21:00:17 <russellb> #startmeeting nova
21:00:17 <openstack> Meeting started Thu Jul 25 21:00:17 2013 UTC.  The chair is russellb. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:21 <openstack> The meeting name has been set to 'nova'
21:00:21 <russellb> hello, everyone!
21:00:24 <mikal> Hi
21:00:24 <mriedem> hi
21:00:26 <dansmith> yo
21:00:29 <cyeoh> hi
21:00:31 * geekinutah waves
21:00:41 <mrodden> hi
21:00:42 <russellb> happy thursday (or friday)
21:00:51 <n0ano> o/
21:00:53 <alaski> hi
21:00:59 <jog0> o/
21:01:05 <russellb> so I figure we'll spend most of the time going through havana-3 stuff, kind of like we did last week
21:01:07 <comstud> o/
21:01:13 <russellb> but one thing i wanted to mention first
21:01:17 <russellb> #topic nova wiki pages
21:01:24 <bnemec> o/
21:01:27 <russellb> i started working on a couple of wiki pages this week
21:01:34 <russellb> and i could use help on them, especially the 2nd one
21:01:43 <russellb> #link https://wiki.openstack.org/wiki/Nova/CoreTeam
21:01:50 <russellb> #link https://wiki.openstack.org/wiki/ReviewChecklist
21:02:03 <russellb> so, first one i'm trying to write down what the core team is, and what's expected of the members
21:02:16 <russellb> because i'm not sure it's really written down anywhere ...
21:02:25 <jog0> ideally that should be very similar to other core teams
21:02:28 <vishy> hi
21:02:33 <russellb> so just trying to be more explicit, but would like feedback on anything missing
21:02:43 <russellb> and the second page i'd like everyone's help with filling in
21:03:01 <russellb> goal being that we try to improve our consistency in reviews by writing down things to look for
21:03:46 <russellb> jog0: yep, and maybe the page becomes more generic once it's fleshed out.  i can post to the -dev list
21:03:58 <jog0> russellb: ++ to all of the above
21:04:07 <sdague> there should also be some extra care when looking at db migrations
21:04:13 <mikal> Sounds like a good idea to me
21:04:14 <sdague> that's probably worth calling out
21:04:17 <russellb> sdague: +1
21:04:36 <russellb> alright, on to havana-3
21:04:40 <russellb> #topic havana-3
21:04:48 <russellb> #link https://launchpad.net/nova/+milestone/havana-3
21:05:04 <russellb> 93 blueprints, 2 marked as implemented so far.
21:05:26 <russellb> so let's just crank through what we can
21:05:47 <russellb> comstud: alaski: think you can get the cells job working by havana?
21:05:57 <russellb> right now we have it as a high priority blueprint
21:06:34 <russellb> hrm
21:06:35 <alaski> I think it's reasonable to get it running a subset of tests
21:06:36 <comstud> yeah
21:06:44 <russellb> k
21:06:46 <dansmith> that would be awesome
21:06:52 <sdague> it would be good to get it over the devstack exercises rsn
21:06:59 <comstud> yeah
21:07:05 <comstud> that's on my list now that i'm back home
21:07:06 <sdague> because we don't really know what the shape of tempest is
21:07:23 <russellb> might end up having to scope this to fixing devstack exercises
21:07:33 <russellb> because at least then, even non-voting with some tempest failures, it's useful
21:07:44 <sdague> realistically cells just aren't being configured in devstack right now from what I can tell
21:07:44 <russellb> whereas with where it fails now, not so much
21:07:49 <comstud> there seems to be a basic problem in getting any build to work
21:07:51 <comstud> from what I can tell
21:08:01 <russellb> yeah, maybe i screwed it up
21:08:02 <sdague> the exercises just die because nova dies on anything
21:08:04 <russellb> i haven't really had a chance to dig
21:08:15 <russellb> cells seemed to work with that config for me
21:08:19 <alaski> I started to dig and got sidetracked.  I'll spend time with it tomorrow
21:08:20 <sdague> I think they'll work once devstack is fixed
21:08:21 <comstud> The config seems okay
21:08:32 <russellb> ok, cool
21:08:43 <sdague> realize there are enough things coming down with devstack-gate that it's not quite a local devstack config
21:08:46 <russellb> i think we should definitely aim for having it get pass devstack exercises and into tempest
21:09:00 <russellb> sdague: yeah, and i didn't test it that way
21:09:21 <russellb> joel-coffman: around?
21:09:30 <jog0> openstack-infra is working on documenting devstack-gate better
21:09:56 <joel-coffman> yep
21:10:24 <russellb> joel-coffman: how goes encrypted cinder volumes?
21:10:31 <russellb> joel-coffman: i marked it as blocked on the cinder side
21:10:43 <joel-coffman> thanks, that's still the current status
21:10:49 <russellb> k
21:10:58 <russellb> next is a volume weighter function ...
21:11:05 <russellb> shouldn't be hard to get in
21:11:05 <joel-coffman> we're flushing out a few bugs and will resubmit the Cinder patch next week
21:11:13 <russellb> joel-coffman: ok cool
21:11:19 <russellb> joel-coffman: let me know if anything changes
21:11:24 <russellb> comstud: direct mysql?
21:11:34 <russellb> still optimistic for h-3 ?
21:11:42 <comstud> optimistic... maybe overly
21:11:47 <comstud> but still optimistic
21:11:58 <russellb> well we have way too much, if it's something we can punt, let's do it
21:12:16 <comstud> let's revisit next week after I can have time to think
21:12:18 <russellb> ok
21:12:19 <russellb> heh
21:12:22 <comstud> :)
21:12:42 <russellb> mikal: any progress on libvirt console logging?
21:12:58 <mikal> russellb: i got bogged down on reviews for instance delete
21:13:02 <mikal> So no
21:13:09 <russellb> k
21:13:30 <russellb> oh, i forgot to link to my havana-3 openstack-dev email
21:13:32 <russellb> #link http://lists.openstack.org/pipermail/openstack-dev/2013-July/012323.html
21:13:56 <russellb> cyeoh: v3 api status?  we moved some of it to Icehouse already
21:14:27 <cyeoh> doing cleanup work post-h2 merge. Most of the focus is moving to testing and docs now
21:14:41 <cyeoh> there are going to be a *lot* of api sample patches going through at some stage
21:14:46 <russellb> all remaining extensions up for review?
21:15:02 <cyeoh> I was thinking of batching them up to help with the review load but they will be largish (just the nature of api samples)
21:15:16 <annegentle> cyeoh: I did hear from Kersten this morning about docs, thanks for that.
21:15:32 <russellb> ok, sounds good
21:15:37 <cyeoh> the extensions are in review, there are a handful of things to demote from core though
21:15:42 <cyeoh> but nothing big there
21:15:59 <cyeoh> annegentle: oh excellent - trying to work out how we can automate the process
21:16:01 <russellb> ok, will try to find the ones demoting things for core
21:16:18 <cyeoh> as there is currently a lot of manual work involved
21:16:27 <annegentle> cyeoh: yes it'll be great to get it automated
21:16:45 <annegentle> russellb: I invited her to Docs boot camp
21:16:48 <cyeoh> russellb: oh I meant there are a handful of ones for demotion from core that need to be written still... (eg they are merged, just as core instead of as an extension)
21:17:03 <russellb> annegentle: awesome
21:17:21 <russellb> annegentle: speaking of docs bootcamp, i reached out to some of the nova folks to see who might want to come ... no bites yet
21:17:27 <russellb> annegentle: s/want/can/
21:17:38 <russellb> would really like to have a nova rep though
21:17:41 <vishy> cyeoh: is the plan to remove some of the extensions that just extend other extensions?
21:17:57 <vishy> * er combine would be a better word
21:17:58 <annegentle> russellb: yep, ok thanks
21:18:03 <cyeoh> vishy: yes I think we've caught all of the ones that are just dummy placeholder ones
21:18:18 <dansmith> well, some function is being promoted too, right?
21:18:22 <cyeoh> vishy: and some of the others which are plainly just there because that was the only way to extend extensions
21:18:33 <vishy> cyeoh: cool I could use a link to the info about how to convert v2 to v3 since i need to do it for volume-swap
21:18:36 <cyeoh> vishy: but if you know of any others that you think should be merged, please let us know
21:18:52 <russellb> vishy: a straight conversion is pretty easy
21:18:58 <russellb> vishy: tons of examples that have been merged
21:19:05 <russellb> vishy: search for "part 1" and "part 2" commits
21:19:13 <cyeoh> vishy, https://wiki.openstack.org/wiki/Nova/PortingAPIExtensionToV3 covers most of it
21:19:14 <vishy> russellb: ok cool thx
21:19:49 <cyeoh> vishy: though yea looking at examples will probably tell you more :-)
21:19:49 <russellb> vishy: not sure if you caught it last week, but we decided we're pushing off finalization of v3 to Icehouse
21:20:08 <russellb> so we'll have it as experimental in Havana, and will tighten it up for Icehouse
21:20:36 <russellb> ok, so before I just keep going down the list, any specific blueprints anyone would like to discuss?
21:20:42 <cyeoh> we kind of need to work out when we will gate on v3 api tempest tests too as just adding them will blow out the test time (unless we have another vm)
21:20:57 <dansmith> boo
21:21:03 <russellb> dansmith: boo what
21:21:11 <dansmith> boo making the gate slower :P
21:21:15 <russellb> oh
21:21:22 <russellb> isn't there a tempest testr job now?  heh
21:21:26 <dansmith> I'm just being silly
21:21:45 * russellb nods
21:21:46 <dansmith> it does suck to have to create and destroy twice as many guests, but ..
21:22:08 <russellb> MOAR jobs!
21:22:43 <russellb> so any specific blueprints or bugs anyone wants to provide an update on?
21:22:45 <mriedem> russellb: just fyi that i'm told a patch will be up for this before end of week: https://blueprints.launchpad.net/nova/+spec/powervm-configdrive
21:22:53 <sdague> so the testr job could also use some help in debug. Most of the races are nova things we've tripped up. And it's a mix of tempest and nova bugs from what we can see.
21:23:07 <russellb> mriedem: ok, blueprint says "not started", could use an update
21:23:45 <mriedem> russellb: does launchpad change status when a patch is pushed for it like it does for bugs?
21:23:51 <russellb> mriedem: only bugs
21:23:53 <russellb> blueprints are manual
21:24:01 <mriedem> russellb: ok, i'll let david know
21:24:03 <russellb> it's thrilling.
21:24:34 <russellb> ok, well if nobody has specific blueprints they want to discuss, we can move to open discussion
21:24:48 <russellb> in general, i'm going to have to start aggressively pushing things out to Icehouse that don't look like they have progress
21:25:00 <russellb> #topic open discussion
21:25:06 <russellb> so what else is going on?
21:25:27 <dansmith> I have seen a bug a couple times,
21:25:33 <dansmith> once locally and in jenkins a bit:
21:25:34 <dansmith> http://logs.openstack.org/63/36363/17/check/gate-nova-python26/29748/console.html
21:25:40 <dansmith> anyone else seeing that?
21:25:48 <jog0> dansmith: yup
21:25:51 <jog0> was gonna say the same thing
21:25:55 <dansmith> okay, didn't see a bug for it
21:25:58 <clarkb> dansmith: there is a bug that was opened against openstack-ci
21:26:03 <clarkb> dansmith: let me find it so it can be moved to nova
21:26:09 <dansmith> ah, okay
21:26:29 <russellb> clarkb: highlight on "jenkins" ?  heh
21:26:32 <clarkb> https://bugs.launchpad.net/openstack-ci/+bug/1205022
21:26:33 <dansmith> heh
21:26:33 <uvirtbot> Launchpad bug 1205022 in openstack-ci "gate-nova-python26 fails build" [Undecided,New]
21:26:44 <jog0> I saw it for py27 to I think
21:27:15 <jog0> yeah http://logs.openstack.org/36/36436/5/check/gate-nova-python27/29937/console.html.gz
21:28:12 <bnemec> I saw that locally too.
21:28:22 * dansmith blames comstud
21:28:25 <russellb> well then
21:28:33 <russellb> who's going to fix it?  :-)
21:28:45 <dansmith> the bug tracker really needs a "to-blame" field
21:29:01 <mrodden> make a tag
21:29:05 <mrodden> blame-comstud
21:29:09 <dansmith> heh
21:29:12 <dansmith> I'll look at it
21:29:17 <clarkb> dansmith: is that so that bugs opened against nova can be blamed on infra?
21:29:32 <dansmith> it's instance-related so it's *possible* it's objects related, but I doubt it, given its location
21:29:44 <dansmith> clarkb: no, we could even hard-code the field to be "comstud" for all I care :P
21:30:39 <jog0> I am unable to produce it locally
21:30:44 <dansmith> I saw it once
21:30:50 <bnemec> It's inconsistent.
21:31:00 <bnemec> One time I got two failures, the next none, the next only one.
21:31:12 <jog0> this may be python version related
21:32:39 <russellb> dragondm: around?  i haven't heard/seen anything on versioning notifications
21:32:53 <russellb> dragondm: have tried to ping a few times :-/
21:33:30 <dragondm> ah, yah, that's been sidelined for awhile, because of other, more pressing work.
21:33:41 <russellb> dragondm: ok, so should I just push it off to Icehouse?
21:34:18 <dragondm> yah, for now
21:34:21 <russellb> dragondm: ok, thanks
21:34:38 <comstud> wait, wut!?  i didn't do nothin'.
21:34:46 <russellb> alright any other topics before i shut down the meeting?
21:35:23 <russellb> alrighty then
21:35:24 <russellb> thanks!
21:35:25 <russellb> #endmeeting