14:00:19 <n0ano> #startmeeting nova-scheduler
14:00:19 <openstack> Meeting started Mon Sep 14 14:00:19 2015 UTC and is due to finish in 60 minutes.  The chair is n0ano. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:20 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:23 <openstack> The meeting name has been set to 'nova_scheduler'
14:00:27 <lxsli> o7
14:00:33 <n0ano> anyone other than lxsli here to talk about the scheduler?
14:00:37 <edleafe> o/
14:02:26 <bauzas> \o
14:02:41 <PaulMurray> \o
14:03:12 <n0ano> #topic Liberty patches – https://etherpad.openstack.org/p/liberty-nova-priorities-tracking
14:03:33 <n0ano> so I guess we didn't get all our patches into Liberty, that's a bit discouraging
14:04:28 <n0ano> I'm guessing everyone will be scrambling to finish the release so there won't be any time for Mitaka work until Nov.
14:05:30 <n0ano> the big miss was the request spec object, do we think we can get that pushed early in Nov?
14:05:37 <edleafe> yeah, mitaka won't even open before RC1
14:07:00 <n0ano> bauzas, was hoping you had some thoughts on your patches
14:07:11 <bauzas> so
14:07:20 <johnthetubaguy> curious if there are any big scheduler bugs folks think we need for Liberty, particularly release blockers
14:07:55 <bauzas> n0ano: a few changes merged, but not really related to the scheduler
14:08:11 <bauzas> johnthetubaguy: looked at the bugs, and no really bugs I can see to be merged
14:08:30 <n0ano> johnthetubaguy, haven't heard of any so I certainly don't think there are any blockers out there
14:08:56 <johnthetubaguy> n0ano: bauzas: good good, do shout if you hear about one, we are actively listening
14:09:13 <bauzas> johnthetubaguy: https://bugs.launchpad.net/nova/+bugs?field.tag=scheduler is just havng 2 bugs in progress
14:09:18 <bauzas> both of them are Low
14:09:49 <bauzas> others are wishlist ones, so I don't want them particularly to be merged during RC1
14:09:56 <johnthetubaguy> ack
14:11:26 <n0ano> so, back to request spec object
14:11:35 <johnthetubaguy> so I have a feeling we could get request spec merged before the summit, are we all lined up to push on things once mitaka opens?
14:12:19 <johnthetubaguy> I should be able to unblock process things once we are open
14:12:48 <n0ano> do we have any idea how bad the merge hell cycle will be once it opens?
14:12:58 <bauzas> your call
14:14:16 <johnthetubaguy> bauzas: we want to get things that were previously approved up for review ASAP, I feel, request spec seems a good candidate
14:14:29 <bauzas> johnthetubaguy: the spec is up for fast-track approval
14:14:55 <bauzas> #link https://review.openstack.org/#/c/220312/
14:14:56 <johnthetubaguy> bauzas: thats a good start
14:15:17 <johnthetubaguy> but I focused on code reviews right this second
14:15:38 <bauzas> okay
14:15:50 <bauzas> reviews are good anyway
14:17:25 <bauzas> well, I can hear crickets
14:17:30 <bauzas> are we moving on ?
14:17:36 <lxsli> lets
14:18:01 <n0ano> OK, no real resolution here, I guess it's just try an push to Mitaka as soon as it opens
14:18:11 <bauzas> n0ano: just trust ourselves:)
14:18:28 <bauzas> n0ano: we're in good progress and we'll be able to land some stuff
14:18:29 <n0ano> bauzas, indeed :-)
14:19:00 <n0ano> I just worry about when will we be ready to make the scheduler split out, it keeps dragging out
14:19:00 <bauzas> that's just a pretty calm period because of $ff  that's it
14:19:08 <johnthetubaguy> n0ano: what resolution are you looking for here?
14:19:40 <n0ano> I guess we're committing to having the code ready to push, will there be core reviewers ready to approve?
14:20:07 <bauzas> n0ano: we have
14:20:30 <johnthetubaguy> it seemed to get a lot of attention before
14:20:42 <bauzas> johnthetubaguy: +1 even +2
14:20:50 <johnthetubaguy> so I would hope it gets similar high levels of attention again
14:20:56 <bauzas> I never felt a miss of traction during that cycle, the prio helped a lot
14:21:16 <n0ano> OK, cross our fingers and let's hope we can get it merged
14:21:38 <johnthetubaguy> it was just a lot of code to review and agree on, I think we broke through the main bits, so hopefully not too bad now
14:22:08 <n0ano> OK, let's move on then
14:22:18 <n0ano> #topic Mitaka planning
14:22:35 <n0ano> I assume we'll have a scheduler session at the summit
14:22:53 <n0ano> I want to talk about removing flavor knowledge from the scheduler
14:22:55 <johnthetubaguy> so one extra request on the last stuff, sorry, really need more non-core reviews on those patches too, if possible
14:23:14 <n0ano> any other hot items we want to talk about there?
14:23:18 <jaypipes> hi guys, sorry for being late
14:23:33 <n0ano> johnthetubaguy, we'll do our best and try and get some more
14:24:15 <johnthetubaguy> n0ano: that would be cool, thank you
14:24:47 <n0ano> jaypipes, NP, we beat up on bauzas request spec patches, were all of yours merged?
14:26:32 <n0ano> jaypipes, YT
14:27:20 <n0ano> guess not, so anyother ideas people want for the summit?
14:29:09 <n0ano> must be too early, think about it, we can talk in more detail next week maybe
14:29:19 <edleafe> n0ano: I still think removing flavors from everything except the API is a worthwhile topic
14:29:36 <n0ano> edleafe, tnx, that's the one I want to raise
14:30:01 <lxsli> +1 but I don't have other ideas right now
14:30:05 <n0ano> #topic opens
14:30:15 <n0ano> so, any new issues for today?
14:31:11 <n0ano> OK, hearing crickets
14:31:14 <edleafe> just curious: is everyone going to be in Tokyo?
14:31:20 <bauzas> I do
14:31:21 <lxsli> aww yiss
14:31:24 <n0ano> I will be, just got approved
14:31:51 <jaypipes> n0ano: no, mine were not merged :(
14:32:08 <PaulMurray> jaypipes, has a very slow link
14:32:16 <PaulMurray> yes I will be ther
14:32:30 <jaypipes> bauzas: I tried to rework the one unit test you had suggested (the context manager one) but it was just pointless.
14:32:33 <n0ano> jaypipes, bummer, scroll back johnthetubaguy thinks we can get some things merged before Tokyo, mayber yours also
14:32:53 <bauzas> jaypipes: okay, we can chat on that later, I don't want to nitpick
14:32:55 <jaypipes> bauzas: I think I'd rather just delete the unit test. I mean, are we *really* testing to see whether a context manager works?
14:33:11 <bauzas> jaypipes: link to refresh my mind?
14:33:16 <johnthetubaguy> we can sure try, as normal, previously approved stuff gets fast track approved when the next release opens
14:33:31 <johnthetubaguy> just ping me once mitaka is open to help make that happen
14:33:40 <johnthetubaguy> (or any other nova-driver)
14:33:58 <n0ano> johnthetubaguy, I'll definitely remember to do that
14:34:20 <bauzas> I honestly wonder if we still need a weekly meeting until Mitaka opens
14:34:36 <edleafe> bauzas: good point
14:34:37 <johnthetubaguy> n0ano: cool, same process as last two releases with all that, basically
14:34:40 <bauzas> given the pretty low benefits of cross-coordination we have
14:34:43 <johnthetubaguy> n0ano: at least thats the current plan
14:34:49 <jaypipes> bauzas: https://review.openstack.org/#/c/213591/
14:35:07 <edleafe> bauzas: maybe one more just bfore tokyo to coordinate discussions at the summit
14:35:22 <bauzas> edleafe: yup, Tokyo is in 4 weeks, hence my point
14:35:45 <n0ano> edleafe, I'm willing to do that, maybe cancel next week, do the one after and then meet at Tokyo?
14:35:56 <edleafe> n0ano: +1
14:36:01 <bauzas> jaypipes: let's discuss that in -nova once we're done, agreed ?
14:36:10 <jaypipes> bauzas: yup, no worries.
14:36:12 <bauzas> n0ano: fair enough
14:36:56 <n0ano> OK, unless there's any last minute items
14:37:23 <n0ano> tnx everyone, talk again soonish
14:37:27 <n0ano> #endmeeting