21:00:03 <mikal> #startmeeting nova
21:00:03 <openstack> Meeting started Thu Jul  2 21:00:03 2015 UTC and is due to finish in 60 minutes.  The chair is mikal. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:04 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:07 <openstack> The meeting name has been set to 'nova'
21:00:12 <alaski> o/
21:00:15 <mriedem> hi
21:00:15 <mikal> abhishekk mikal tjones cburgess jgrimm adrian_otto funzo mjturek jcookekhugen irina_pov krtaylor danpb alexpilotti flip214 jaypipes gilliard garyk edleafe dims moshele anteaya Nisha sileht claudiub lxsli neiljerram markus_z swamireddy alevine tonyb andreykurilin ndipanov sc68cal akuriata artom jlvillal mnestratov kashyap aloga rgeragnov: ping
21:00:15 <beagles> o/
21:00:17 <tpatil> o/
21:00:20 <dansmith> o/*  <-- holding a firework
21:00:20 <alex_xu> o/
21:00:20 <sc68cal> o/
21:00:24 <mikal> Heh
21:00:25 <melwitt_> o/
21:00:28 <sc68cal> dansmith: heh
21:00:36 <tjones1> o/
21:00:44 <mikal> So, I am under orders to keep this short so the Americans can go and blow stuff up
21:00:48 <mikal> So let's do that thing
21:00:48 <llu-laptop> o/
21:00:52 <dansmith> hehe
21:00:53 <mikal> #topic Release Status
21:01:08 <mikal> July 16 is non-priority feature proposal freeze
21:01:26 <mikal> Which as well as being a mouthful means if you are working on a non-priority non-bug fix, the code needs to be proposed by then
21:01:46 <mikal> We're also working through spec freeze exception requests, for which there is an etherpad
21:01:53 <mikal> #link https://etherpad.openstack.org/p/liberty-spec-freeze-exceptions
21:02:26 <mikal> We're under orders to discuss blocked spec freeze exceptions here
21:02:29 * mikal scans the etherpad
21:02:58 <mikal> Its not clear to me how blocked is defined in this context
21:03:01 <mikal> Thoughts people?
21:03:14 <dansmith> I honestly wasn't sure there was anything going on in there other than proposals
21:03:31 <mikal> Yeah, I thought this was people asking for review "votes"
21:03:37 <mikal> There is a lot there, which is a bit scary
21:03:38 <dansmith> I guess john +2'd something in there
21:04:01 <mikal> So, I guess take a look at the listed reviews ASAP
21:04:07 <mikal> And we can discuss them in gerrit as required
21:04:22 <dansmith> I see precisely one vote for one thing
21:04:25 <alaski> is there an exception deadline?
21:04:29 <dansmith> aside from anonymous +1s
21:04:31 <dansmith> alaski: today
21:04:43 <alaski> for voting, or proposing?
21:04:44 <claudiub|2> o/
21:04:44 <mikal> alaski: final review is "tomorrow" according to the agenda
21:04:48 <alaski> ahh, okay
21:04:52 <mikal> So yes, I guess the close of this week
21:04:52 <dansmith> I thought it was Jul 2?
21:04:56 <mikal> Which kind of sucks for Americans
21:05:15 <mikal> dansmith: I just obey the agenda... Not sure if its been extended deliberately or not.
21:05:34 <mikal> Unless anyone wants to discuss a specific proposal I think we should move on?
21:05:45 <tpatil> I would like to discuss about Improving performance of UnShelve API specs
21:05:49 <mikal> And promise to just keep reviewing during gaps in our Freedom Explosions?
21:05:51 <dansmith> Ideally we would have all spec exceptions voted on by 2nd July, and
21:05:56 <dansmith> from the email ^
21:06:00 <dansmith> so today is the proposal deadline
21:06:06 <dansmith> tomorrow is the merge deadline I guess
21:06:17 <alex_xu> I hope get some review for custom cpu model https://review.openstack.org/#/c/168982/, it already get agreement on the direction, just need some review
21:06:19 <mikal> tpatil: what's the review URL?
21:06:28 <tpatil> #link : I would like to discuss about Improving performance of UnShelve API specs
21:06:36 <tpatil> sorry
21:06:50 <dansmith> the shelve things are really not in the critical path, IMHO
21:06:54 <tpatil> #link: https://review.openstack.org/#/c/135387/13
21:07:13 <tpatil> we have been trying to get this thing from last release cycle
21:07:19 <dansmith> alex_xu: I don't think I'd characterize that as "agreed on direction"
21:07:29 <tpatil> I would appreciate if it's consider for this release cycle atleast
21:07:42 <tpatil> considered
21:07:46 <alex_xu> dansmith: emm...no...how can I get agreed on that
21:08:14 <mikal> tpatil: well, it has been considered right? I see heaps of discussion on it in gerrit
21:08:26 <mikal> tpatil: so much in fact that I can't immediately see what the current concerns are
21:08:35 <dansmith> alex_xu: the last reviewer was me on Jun 24.. I guess it feels to me like it's a sticky situation and there isn't a lot of real support for it :/
21:08:49 <tpatil> mikal: right, we have addressed alaski's concern in PS 13.
21:09:01 <tpatil> Want people to review specs and give some feedback
21:09:15 <alaski> I will review the shelving one again before I leave today
21:09:27 <tpatil> Thanks, alaski
21:09:28 <mikal> alaski: thanks
21:09:30 <mikal> I'll read it too
21:09:37 <alex_xu> dansmith: emm...I think all the reviewer think we need abstract cpu capabilities, that is the direction we agreed? The rest is how to implement it.
21:09:40 <mikal> Its already on the request for exception list, so we should be doing that thing anyways
21:09:40 <tpatil> Thanks, mikal
21:09:50 <llu-laptop> spec 'Add new API to list compute node metric names' wants reviews. It was discussed during Vancouver summit and take-out then was to wait for Jay Pipe's input. Now Jay has given his +1 on this spec.
21:10:01 <tpatil> mikal: great
21:10:02 <llu-laptop> #link https://review.openstack.org/#/c/180049/
21:10:17 <dansmith> alex_xu: I'm not so sure about that
21:10:22 <melwitt_> the service group api spec had a lot of discussion and has one +2 already so I was thinking probably that one is good to go
21:10:30 <mikal> llu-laptop: ahhh, API ones are interesting in that the freeze might not apply if its a bug fix
21:10:39 <mikal> llu-laptop: but this one looks like a feature to me?
21:10:49 <malini> @dansmith,  there is a need for cpu-info, an improvement on what is there today, pave the way for something better
21:10:54 <llu-laptop> mikal: it's not a bug fix, it's adding new api
21:11:42 <malini> In Kilo it was heavily debated, if there is improvement and a plan to make it robust, better than analysis paralysis
21:11:43 <mikal> llu-laptop: can you address Jay's nits, and I'll add it to the list I try to review today
21:12:00 <llu-laptop> mikal: sure
21:12:14 <dansmith> mikal: that one is pretty small,
21:12:22 <dansmith> mikal: and probably the best candidate I've seen so far
21:12:32 <malini> Alex_xu, Chris (windriver) and others at Intel really are working towards addressing concerns and making this happen
21:12:36 <mikal> dansmith: cool, I havne't read it yet to be hoenst
21:12:38 <malini> we need your help
21:12:42 <mikal> So yeah, we have a list of these things
21:12:53 <mikal> And specs-core is meant to be reviewing based on that list
21:13:01 <mikal> And all the ones that have been brought up are correctly listed
21:13:14 <mikal> So I feel like we just need to give the process more time to do those reviews?
21:13:57 <beagles> I'd like to point out https://review.openstack.org/#/c/193668/ - judging by the notes on the etherpad there may have been conversations or decisions on this already that I'm not aware of
21:14:38 <dansmith> this one terrifies me,
21:14:45 <dansmith> in that it's big, complicated, cross-project, and very very new
21:14:46 <mikal> beagles: I am unaware of any such conversations, but that doesn't mean they didn't happen
21:15:02 * beagles nods
21:15:06 <dansmith> but, I think it's also something that will mostly be not a nova thing until the end when we go try to merge nova changes to use it
21:15:10 <mikal> If that one is big and complicated its it a good candidate for discussion at the mid-cycle?
21:15:23 <mikal> Sorry, is it?
21:15:25 <dansmith> well, midcycle is too late
21:15:40 <mikal> The bits outside nova are already done?
21:15:41 <mriedem> the library can always be worked as POC on the side
21:15:47 <dansmith> no
21:15:54 <dansmith> mikal: this has only been a thought for like 1.5 weeks or so
21:16:06 <dansmith> mikal: jaypipes rage-coded something over a weekend
21:16:14 <mikal> Yeah, so it seems premature to be landing the nova bits then
21:16:14 <dansmith> to go with the spec
21:16:27 <mriedem> it's also neutron changes
21:16:37 <dansmith> right, it's just a large scope
21:16:48 <dansmith> so like I said above,
21:16:52 <beagles> yeah... it's kind of tough. We *could* say land one side and then the next but that worries me
21:17:01 <beagles> it isn't really done until both sides work
21:17:05 <dansmith> the library work seems like it could go and and mature this cycle and the spec is actually for M to make nova _use_ it
21:17:22 <dansmith> s/and the/and then the/
21:17:25 <mikal> Yeah, that's kind of what I meant by the mid-cycle thing
21:17:32 <mikal> If this is happening outside nova, and aimed at M
21:17:37 <dansmith> mikal: cool, I thought you meant midcycle for discussion of the spec
21:17:42 <mikal> Then what we need to do is talk through what we'd want to see from it
21:17:50 <mikal> And make sure we're on the same page as them
21:18:02 <mikal> dansmith: nah, just check pointing the direction of external work
21:18:10 <dansmith> then..yes
21:18:17 <mriedem> and mestery will be at our midcycle i think
21:18:18 <mriedem> so yeah
21:18:19 <mriedem> punt
21:18:35 <mikal> Can we put something to that effect into the etherpad then please?
21:18:47 <mikal> Given people want to leave early I feel like we should move on
21:18:50 <mikal> Anything else here?
21:18:55 <mriedem> i'll say something in the etherpad
21:18:59 <mikal> Ta
21:19:04 <beagles> thanks
21:19:46 <mikal> #topic Traditional Reminders
21:19:53 <mikal> So, the agenda asks me to remind you that...
21:19:54 <malini> sorry to raise https://review.openstack.org/#/c/168982/ again, but no further discussion. it has been heavily debated, do not want this to fall into analysis paralysis but allow to implement
21:21:11 <malini> mikal, pinging again for the cpu-info BP that Alex xu requested feature free exception on
21:21:31 <tonyb> malini: that was discussed here already
21:21:35 <mikal> malini: yep, its on the list for consideration, I don't think there's a lot else I can do for it in this meeting
21:21:53 <mikal> So, reminders
21:22:09 <mikal> Yuo should be reviewing with the priority review list in mind
21:22:12 <mikal> #link https://etherpad.openstack.org/p/liberty-nova-priorities-tracking
21:22:25 <mikal> You probably have action items from the summit you should feel bad about
21:22:29 <mikal> #link https://etherpad.openstack.org/p/YVR-nova-liberty-summit-action-items
21:22:39 <mikal> The registration deadline for the mid-cycle is July 7 (next week)
21:22:47 <mikal> Any other reminder things?
21:22:54 <malini> thanks tonyb and mikal for the ack
21:23:27 <mikal> #topic Stable branch
21:23:38 <mriedem> stable is hunky dory
21:23:39 <mikal> Icehouse has now been eol'ed I believe?
21:23:44 <mriedem> yeah, working toward
21:23:48 <mikal> But that's informational
21:24:00 <mikal> malini: oh, I thought fungi has actually done it now?
21:24:05 <mriedem> yeah it's done
21:24:07 <mriedem> just looked
21:24:08 <mikal> Cool
21:24:17 <mikal> #topic Gate status
21:24:20 <mriedem> hunky dory
21:24:23 <mikal> The gate has been perfect!
21:24:26 <mikal> Perfectly broken
21:24:29 <mriedem> if you're not neuitron or horizon
21:24:42 <mikal> Well, it was exciting earlier in the week
21:24:45 <mriedem> gate is fine excluding the first half of this week
21:24:46 <mikal> But its sorted now right?
21:24:47 <mriedem> yes
21:24:51 <mriedem> no recent oslo releases
21:24:58 <mikal> Thanks for chasing that when it was all wonky
21:25:00 <mikal> Heh
21:25:08 <mikal> #topic Open Discussion
21:25:15 <mikal> Ok, we can talk, or we can go an explode things.
21:25:17 <mikal> Thoughts?
21:25:19 <malini> mikal -- :-) no no, ack in my remark here in meeting, would very much like the +2s on the BP, not there yet!
21:25:33 <shhfeng> sorry to raise https://review.openstack.org/#/c/192622 , it enable pci api
21:25:58 <claudiub|2> yeah, I have a few spec exceptions that have +2s already.
21:26:02 <mikal> malini: yep, people are reviewing it. It might take some time though.
21:26:17 <mikal> claudiub|2: and they're on the etherpad, yes?
21:26:24 <claudiub|2> mikal: yeah.
21:26:27 <mikal> claudiub|2: cool
21:27:21 <shhfeng> we have implement the underly pci logical and API code, just need to enable this api.
21:27:53 <mikal> shhfeng: is it on the spec freeze exception etherpad?
21:28:30 <mriedem> early mark
21:28:30 <alex_xu> all the code for pci api already in the nova tree, just need enable, so coding should be easy
21:28:40 <shhfeng> mikal, yes.
21:28:53 <mikal> shhfeng: cool, we it should get considered as part of the process then
21:28:54 <mriedem> testing?
21:29:03 <mikal> I think we could let people go do 4th of July prep now, yes?
21:29:09 <mriedem> sure
21:29:28 <shhfeng> mikal, great.
21:29:30 <mikal> I insist that all fireworks are renamed Freedom Explosions though
21:29:49 <mriedem> freedomcrackers
21:29:50 <mikal> Its like a little burst of freedom in the sky
21:29:56 <mikal> And with that...
21:29:56 <mriedem> or your hand
21:29:58 <mikal> #endmeeting