17:00:31 <dansmith> #startmeeting nova_cells
17:00:32 <openstack> Meeting started Wed Jan  4 17:00:31 2017 UTC and is due to finish in 60 minutes.  The chair is dansmith. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:33 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:36 <openstack> The meeting name has been set to 'nova_cells'
17:00:48 <mriedem> o/
17:00:52 <dansmith> ohai
17:00:53 <dtp> here
17:00:56 <melwitt> o/
17:01:13 <dansmith> cool
17:01:23 <dansmith> #topic testing/bugs
17:01:58 <mriedem> so,
17:01:59 <dansmith> anything here? I've been disconnected for >2 weeks, so I wouldn't know if we have any fires burning,
17:02:05 <mriedem> there aren't fires
17:02:20 <mriedem> just a general nod that i need to get the nova-status upgrade check stuff into our ci system,
17:02:28 <mriedem> i haven't yet dug into where to place that in the flow,
17:02:40 <dansmith> okay
17:02:40 <mriedem> i think it could be in grenade and/or devstack, because it also works for base installs
17:03:06 <mriedem> i'm not sure at what point grenade has the ocata code pip installed but not started - i think that's devstack
17:03:12 <mriedem> unless there is a post-install / pre-start hook
17:03:27 <mriedem> if not, i might have to use a venv?
17:03:32 <dansmith> hmm
17:03:58 <dansmith> can't imagine people being super excited about that.. I was thinking it would be in the pre-upgrade stuff in grenade where we have installed, but not started things yet
17:04:04 <dansmith> and if that isn't actually a phase, then we probably need it to be
17:04:25 <mriedem> heh yeah i'm not sure if that exists
17:04:30 <mriedem> but i haven't dug into it yet either
17:04:31 <dansmith> okay
17:04:38 <mriedem> i can, and if i get stuck i'll bug sdague
17:04:48 <dansmith> okay
17:05:32 <dansmith> anything else here?
17:05:59 <dansmith> #topic open reviews
17:06:20 <dansmith> so, my mega series is becoming less mega, which is cool: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/cells-scheduling-interaction
17:06:25 <dansmith> merged another thing yesterday
17:06:37 <dansmith> the bottom patch is bumping up against the cellsv1 job not having cellsv2 mappings setu
17:06:47 <mriedem> yeah....
17:06:48 <dansmith> I've got a WIP against devstack to do that and see how it goes
17:06:52 <mriedem> cool
17:07:04 <dansmith> I *think* we left that unconverted just to have one job that wasn't or something
17:07:12 <dansmith> hoping there's not something more complicated needed, but we'll see
17:07:21 <mriedem> i don't think we put that much thought into it
17:07:23 <mriedem> unless laski did
17:07:33 <mriedem> i think we just did it because 'this probably blows up with cells v1 so let's not anger that troll yet'
17:07:46 <dansmith> yeah, maybe
17:07:50 <melwitt> heh
17:08:10 <dansmith> so, that's all from me.. who else has stuff up that needs calling out?
17:08:18 <dtp> https://review.openstack.org/#/c/415922/
17:08:49 <dansmith> sweet
17:08:50 <melwitt> I put up a first draft of the quota resource counting at https://review.openstack.org/#/c/416521/ this morning
17:09:06 <dansmith> dtp: I'll take a look at that later today.. hopefully melwitt will as well
17:09:12 <dtp> thanks
17:09:22 <melwitt> yeah, will do
17:09:22 <mriedem> dtp: you need to fix the commit message :)
17:09:26 <mriedem> that's my review contribution
17:09:44 <dtp> can i edit this UI to word wrap? :P
17:09:57 <mriedem> dtp: i'd do it locally
17:10:10 <dtp> will do
17:10:16 <dtp> is that 80 chars too?
17:10:19 <dansmith> okay, got both of those things on the priorities tracking etherpad
17:10:38 <mriedem> dtp: whatever this says https://wiki.openstack.org/wiki/GitCommitMessages
17:10:44 <mriedem> wrap at 79
17:10:55 <melwitt> I think it's more like 72 I think but I don't think anything checks
17:11:00 <mriedem> yeah ^ says 72
17:11:09 <dtp> okey dokey, thanks
17:11:36 <mriedem> melwitt: you had some WIP patches for moving quotas to the API DB
17:11:54 <mriedem> i think lacking tests, was that all that was making them WIPs?
17:12:27 <melwitt> mriedem: yeah, still do. need to add a bunch o tests, and I think trim down the quota class migration to be only the default class maybe
17:12:52 <mriedem> i guess we never settled on that
17:13:06 <mriedem> but since custom quota classes aren't used anyway, i think that's fine
17:13:27 <melwitt> I also have this patch for adding message queue switching back to target_cell https://review.openstack.org/#/c/411543/
17:13:28 <mriedem> but i assume the migration code is rather generic anyway so i'm not sure how much it matters
17:14:02 <melwitt> yeah, the way I have it right now it'll migrate all quota classes the user has, which is maybe better anyway
17:16:25 <mriedem> dansmith: another thing i thought about last night
17:16:31 <mriedem> on your bottom patch for listing instances across cells,
17:16:41 <mriedem> we talked about a dumb merge sort, is that later in the series?
17:17:07 <dansmith> mriedem: not yet
17:17:18 <mriedem> ok. it also got me thinking,
17:17:21 <dansmith> mriedem: figured we'd add that before we add a release note saying that multiple cells is technically supported
17:17:32 <mriedem> since we stack build requests then cell0 then cells instances,
17:17:40 <mriedem> any sort order thing is out of whack today anyway
17:18:07 <mriedem> so i guess it could just be considered a bug
17:18:15 <mriedem> idk, my point is i'm not terribly concerned about it right now
17:18:29 <dansmith> so today you'll get the barber pole results across all cells, which I think is what you're getting at
17:18:42 <mriedem> right
17:18:49 <dansmith> presumably some people will want to just disable the dumb sorting thing anyway
17:18:53 <mriedem> but even with a single cell,
17:18:58 <dansmith> right, I know
17:19:01 <mriedem> ok
17:19:10 <mriedem> yeah my point being it's already 'busted'
17:19:13 <dansmith> yeah
17:19:13 <mriedem> since newton
17:19:15 <mriedem> ok
17:19:32 <mriedem> lucky for us everyone is still on <=mitaka
17:19:35 <dansmith> heh
17:19:58 <mriedem> ok i don't have anything else
17:19:59 <melwitt> "lucky" :)
17:20:00 <dansmith> okay any other reviews up that need visibility?
17:20:43 <dansmith> #topic open discussion
17:20:52 <dansmith> anything else about anything?
17:22:21 <mriedem> nope
17:22:39 <dtp> not me
17:22:45 <dansmith> alrighty then
17:22:48 <dansmith> #endmeeting