08:00:21 <therve> #startmeeting heat
08:00:22 <openstack> Meeting started Wed Jul 27 08:00:21 2016 UTC and is due to finish in 60 minutes.  The chair is therve. Information about MeetBot at http://wiki.debian.org/MeetBot.
08:00:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
08:00:27 <openstack> The meeting name has been set to 'heat'
08:00:27 <therve> #topic Roll call
08:00:49 <stevebaker> heya
08:00:58 <elynn_> o/
08:02:10 <therve> ramishra, Around?
08:02:13 <tiantian> hi
08:02:17 <ramishra> hi
08:03:11 <therve> That'll be a quick one :)
08:03:20 <therve> #topic Adding items to agenda
08:03:28 <therve> #link https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282016-07-27_0800_UTC.29
08:04:54 <therve> #topic Gate status
08:05:20 <therve> It seems relatively bad these past couple of days, but it doesn't look like our fault
08:05:52 <therve> Infra seems particularly slow
08:05:53 <stevebaker> http://status.openstack.org/elastic-recheck/
08:06:00 <therve> ramishra, Noticed anything?
08:06:29 <therve> stevebaker, Yeah not sure there is a bug in there per se
08:06:41 <therve> Nodes are just super super slow and we timeout a lot
08:06:48 <stevebaker> Bug 1546431 may have gotten worse after ramishra's last change landed, but it may be completely unrelated
08:06:48 <openstack> bug 1546431 in heat "Tests fails with SQL error 'Command Out of Sync'" [High,In progress] https://launchpad.net/bugs/1546431 - Assigned to Rabi Mishra (rabi)
08:07:06 <therve> stevebaker, Where have you seen it?
08:07:09 <stevebaker> therve: I've seen a couple of fedora image download timeouts
08:07:25 <therve> Yeah... Still better than the previous download I think
08:07:27 <stevebaker> therve: in the chart on http://status.openstack.org/elastic-recheck/
08:07:31 <therve> And at least it fails fast :)
08:07:40 <ramishra> Are we on to the first topic?
08:07:46 <therve> Ah :/
08:07:56 <stevebaker> therve: we could get the fedora image cached on AFS
08:08:04 <therve> ramishra, The topic should be in the topic :)
08:10:14 <therve> stevebaker, I've seen that error somewhat related to infra performance, so hopefully that's just the link
08:10:54 <stevebaker> therve: you mean the image download?
08:11:08 <therve> stevebaker, No the out of sync error
08:11:33 <stevebaker> hmm, I guess a slow node could make the race more likely
08:11:57 <therve> I think so yeah
08:12:19 <ramishra> sorry guys, my network is terrible today.
08:12:37 <therve> Anyway there is not much we can do when a node takes 45 mins to setup
08:13:25 <therve> #topic N3 status
08:13:54 <therve> We're moving sloooowwwwwly
08:14:39 <therve> Please review more stuff :/
08:15:12 <ramishra> therve: should we maintain priority reviews etherpad like we used to do in ealrier cycles?
08:15:26 <stevebaker> #link https://launchpad.net/heat/+milestone/newton-3
08:15:38 <therve> ramishra, I'm not a giant fan, but maybe we'll have to do it
08:17:47 <therve> #topic Open discussion
08:18:33 <stevebaker> I've started playing with gabbi today, and have it doing some basic stack operations
08:19:35 <therve> stevebaker, Is that... useful?
08:19:39 <stevebaker> I'd like to propose that we write a full REST API test suite in gabbi. therve what do you think that would need to discuss? a blueprint? a wishlist bug?
08:20:37 <stevebaker> therve: having decarative REST API tests will be nice I think. And I suspect it will be the easiest thing to get into tempest core for defcore
08:20:37 <therve> stevebaker, What's the point exactly?
08:20:54 <therve> I though using tempest was the best for defcore
08:22:05 <stevebaker> therve: nobody maintains the REST API tests in tempest core, and they're incomplete so they're not a great start for defcore
08:22:13 <therve> I guess I have enough questions that he deserves a spec
08:22:27 <therve> Also, we should probably do that in the next cycle unless there is a real rush
08:22:49 <stevebaker> no rush I think at this point, I'm just seeing if its feasible
08:23:15 <therve> stevebaker, Yeah but I wonder why rewriting stuff with another tool will make it maintenable
08:23:25 <therve> Instead of improving what's already there
08:24:06 <stevebaker> testing REST APIs with declaritive yaml is *really* nice
08:24:42 <stevebaker> it will likely show some issues with our api-ref because its so clear to see what is going on
08:25:40 <therve> Okay
08:26:07 <therve> Anything else?
08:28:16 <therve> Alright, thanks
08:28:19 <therve> #endmeeting