14:00:10 <gibi> #startmeeting nova
14:00:11 <openstack> Meeting started Thu Nov 30 14:00:10 2017 UTC and is due to finish in 60 minutes.  The chair is gibi. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:14 <openstack> The meeting name has been set to 'nova'
14:00:24 <gibi> hi! I will be your host today
14:00:28 <tssurya_> o/
14:00:36 <takashin> o/
14:00:44 <gmann> o/
14:00:46 <edleafe> \o
14:01:08 <efried> \o
14:01:13 <stephenfin> o/
14:01:23 <ttsiouts> o/
14:01:36 <johnthetubaguy> o/
14:02:01 <gibi> let's start
14:02:04 <gibi> #link agenda https://wiki.openstack.org/wiki/Meetings/Nova#Agenda_for_next_meeting
14:02:11 <gibi> #topic release news
14:02:21 <gibi> #link Queens release schedule: https://wiki.openstack.org/wiki/Nova/Queens_Release_Schedule
14:02:27 <gibi> #info Next week (December 7) is the q-2 milestone
14:02:34 <gibi> #info Blueprints: 51 targeted, 51 approved, 9 complete (7 completed since the last meeting)
14:03:22 <gibi> is there anything to discuss about the milestone?
14:03:51 <gibi> OK, moving forward
14:03:53 <gibi> #topic bugs
14:04:06 <gibi> I don't see any critical bugs on launchpad
14:04:16 <gibi> #info 28 new untriaged bugs (up 6 from two weeks ago)
14:05:07 <gibi> the health of the gate looks good #link check queue gate status http://status.openstack.org/elastic-recheck/index.html
14:05:29 <gibi> #link 3rd party CI status http://ci-watch.tintri.com/project?project=nova&time=7+days
14:05:59 <gibi> I don't see any major problem there
14:06:24 <gibi> #topic Reminders
14:06:46 <gibi> There is an ML thread discussing the PTG format
14:06:48 <gibi> #link ML thread on the Rocky PTG format: http://lists.openstack.org/pipermail/openstack-dev/2017-November/124898.html
14:07:31 <gmann> gibi: yea, old format more better and seems like majority on that.
14:07:47 <gibi> gmann: yeah, I had the same feeling when I read the thread
14:08:04 <gibi> anyhow it is the good time to express opinions and send in ideas
14:08:25 <gibi> #topic Stable branch status
14:08:34 <gibi> #link stable/pike: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/pike,n,z
14:08:46 <gibi> #link Etherpad for bugs tracked for Pike backports: https://etherpad.openstack.org/p/nova-pike-bug-fix-backports
14:09:01 <gibi> #link stable/ocata: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/ocata,n,z
14:09:09 <gibi> #link stable/newton: https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:stable/newton,n,z
14:09:19 <gibi> any question about the stable branches?
14:09:44 <gibi> It seems we still want to land some fix on newton before the EOL
14:09:56 <cdent> which is when?
14:10:22 <gibi> cdent: which is already happened in other projects I think
14:10:45 <gibi> I don't know if we have a hard deadline to tag EOL
14:11:14 <gmann> seems like not yet as far as i know
14:11:42 <gibi> #link https://releases.openstack.org/ shows EOL 11th of October
14:12:37 <edmondsw> I believe mriedem wanted to mark it EOL ASAP, just had to wait for a couple things to land
14:12:39 <gibi> cdent: I guess we have to ask somebody more familiar with the stable releases
14:12:49 <gibi> edmondsw: that is my view as well
14:13:20 <gibi> anyhow moving forward
14:13:25 <gibi> #topic Subteam Highlights
14:13:40 <gibi> dansmith: any news from the cells meeting?
14:13:58 <mriedem> early for him,
14:14:12 <mriedem> we mostly talked about the patch to page over migrations in the api and it's multi-cell implications
14:14:16 * cdent hopes dansmith put away his grenade
14:14:28 <mriedem> and started an etherpad (link in the ML) about cells v1 to v2 migration notes
14:14:39 <mriedem> that's it
14:14:45 * edleafe hopes dansmith put the pin back first
14:14:53 <gibi> mriedem: thanks
14:15:02 * gibi is sorry for the early ping on dansmith
14:15:25 <gibi> edleafe: any scheduler news?
14:15:30 <edleafe> yep
14:15:34 <edleafe> Noted the the nested resource provider series was showing signs of maturity
14:15:37 <edleafe> Alternate host series is also ripe
14:15:38 <edleafe> that's it
14:15:40 <efried> Regarding the review push (http://lists.openstack.org/pipermail/openstack-dev/2017-November/124909.html) -- Several patches have merged.  Thanks to gibi bauzas cdent jaypipes takashin tetsuro and others for jumping on that.  Let's keep the momentum going so we're not stuck in the mud when everybody disappears in a couple weeks.
14:15:40 <edleafe> cdent concurred a lot
14:15:43 <edleafe> Declared that this is officially "push this shit" week
14:15:45 <edleafe> So maybe remove the -2 hold on https://review.openstack.org/499239 so the bottom patches can merge, and put it on where the RPC call changes?
14:16:46 <mriedem> i don't think we're ready for that until we're comfortable with the changes in conductor
14:17:11 <edleafe> mriedem: I'm not following.
14:17:25 <edleafe> The earlier patches don't affect conductor
14:17:26 <mriedem> i don't really have time to get into details right now
14:18:04 <gibi> let's discuss that then later on #openstack-nova
14:18:20 <gibi> next subteam is API
14:18:31 <gibi> alex_xu, sdague: any news?
14:19:09 <gmann> gibi: no meeting yesterday
14:19:31 <gmann> no updates on API side as such
14:19:32 <gibi> OK
14:19:34 <gibi> thanks
14:19:38 <gibi> next is notifications
14:19:40 <gibi> that is me
14:19:49 <gibi> only one news
14:20:01 <gibi> the main patch for the bp service-create-destroy-notification has been merged
14:20:12 <gibi> there is one followup fixing nits
14:20:38 <gibi> OK, Next is Cinder
14:20:53 <gibi> mriedem left a the note #help Need reviews on Cinder v3 attach patch https://review.openstack.org/#/c/330285
14:21:07 <ildikov> yep, pretty please
14:21:28 <ildikov> in theory all the issues are fixed that we've found with that flow so far
14:21:44 <ildikov> the patch is large because of the tests
14:22:00 <ildikov> the code change is rather small, and I'm happy to chat with anyone who has questions
14:22:20 <gibi> ildikov: thanks
14:22:25 <ildikov> there's also a multi-attach related patch on top of it with a libvirt change
14:22:36 * ildikov would love to get to that one soon :)
14:22:53 <gibi> OK, moving forward
14:22:57 <gibi> #topic Stuck Reviews
14:23:07 <gibi> nothing on the agenda
14:23:27 <gibi> does anybody has any stuck review to mention?
14:24:01 <gibi> #topic Open discussion
14:24:04 <gmann> not urgent but i have 1
14:24:09 <gmann> will be happy to get review on novaclient zuulv3 job migration  - https://review.openstack.org/#/c/500347/12
14:24:29 <stephenfin> oh, I've a couple of those for nova too
14:24:32 <gibi> gmann: is this the correct link?
14:24:38 <gmann> ops
14:24:40 <gmann> opps
14:24:48 <gmann> #link https://review.openstack.org/#/q/topic:zuulv3-novaclient
14:24:55 <gmann> ^^ this one
14:25:10 <stephenfin> #link https://review.openstack.org/#/q/status:open+project:openstack/nova+branch:master+topic:zuulv3-jobs
14:25:14 <stephenfin> that's the equivalent for nova
14:25:51 <stephenfin> gmann: I'll take a look at the novaclient ones but I'm no pro here. sdague might be better?
14:26:09 * gibi opened the links as well now
14:26:28 <gmann> stephenfin: thanks
14:26:47 <gibi> cool, anything else?
14:26:48 <claudiub> o/
14:26:50 <efried> Meant to mention the alternate hosts series under "stuck reviews".  That's been wedged for a while.  We ought to come up with a hard time for the key players to get together and hammer out the remaining issues to get it moving.
14:27:02 <claudiub> I have an item for open discussion as well.
14:27:11 <efried> especially since it's a key priority for Q
14:27:29 <gibi> efried: I guess the next push this stuff week can be for alternate hosts
14:27:42 <gibi> as it worked for nested r-p
14:27:57 <efried> ..."is working" <= ftfy
14:28:07 <efried> we ain't done yet.
14:28:19 <cdent> miles to go before we sleep
14:28:24 <gibi> yeah, it is just my bad English
14:28:27 * efried cedes the floor te claudiub
14:28:31 <gibi> claudiub: go ahead
14:28:44 <claudiub> \o/
14:28:47 <claudiub> So, I've been fixing the mock autospec usage in nova. There were quite a few issues that I've found.  https://review.openstack.org/#/c/447505 https://review.openstack.org/#/c/470775
14:29:23 <claudiub> And also, I've enabled by default mock.patch with autospec. meaning that pretty much every mock.patch'ed method is autospeced, unless specified otherwise
14:29:48 <claudiub> anyways, the main issue is that there are a couple of bugs with mock autospec, which i've "patched" in the links above
14:30:24 <claudiub> primarely this: https://review.openstack.org/#/c/447505/27/nova/tests/fixtures.py
14:30:41 <gibi> claudiub: so eventually these patches will go away when the mock lib fixes the problem internally?
14:31:06 <claudiub> ofc, I've sent bug reports and bug fixes to the mock library, and upstream python, since mock became a native library in py3, but they haven't merged
14:31:45 <claudiub> gibi: the mock fixes i've done in those patches, yes. they're temporary. the rest of the fixes are good.
14:32:07 <gibi> claudiub: OK
14:32:20 <claudiub> but I don't know what are the chances of my bug fixes being merged in the mock library
14:32:29 <claudiub> the last patch merged in mock was 1 year ago.
14:32:59 <gibi> I have never sent fixes python upstream so I don't have experience how reactive they are
14:33:00 <cdent> stdlib is where modules go to freeze
14:33:15 <edmondsw> claudiub should you split these patches into a) things that could merge now and b) things that can't?
14:33:29 <claudiub> anyways. I would like to know if there's any chance we're going to merge these, or not. :D
14:33:52 <stephenfin> claudiub: I was avoiding them because they were marked WIP
14:34:08 <stephenfin> If that's not the case, I can take a shot next week
14:34:11 <claudiub> edmondsw: well, IMO, we could merge them as is.
14:34:23 <edmondsw> claudiub oh, ok... then remove "POC" ?
14:34:25 <claudiub> stephenfin: well, yeah. because until recently tests were still failing on them
14:35:05 <claudiub> edmondsw: sure. will do, it was pretty experimental, and I had no idea if I could get it to work properly, which is why it was POC. :D
14:35:22 <claudiub> and also, IMO, something like this could be used in other openstack projects.
14:35:37 <gibi> So I think there is a way forward
14:35:45 <claudiub> I was thinking if the mock related "fixes" should be merged in oslotest, or something
14:35:56 <efried> Agree proper autospeccing is useful in general
14:36:04 <claudiub> otherwise, I'd have to send the same fixes to other projects, meaning duplicate code.
14:36:29 <claudiub> not a problem in my perspective, but eh. just a thought.
14:36:34 <gibi> claudiub: if it would be the _same_ patch then lets move it to a common place
14:37:00 <claudiub> sure. oslotest?
14:37:01 <gibi> I guess that would make the nova patch smaller
14:37:06 <claudiub> indeed
14:37:24 <gibi> I think oslotest is a good target
14:37:39 <claudiub> cool.
14:37:45 * cdent concurs
14:37:50 <gibi> OK, anything else to discuss?
14:37:59 <claudiub> well then, I'll also send a bug report regarding this, and update the patches accordingly.
14:38:05 <gibi> claudiub: thanks
14:38:09 <claudiub> ty as well.
14:38:17 <efried> Thanks for sticking with this claudiub -- this is a good thing.
14:38:27 <claudiub> until then, if you have some feedback on them, it'd greatly appreciate it. :D
14:38:33 <claudiub> efried: np. :)
14:38:50 <gibi> If nothing else then let's close this meeting
14:39:03 <claudiub> we could also create a general unit test guide - to make sure autospec is used as often as possible.
14:39:23 <gibi> claudiub: good idea
14:39:51 <gibi> Ok, let's close this. thanks for the meeting
14:39:53 <gibi> #endmeeting