07:05:29 <pas-ha> #startmeeting heat 07:05:30 <openstack> Meeting started Wed Aug 5 07:05:29 2015 UTC and is due to finish in 60 minutes. The chair is pas-ha. Information about MeetBot at http://wiki.debian.org/MeetBot. 07:05:32 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 07:05:34 <openstack> The meeting name has been set to 'heat' 07:05:45 <pas-ha> #topic rallcall 07:05:51 <Qiming> o/ 07:05:54 <xek> o/ 07:05:54 <sirushti> o/ 07:05:55 <pas-ha> #topic rollcall 07:05:55 <tspatzier> hi 07:05:57 <ramishra> o/ 07:05:58 <KanagarajM> Hi 07:05:58 <rakesh_hs> o/ 07:05:58 <huangtianhua> o/ 07:05:59 <skraynev> pas-ha : I will help ;) 07:06:00 <ananta_> hi 07:06:03 <zengyingzhe_> hi 07:06:23 <ramishra> do we have an agenda? 07:06:27 <kairat> o/ 07:06:30 <skraynev> #link https://wiki.openstack.org/wiki/Meetings/HeatAgenda 07:06:40 <pas-ha> #topic Adding items to agenda https://wiki.openstack.org/wiki/Meetings/HeatAgenda#Agenda_.282015-08-05_0700_UTC.29 07:06:59 <pas-ha> nothing special from my side 07:07:06 <skraynev> from me too. 07:07:14 <skraynev> any suggestions from other people ? :) 07:07:14 <pas-ha> have one item for later 07:07:52 <skraynev> hm. probably we can discuss short update about convergence work ? 07:07:58 <Qiming> need some help on review, not a big item though 07:08:12 <skraynev> sirushti: ananta_ ^ 07:08:18 <pas-ha> #topic heat reviews https://etherpad.openstack.org/p/heat-reviews 07:08:19 <ananta_> skraynev: yes 07:08:24 <skraynev> do you want to make some update :) 07:08:30 <sirushti> skraynev, sure 07:08:37 <sirushti> we have a voting convergence gate now 07:08:48 <pas-ha> #link convergence gerrit topic https://review.openstack.org/#/q/message:convergence+project:openstack/heat+status:open,n,z 07:09:00 <ananta_> we are almost done with phase 1 of convergence 07:09:07 <sirushti> *non-voting 07:09:20 <ananta_> there are few WIP for coancel update and timeout 07:09:37 <asalkeld> o/ 07:10:02 <skraynev> all these patches https://review.openstack.org/#/q/message:convergence+project:openstack/heat+status:open,n,z are related with phase 1 ? 07:10:10 <skraynev> or only some of them? 07:10:19 <pas-ha> skraynev, IMO not all 07:10:27 <ananta_> not all 07:10:32 <pas-ha> surely the one backported is not ^) 07:10:38 <asalkeld> the input_data will take a little time 07:10:45 <skraynev> So, could you point on major patches ? 07:10:50 <pas-ha> the gerrit query searches for "convergence" in commit message 07:10:54 <skraynev> and add to etherpad 07:11:03 <skraynev> with topic : phase1 07:11:14 <ananta_> the DNMs are only for testing 07:11:27 <asalkeld> we can add to review etherpad 07:11:29 <ananta_> there are two by stevebaker which I am not sure if they are for convergence 07:11:39 <sirushti> ananta_, they're not 07:11:50 <asalkeld> ananta_: they have convregence in the message somewhere 07:12:12 <ananta_> okay 07:12:29 <rakesh_hs> skraynev: this is one of the major patch https://review.openstack.org/#/c/203661/ 07:12:45 <rakesh_hs> it provides scenario tests for convergence 07:13:08 <rakesh_hs> testing almost the whole of convergence developed till now 07:13:15 <ananta_> the one for Autoscaling by shardy is also not related to convergence 07:13:24 <skraynev> rakesh_hs: nice :) 07:13:37 <asalkeld> rakesh_hs: still failing a test 07:13:50 <ananta_> so, all patches except DNMs, by stevebaker and shardy ;) 07:13:51 <pas-ha> py27, not the functioal 07:13:56 <rakesh_hs> asalkeld: yes, it will be fixed once your patch is done 07:13:59 <sirushti> asalkeld, that depends on the cache_data patch 07:14:07 <asalkeld> ok 07:14:23 <asalkeld> sirushti: can't we skip that test for now? 07:14:27 <pas-ha> oh, those are unit ones.. taking my note back 07:14:31 <skraynev> ananta_: https://etherpad.openstack.org/p/heat-reviews 07:14:42 <asalkeld> merge, then enable it once my stuff gets in 07:14:49 <sirushti> asalkeld, ++ 07:14:54 <skraynev> I have added it with special header ^ 07:15:05 <asalkeld> nice to have that in 07:15:10 <ananta_> skraynev: looking 07:15:38 <skraynev> ananta_: ps. only two last mentioned , so add other necessary, please ;) 07:15:49 <ananta_> skraynev: I will do that 07:16:33 <stevebaker> sorry I'm late 07:16:44 <asalkeld> hi 07:16:50 <skraynev> stevebaker: np :) hi 07:16:56 <pas-ha> anything else on review topic? 07:16:59 <stevebaker> pas-ha: keep chairing by all means 07:17:03 <skraynev> we just discussed convergence progress 07:17:06 <pas-ha> stevebaker, sure 07:17:16 <skraynev> stevebaker: and updated https://etherpad.openstack.org/p/heat-reviews 07:18:12 <asalkeld> stevebaker: is the talk votes finished? 07:18:13 <stevebaker> I've put the OpenStackClient review on that. I think it would be nice to have full osc commands available for liberty release 07:18:13 <pas-ha> stevebaker, anything specific you wanted to discuss? cuz we are out of topics :) 07:18:30 <asalkeld> you wanted to do a "share your talk topic" 07:18:41 <stevebaker> asalkeld: I think the votes ended last week 07:18:46 <pas-ha> yes 07:18:47 <asalkeld> ok, np 07:18:49 <pas-ha> friday 07:18:51 <ramishra> can I have some more cores review the RG/ASG patches? 07:19:05 <stevebaker> having a full set of osc commands might justify releasing python-heatclient 1.0.0 07:19:07 <skraynev> stevebaker: I leaved couple comment for this patch:) 07:19:09 <stevebaker> #link https://review.openstack.org/#/c/195867/ 07:19:16 <stevebaker> skraynev: I saw those, I'll get to that 07:19:20 <sirushti> stevebaker, it would be nice to have functional tests for that, I guess 07:19:23 <skraynev> stevebaker: ok 07:19:42 <asalkeld> stevebaker: does osc support aliases like git? 07:19:48 <stevebaker> I'm not that keen on implementing all the commands, if anyone is keen to help 07:19:53 * asalkeld would love that 07:19:57 <stevebaker> asalkeld: I've aliased openstack to os 07:20:01 <stevebaker> os stack list 07:20:03 <stevebaker> os stack show 07:20:41 <stevebaker> I'll raise a feature bug on python-heatclient for each command so we can discuss specifics and claim tasks 07:20:49 <skraynev> stevebaker: btw, do we plan to use openstack-sdk when it will be released ? 07:20:51 <asalkeld> stevebaker: more like "os stack cleanup" -> bash script to delete all stacks (or whatever) 07:21:15 <stevebaker> asalkeld: I'd like all sorts of higher-level stack commands, like os stack failures 07:21:30 <stevebaker> asalkeld: which drills down to the resource which failed 07:21:59 <asalkeld> yeah 07:22:00 <pas-ha> +1, that would be nice 07:23:00 <pas-ha> anything else? 07:23:08 <asalkeld> not from me 07:23:13 <pas-ha> #topic open discussion 07:23:30 <stevebaker> skraynev: good question. we'd likely take the lead from openstackclient project. If someone had some effort to ensure the heat support in the sdk is complete that would be good anyway 07:23:36 <xek> I have a problem with failing integration tests in https://review.openstack.org/#/c/190183/ 07:23:47 <pas-ha> I have one small question - who is most knowledgable with parameters/properties encryption? 07:23:58 <Qiming> stevebaker, I'm been working on that 07:24:06 <stevebaker> pas-ha: jasond? 07:24:17 <stevebaker> Qiming: oh, nice 07:24:24 <pas-ha> also therve? 07:24:40 <skraynev> stevebaker: ok. I will try to hold the hand on pulse (about sdk) 07:24:46 <therve> pas-ha, Sure somewhat 07:24:53 <stevebaker> btw, this bug is nasty, could it get reviews please? https://review.openstack.org/#/c/206256/ and the backport https://review.openstack.org/#/c/208777/ 07:24:56 <xek> it's consistently failing on heat_integrationtests.functional.test_notifications.NotificationTest.test_basic_notifications but I run the test locally without any problems 07:25:03 <pas-ha> therve, https://review.openstack.org/#/c/208547/ 07:25:23 <pas-ha> I "actually" enabled encruption om gate and broke half the tests 07:25:50 <pas-ha> xek, there is a patch somewhere disabling notif tests for now 07:25:59 <pas-ha> afir 07:26:25 <asalkeld> pas-ha: yeah, but his problem is in the gate 07:26:39 <xek> pas-ha, ok, I'll search for it and add it as a dependency 07:26:42 <therve> pas-ha, Awesome 07:27:27 <stevebaker> I don't have anything else 07:27:35 <pas-ha> therve, could you check it please ? I would also ping jasond and add ypu both to review 07:27:48 <therve> pas-ha, I'll look into it. It's possible it's very broken, or we just make wrong assumptions in the tests 07:27:52 <skraynev> asalkeld: also we with ramishra had a question about https://review.openstack.org/#/c/208165/ 07:28:12 <skraynev> do you plan to use or already use it somewhere ? 07:28:19 <pas-ha> therve, it seems first :( as tests break that are not related 07:28:32 <skraynev> and how about moving to https://review.openstack.org/#/c/208165/ 07:28:54 <asalkeld> i like it 07:29:08 <asalkeld> i run standalone and always have to turn some tests off 07:29:17 <ramishra> skraynev: we discussed that this morning 07:29:20 <asalkeld> and it's a pain to come up with the regex 07:30:06 <skraynev> ramishra: oops. I have not know about it :) and what results of discussion? 07:30:30 <skraynev> ramishra: removing or revert? 07:30:40 <asalkeld> just remove is fine 07:30:48 <ramishra> skraynev: removed in the same patch 07:31:02 <xek> pas-ha, I can't find the patch 07:31:22 <skraynev> ramishra: got it. thx :) 07:31:44 <stevebaker> ramishra: its a big change but it would be useful. 07:32:01 <stevebaker> ramishra: it beats adding a config option for every test 07:32:15 <pas-ha> xek, i'll search and ping you 07:32:24 <xek> pas-ha, ok, thanks! 07:32:37 <ramishra> I've been rebasing the RG/ASG patches forever:( have taken care of most of zaneb's concerns 07:33:04 <ramishra> can I've some more core eyes? 07:33:22 <asalkeld> ramishra: sure 07:33:27 <stevebaker> ramishra: heh, i need to look at those. thanks for keeping it up 07:33:39 <asalkeld> they are complex, so people avoid :-O 07:34:08 <ramishra> yeah I know. now it supports both random and indexed resources 07:34:20 <stevebaker> sweet 07:34:28 <ramishra> RG by default indexed and ASG by default random names 07:34:47 <stevebaker> ramishra: what is the chance existing templates could switch from rg to asg on stack update without replacing everything? 07:36:19 <ramishra> stevebaker: that's not possible atm. but there can be mix of indexed and random name(exisiting) resources with hidden indexes for existing resources 07:36:39 <stevebaker> ok 07:37:39 <Qiming> need some helps on reviewing this: https://review.openstack.org/#/c/201079/ 07:37:47 <Qiming> it is a pure refactor work 07:38:03 <Qiming> I'd doing it to clear the way for some more interesting patches 07:38:23 <Qiming> say, stricter checking of filter/sort_keys 07:38:28 <stevebaker> a ~5000 line test file, yeesh 07:38:44 <Qiming> too big to maintain 07:39:39 <asalkeld> we all done? 07:39:44 * asalkeld getting hungry 07:39:48 <stevebaker> I reckon so 07:40:40 <pas-ha> ok then, let's finish early 07:40:45 <pas-ha> thanks everyone 07:40:49 <pas-ha> #endmeeting