14:00:06 <sdague> #startmeeting nova
14:00:06 <edleafe> \o
14:00:06 <openstack> Meeting started Thu Sep 22 14:00:06 2016 UTC and is due to finish in 60 minutes.  The chair is sdague. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:07 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:09 <openstack> The meeting name has been set to 'nova'
14:00:20 <alex_xu> o/
14:00:23 <takashin> o/
14:00:24 <BobBall> o/
14:00:27 <raj_singh> o/
14:00:29 <scottda> hi
14:00:30 <sdague> you are getting a non regularly scheduled chair today
14:00:38 <dansmith> o/
14:00:39 <sdague> #link https://wiki.openstack.org/wiki/Meetings/Nova
14:00:39 <bauzas> yet again \o
14:00:44 <sdague> agenda for today
14:00:47 <gibi> o/
14:01:04 <sdague> #topic Release News
14:01:13 <PaulMurray> o/
14:01:33 <johnthetubaguy> o/
14:01:46 <sdague> we're in rc phase now, as was just said in the nova channel, there will be an rc2 next week
14:01:52 <sdague> just waiting for translation imports
14:02:09 <sdague> and any critical bugs found could be nominated for inclusion
14:02:31 <sdague> the ocata proposed schedule is also listed - https://wiki.openstack.org/wiki/Nova/Ocata_Release_Schedule
14:02:36 <alaski> o/
14:02:45 <dane-fichter> o/ hi folks
14:02:48 <sdague> any other Release items from folks?
14:02:51 <sdague> or questions?
14:03:11 <sdague> #topic Bugs (stuck/critical)
14:03:37 <sdague> rc potential bug list -  https://etherpad.openstack.org/p/nova-newton-rc-potential
14:03:50 <sdague> currently, nothing is tagged for rc2 that is outstanding, which is good
14:04:07 <sdague> if you hit something, please tag it with newton-rc-potential
14:04:29 <sdague> gate's been pretty good
14:04:40 <sdague> #topic Summit planning
14:04:48 <sdague> #link Etherpad for Ocata summit topics: https://etherpad.openstack.org/p/ocata-nova-summit-ideas
14:04:55 <sdague> #info Cross project summit session ideas are due by 10/1: http://lists.openstack.org/pipermail/openstack-dev/2016-September/103657.html
14:05:19 <sdague> feedback for both of those welcome in their respective locations
14:05:36 <sdague> #topic Reminders
14:05:43 <sdague> #help https://wiki.openstack.org/wiki/Nova/BugTriage#Weekly_bug_skimming_duty Volunteers for 1 week of bug skimming duty?
14:05:56 <sdague> we're definitely in need of more bug skim volunteers
14:06:17 <sdague> it's a good way to get a feel of the incoming bugs to nova and see lots of bits of the project
14:06:44 <sdague> so, if you are looking to help out, please consider doing a rotation
14:06:54 <sdague> #topic Stable branch status
14:07:04 <sdague> #link https://etherpad.openstack.org/p/stable-tracker
14:07:23 <sdague> the biggest thing is the liberty qemu sec bug finally got resolved and landed
14:07:33 <kashyap> Yay!
14:07:38 <sdague> #topic Subteam Highlights
14:07:45 <jaypipes> o/ sorry for being late
14:08:02 <sdague> skipping cells team per request
14:08:11 <sdague> jaypipes: just in time, scheduler team highlights?
14:08:19 <edleafe> The discussion was mostly about how to better balance the workloads in Ocata
14:08:37 <edleafe> IOW, avoiding the bottlenecks we had in Newton and prior
14:08:46 <jaypipes> yeah, I think the retrospective is going to be... interesting. :)
14:09:32 <sdague> yes, reminder please consider providing input to the retrospective
14:09:35 <sdague> #link https://etherpad.openstack.org/p/nova-newton-retrospective
14:09:54 <sdague> ok, Live Migration team? PaulMurray ?
14:10:00 <jaypipes> sdague: yeah, about halfway through reading it. will try to add some comments today if I find something that is not already covered in depth
14:10:01 <PaulMurray> nope - nothing
14:10:11 <PaulMurray> nothing special anyway
14:10:29 <sdague> #info Scheduler team highlights - discussing how to better balance workloads in Ocata
14:10:39 <sdague> #info Live Migration team highlights - nothing special
14:10:58 <sdague> API subteam
14:11:34 <sdague> we're mostly doing unwind of the missed network api deprecations, as well as trying to make security groups consistent with neutron
14:12:02 <sdague> as the model is very different than in n-net, and our current API actually maps poorly to how they work in neutron
14:12:11 <sdague> alex_xu: anything else you want to add?
14:12:31 <alex_xu> sdague: no more from me, that is all thing about also
14:12:50 <sdague> #info API team highlights - continuing with network api deprecations, as well as trying to make security group interaction more sane with neutron backend
14:13:01 <sdague> SR-IOV/PCI - moshe?
14:13:06 <sdague> SR-IOV/PCI - moshele?
14:13:18 <sdague> Notification - gibi ?
14:13:22 <gibi> hi
14:13:27 <jaypipes> sdague: the Moshe VF is currently consumed by another instance.
14:13:28 <gibi> the ocata bp for the notification transformation has been approved
14:13:46 <gibi> so we are ramping up the activity in that area again
14:13:54 <sdague> #info SR-IOV/PCI team highlights - no one here
14:13:55 <gibi> I notified those who had patches which missed Newton to restore them
14:14:07 <gibi> The wiki is up to date about what to transform
14:14:14 <gibi> https://wiki.openstack.org/wiki/Nova/VersionedNotificationTransformation
14:14:19 <jaypipes> gibi: would be awesome to have a sdague-like checklist for notifications...
14:14:25 <jaypipes> gibi: think you can do that?
14:14:41 <sdague> #info Notification team highlights - ocata bp approved, work started in earnest to get code merging again
14:14:45 <gibi> jaypipes: yeah, I can do that, could you point me to the sdague-like checklist ?
14:14:59 <sdague> gibi: hit me up after the meeting
14:14:59 <gibi> sdague: thats all from the subteam
14:15:06 <gibi> sdague: ok
14:15:09 <sdague> ok, nothing on stuck reviews
14:15:17 <sdague> #topic Open discussion
14:15:33 <sdague> dave-johnston ?
14:15:48 <sdague> don't see him, on, moving on
14:16:00 <sdague> gibi: you are up next with - Does this bp needs a spec? https://blueprints.launchpad.net/nova/+spec/json-schema-for-versioned-notifications
14:16:07 <gibi> that is me again
14:16:14 <gibi> the main goal of this bp is to provide json schema for the versioned notification
14:16:21 <gibi> payload based on the work that is pretty much finished in ovo
14:16:28 <woodster_> o/
14:16:30 <gibi> the ovo feature makes it possible to generate json schema for any VersionedObject.
14:16:38 <gibi> so we would just build on that
14:17:22 <gibi> what do you think?
14:17:27 <sdague> it seems like we probably still want a basic spec to keep an eye on testing and the like, but I would think it would be pretty quick to approve
14:17:47 <sdague> dansmith / alaski - opinions?
14:18:06 <dansmith> not just a specless bp?
14:18:08 <alaski> agreed. most of the info is in the bp already, but it would be nice to have the spec
14:18:21 <dansmith> meh
14:18:35 <mriedem> so this sends a schema with the versioned notification?
14:18:36 <mriedem> is that the change?
14:18:52 <gibi> it does not send the schema on the message bus
14:18:59 <alaski> looks like the Nova side of this is a new CLI tool to output the schemas
14:19:01 <gibi> it just give a way to  generate the schema ofline
14:19:01 <mriedem> so write a tool to generate the schemas?
14:19:08 <gibi> mriedem: yes
14:19:14 <mriedem> ok, so i'm a consumer and run the tool to get the schemas,
14:19:21 <mriedem> and then compare my incoming notifications against those
14:19:26 <gibi> yes
14:19:37 <mriedem> if it's just a cli tool, i think specless is fine
14:19:42 <sdague> ok
14:20:02 <gibi> it is just a cli tool right now
14:21:11 <sdague> ok, I guess we're saying fine with specless bp
14:21:23 <gibi> thanks
14:21:43 <sdague> panbalag - Tempest plugin for Nova (https://review.openstack.org/#/c/366091/) - a place holder for nova tests that work with internals like libvirt XML, database, etc.,
14:22:13 <panbalag> The goal of this patch is to start a repo where we can place integration tests that run against an active deployment (devstack). I would like to get suggestions/comments on this effort.
14:22:45 <sdague> panbalag: there isn't much to evaluate in that patch, and the last time I looked at the repo it was pretty empty
14:23:26 <sdague> personally, it seems like a discussion on the mailing list might be appropriate, especially as a related effort was pushed out by mirantis qa folks to ML for discussion as well
14:23:52 <panbalag> So do you want to have sample tests to see what kind of scripts would go in eventually?
14:24:23 <sdague> explain the grand vision, list some examples, and what the setup would look like
14:24:58 <sdague> because based on #openstack-nova conversation yesterday, there seemed to be some confusion about why this was going to be tempest plugins
14:26:16 <sdague> ok, moving on
14:26:21 <sdague> (BobBall): Spec-less BP: VIF hotplug for XenAPI https://blueprints.launchpad.net/nova/+spec/xenapi-vif-hotplug
14:26:26 <panbalag> as I had replied in the patch, we wanted to use tempest infrastructure
14:26:32 <BobBall> We've got two feature parity changes we'd like as specless BPs - but only one has a BP so far
14:26:39 <BobBall> XenAPI VIF hotplug: https://blueprints.launchpad.net/nova/+spec/xenapi-vif-hotplug (Code at https://review.openstack.org/#/c/365505/)
14:26:53 <sdague> panbalag: ok, how about we take this out to the ML, as suggested, and we can discuss there
14:26:55 <panbalag> ok. I'll probably start the discussion on the mailing list.
14:27:15 <sdague> BobBall: right, feature parity, I think that's pretty clearly specless BP
14:27:37 <BobBall> Awesome - and the other one I hope is the same, but no BP yet
14:27:41 <BobBall> Device tagging: Following on from https://blueprints.launchpad.net/nova/+spec/virt-device-role-tagging - Needs new BP (Code at https://review.openstack.org/#/c/333781/)
14:28:16 <sdague> cool, related to xen, we still have the outstanding issue that the xenplugins don't even get syntax checked
14:28:22 <sdague> is that on the xen team radar
14:28:40 <sdague> especially as someone was trying to make them python3 the other day (and not realizing they live in a python 2.4 env)
14:28:42 <BobBall> Yes - that's fixed in https://review.openstack.org/#/c/289431/ which is pending review
14:29:00 <sdague> BobBall: ok, great
14:29:21 <sdague> BobBall: anything else?
14:29:33 <sdague> (dane-fichter): Image signing issues. Certificate validation (https://review.openstack.org/#/c/357151/), Cursive integration (https://review.openstack.org/#/c/351232/)
14:29:33 <sdague> 
14:29:37 <BobBall> Absolutely.  But not for this meeting :) We can talk about it next meeting :)
14:29:45 <dane-fichter> hey all
14:30:18 <dane-fichter> So after talking w mriedem yesterday, look like wqe're going to try to create a gate job for testing image signature verification
14:30:32 <sdague> dane-fichter: great
14:30:38 <dane-fichter> So I'm looking for people who want to help / have opinions on that
14:31:06 <dane-fichter> I'd also like some feedback / discussion on this: https://review.openstack.org/#/c/357151/
14:31:22 <sdague> #help looking for people to help dane-fichter on gate job setup that does fullstack verification with image signature verification
14:31:34 <dane-fichter> Since it's pretty necessary for ensuring that the feature actually accomplishes what it's supposed to do
14:31:55 <sdague> #link https://review.openstack.org/#/c/357151/5/specs/ocata/approved/nova-certificate-validation.rst - spec for cert validation
14:32:07 <dane-fichter> thanks sdague
14:32:22 <sdague> ok, I think that's the whole agenda
14:32:27 <sdague> anything else people want to bring up?
14:33:01 <sdague> alright, thanks folks
14:33:07 <sdague> happy rc testing
14:33:11 <sdague> #endmeeting