17:00:40 <mugsie> #startmeeting Designate
17:00:41 <openstack> Meeting started Wed May 18 17:00:40 2016 UTC and is due to finish in 60 minutes.  The chair is mugsie. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:42 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:45 <openstack> The meeting name has been set to 'designate'
17:00:47 <mugsie> #topic Roll Call
17:00:54 <mugsie> so, whos here?
17:01:00 <pglass> o/
17:01:25 <mugsie> timsim: Kiall federico3 elarson ping
17:01:34 <Kiall> o/
17:01:47 <mugsie> rsyed:
17:02:01 <mugsie> Kiall: we need 4 weeks for the backports btw :P
17:02:10 <Kiall> was just checking how many ;)
17:02:31 <timsim> o/
17:02:38 <mugsie> #topic Announcements
17:02:50 <mugsie> API-Ref docs have started to get worked on
17:02:54 <rsyed2> o/
17:03:15 <mugsie> #link http://developer.openstack.org/api-ref/dns
17:03:23 <mugsie> will have them in a few hours hopefully
17:03:30 <federico3> o/
17:03:35 <mugsie> #topic Action items
17:03:52 <elarson> o/
17:03:59 <mugsie> mugsie to do stable bp for 4 weeks
17:04:06 <mugsie> well, that was delegated :P
17:04:15 <mugsie> #topic Bug Triaged
17:04:19 <mugsie> #topic Bug Triage
17:04:25 <mugsie> timsim:
17:04:39 <timsim> https://bugs.launchpad.net/designate/+bug/1581090
17:04:40 <openstack> Launchpad bug 1581090 in Designate " Expose /v2/recordsets api endpoint" [Undecided,New]
17:04:47 <timsim> looks like it was auto-filed to make sure we write docs?
17:05:00 <mugsie> yeap
17:05:10 <mugsie> there is docs afaik?
17:05:22 <mugsie> I know there is API-Ref docs anyway :P
17:05:24 <Kiall> mugsie's got it covered in the API docs too
17:05:33 <mugsie> fix released
17:05:46 <timsim> k
17:05:58 <timsim> That's it :) (there's one that I forgot to respond to last week, will do that now
17:06:03 <mugsie> cool
17:06:09 <Kiall> quiet bug week
17:06:11 <mugsie> #topic Stable Backport Triage (kiall - recurring)
17:06:13 <Kiall> #link http://paste.openstack.org/show/497560/
17:06:21 <Kiall> as usual, take a few to nominate backports
17:06:27 <Kiall> 4 weeeks work.. So.. tis a tad long
17:06:54 <mugsie> ca7ffba Support both olso.m v4 and v5 ?
17:06:58 <mugsie> did that go back?
17:07:09 <federico3> ed51f49 Ensure services shutdown during unit testing  ?
17:07:21 <timsim> mugsie yeah
17:07:42 <timsim> https://github.com/openstack/designate/commits/stable/mitaka
17:08:04 <rsyed2> 699468c DevStack: Allow disabling of API versions
17:08:04 <rsyed2> ?
17:08:06 <Kiall> federico3: I dunno, it wasn't the CI issue in the end.. so, probably not give it's test only changes and not 100% needed
17:08:13 <Kiall> rsyed2: I believe that went back already
17:08:16 <rsyed2> cool
17:08:21 <Kiall> yep - it did :)
17:08:37 <Kiall> (was needed to get tempest going against stable/mitaka
17:08:40 <federico3> Kiall: do you think it mitigated the CI issues in part?
17:08:43 <rsyed2> ah
17:09:22 <Kiall> federico3: I don't think it had an effect in the end
17:09:27 <Kiall> 7ee8501 Merge "Allow api_export_size to be updated"
17:09:34 <mugsie> api change
17:09:53 <mugsie> so, not without an exception
17:10:06 <Kiall> True...
17:10:14 <Kiall> then - nothing to BP?
17:10:25 <mugsie> I dont think so
17:10:33 * Kiall doesn't believe that
17:10:46 <mugsie> we have been pretty good with proposing to all brnachs recently
17:10:54 <timsim> nice
17:11:16 <mugsie> anyone else see any potential commits?
17:11:21 <mugsie> or are we good to move on?
17:11:51 <Kiall> actually. CI fix
17:11:55 <federico3> https://bugs.launchpad.net/designate/+bug/1550441 ?
17:11:56 <openstack> Launchpad bug 1550441 in Designate "SerializationMiddleware: rrset can be None" [Undecided,Fix released] - Assigned to Federico Ceratto (federico-ceratto)
17:12:02 <Kiall> that didn't go back, but somehow isn't causing an issue/
17:12:29 <Kiall> it was 96de5c139bcccf6cc6314bc9a8c579b996f2cb27
17:12:30 <Kiall> https://github.com/openstack/designate/commit/96de5c139bcccf6cc6314bc9a8c579b996f2cb27
17:12:41 <mugsie> federico3: what commit was that?
17:12:42 <Kiall> (I use the word "fix" loosly)
17:12:59 <Kiall> federico3's one was https://git.openstack.org/cgit/openstack/designate/commit/?id=4bc65992ce98e1146803ec6befdb87315bd61b1b
17:13:05 <federico3> 4bc6599 Fix rrset serialization, improve mdns tests
17:13:17 <Kiall> The refactor scares me ;)
17:13:29 <Kiall> It's really too big to backport at 6 files changed, 274 insertions, 44 deletions
17:13:33 <mugsie> yeah
17:13:55 <federico3> we can backport only teh fix
17:14:42 <mugsie> if we can get just the fix extracted, maybe.
17:14:53 <mugsie> but, it is not a clean backport / cherry-pick
17:14:54 <Kiall> possibly - hard to know without seeing it
17:15:02 <mugsie> so, it is not ideal
17:15:35 <federico3> OTOH the impact of the bug is not huge
17:15:45 <mugsie> yeah, we may be better leaving it be
17:16:06 <mugsie> opinions?
17:16:35 <Kiall> We're probably OK to leave it, and can re-eval later if people are hitting it
17:16:41 <timsim> That ^
17:16:41 <mugsie> OK, if there is no disent, we should leave it
17:16:46 <mugsie> ok
17:16:49 <federico3> if some user ping us on the bug report in future we can backport it
17:16:52 <mugsie> #topic Open Discussion
17:17:01 <mugsie> I have 2 items
17:17:07 * Kiall has 1 too
17:17:11 <mugsie> one - we are about to EOL Kilo
17:17:18 <Kiall> :)
17:17:29 <mugsie> so, if there is any vital backports, they need to be done now
17:17:34 <timsim> bye felicia
17:17:36 <mugsie> (like in the next few hours)
17:18:02 <mugsie> 2 - Requrements Updates. I suggest that we just +2+W them
17:18:06 <mugsie> as long as they pass gate
17:18:19 <mugsie> (i.e. no need for 2 cores)
17:18:20 <Kiall> ++ - translation updates too
17:18:27 <mugsie> yup - good point
17:18:35 <mugsie> but I think they may be a thing of the past?
17:18:45 <timsim> Seems legit
17:18:46 <mugsie> need to go look at the new service
17:18:52 <Kiall> True, haven't seen them in a while
17:19:10 <mugsie> everyone OK with that?
17:19:18 <mugsie> if not speak now :)
17:19:30 <timsim> I think that's a good idea
17:19:39 <mugsie> #info Bot Patches only require one core to review from now on
17:19:47 <mugsie> and one last item -
17:19:49 <elarson> +1
17:20:03 <mugsie> I am on vacation from Friday morn
17:20:20 <mugsie> Kiall is my nominated "emergancy" stand in
17:20:29 <mugsie> for releases etc
17:20:43 <elarson> +1
17:20:46 <Kiall> what's an emergancy? ;)
17:21:03 <elarson> Lol
17:21:03 <timsim> It's depricated, don't worry about it
17:21:06 <mugsie> a lack of spell check
17:21:14 <Kiall> Wasn't it dipricated? :D
17:21:17 <mugsie> damn word didnt auto complete
17:21:17 <rsyed2> dipricated i thought
17:21:25 <Kiall> lol
17:21:27 <timsim> something something writing on the wall
17:21:39 <Kiall> timsim: highfive
17:21:45 <timsim> #neverforget
17:21:46 <mugsie> peeing in a flood
17:21:47 <elarson> Derprercerte
17:22:03 <timsim> mugsie: lol
17:22:22 <timsim> ANYWAY
17:22:25 <Kiall> Ok - FYI re Tempest - We're now using master of designate-tempest-plugin as a gate on stable/mitaka, and have a gate on master of designate-tempest-plugin that tests against stable/mitaka of designate. This means we forced to avoid making breaking API changes (at least for stuff covered by tempest)....
17:23:14 <Kiall> New APIs etc require a small dance in order to test with tempest, you must A) have a config flag in tempest defaulting to off... and then B) you must update our devstack plugin to set the option to True so the tests run on the master etc
17:23:20 <mugsie> 👍
17:23:53 <Kiall> https://review.openstack.org/#/c/316649/ is a good example of where this is needed
17:24:10 <timsim> Seems easy enough. Could we not get to a point with the tests that they cover Mitaka and explicitly check out that point in the mitaka gate?
17:24:12 <Kiall> the /v2/recordsets endpoint is in master, but not mitaka - so the *-mitaka jobs failed..
17:24:28 <mugsie> timsim: tempest is branchless
17:24:45 <mugsie> I.e. it should run on all versions of the API, with some features disabled
17:24:47 <Kiall> timsim: that means you can make breaking changes in Newton and update the tests without them exploding ;)
17:25:03 <mugsie> it stops us breaking APIs in N for people who use M
17:25:04 <timsim> Alright, makes sense.
17:25:23 <Kiall> If we had microversions, like some of the other projects, conf flags aren't needed..
17:25:24 <mugsie> it makes some dev slower, but much better for people who actually use our stuff
17:25:31 <Kiall> but that's a problem for another year
17:25:42 * mugsie will never implement nova style microversions
17:26:17 <Kiall> Anyway - Just FYI in case anyone hits it and is wondering what's going on
17:26:20 <mugsie> any other topics?
17:26:50 <mugsie> goign once
17:26:51 <mugsie> twice
17:26:52 <Kiall> Not from me
17:27:00 <timsim> I'm good
17:27:06 <rsyed2> nope
17:27:09 <mugsie> timsim: no your not :P
17:27:15 <timsim> oh wait
17:27:18 <timsim> review the worker model :)
17:27:24 <timsim> i wasn't, actually
17:27:39 <mugsie> I knew
17:27:51 <mugsie> ok - everyone enjoy the 30 mins back
17:27:54 <mugsie> #endmeeting