15:00:12 <ramishra> #startmeeting heat
15:00:12 <openstack> Meeting started Wed Jan  4 15:00:12 2017 UTC and is due to finish in 60 minutes.  The chair is ramishra. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:14 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:16 <openstack> The meeting name has been set to 'heat'
15:00:21 <ramishra> #topic roll call
15:00:30 <therve> Heya
15:00:47 <ricolin> O/
15:00:49 <ramishra> happy new year all
15:01:01 <zaneb> happy new year!
15:01:02 <ricolin> happy new year!
15:01:47 <cwolferh> happy new year!
15:02:52 <ramishra> #topic adding items to agenda
15:02:59 <ramishra> #link https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282017-01-04_1500_UTC.29
15:03:29 <ramishra> I've added a few topics, anything else we would like to discuss?
15:03:58 <therve> ramishra, Adding one about heat-agents
15:04:16 <ramishra> therve: yep
15:05:05 <ramishra> #topic gate failures
15:05:24 <ramishra> So the new year has not been good;)
15:05:40 <ramishra> failure rate is very high.
15:05:59 <therve> We had a barbican issue?
15:06:05 <therve> Then that cfn weird stuff
15:06:12 <ramishra> therve: yeah I fixed that
15:06:16 <therve> Anything else identified?
15:06:30 <ramishra> I noticed that we get slow nodes for the jobs.
15:06:45 <ramishra> and devstack build takes 30-40 mins
15:06:55 <ramishra> from the total 80 mins for the job.
15:07:35 <ramishra> I propose this https://review.openstack.org/#/c/414365/
15:07:51 <ramishra> comment in that patch has the logstash query too.
15:08:49 <ramishra> To me it's mostly infra issue, tests with waitconditions fail mostly.
15:09:49 <therve> That sounds like a good workaround
15:10:04 <ramishra> One more issue I noticed today, raised a bug 1653990
15:10:05 <openstack> bug 1653990 in heat "test_asg_scale_up_max_size fails with KeyError" [Undecided,New] https://launchpad.net/bugs/1653990
15:11:50 <ramishra> weird kind of failure.
15:12:30 <therve> Some kind of race with metadata maybe?
15:13:33 * zaneb groans
15:13:38 <tiantian> ;)
15:13:41 <ramishra> therve: it seems to me like the resource is in db but not in template.
15:13:55 <ramishra> not looked in detail though.
15:15:06 <ramishra> Any help to fix these is appreciated:)
15:16:24 <therve> ramishra, Ah probably just some transaction boundaries not handled properly
15:17:01 <ramishra> therve: yeah, possibly had never seen this before.
15:17:15 <ramishra> ok, lets move on
15:17:47 <ramishra> else we would only be discussing gate issues;)
15:17:57 <ramishra> #topic Status of Blueprints
15:18:29 <ramishra> #link https://launchpad.net/heat/+milestone/ocata-3
15:19:07 <ramishra> can we identify the bps we plan to land in ocata? We don't have much time.
15:19:27 <ramishra> None of them seem super important to me.
15:20:23 <tiantian> https://blueprints.launchpad.net/heat/+spec/custom-update-replace-for-failed-resources hope this can be landed in ocata
15:20:44 <tiantian> acturally it's simple
15:21:14 <ramishra> tiantian: ok
15:21:17 <tiantian> one or two patches have been merged already
15:21:28 <tiantian> thank you :)
15:22:08 <ricolin> I think we should define the Priority of undefined one and I feel free to land till next release if no high Priority there
15:22:23 <ramishra> it seems we would not be able to do much translation stuff.
15:22:28 <ramishra> Not seen any patches.
15:23:41 <ramishra> ok, I'll talk to the respective owners and change the target release accoridingly.
15:24:39 <ramishra> moving on...
15:24:46 <ramishra> #topic User survey question
15:25:00 <ramishra> #link http://lists.openstack.org/pipermail/openstack-dev/2017-January/109511.html
15:25:29 <therve> Yeah that's kind of tough
15:25:41 <ramishra> can we've some good adoption questions? I've not seen many replies to my mail;)
15:25:58 <zaneb> would be interesting to know if people are enabling convergence (phase 1)
15:26:20 <zaneb> although that'd *more* interesting in 6 months or a year from now
15:27:03 <zaneb> so I'm inclined to leave it out for now, because I can't think of anything we'd do differently at _this_ point based on the results
15:27:28 <ricolin> ramishra: how about put throughs here https://etherpad.openstack.org/p/ocata-heat-user-survey
15:27:40 <ramishra> ricolin: sure
15:27:56 <ramishra> actually we need one good/relevant question:)
15:28:31 <ricolin> #link https://etherpad.openstack.org/p/ocata-heat-user-survey
15:28:38 <cwolferh> how often do you purge your stacks, with what parameters? :)
15:29:29 <ramishra> How do we decide the *one* question?
15:30:56 <ramishra> some projects are using voting/scoring system etc. We should not make it complicated;)
15:31:25 <zaneb> cwolferh: how many operators would even *know* the answer to that? ;)
15:32:19 <therve> ramishra, Some kind of rough estimate of stack size would be interesting
15:32:22 <therve> stack lifetime, too
15:33:48 <ramishra> therve: yeah
15:34:37 <ramishra> Someone would like to help finalize the question? we've few more days to decide.
15:35:30 <ricolin> ramishra: I can help:)
15:35:53 <ramishra> ricolin: great:)
15:36:27 <cwolferh> zaneb, i'll just go with "not enough" :)
15:37:50 <ramishra> moving on..
15:38:03 <ricolin> Let's put some random questions list in etherpad, I will help to finalize it (probility by voting)
15:38:03 <ramishra> #topic heat-agents repository
15:38:04 <zaneb> cwolferh: in an ideal world installers just do the Right Thing my default and 90% of operators never have to think about it. (I don't think we live in that world yet.)
15:38:25 <zaneb> s/my/by/
15:38:38 <ramishra> therve: your topic:)
15:38:45 <therve> Yes!
15:38:54 <therve> So I started https://review.openstack.org/416244
15:39:08 <therve> To create the new repository out of heat-templates
15:39:36 <therve> https://github.com/therve/heat-agents is the source, I used some git filter-branch invocation to get the content
15:39:47 <therve> Plus one commit to fix the folder moves
15:40:14 <therve> Depending on when it's created I may have to catch up some commits manually later on
15:40:24 <therve> Reviewing that repo can be helpful
15:40:36 <therve> I hope we'll get infra feedback quickly afterwards
15:40:46 <ramishra> #link http://lists.openstack.org/pipermail/openstack-dev/2016-November/107306.html
15:40:57 <ramishra> that's the ML thread on it.
15:40:58 <therve> It'd be nice if someone contacted the rdo package maintainer
15:41:23 <tiantian> good start:)
15:42:12 <therve> That's pretty much it, unless someone has questions :)
15:43:21 <ramishra> therve: are we going to able to keep the history after the move?
15:43:39 <therve> ramishra, Yeah that's the point of that github repo
15:43:49 <ramishra> ok great:)
15:44:06 <zaneb> therve: did I just hear you volunteer to be the rdo package maintainer?
15:44:21 <therve> ramishra, Actually, if you know how to improve the history that'd be great actually
15:44:21 <ramishra> :)
15:44:33 <therve> It contains empty files changes currently
15:44:47 <therve> zaneb, Seems like twisting my word quite heavily :)
15:45:58 <zaneb> therve: is it working? ;)
15:46:07 <therve> I don't think so
15:46:28 <zaneb> dang
15:46:43 <ricolin> therve: will `upstream: ` change after?
15:47:02 <zaneb> therve: I think stevebaker is the current maintainer for that
15:47:03 <therve> ricolin, It doesn't need to, from what I understand
15:47:17 <ramishra> I think someone is already maintaining the rdo packages(python-heat-agent*) as per stevebaker in that ML thread
15:47:25 <therve> ricolin, It's only used for initial import if we don't set that other keyword
15:47:46 <ricolin> therve: Okey, thx
15:47:55 <therve> hguemar and apevec seem to make the uploads
15:49:47 <ramishra> therve: anything else you want to add?
15:49:53 <therve> Nope
15:50:18 <ramishra> #topic open discussion
15:51:17 <therve> Should we try to write down some PTG schedule?
15:51:42 <ramishra> therve: yeah, I forgot that completely. Thanks for reminding.
15:52:54 <ramishra> Would be it design session like the earlier summits or some concrete use cases that we can work on there?
15:54:18 <ricolin> Will be great to strong cross project discussion
15:54:23 <ramishra> I think some teams are doing both.
15:55:04 <ramishra> Will create an etherpad and let's capture ideas first.
15:55:09 <zaneb> therve: if you run "git rebase 40e6c0874dc3877666a92c706079e7b74cc67601" on the heat-agents repo it'll delete all the merge commits
15:55:17 * zaneb discovered that just now
15:55:36 <therve> ramishra, I was thinking a bit of both? I'd like to do some real implementation if we can
15:55:45 <therve> zaneb, Does that remove commits with no files?
15:56:05 * zaneb looks
15:57:34 <therve> zaneb, Looks like a winner, thanks :)
15:57:35 <zaneb> therve: gonna say yes. I don't see any commits left with no files, except the first one, 40e6c0874dc3877666a92c706079e7b74cc67601, "Added .gitreview"
15:57:47 <zaneb> therve: speaking of which, we need to add a .gitreview :)
15:58:02 <therve> zaneb, It's there?
15:58:43 <zaneb> therve: oh, you added it in the last commit. great :)
15:59:07 <zaneb> there's probably a way to delete that first empty commit, but not sure what it is
15:59:28 <therve> ramishra, end meeting!
15:59:46 <ramishra> yep
15:59:49 <zaneb> btw +1 on "a bit of both" for the PTG
15:59:52 <therve> zaneb, Thanks that look much nicer
16:00:10 <ramishra> thanks all for joining
16:00:15 <ramishra> #endmeeting heat