21:05:17 <mriedem> #startmeeting nova
21:05:17 <openstack> Meeting started Thu Jan 21 21:05:17 2016 UTC and is due to finish in 60 minutes.  The chair is mriedem. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:05:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:05:20 <openstack> The meeting name has been set to 'nova'
21:05:28 <ctrath> o/
21:05:31 <edleafe> \o
21:05:33 * Vek waves
21:05:39 <alaski> o/
21:05:44 <cdent> o/
21:05:44 <markus_z> o/
21:05:45 <thomasem> o/!
21:05:45 <rlrossit_> o/
21:05:47 <takashin> o/
21:05:50 <scottda> Hi
21:05:56 <melwitt> how do you like it at o/
21:05:59 <jlvillal> o/
21:06:02 <jroll> ohai
21:06:09 <mriedem> so we're winging it today
21:06:16 <mriedem> #link agenda https://wiki.openstack.org/wiki/Meetings/Nova
21:06:26 <mriedem> #topic release status
21:06:31 <mriedem> Jan 21: Nova non-priority feature freeze
21:06:34 <mriedem> as most people know
21:06:35 <claudiub> o/
21:06:42 <mriedem> today is the day to get non-priority BPs in
21:06:58 <mriedem> i was told this morning that anything that has 2 +2s on it will still get in despite the gate backup
21:07:04 <mriedem> which will probably last over the weekend
21:07:16 <mriedem> for the rest, i'm assuming there will be an exception process/period
21:07:23 <mriedem> Jan 19-21: mitaka-2
21:07:37 <jaypipes> o/
21:07:39 <mriedem> i'm not sure if m-2 happened yet
21:07:44 <mriedem> we were waiting on some fixes to merge
21:07:49 <mriedem> in here https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking
21:07:55 <mriedem> #link mitaka priorities tracking https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking
21:08:21 <mriedem> 1 of 3 in that list are merged
21:08:23 <mriedem> one is approved
21:08:34 <mriedem> the other is not https://review.openstack.org/#/c/265601/
21:08:51 <mriedem> oh wait yeah it is
21:09:09 <mriedem> so i guess m-2 will wait until those are all merged, so probably next week
21:09:20 <mriedem> questions about any of that?
21:09:45 <mriedem> #topic regular reminders
21:09:48 <mriedem> https://etherpad.openstack.org/p/mitaka-nova-priorities-tracking
21:09:50 <mriedem> already posted that once
21:10:20 <mriedem> so for the rest of the release it's going to be focusing on getting the priorities done (as much as possible) and bug fixing/stabilization
21:10:34 <mriedem> #topic bugs
21:10:44 <mriedem> gate status....
21:10:46 <mriedem> so, it's bad
21:10:57 <thomasem> :(
21:10:58 <mriedem> the grenade multinode job was made non-voting
21:11:03 <edleafe> "How bad is it?
21:11:15 <mriedem> which should help, but there was a gate reset so everything was re-queued
21:11:27 * cdent gently slaps edleafe
21:11:34 <mriedem> hence 30+ hour gate times
21:11:43 <anteaya> well that and pip
21:12:02 <anteaya> still waiting on travis for release of virtualenv 14.0.1
21:12:13 <mriedem> yeah, that was a few days ago though
21:12:18 <mriedem> one of many disasters this week
21:12:31 <mriedem> anyway, that's what i know of at least so far, nothing new beyond that
21:12:46 <mriedem> #topic 3rd party ci status
21:12:55 <mriedem> i guess 2 things here
21:13:03 <mriedem> 1. vmware nsx ci was taking 3.5+ days to report on changes
21:13:14 <mriedem> so something snuck in which broke spawn for vcenter, which is not good
21:13:19 <mriedem> i've added that patch to the m-2 list
21:13:26 <mriedem> https://review.openstack.org/#/c/270867/
21:13:50 <mriedem> anyway, i'm hoping tjones and the vmware people can figure out why it's taking nearly 4 days for a reply
21:14:05 <mriedem> 2. hyper-v ci was failing a lot this week
21:14:07 <tjones> yes we are looking at it.  trying to add more resources to it
21:14:19 <mriedem> claudiub: do you know anything about the hyper-v ci status?
21:14:37 <mriedem> taking that as a no
21:14:45 <claudiub> mriedem: hello, yeah, I am investigating it. I've fixed quite a few of them
21:14:46 <mriedem> anyway, it's not gating so it's not blocking anything, just fyi
21:14:53 <mriedem> claudiub: oh hi
21:15:02 <mriedem> claudiub: anything you want to share or just 'we're working on it?'
21:15:12 <claudiub> mriedem: there are still a couple of them, plus, there are some neutron related issues that we have to deal with
21:15:47 <mriedem> ok, i guess let us know if there is anything needed on the nova side
21:15:56 <claudiub> mriedem: plus, there seems to be some tests that are always failing due to some weird circumstances.
21:16:16 <claudiub> mriedem: yeah, there's a mini-patch that would really help me on fixing the hyper-v ci
21:16:20 * claudiub fetching link
21:16:58 <claudiub> this: https://review.openstack.org/#/c/265239/
21:17:13 <claudiub> basically, it keeps the original exception's trace
21:17:20 <mriedem> #note please help review https://review.openstack.org/#/c/265239/
21:17:29 * mriedem assumes that is a valid meeting tag
21:17:41 <mriedem> #help review https://review.openstack.org/#/c/265239/
21:17:48 <anteaya> I think #info is
21:17:54 <anteaya> I don't know about #note
21:17:56 <mriedem> help is too
21:17:58 <claudiub> otherwise, the original trace is lost, losing valuable debugging information.
21:18:07 <mriedem> claudiub: ok, i'll review after the meeting, it's small
21:18:09 <mriedem> shouldn't be a problem
21:18:13 <mriedem> moving on
21:18:16 <mriedem> #topic critical bugs
21:18:24 <mriedem> all i know of is https://bugs.launchpad.net/nova/+bug/1536718
21:18:26 <openstack> Launchpad bug 1536718 in OpenStack Compute (nova) "VMware: unable to boot an instance" [High,In progress] - Assigned to Gary Kotton (garyk)
21:18:27 <mriedem> which the change is approved
21:18:36 <mriedem> anyone else have any critical bugs?
21:18:39 <markus_z> the reno one is also done?
21:18:47 <mriedem> markus_z: link?
21:18:56 <markus_z> mriedem, https://bugs.launchpad.net/nova/+bug/1534613
21:18:57 <openstack> Launchpad bug 1534613 in OpenStack Compute (nova) "warnings in stable/liberty release notes are breaking releasenotes job in master" [Critical,In progress] - Assigned to Matt Riedemann (mriedem)
21:19:06 <mriedem> markus_z: that was fixed earlier in the week
21:19:15 <mriedem> dhellmann graciously held my hand through that one
21:19:23 <markus_z> ok, cool, it wasn't totally clear to me
21:19:26 <mriedem> #info dhellmann is the man
21:19:31 <mriedem> anteaya: ^ there
21:19:45 <anteaya> :)
21:19:53 <mriedem> markus_z: yeah https://review.openstack.org/#/c/268669/ closes it
21:19:58 <anteaya> yes #help and #info are commands, #note is not
21:20:30 <edleafe> #note '#note' is not a command
21:20:30 <markus_z> ack
21:20:32 <mriedem> #topic bug reminders
21:20:35 <mriedem> volunteers for 1 week of bug skimming duty? Weekly_bug_skimming_duty
21:20:38 <anteaya> and dhellman is the man
21:21:02 <markus_z> #link https://wiki.openstack.org/wiki/Nova/BugTriage#Weekly_bug_skimming_duty
21:21:03 <mriedem> we need a bug skimmer for next week if anyone wants to volunteer
21:21:21 <mriedem> doesn't have to happen here
21:21:30 <markus_z> yeah, there is a ML post too
21:21:32 <mriedem> Nova bugs team IRC meeting will start at February the 9th http://lists.openstack.org/pipermail/openstack-dev/2016-January/084543.html
21:21:43 <mriedem> #info Nova bugs team IRC meeting will start at February the 9th http://lists.openstack.org/pipermail/openstack-dev/2016-January/084543.html
21:21:44 <markus_z> just as a heads up
21:21:56 <mriedem> #topic stable branch status
21:22:05 <mriedem> we released liberty 12.0.1 earlier in the week
21:22:14 <mriedem> to much fanfair
21:22:36 <mriedem> also, tonyb is now nova stable core, so everyone can bug him now
21:22:47 <anteaya> yay tonyb
21:22:50 <markus_z> hooray for tonyb :)
21:22:52 <jaypipes> if we can understand his funny words.
21:22:55 <claudiub> congrats!
21:23:02 <Vek> heh
21:23:10 <mriedem> stable/kilo will release soonish, i'm told
21:23:12 <jroll> you don't need to understand his words to ping him incessantly :)
21:23:20 <mriedem> Daviey is running the stable/kilo release
21:23:39 <mriedem> once stable/kilo is released, we'll need to scrub through https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/kilo,n,z and filter out anything that's not a security or critical bug fix
21:23:52 <mrda> tonyb: does sound funny, even to Australians
21:24:01 <Vek> haha :)
21:24:13 <mriedem> #topic stuck reviews
21:24:18 <mriedem> there was nothing on the agenda
21:24:26 <mriedem> so i take it we're all in agreement about all things
21:24:41 <mriedem> except cinder multiattach, which we'll discuss at the midcycle
21:24:53 <mriedem> anyone else for stuck reviews?
21:25:00 <vladikr> mriedem, I have a libosinfo series : https://review.openstack.org/#/q/topic:bp/libvirt-hardware-policy-from-libosinfo
21:25:13 <vladikr> mriedem, it has danpbs +2
21:25:14 <mriedem> vladikr: stuck does not mean not getting reviewed
21:25:20 <vladikr> ok :)
21:25:22 <mriedem> so please don't solicit for reviews here
21:25:26 <Vek> I've sent a couple of novaclient reviews over to the -dev list, but I haven't seen anyone respond to those emails...
21:25:43 <thomasem> mriedem: Sorry for my ignorance... how would you define a stuck review?
21:25:58 <mriedem> thomasem: the same way the meeting agenda does
21:25:59 <mriedem> "Please note "stuck review" means a review where there is some disagreement that needs resolving. Its not for reviews that just haven't had attention, except for exceptional cases. Where you see "?" feel free to just edit the wiki and add your item."
21:26:06 <Vek> thomasem: a review where two reviewers are pulling in opposite directions, I think...
21:26:21 <thomasem> Oh, crud. Sorry. Didn't notice that on the page. :(
21:26:29 <mriedem> :)
21:26:34 <mriedem> walked right into that trap
21:26:37 <thomasem> Okay, yeah, not having that problem with my review.
21:26:39 <thomasem> LOL right
21:26:57 <thomasem> Well, fyi for anyone else who didn't catch it so they don't look stupid like me ^^
21:26:58 <mriedem> vladikr: ok, i guess you still technically have the rest of the day on those for non-priority bp feature freeze
21:27:31 <vladikr> mriedem, ok
21:27:43 <mriedem> #topic open discussion
21:27:57 <mriedem> midcycle details
21:28:02 <mriedem> #link midcycle wiki https://wiki.openstack.org/wiki/Sprints/NovaMitakaSprint
21:28:10 <mriedem> #link midcycle topics https://etherpad.openstack.org/p/mitaka-nova-midcycle
21:28:17 <mriedem> if you have topics for the midcycle ^ is where they go
21:28:36 <mriedem> i think wed we have some time with the cinder midcycle
21:28:38 <mriedem> over hangout
21:28:47 <mriedem> wed afternoon in bristol, wed morning in rtp
21:28:52 <mriedem> it should be fun
21:29:01 <mriedem> anything else about the midcycle?
21:29:02 <scottda> mriedem: yup, that's the plan
21:29:15 <ildikov> +1
21:29:20 <anteaya> neutron items seem to be in place
21:29:29 <mriedem> i'm going to skip the TC review for containers in the compute mission that was in the agenda
21:29:38 <mriedem> (thomasem) Libvirt/LXC experimental gate test progress
21:29:46 <mriedem> ^ is old too i think
21:29:48 <edleafe> any chance of listening in for those of us who can't make it to the UK?
21:29:55 <thomasem> So, yes
21:30:05 <mriedem> edleafe: maybe?
21:30:06 <mriedem> probably
21:30:12 <mriedem> edleafe: talk to PaulMurray
21:30:13 <thomasem> Got a bug filed upstream. Found it's an issue with the kernel being used in gate
21:30:22 <edleafe> mriedem: will do
21:30:23 <mriedem> thomasem: nice
21:30:30 <thomasem> the newest Trusty kernel has issues with Libvirt's virDomainShutdown
21:30:40 <mriedem> and releasing things?
21:30:43 <thomasem> While upstream 3.13.x does not
21:31:04 <anteaya> I think .58 was what you said worked for you
21:31:07 <mriedem> thomasem: ok, so that probably affects all tests in the job since they rely on instances to be deleted
21:31:10 <thomasem> So, we have this patch https://review.openstack.org/#/c/267724/7 up to skip those tests while the bug is being worked. And here's the bug: https://bugs.launchpad.net/ubuntu/+source/libvirt/+bug/1536280
21:31:12 <openstack> Launchpad bug 1536280 in linux (Ubuntu) "domain shutdown fails for libvirt/lxc" [Undecided,Confirmed]
21:31:18 <anteaya> where as we have .76 everywhere do to last tuesday's cve
21:31:19 <thomasem> mriedem: It's only with Libvirt/LXC
21:31:34 <mriedem> thomasem: yeah, but if you can't delete an instance you eventually go over quota
21:31:38 <mriedem> or is this just stop?
21:31:41 <thomasem> It's only the way libvirt/LXC does the shutdown operation and how that interacts with the kernel that appears to be an issue.
21:31:54 <thomasem> mriedem: just stop
21:31:56 <thomasem> you can still delete it
21:32:04 <thomasem> so virsh -c lxc:/// destroy still works
21:32:09 <mriedem> ah, ok
21:32:09 <thomasem> it's just virsh -c lxc:/// stop <instance>
21:32:11 <thomasem> that fails
21:32:11 <thomasem> yep
21:32:32 <mriedem> we should probably put a note in our hypervisor support matrix for that
21:32:35 <thomasem> So, filed that bug, skipping the tests in that patch. Now gates just being painful and having resource issues on the slave it looks like... not sure where to go from there.
21:32:47 <mriedem> i'll review the rc change at some point
21:33:10 <thomasem> That'd be very helpful. Thank you. I don't mind adding a note to the hypervisor matrix. Just need to find it. I haven't looked at it in a long, long time.
21:33:11 <thomasem> :)
21:33:23 <mriedem> thomasem: thanks for the work on this, it's cool to see it progressing
21:33:30 <mriedem> (tpatzig) Flavor with volume option (https://blueprints.launchpad.net/nova/+spec/flavor-with-volume)
21:33:39 <mriedem> i'm assuming ^ is old b/c tpatzig isn't here
21:33:47 <thomasem> mriedem: You bet! Thanks for the support. :)
21:33:50 <alaski> I don't think that's old
21:33:52 <mriedem> it's also a spec and we're long past spec freeze
21:33:56 <alaski> yeah
21:34:08 <alaski> I've commented on it
21:34:35 <mriedem> - AL
21:34:39 <mriedem> :)
21:34:46 <mriedem> MLB AL has taken a stance
21:34:56 <alaski> heh
21:35:18 <mriedem> (sdague) Unit tests with constraints - https://review.openstack.org/#/c/267096/
21:35:20 <mriedem> ^ is old
21:36:01 <mriedem> it looks like the 2 dependent changes merged for that
21:36:07 <mriedem> so maybe sdague will resurrect it next week
21:36:26 <mriedem> jenkins hates it right now though
21:36:36 <mriedem> (sdague) What are acceptable constraints on project_id in url - http://lists.openstack.org/pipermail/openstack-dev/2016-January/083976.html
21:36:39 <mriedem> i think ^ is old?
21:36:45 <Vek> jenkins hates just about everything right now, I think...
21:36:51 <anteaya> yeah :(
21:37:03 <anteaya> jenkins has had better weeks
21:37:09 <mriedem> i think that ML was talked about last week and we had the change for a configurable project_id
21:37:14 <mriedem> alaski: ^ right?
21:37:27 * Vek thinks it heard about me threatening to rewrite jenkins from scratch and is quaking in its boots...
21:37:31 <alaski> we did discuss it
21:37:33 <alaski> I
21:37:38 <alaski> I'm not sure if a change went up
21:37:39 <mriedem> ok, and i'm pretty sure the change landed
21:37:45 <alaski> okay
21:37:56 <mriedem> yeah, i remember we nit picked all of sean's spelling issues
21:37:59 <mriedem> it was good fun
21:38:03 <Vek> haha :)
21:38:10 <mriedem> and finally
21:38:11 <mriedem> (thomasem, dimtruck) Update on Libvirt/LXC gate test progress
21:38:16 <mriedem> which is a duplicate
21:38:18 <mriedem> so ignoring
21:38:23 <mriedem> anything else for open discussion?
21:38:38 <mriedem> i'll clean up the meeting agenda after this
21:38:41 <mriedem> oh, also
21:38:46 <mriedem> #info no nova meeting the week of 1/21
21:38:49 <mriedem> oops
21:38:50 <mriedem> 1/24
21:38:53 <mriedem> #undo
21:38:54 <openstack> Removing item from minutes: <ircmeeting.items.Info object at 0x982ef90>
21:39:03 <mriedem> #info no nova meeting on 1/28
21:39:20 <mriedem> ok, i don't hear anything else, will shut this down, thanks all
21:39:21 <claudiub> question: can we still merge teeny patches related to os-brick in nova after m-2?
21:39:31 <mriedem> claudiub: bug fixes?
21:39:43 <mriedem> yes to bug fixes
21:39:51 <mriedem> anything else will have to be evaluated
21:40:01 <claudiub> more like, using os-brick for volume related operations in hyper-v? :D
21:40:08 <mriedem> that's a bp
21:40:12 <mriedem> at least in my mind
21:40:16 <mriedem> it doesn't sound teeny
21:40:23 <alaski> agreed
21:40:24 <claudiub> it would be < 100 loc, IMO
21:40:35 <claudiub> added loc
21:40:36 <mriedem> it's still a new dependency after feature freeze and not trivial
21:40:37 <alaski> it's a functional change
21:40:44 <mriedem> *new dep for hyper-v
21:41:06 <mriedem> so i think newton
21:41:18 <mriedem> anyone for anything else?
21:41:19 <claudiub> meh, ok
21:41:31 <mriedem> alright, thanks all, ttyl
21:41:33 <mriedem> #endmeeting