21:00:00 <dansmith> #startmeeting nova
21:00:01 <openstack> Meeting started Thu Mar 30 21:00:00 2017 UTC and is due to finish in 60 minutes.  The chair is dansmith. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:02 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:04 <openstack> The meeting name has been set to 'nova'
21:00:07 <mriedem> o/
21:00:08 <takashin> o/
21:00:12 <edleafe> \o
21:00:22 <melwitt> o/
21:00:23 * edleafe has to run out in a few minutes
21:00:31 <dtp> o/
21:01:02 <dansmith> sadly, you have to deal with me today because mriedem is in the air
21:01:03 <dansmith> #topic Release News
21:01:10 <dansmith> #link Pike release schedule: https://wiki.openstack.org/wiki/Nova/Pike_Release_Schedule
21:01:45 <dansmith> The next thing on the actual release schedule is April 13: P-1
21:01:54 * bauzas waves
21:01:58 <dansmith> #info Blueprints: 59 targeted, 39 approved, 1 completed
21:02:09 <dansmith> and one more stat for you stat lovers:
21:02:14 <dansmith> #info Open spec reviews: 109 (down 2 from last week)
21:02:23 <mriedem> god i love those stats
21:02:27 <mriedem> thanks dan!
21:02:30 <dansmith> I know you do
21:02:44 <dansmith> #topic Bugs
21:03:02 <dansmith> #help Our backlog of new untriaged bugs is high. Need help with triage of new bugs.
21:03:13 * dansmith makes a note to silently close some bugs to help the ratios
21:03:26 <bauzas> roger.
21:03:34 <dansmith> I see no critical bugs in the query
21:03:36 <dansmith> so I guess that's good
21:03:43 <jaypipes> hey, I'm here.
21:04:03 <dansmith> #link check queue gate status http://status.openstack.org/elastic-recheck/index.html
21:04:15 <mriedem> all's well in the gate besides libvirt
21:04:16 <dansmith> I know clarkb was looking for help on that top libvirt crash bug
21:04:32 <dansmith> cool
21:04:38 <dansmith> #link 3rd party CI status http://ci-watch.tintri.com/project?project=nova&time=7+days
21:04:45 <dansmith> can we say the same about 3rd party CI?
21:05:00 <clarkb> yes I was. Seemed that from libvirt side answer was "stop using old libvirt" so I've got a patch up to use UCA in devstack (thats still not latest or a supported stable version? but at least its newer)
21:05:12 <mriedem> dansmith: yeah
21:05:15 <dansmith> clarkb: ah, awesome
21:05:22 <clarkb> ianw was also working on getting core dumps better supported by devstack so we can collect them
21:05:27 <mriedem> that's always the answer from libvirt
21:05:59 <dansmith> well, the third party CI dashboard won't even load for me
21:06:04 <dansmith> so I'll just assume it's all rosy
21:06:24 <dansmith> #topic Reminders
21:06:31 <dansmith> #info Spec review sprint is April 4
21:06:37 <dansmith> That is next tuesday
21:06:45 * dansmith plans a sudden medical incident for tuesday
21:06:53 <mriedem> severed nips
21:06:58 <mriedem> classic excuse
21:07:11 <dansmith> "honest doc, I was using the recommended setting on the clamps"
21:07:18 <dansmith> er, anyway
21:07:24 <dansmith> #info Spec freeze is April 13
21:07:33 <dansmith> #link Pike Review Priorities etherpad: https://etherpad.openstack.org/p/pike-nova-priorities-tracking
21:07:37 <ianw> (https://review.openstack.org/#/c/451128/ devstack libvirt coredumps, sorry, from previous topic)
21:07:46 <dansmith> I know I'm a bit behind on updating the cells section of the prio etherpad,
21:07:51 <dansmith> but do as I say not as I do, and stuff
21:07:56 <dansmith> ianw: thanks
21:08:13 <dansmith> #link Forum planning: https://wiki.openstack.org/wiki/Forum/Boston2017
21:08:23 <dansmith> everyone should be aware of this by now, but if not, ^
21:08:37 <dansmith> there's a nova-specific place for brainstorming and smaller topics:
21:08:42 <dansmith> #link https://etherpad.openstack.org/p/BOS-Nova-brainstorming Forum discussion planning for nova (add your name if you are going)
21:08:45 <dansmith> deadline is:
21:08:49 <dansmith> #info EOD April 2: deadline for Forum topic submission: http://forumtopics.openstack.org/
21:09:02 * dansmith notes mriedem has made running the meeting as easy as cut-n-paste
21:09:09 <mriedem> \o/
21:09:10 <dansmith> any other reminders here?
21:09:40 <dansmith> #topic Stable Branch Status
21:09:48 <mriedem> no news really
21:09:48 <bauzas> maybe the new contributors session
21:09:57 <bauzas> but meh, let's discuss it offline
21:10:01 <dansmith> #link https://etherpad.openstack.org/p/stable-tracker
21:10:02 <mriedem> we have one ocata regression to get fixed yet,
21:10:03 <mriedem> and backporte
21:10:07 <mriedem> *backported
21:10:23 <dansmith> sweet
21:10:50 <dansmith> #topic Subteam Highlights
21:11:02 <dansmith> dansmith: go
21:11:08 <dansmith> dansmith: well, we didn't really have a meeting this week because only myself and dtp showed up
21:11:14 <dansmith> dansmith: so we punted and got back to work
21:11:16 <dansmith> dansmith: that is all
21:11:30 <dansmith> dansmith: thanks for that
21:11:37 <dansmith> scheduler
21:11:52 <danased> Discussed the os-traits work, and the change to re-organize the library: https://review.openstack.org/#/c/448282
21:12:01 <danased> Discussion about the use-local-scheduler patch took up most of the rest of the meeting: https://review.openstack.org/#/c/438936
21:12:08 <danased> EOM
21:12:17 <dansmith> thanks, ed
21:12:23 <mriedem> oh dan as ed
21:12:25 <mriedem> i get it now
21:12:29 <mriedem> i thought dana sed
21:12:32 <dansmith> tdurakov: around for live migration?
21:12:33 <mriedem> and who's dana?
21:12:33 <dansmith> heh
21:12:53 <melwitt> I thought da nased
21:13:02 <dansmith> guess not
21:13:05 <dansmith> alex_xu: sdague: ?
21:13:07 <dansmith> api?
21:13:33 <mriedem> it's in the agenda
21:13:34 <dansmith> Looks like there is some stuff in the agenda, so:
21:13:36 <mriedem> same for live migration
21:13:39 <mriedem> refresh
21:13:40 <dansmith> Good progress on policy doc https://review.openstack.org/#/q/project:openstack/nova+branch:master+topic:bp/policy-docs
21:13:46 <dansmith> Other two specs still need more review https://review.openstack.org/433037 and https://review.openstack.org/427872
21:13:51 <dansmith> The patch for removing stevedore is ready https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:bp/api-no-more-extensions-pike. The whole plan is in the etherpad https://etherpad.openstack.org/p/api-no-more-extensions-pike
21:13:57 <dansmith> Discuss the nova api via web-servers. sdague think the uwsgi stuff should be sorted in devstack before cutting over anyway
21:14:07 <dansmith> ack
21:14:14 <dansmith> live migration stuff is:
21:14:18 <dansmith> Live migration completion timeout(config part): https://review.openstack.org/#/c/438467/
21:14:22 <dansmith> Per instance API timeout: https://review.openstack.org/#/c/442809/
21:14:51 <dansmith> no moshele and it's late
21:15:00 <mriedem> there isn't an sriov meeting anymore
21:15:04 <mriedem> since january
21:15:07 <mriedem> i should just remove that
21:15:14 <dansmith> mriedem: well, who's fault is that?
21:15:20 <mriedem> mine?
21:15:37 <dansmith> update from gibi on notifications:
21:15:37 <dansmith> We spent most of the meeting time on https://review.openstack.org/#/c/448779 Add BDM to InstancePayload which is needed for Searchlight. It would be good to get some eyes at least on the extra db query part https://review.openstack.org/#/c/448779/9/nova/compute/utils.py@217
21:15:42 <dansmith> https://review.openstack.org/#/c/428260/ Short circuit notifications when not enabled patch finally got the missing test coverage so it is complete now.
21:15:44 <mriedem> oh wait, it's obama's fault
21:15:45 <mriedem> sorry
21:15:51 <dansmith> At some point we need to talk about the soft- and force-delete complication in https://review.openstack.org/#/c/443764/1/nova/compute/api.py@1868
21:16:30 <dansmith> efried: thorst: powervm apparently gets its own section, so .. /
21:16:53 <efried> Right
21:16:55 <mriedem> same patches up for review
21:17:03 <efried> A growing list, yes.
21:17:25 <mriedem> i'd advise not letting the list grow too much
21:17:28 <mriedem> before you get reviews
21:17:40 <efried> Well, yeah, we're kinda catch-22ed there.
21:17:41 <mriedem> half a dozen at most maybe if they aren't small
21:18:10 <efried> Sokay, we're running into some problems with the SSP ephemeral disk driver that we're going to need some time to solve.
21:18:29 <efried> Something (maybe?) changed in threading that freezes the whole compute process for us.
21:18:36 <efried> Haven't nailed it down yet.
21:18:52 <efried> Anyway, that's about it for powervm.  Begging for reviews ;-)
21:19:02 <efried> Will resort to bribery if necessary.
21:19:02 <dansmith> ack.. mriedem wanna do cinder?
21:19:11 <mriedem> dansmith: sure
21:19:13 <dansmith> so we have something other than me pasting stuff in here and impersonating others/
21:19:26 <mriedem> John's spec is merged; review focus is on detach refactor https://review.openstack.org/#/c/439520/
21:19:38 <mriedem> which should be merged tomorrow since johnthetubaguy was +2 on it before
21:19:50 <mriedem> so next week we should be starting on reviewing the actual detach new flow
21:20:03 <mriedem> otherwise i wasn't in the weekly meeting today so...
21:20:13 <mriedem> i'm assuming they just didn't hold it b/c i wasn't there
21:20:21 <mriedem> ildikov: ^?
21:20:22 <dansmith> nauturally
21:20:30 <dansmith> wow, that was .. not right
21:20:47 <mriedem> we can move on
21:20:59 <ildikov> mriedem: jgriffith's patch is now on top of the detach refactor
21:21:01 <dansmith> #topic Stuck reviews
21:21:06 <dansmith> #undo
21:21:07 <openstack> Removing item from minutes: #topic Stuck reviews
21:21:32 <ildikov> mriedem: rebased just now should be in better shape by next week, eyes and suggestions on it are already welcomed, tnx :)
21:22:13 <mriedem> ack
21:22:15 <dansmith> #topic Stuck reviews
21:22:22 <dansmith> I see nothing in here..
21:22:33 * dansmith pauses cautiously
21:23:03 <dansmith> #topic Open Discussion
21:23:13 <dansmith> who has something to discuss openly?
21:23:57 <clarkb> I will also mention that the other issue we have had recently is memory pressure. It might be nice if someone from within nova could poke around and see if nova can go on a cheap diet
21:24:13 <clarkb> (I know python makes this hard, if only we could mark the memory pages as mergeable)
21:24:22 <dansmith> clarkb: wait until I'm running 3x separate conductors for cells
21:24:36 <dansmith> although I'm going to do that on the subnode for the multinode case, which will help things
21:24:46 <dansmith> but the single node one may get a little fatter, depending
21:25:05 <clarkb> heat apparently had good results putting heat on a diet so I figure chances are there are things we can improve elsewhere
21:25:07 <dansmith> #help Infra looking for ways to get nova to hit the gym and eat right to slim down a bit
21:25:48 <clarkb> remember when you could run devstack with 1GB of memory?
21:25:51 * clarkb remembers
21:25:52 <dansmith> heh
21:26:22 <dansmith> okay anything else for open discussion?
21:26:52 <mriedem> end it
21:27:02 <dansmith> #endmeeting