15:00:10 <therve> #startmeeting heat
15:00:11 <openstack> Meeting started Wed Jul 20 15:00:10 2016 UTC and is due to finish in 60 minutes.  The chair is therve. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:12 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:15 <openstack> The meeting name has been set to 'heat'
15:00:17 <therve> #topic Roll call
15:00:21 <prazumovsky> hello
15:00:34 <duvarenkov> hi
15:00:37 <cwolferh> o/
15:01:06 <ramishra> hi
15:01:08 <spzala_> Hi
15:01:16 <therve> skraynev around?
15:01:26 <zaneb> o/
15:01:34 <mfedosin> o/
15:01:56 <therve> #topic Adding items to agenda
15:01:57 <kairat> o/
15:02:01 <therve> #link https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282016-07-20_1500_UTC.29
15:02:04 <therve> Feel free to add stuff
15:02:16 <skraynev> yeap
15:02:18 <skraynev> hi all
15:02:47 <jmiu> o/ hi
15:03:09 <Drago> o/
15:03:23 <therve> #topic Gate failures
15:03:35 <therve> So, it seems we're a bit better recently
15:03:43 <therve> The lbaas test should be fixed
15:03:52 <therve> ramishra, Could you re-enable it btw?
15:04:15 <therve> We're still seeing failures with image download, but a bit less I feel since I changed to the fedora image
15:04:23 <ramishra> yeah I can, I think you had a patch already should remove the depends-on
15:04:43 <therve> I made a stupid code change in it too
15:04:50 <therve> I don't remember if you need it to trigger builds
15:05:53 <ramishra> I don't see any db ResourceClosed failures off late. May be some of the patches by stevebaker fixed them:)
15:06:09 <therve> Yeah. It's a bit bad we don't know which one :)
15:06:21 <therve> Anyway, we should merge things a bit quicker, let's try to keep it this way. Please continue to report intermittent errors
15:06:34 <therve> #topic n3 status
15:06:46 <therve> So I wanted to make a quick status update, as n2 went by fast :)
15:07:19 <therve> We're still behind on convergence, with cancel update still in progress
15:07:32 <therve> It seems to still WIP :/
15:07:50 <therve> We may have to revert back to disable it if it's not in by n3
15:08:24 <therve> external resources is still pending, patch not reviewable that I can see
15:08:56 <therve> conditions support is moving, though the last patch is a bit big so getting so renewed feedback
15:09:36 <therve> zaneb, You have that series about cancel update, but it seems to be merging nicely?
15:09:53 <therve> And stevebaker's db patches are moving too, please review that as well
15:10:07 <therve> #link https://launchpad.net/heat/+milestone/newton-3
15:10:09 <therve> BTW :)
15:10:10 <zaneb> therve: yeah, just need reviews on the last 3 patches that actually fix the bugs :)
15:11:26 <therve> prazumovsky, What's the status on get-reality-for-resources? Just need a billion reviews?
15:11:33 <therve> https://review.openstack.org/#/q/status:open+project:openstack/heat+branch:master+topic:bp/get-reality-for-resources ouch :)
15:12:14 <skraynev> therve: I worried about Migration to glance v2, as I see it's not done yet. However AFAIK, v1 was deprecated.
15:12:28 <skraynev> so we need to do it during Newton
15:12:44 <mfedosin> glance v1 is not deprecated yet
15:12:45 <therve> Do we?
15:12:56 <mfedosin> patch is on review but it's not merged
15:12:58 <prazumovsky> therve: exactly. think to reduce patches to one per folder, e.g. one for sahara, one for neutron, etc.
15:13:15 <therve> prazumovsky, I think it's nice as it is
15:13:22 <therve> prazumovsky, We just need to bug people :)
15:13:36 <therve> skraynev, Even if it is, it's not going away anytime soon
15:13:41 <therve> I don't think we're in a hurry
15:14:00 <skraynev> therve: also about get_reality ... I thought, that it should be on hold, while convergence has not good feedback (during one release)
15:14:13 <mfedosin> I think I'll send a message to ML and add Heat tag there - because Horizon requires setting custom locations too
15:14:20 <therve> skraynev, I think that's fine, those are independent
15:14:27 <skraynev> mfedosin: I thought, that nit was deprecated already...
15:15:08 <skraynev> s/nit/it
15:15:21 <zaneb> deprecated is not the same as removed
15:15:22 <mfedosin> until all projects support v2 we can't deprecate v1 :(
15:15:25 <skraynev> therve: ok, in that case we may do review
15:15:33 <prazumovsky> therve: OK, will spam with ask for review. I think so, get_reality is deep phase 2, we should successfully complete all phases before
15:15:42 <zaneb> deprecated means you should start migrating because it's going to get removed in the future
15:15:54 <mfedosin> here's the patch https://review.openstack.org/#/c/328390/
15:15:58 <therve> prazumovsky, Your call. I don't want to lose those patches
15:16:03 <skraynev> zaneb: true, but be prepared, better ,than do it with fired head :)
15:16:20 <therve> mfedosin, That's the other way around, you should deprecate to motivate people to move :)
15:16:35 <therve> Anything else on n3?
15:16:54 <skraynev> zaneb: s/fired/ fire on the head ;)
15:16:56 <therve> FWIW, I'll be away most of august, just to be back to make the release.
15:16:57 <zaneb> skraynev: right, but just because it might be deprecated in this release doesn't mean we have to stop using it in this release. it does mean we should stop using it in the next release
15:16:57 <mfedosin> therve: tell it to our ptl
15:17:24 <therve> So don't count on me for reviews and stuff
15:17:27 <therve> mfedosin, Ah :)
15:17:49 <therve> #topic Integration with Glare service
15:17:52 <skraynev> zaneb: agree. I just thought, that it was deprecated in mitaka and be removed in O
15:17:57 <therve> mfedosin, I think that's on you
15:18:05 <mfedosin> therve: yeah
15:18:09 <skraynev> zaneb: so right now it's not really big issue
15:18:42 <mfedosin> so, we developed a service that allows to store various structures
15:18:46 <therve> I looked at the proposition, and I think we need both
15:18:53 <zaneb> skraynev: fair enough :)
15:18:54 <therve> mfedosin, Yeah assume people know that :)
15:19:00 <mfedosin> we call these structures artifacts :)
15:19:13 <ramishra> I agree, we should start with option 1 and support both
15:19:19 <therve> ramishra, +1
15:19:22 <mfedosin> and now we want to define how heat artifact type will look like
15:19:59 <therve> But if at some point we can't change the environment of a template, that's a bummer
15:20:02 <therve> That's the whole point
15:20:52 <mfedosin> so, here's my vision of artifact type for option 1 https://etherpad.openstack.org/p/heat-artifact-type
15:21:40 <therve> mfedosin, env ought to be a list, I think
15:21:58 <therve> A list of dict I mean
15:21:59 <mfedosin> list of blobs?
15:22:22 <mfedosin> I thought that env is a yaml file
15:22:27 <therve> Ah yeah
15:22:28 <skraynev> therve: ideally we wanted to involve more people and especially guys from Tripleo to get final decision. which of two approaches should be used
15:22:53 <skraynev> So it's the main question which kind of artifact is more preferable
15:23:04 <therve> skraynev, In the long term, just having option 1 is wrong
15:23:15 <therve> You want to be able to say "use that template with that environment"
15:23:22 <kairat> it depends how you are going to use glare
15:23:42 <therve> But, we can build it later once we see how it goes
15:23:53 <duvarenkov> There is also a spec for integration of Heat with Glare - https://review.openstack.org/#/c/315439/ - still not final
15:24:03 <therve> kairat, Well otherwise we copy the templates in several places
15:24:13 <skraynev> kairat: is it difficult to add new artifact type later ?
15:24:50 <mfedosin> I think we should start with option 1
15:24:51 <therve> Randall just responded about that one too
15:25:13 <mfedosin> and then we can add another artifact type for ens if it's required
15:25:19 <kairat> skraynev, you can do that but once it published you have to deal with compatibility
15:25:48 <kairat> glare will have versioning for artifact types in roadmap so it is feasible
15:26:16 <mfedosin> kairat: +1
15:26:19 <therve> kairat, I think that's fine that env can be in the template artifact
15:26:27 <therve> Ultimately, it should exist outside though
15:26:37 <therve> Independently I should say
15:28:03 <therve> mfedosin, Do you have your answers, or do we need to continue to discuss this?
15:28:09 <ramishra> Are we the first ones to start experimenting with glare? not a bad proposition though;)
15:28:42 <mfedosin> ramishra: murano has been using glare for some time
15:28:44 <skraynev> ramishra:honestly we are the second
15:28:47 <skraynev> ;)
15:28:50 <kairat> not only
15:28:59 <kairat> we got response from Tacker team AFAIK
15:29:01 <mfedosin> also tacker and app-catalog
15:29:05 <kairat> they are willing to user
15:29:10 <kairat> and of course app-catalog
15:29:35 <mfedosin> Now I have a perfect vision how to implement heat artifact type
15:29:35 <kairat> the final idea is to build app-catalog that can be integrated with global app-catalog
15:29:50 <kairat> so you can spread your work across OS
15:29:54 <mfedosin> I'll add my thoughts to the related spec
15:30:34 <mfedosin> btw, can you answer my message in ML and confirm that you're going to use glare
15:30:36 <mfedosin> ?
15:31:01 <kairat> We also heard some requests from magnum
15:31:04 <mfedosin> #link http://lists.openstack.org/pipermail/openstack-dev/2016-July/099553.html
15:31:30 <kairat> but nothing concrete TBH
15:32:52 <therve> Okay!
15:32:59 <therve> #topic Open discussion
15:34:09 <therve> Well that's calm :)
15:34:19 <zaneb> I added an agenda item :)
15:34:28 <zaneb> are we going to get a mascot?
15:34:35 <zaneb> can we keep our flame logo?
15:34:47 <ramishra> +1
15:34:49 <zaneb> (and get it redrawn to match the others)
15:34:53 <therve> Ah yeah I got an email about that
15:35:07 <Drago> redrawn +1
15:35:13 <zaneb> flames are natural
15:35:18 <therve> zaneb, That's what I answered, yes. We'll see how it goes.
15:35:32 <zaneb> so I'm hoping we won't run into the same issues as other projects that already have logos
15:35:45 <zaneb> therve: cool, thanks
15:36:07 <therve> Apparently they were keen on keeping that theme
15:37:22 <therve> But yeah I haven't send something specific for Heat. If you feel we should have another mascot, don't hesitate to send a message to the ML
15:38:33 <therve> zaneb, Don't click on "minor change" otherwise I don't get the email :p
15:39:02 <therve> Anything else?
15:39:08 <zaneb> therve: didn't think I had, but I did it after the meeting started ;)
15:39:25 <therve> 3
15:39:31 <therve> 2
15:39:38 <therve> 1
15:39:44 <skraynev> nothing from me
15:39:51 <therve> #endmeeting