21:04:15 <dansmith> #startmeeting nova 21:04:17 <openstack> Meeting started Thu Sep 25 21:04:15 2014 UTC and is due to finish in 60 minutes. The chair is dansmith. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:04:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:04:20 <openstack> The meeting name has been set to 'nova' 21:04:21 <dansmith> #chair mriedem 21:04:22 <openstack> Current chairs: dansmith mriedem 21:04:28 <mriedem> #link https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting 21:04:30 <mriedem> does that work? 21:04:33 <dansmith> I'm on a cell connection, so I can't run it 21:04:35 <dansmith> mriedem: yep, go 21:04:45 <mriedem> ok, juno rc1 21:04:51 <leifz> o/ 21:04:53 <mriedem> sdague just dumped 3 bugs on the list, those are in the agenda now 21:04:56 <mriedem> oh yeah who's here? 21:05:02 <melwitt> o/ 21:05:07 <alaski> o/ 21:05:09 <dansmith> o/ 21:05:10 <n0ano> o/ 21:05:15 <edleafe> o/ 21:05:38 <mriedem> cool so juno rc1 is supposed to be today 21:05:45 <mriedem> this is the rc1 bug list https://launchpad.net/nova/+milestone/juno-rc1 21:05:48 <mriedem> #link https://launchpad.net/nova/+milestone/juno-rc1 21:06:15 <jogo> o/ 21:06:18 <mriedem> 9 open, 6 of those in progress 21:06:44 <mriedem> sounds like we might have a patch for something similar to https://bugs.launchpad.net/nova/+bug/1373962 21:06:45 <uvirtbot> Launchpad bug 1373962 in nova "LVM backed VM fails to launch" [High,Confirmed] 21:06:55 <mriedem> so if someone can dig that up (ML?) it'd be good 21:07:15 <mriedem> another confirmed is https://bugs.launchpad.net/neutron/+bug/1357476 21:07:19 <uvirtbot> Launchpad bug 1357476 in nova "Timeout waiting for vif plugging callback for instance" [Critical,Confirmed] 21:07:25 <mriedem> dansmith: you improved logging for that, 21:07:28 <mriedem> but not sure if anything has changed 21:07:32 <mriedem> or if we've noticed 21:07:43 <dansmith> ah, yeah, was waiting to get a post-patch hit on that 21:07:45 <mriedem> mikal noted we should untarget from rc1 if it's not being worked, i'd second that 21:08:02 <dansmith> we should try to see if it's still getting hit before we do I think 21:08:10 <mriedem> ok 21:08:12 <dansmith> at least so we know if it is a thing we're capable of looking at 21:08:19 <mriedem> do we have a query? 21:08:19 <dansmith> I'm out tomorrow, otherwise I'd do that tomorrow 21:08:21 * mriedem looks 21:08:24 <dansmith> we must 21:08:37 <mriedem> running it 21:08:47 <dansmith> #link http://lists.openstack.org/pipermail/openstack-dev/2014-September/046741.html 21:08:56 <dansmith> that's the ML post about the LVM issue with config drive 21:08:57 <mriedem> #action see if https://bugs.launchpad.net/neutron/+bug/1357476 is still a problem 21:08:59 <uvirtbot> Launchpad bug 1357476 in nova "Timeout waiting for vif plugging callback for instance" [Critical,Confirmed] 21:09:08 <mriedem> 0 hits in 7 days 21:09:28 <dansmith> heh, well, can't be that big of a deal then 21:09:32 <dansmith> I say untarget 21:09:35 <mriedem> ok i'll untarget 21:09:48 <dansmith> so, on the LVM thing 21:10:07 <dansmith> looks like maybe mikal thought that the only consequence was not being able to have an LVM-formatted configdrive 21:10:19 <mriedem> is this flwang's patch for rbd then? 21:10:22 <dansmith> but I think the bug is about using LVM breaking configdrive incidentally, right? 21:10:38 <dansmith> I don't think so, that's for evacuate and unrelated AFAICT 21:10:41 <mriedem> ok 21:11:35 <mriedem> it's not popping out to me if there is a related patch up for that issue from the ML 21:11:38 <mriedem> anyway, moving on 21:12:05 <mriedem> this is the other rc1 bug https://bugs.launchpad.net/nova/+bug/1323658 21:12:07 <uvirtbot> Launchpad bug 1323658 in nova "Nova resize/restart results in guest ending up in inconsistent state with Neutron" [Critical,Confirmed] 21:12:10 <dansmith> yeah, don't see it 21:12:27 <mriedem> arosen isn't here 21:12:35 <dansmith> so, on that 21:12:44 <dansmith> I worked with him for a few days to try to get to the bottom of it 21:12:55 <dansmith> he actually thinks that it's a sequencing thing in the tempest test, 21:13:02 <dansmith> and not something that actually matters in the real world 21:13:12 <mriedem> ok, still top gate bug 21:13:17 <mriedem> but tempest... 21:13:22 <dansmith> yeah, but not critical for the release as a result 21:13:27 <dansmith> based on what I heard last 21:13:33 <dansmith> so, painful, but not something we need to hold up for 21:13:48 <mriedem> ok, i'll untarget rc1 21:14:28 <mriedem> ok so that leaves the lvm thing for rc1 21:14:36 <mriedem> anything else on rc1? 21:14:40 <dansmith> When mikal pops up, 21:14:50 <dansmith> we should make sure he looks at the LVM thing, since we think he might've broke it 21:15:16 <dansmith> then he can decide if it can wait for rc2 or if it's worth delaying rc1 21:15:27 <mriedem> sure, we have mikal's revert linked into the bug 21:15:48 <mriedem> what's the tag for a new topic? topic? 21:15:59 <dansmith> I think 21:16:04 <mriedem> #topic bugs 21:16:14 <dansmith> like a pro 21:16:19 <mriedem> \o/ 21:16:34 <mriedem> i added 3 bugs here https://wiki.openstack.org/wiki/Meetings/Nova#Weekly_Nova_team_meeting 21:16:52 <mriedem> one is the lvm config drive rc1 bug 21:17:07 <mriedem> one is https://bugs.launchpad.net/nova/+bug/1371677 which is our happy periodic task / resize timeout bug 21:17:09 <uvirtbot> Launchpad bug 1371677 in nova "Race in resource tracker causes 500 response on deleting during verify_resize state " [Critical,Fix committed] 21:17:24 <mriedem> since e-r is backed up to hell we don't know if dansmith's latest pach for that has resolved it 21:17:30 <dansmith> still like five hours away from e-r showing us anything 21:17:54 <mriedem> we suspect we still need some logic in the libvirt driver to check if the volume exists before calling (blocking) blockdev to get the volume size, because danpb said that'll hang if the volume is gone 21:17:57 <mriedem> which is causing the timeout 21:18:07 <dansmith> well, 21:18:09 <mriedem> i didn't get a chance to work on that today 21:18:13 <dansmith> but we'll be doing that without the lock, 21:18:22 <dansmith> so it will only delay the periodic, but not really a probem right? 21:18:32 <mriedem> i'd have to double check 21:18:39 <dansmith> that's my understanding, FWIW 21:18:42 <jogo> Its also odd that we have an RPC CALL (versus cast) in nova-delete in this case 21:18:55 <mriedem> jogo: cyeoh said there were some notes in the code on why that is 21:18:59 <mriedem> i didn't look last night though 21:19:05 <dansmith> yeah, there is some reson 21:19:08 <jogo> mriedem: yeah but I think we can do better 21:19:08 <dansmith> I forget why 21:19:15 <mriedem> cyeoh said some other race 21:19:24 <jogo> dansmith: it has to check the source of the migrate 21:19:30 <jogo> before moving on or something 21:19:37 <mriedem> dansmith: ok, so you moved the lock past this now: 21:19:37 <mriedem> resources = self.driver.get_available_resource(self.nodename) 21:19:39 <jogo> but maybe long term we can move this logic to the conductor so it can be async 21:19:44 <dansmith> mriedem: right 21:19:45 <mriedem> that's what would be eventually hitting the blockdev call 21:19:45 <mriedem> ok 21:19:48 <jogo> from API 21:20:07 <mriedem> alright, so thinking dansmith's fix is good 21:20:13 <mriedem> we'll find out when e-r flushes 21:20:31 <mriedem> the 3rd bug that sdague had listed was https://bugs.launchpad.net/keystone/+bug/1188189 21:20:33 <uvirtbot> Launchpad bug 1188189 in cinder "Some server-side 'SSL' communication fails to check certificates (use of HTTPSConnection)" [High,In progress] 21:20:44 <mriedem> that's an ssl issue, which is really like 5 bugs in nova apparently 21:20:51 <mriedem> sounds like there are separate bugs opened for each 21:21:02 <mriedem> so if people want something to work on, pick those up 21:21:27 <mriedem> #link https://bugs.launchpad.net/keystone/+bug/1188189/comments/54 21:21:31 <mriedem> those are the separate bugs 21:21:55 <mriedem> anything else on bugs? i guess checkout the 'in progress' rc1 bugs for code reviews 21:22:11 <mriedem> since we can't tag rc1 until those are closed out probably 21:22:26 <mriedem> moving on 21:22:32 <mriedem> #topic gate status 21:22:45 <mriedem> #link http://status.openstack.org/elastic-recheck/ 21:22:49 <mriedem> as noted e-r is backed up 21:22:53 <mriedem> so we don't know the true state of the gate right now 21:23:19 <mriedem> i think we've got fixes out for the top nova bugs in there right now though 21:23:32 <mriedem> hell, the wsgi test was deleted :) 21:23:37 <dansmith> heh 21:23:44 <mriedem> jogo: anything on gate? 21:23:54 <mriedem> categorization rate is high http://status.openstack.org/elastic-recheck/data/uncategorized.html 21:23:59 <mriedem> but that will also change when e-r flushes 21:24:21 <jogo> mriedem: http://jogo.github.io/gate/ is looking decent 21:24:24 <mriedem> and looks like https://review.openstack.org/#/c/122291/ was approved, that was a gate issue 21:24:52 <mriedem> moving on 21:24:53 <jogo> mriedem: that page uses graphite so its up to date 21:25:02 <mriedem> ok 21:25:10 <mriedem> yeah things seem to be moving at least 21:25:34 <mriedem> #topic open discussion 21:25:48 <mriedem> agenda says summit topics 21:25:52 <mriedem> #link https://etherpad.openstack.org/p/kilo-nova-summit-topics 21:26:02 <mriedem> like most etherpads it's a mess 21:26:04 <mriedem> :) 21:26:25 <mriedem> anyone have anything on summit topics? 21:26:58 <mriedem> regarding kilo and the bp/spec process, 21:27:19 <mriedem> we approved jogo's https://review.openstack.org/#/c/116699/ and https://review.openstack.org/#/c/112733/ will probably be approved when johnthetubaguy wakes up 21:27:33 <mriedem> still need to work out some of the runway stuff https://review.openstack.org/#/c/115410/ 21:28:00 <mriedem> any other topics? sub-teams? 21:28:10 <dansmith> please no. 21:28:16 <mriedem> end early? 21:28:19 <dansmith> +1 21:28:23 <jogo> +1 21:28:23 <tjones> :-) 21:28:28 <mriedem> #endmeeting