:45 <e0ne> :(
16:56:53 <jgriffith> e0ne: and not the source of problems that I'm seeing currently
16:57:07 <jgriffith> current issue is failure to launch the nodes in most cases
16:57:10 <jgriffith> ok...
16:57:14 <jgriffith> stable release
16:57:23 <jgriffith> we've all talked about this at one point or another
16:57:44 <jgriffith> the question is do we want to pursue it, and how should we decide what it *means*
16:57:57 <hemna> what do you think it means?
16:58:00 <hemna> I'm unclear really
16:58:09 <jgriffith> Well there's gray area :)
16:58:18 <jgriffith> so one proposal is "no new features"
16:58:22 <jgriffith> period, full stop
16:58:44 <jgriffith> the entire release is bug fixes and improvements to the code
16:58:58 <hemna> I'm not against that at this point.
16:58:59 <avishay> i personally feel that the foundation of cinder is rather weak (DB, no recovery, limited scalability, etc), and the new fancy features are of limited value do to that.  take a break, fix up the base, and then continue.
16:58:59 <jgriffith> gray areas are things like adding features for HA etc (IMO)
16:59:20 <jgriffith> avishay: I'm not necessarily disagreeing
16:59:30 <hemna> yah I think that would be awesome IMHO
16:59:30 <jgriffith> hemna: avishay the other end of the spectrum.....
16:59:34 <jgriffith> No new drivers
16:59:46 <winston-d> or maybe we can set up virtual team, task force, to focus on tackling those performance/scalability/HA issues.
16:59:55 <avishay> winston-d: +1
17:00:04 <jgriffith> winston-d: sure
17:00:08 <hemna> we've been ratholing to a certain extent on adding new capabilities to cinder and not facing some issues, like HA, states, locks, etc.
17:00:12 <avishay> winston-d: maybe these issues never get done because they're too much for one person
17:00:18 <jgriffith> hemna: +1
17:00:19 <bswartz> time check
17:00:26 <jgriffith> vendor driven focus
17:00:27 <DuncanT> avishay: +1
17:00:31 <avishay> jgriffith: +10
17:00:37 <jgriffith> yeah... we're out of time
17:00:38 <avishay> very vendor-driven
17:00:44 <jgriffith> can we continue in cinder channel?
17:00:51 <avishay> i can for a little while
17:00:52 <jgriffith> and maybe have another talk next week?
17:00:55 <DuncanT> I certainly can
17:00:55 <winston-d> sure
17:00:58 <jgriffith> ok.. let's go over there
17:01:01 <jgriffith> thanks everyone
17:01:05 <jgriffith> #endmeeting cinder