21:00:54 <mriedem> #startmeeting nova 21:00:55 <openstack> Meeting started Thu Aug 18 21:00:54 2016 UTC and is due to finish in 60 minutes. The chair is mriedem. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:00:56 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:00:58 <openstack> The meeting name has been set to 'nova' 21:01:06 <alaski> o/ 21:01:06 <edleafe> \o 21:01:07 <takashin> o/ 21:01:15 <mikal> Hi 21:01:17 <scottda> hi 21:01:19 <melwitt> o/ 21:01:30 <cdent> o/ 21:01:38 <lbeliveau> o/ 21:01:40 * Vek waves 21:01:50 <rlrossit> o/ 21:02:04 <raj_singh> o/ 21:02:05 <mriedem> let's start shall we 21:02:06 <jroll> \o 21:02:07 <macsz> \o 21:02:12 <mriedem> #link agenda https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting 21:02:18 <mriedem> #topic release news 21:02:25 <mriedem> #link Newton release schedule: https://wiki.openstack.org/wiki/Nova/Newton_Release_Schedule 21:02:33 <mriedem> #info Aug 22-26: final non-client library release (oslo) - so if you need things released from dependent libraries get them done soon 21:02:43 <mriedem> ^ is next week 21:03:07 <mriedem> we're 2 weeks from FF 21:03:18 <mriedem> and tagging n-3 21:03:30 <mriedem> any questions? 21:03:44 <mriedem> ok moving on 21:03:46 <mriedem> #topic bugs 21:04:06 <mriedem> the gate... 21:04:13 <mriedem> there were a few issues earlier in the week 21:04:24 <mriedem> http://status.openstack.org//elastic-recheck/index.html 21:04:31 <mriedem> i think the cells upgrade stuff was all reverted 21:04:51 <mriedem> looks like the image test timeouts have dropped http://status.openstack.org//elastic-recheck/index.html#1613749 21:05:11 <alaski> yes, the cells things were reverted 21:05:18 <mriedem> so it looks like we should be in decent shape as far as CI jobs queue, it's just a long haul to get merged 21:05:40 <mriedem> related to that: https://review.openstack.org/#/c/357444/ 21:05:49 <mriedem> ^ proposes to remove the postgresql job from the integrated gate 21:05:53 <mriedem> there is a thread in the dev list about that also 21:06:27 <mriedem> on the topic of the gate, i've proposed that we make the live migration job voting https://review.openstack.org/#/c/356043/ 21:06:39 <mriedem> the question is if it should be check queue only, since it doesn't run on any other project's gate 21:06:52 <mriedem> so running it in our gate queue might be kind of pointless, but i'm not an expert on that 21:06:59 <jroll> infra sort of has a policy that anything voting in check needs to run in gate 21:07:12 <jroll> there's things out there that violate it, but they've been enforcing it more lately 21:07:31 <mriedem> yeah i'm experiencing that here https://review.openstack.org/#/c/354402/ 21:07:44 <mriedem> which proposes to run the neutron LB job in nova check as voting but not gate 21:07:58 <mriedem> b/c we regressed that and it broke neutron for a day 21:08:04 <mriedem> so we should probably co-gate on that one 21:08:22 <mriedem> it's conditional on the files that changed though, so won't run all of hte time 21:08:23 <jroll> I think it's a decent policy, but it does seem like it could be a bit wasteful of gate nodes 21:08:29 <woodster_> o/ 21:08:35 <mriedem> jroll: yup, that's why i proposed check only for that one 21:09:05 <mriedem> as for 3rd party ci status 21:09:12 <mriedem> hyper-v was regressed recently by a fix i made 21:09:13 <mriedem> fixed here https://review.openstack.org/#/c/352837/ 21:09:37 <mriedem> i haven't personally noticed any other major 3rd party CI issues this week 21:09:59 <mriedem> and one more for gating, https://review.openstack.org/#/c/355148/ 21:10:06 <mriedem> ^ proposes a dsvm neutron job for novaclient changes 21:10:12 <mriedem> s/dsvm/functional./ 21:10:22 <mriedem> since we have novaclient functional tests working with a neutron backend now too 21:10:39 <mriedem> we kind of need that given devstack defaults to neutron now 21:11:12 <mriedem> i'm not aware of any critical bugs 21:11:24 <mriedem> https://bugs.launchpad.net/nova/+bug/1608601 was mentioned in -nova today as a possible regression from mitaka 21:11:24 <openstack> Launchpad bug 1608601 in OpenStack Compute (nova) "create_port is failing for non admin context as Nova neutronclient interactions are not using admin context consistently" [Undecided,New] 21:11:28 <mriedem> i haven't had the time to dig in though 21:11:34 <mriedem> #help triage https://bugs.launchpad.net/nova/+bug/1608601 21:11:50 <mriedem> #topic reminders 21:11:56 <mriedem> #info mriedem is out next week (8/22), dansmith is the backup 21:12:06 <mriedem> #info Stop proposing centralize / cleanup config option changes until at least after feature freeze on 9/2. 21:12:17 <mriedem> i've gone through and -2ed anything in ^ that doesn't have a +2 already 21:12:37 <mriedem> #link Newton review focus list: https://etherpad.openstack.org/p/newton-nova-priorities-tracking 21:12:56 <mriedem> ^ should really be placement API, cells v2 and libvirt imagebackend at this point 21:13:11 <mriedem> #help https://wiki.openstack.org/wiki/Nova/BugTriage#Weekly_bug_skimming_duty Volunteers for 1 week of bug skimming duty? 21:13:37 <mriedem> #topic Stable branch status: https://etherpad.openstack.org/p/stable-tracker 21:13:45 <mriedem> stable branches are looking ok 21:13:59 <mriedem> #topic subteam highlights 21:14:04 <mriedem> alaski: cellsv2 - go! 21:14:21 <alaski> we discussed the grenade/devstack changes and what they broke 21:14:29 <alaski> which have now been reverted 21:14:39 <alaski> but dansmith has patches up which appear to work properly 21:15:03 <alaski> and as always there are reviews to do 21:15:13 <alaski> I'll update the priority etherpad, but I didn't think anyone used it anymore 21:15:25 <alaski> and we're going to bi-weekly meetings for a bit 21:15:28 <mriedem> alaski: are there patches up for listing instances in the API now which pull from instances and build requests? 21:15:31 <melwitt> I cleared out merged stuff from the etherpad a little while ago 21:15:33 <alaski> until we pick up for next cycle 21:15:38 <alaski> melwitt: awesome 21:15:51 <alaski> mriedem: there are 21:15:55 <mriedem> sweet 21:16:01 <melwitt> but do add other patches that aren't on there yet, like dansmith's 21:16:30 <mriedem> ok anything else? 21:16:31 <alaski> melwitt: cool, will do 21:16:33 <alaski> mriedem: that's it 21:16:36 <mriedem> thanks 21:16:43 <mriedem> edleafe: scheduler meeting? 21:16:47 <edleafe> Discussed Yinxin's spec on adding qualitative values in the request_spec object. For now they will be requirements only, due to the requirement of one flavor per permuation, and adding optional preferences would be unworkable. 21:16:51 <edleafe> https://memegenerator.net/instance/71130527 21:16:58 <edleafe> Realized that our stuff in the priorities etherpad was horribly out-of-date, so I updated it. 21:17:01 <edleafe> Decided to call these qualitative things "traits" instead of "capabilities". Was waiting for the bikeshedding fest, but for some reason we all agreed. 21:17:06 <edleafe> Focus now is on cdent's placement API stuff. 21:17:16 <edleafe> And the remaining resource provider patches 21:17:18 <edleafe> that's it 21:17:45 <mriedem> ok thanks 21:18:05 <mriedem> cdent: edleafe: are the placement api/rt changes looking OK, no blockers? 21:18:10 <mriedem> just needing review? 21:18:17 <cdent> situation is under control 21:18:38 <mriedem> ok cool, just wanting to avoid late surprises like we did this week with novaclient and 2.36 21:19:11 <cdent> mriedem: the resource tracker stuff is vague at the moment, so that's one to watch, but ought to be straightforward 21:19:18 <mriedem> ok 21:19:32 <mriedem> live migration meeting happened this week, i didn't pay full attention, and tdurakov wasn't there 21:19:45 <mriedem> the one take away i had was that live migration voting patch 21:20:25 <mriedem> sdague: anything to say about the api meeting? 21:20:45 <mriedem> mostly talked about the user_id policy changes 21:21:08 <mriedem> https://review.openstack.org/#/q/topic:bp/user_id_based_policy_enforcement 21:21:33 <mriedem> looks like 2 patches left on that and then it's done 21:21:58 <mriedem> i don't think there was an sriov/pci meeting this week 21:22:06 <lbeliveau> nope 21:22:14 <mriedem> #topic stuck reviews 21:22:23 <lbeliveau> but we have two stuck reviews 21:22:23 <mriedem> nothing on the agenda, did anyone have anything? 21:22:59 <lbeliveau> got +2 from jaypipes, need another one ... https://review.openstack.org/#/c/349060/ and https://review.openstack.org/#/c/347444/ 21:23:01 <mriedem> stuck as in gridlocked or stuck as in no reviews? 21:23:06 <mriedem> ok, that's not really stuck :) 21:23:22 <lbeliveau> ok, then waiting for core reviews :) 21:24:03 <mriedem> #topic open discussion 21:24:19 <mriedem> (markus_z) There are some bug reports emerging regarding "image cache" + "ephemeral disks" + "resize" which could need some attention: 21:25:12 <rlrossit> mriedem's laptop became frozen 21:25:21 <rlrossit> we're having some technical difficulties here 21:25:48 <edleafe> rlrossit: little early for freezes in MN, huh? 21:26:10 <mriedem> i'm back 21:26:14 <mriedem> how embarrassing 21:26:17 <mriedem> https://bugs.launchpad.net/nova/+bug/1608934 ephemeral disk creation fails for local storage with image type raw/lvm 21:26:17 <openstack> Launchpad bug 1608934 in OpenStack Compute (nova) "ephemeral disk creation fails for local storage with image type raw/lvm" [Undecided,In progress] - Assigned to Dr. Jens Rosenboom (j-rosenboom-j) 21:26:22 <mriedem> https://bugs.launchpad.net/nova/+bug/1610015 Creating an instance using images_type=lvm fails on ephemeral volume creation 21:26:22 <openstack> Launchpad bug 1610015 in OpenStack Compute (nova) "Creating an instance using images_type=lvm fails on ephemeral volume creation" [Undecided,New] 21:26:28 <mriedem> https://bugs.launchpad.net/nova/+bug/1558880 instance can not resize ephemeral in mitaka 21:26:28 <openstack> Launchpad bug 1558880 in OpenStack Compute (nova) "instance can not resize ephemeral in mitaka" [High,In progress] - Assigned to Tina Kevin (song-ruixia) 21:26:32 <mriedem> https://bugs.launchpad.net/nova/+bug/1605720 backing store missing for ephemeral disk on cold migration 21:26:33 <openstack> Launchpad bug 1605720 in OpenStack Compute (nova) "backing store missing for ephemeral disk on cold migration" [Undecided,In progress] - Assigned to Chris Friesen (cbf123) 21:26:36 <mriedem> fin 21:26:38 <rlrossit> https://kimfalconer.files.wordpress.com/2012/07/technical_difficulties.jpg 21:26:51 <mriedem> so, i don't know if these are regressions, or we've just always sucked with ephemeral BDMs 21:27:04 <mriedem> i assume mdbooth might know 21:27:28 <mriedem> as this in his wheelhouse right now 21:27:51 <mriedem> anything else for open discussion? 21:28:21 <mriedem> i'll take that as everyone is sleeping 21:28:24 <mriedem> thanks all 21:28:28 <mriedem> #endmeeting