17:00:08 <mtreinish> #startmeeting qa 17:00:12 <openstack> Meeting started Thu Aug 14 17:00:08 2014 UTC and is due to finish in 60 minutes. The chair is mtreinish. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:13 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:15 <openstack> The meeting name has been set to 'qa' 17:00:24 <mtreinish> hi who do we have here today? 17:00:27 <afazekas> o/ 17:00:29 <mtreinish> #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Proposed_Agenda_for_August_14_2014_.281700_UTC.29 17:00:29 <mlavalle> hi 17:00:31 <maurosr> hey 17:00:41 <mtreinish> ^^^ today's agenda 17:00:48 <psedlak> hi 17:01:09 <andreaf> o/ 17:01:19 <asselin> o/ 17:01:39 <mtreinish> ok let's get started 17:01:51 <mtreinish> #topic Name for tempest library (mtreinish) 17:01:56 <dpaterson> o/ 17:02:25 <mtreinish> so as part of splitting off parts of tempest into a library 17:02:31 <mtreinish> we need to make a new repo 17:02:47 <mkoderer> o/ 17:02:51 <mtreinish> I wanted to call it mesocyclone 17:02:53 <jlanoux> hi 17:03:09 <mtreinish> but some people have expressed issue with that name 17:03:17 <andreaf> mtreinish: seriously? :D 17:03:41 <mtreinish> I've heard from some people that since it's a library a more functional name is more appropriate 17:03:42 <mlavalle> mtreinish: what's the rationale behind that name? 17:03:54 <mtreinish> mlavalle: mesocyclone is a precursor to a tornado 17:03:57 <dkranz> o/ 17:04:21 <mlavalle> mtreinish: I like it. + 1 fro me 17:04:31 <dpaterson> twister 17:04:48 <mtreinish> I'm planning to take it to the ml before I do the split 17:04:54 <mtreinish> but I wanted to bring up here first 17:05:06 <mtreinish> to get people thinking about it :) 17:05:07 <dkranz> mtreinish: I like it but maybe it is too long 17:05:11 <jlanoux> blackstorm? 17:05:16 <mkoderer> ahm sry.. maybe I missed that... why are we doing that? 17:05:43 <mtreinish> mkoderer: http://specs.openstack.org/openstack/qa-specs/specs/tempest-library.html 17:06:06 <psedlak> tempestish? :) 17:06:11 <afazekas> tsepmet - tempest in reverse order 17:06:14 <mkoderer> lol 17:06:19 <maurosr> lol 17:06:20 <dkranz> mtreinish: Is there a proposed list of what will be moved? The spec doesn't really say. 17:06:28 <mtreinish> dkranz: not yet 17:06:38 <mtreinish> I was going to start with the cli 17:06:44 <mtreinish> and the rest client 17:06:48 <mtreinish> and go from there 17:06:54 <mtreinish> cli tests that is 17:07:24 <mtreinish> I have to do a list of everything that makes sense to split off still 17:07:30 <dkranz> mtreinish: rest client drags auth and conf and ... 17:07:38 <mtreinish> auth yes 17:07:49 <mtreinish> but I'm going to refactor it to take in args instead of a conf 17:07:56 <mtreinish> to make it portable 17:08:01 <mtreinish> I think I put that part in the spec 17:08:03 <dkranz> mtreinish: great. we needed to do that anyway 17:08:26 <mtreinish> ok is there anything else on this topic? 17:08:50 <psedlak> mtreinish: are there already proposed names? something to pick from? or is this first round? 17:09:03 <mtreinish> psedlak: this is the first round 17:09:05 <afazekas> tempest-lib / lib-tempest 17:09:15 <mkoderer> libtemp ;) 17:09:17 <andreaf> what about calm? the calm before the storm 17:09:24 <psedlak> andreaf: ++ 17:09:27 <mtreinish> afazekas: yeah that would be a more functional name 17:09:49 <mtreinish> anyway I'll put it on the ML 17:10:08 <andreaf> mtreinish: a really functional name would be something like integration-test-lib which is ugly :D 17:10:24 <mtreinish> andreaf: yeah that was my concern :) 17:10:33 <mkoderer> libit then 17:10:33 <afazekas> oepnstack-pythonclient 17:10:37 <jlanoux> mesocyclone is fine for me ;) 17:10:45 <mtreinish> jlanoux: heh, yeah I like it too 17:11:02 <mtreinish> plus google a picture of it, it's cool looking 17:11:14 <mtreinish> anyway lets move on 17:11:30 <mtreinish> #topic Specs Review 17:11:51 <mtreinish> so I haven't had a chance to review any specs lately 17:12:02 <mtreinish> are there any that people would like to get eyes on 17:12:05 <mtreinish> #link https://review.openstack.org/#/q/status:open+project:openstack/qa-specs,n,z 17:12:09 <andreaf> I have one 17:12:13 <andreaf> #link https://review.openstack.org/#/c/94741/ 17:12:25 <andreaf> I refreshed the ssh auth strategy one 17:12:28 <dkranz> mtreinish: https://review.openstack.org/#/c/104098/ is waiting for +A from you 17:12:59 <dkranz> https://review.openstack.org/#/c/86967/ is looking for another +2 17:13:08 <mtreinish> dkranz: hmm, ok I'll take a look 17:13:41 <mtreinish> but for the provider networking one I'm wondering if it belongs under the existing neutron testing bp 17:14:22 <mtreinish> dkranz: the seond one should be fine, but I'll take a closer look after the meeting 17:14:28 <dkranz> mtreinish: It might but I'm not sure we need more spec churn 17:14:48 <mtreinish> andreaf: oh, yeah that one, I need to take a close look 17:14:56 <mtreinish> dkranz: fair enough 17:15:15 <mtreinish> andreaf: cleaning up the ssh code is definitely something we need to do 17:15:20 <andreaf> mtreinish: yes I updated it - included the idea of wait_for_sshable that sdague mentioned in some meeting 17:15:40 <andreaf> mtreinish: and it's something that should probably go in the library as well 17:15:41 <mtreinish> cool 17:15:47 <mtreinish> yeah definitely 17:16:01 <mtreinish> but we don't necessarily have to hold up work on it until the library is a thing 17:16:15 <mtreinish> it'll all depend on the timing I guess 17:16:24 <andreaf> mtreinish: right it can be done in tempest and moved across 17:16:36 <andreaf> to mesocyclone - if I can spell it :P 17:16:48 <mtreinish> ok, are there any other specs? 17:17:34 <mtreinish> then let's move on 17:17:42 <mtreinish> #topic Blueprints 17:17:52 <mtreinish> #link https://blueprints.launchpad.net/tempest 17:18:06 <mtreinish> so we've still got a large number of open blueprints 17:18:20 <mtreinish> does anyone have a status update on an open bp? 17:18:29 <mkoderer> I nearly finished with the api schema unification 17:18:40 <mkoderer> but I need reviews... https://review.openstack.org/#/q/status:open+project:openstack/tempest+branch:master+topic:bp/api-schema-unification,n,z 17:18:47 <mtreinish> mkoderer: awesome 17:19:01 <andreaf> mkoderer: I'll check those 17:19:22 <dpaterson> I sumbitted pull req for review. 17:19:23 <mkoderer> and I started to work on a porting guide for negative tests 17:19:27 <dpaterson> #link https://review.openstack.org/#/c/112581/ 17:19:57 <mtreinish> dpaterson: ok, cool 17:20:10 <mkoderer> dpaterson: nice 17:20:11 <mtreinish> it's good to see progress being made on things 17:20:43 <andreaf> mtreinish: regarding scenario migration to tempest client 17:21:28 <andreaf> mtreinish: we have *a lot* of changes up, but they all depend on the initial one still not merged - so I need reviews on these three 17:22:20 <andreaf> https://review.openstack.org/#/c/107428 https://review.openstack.org/#/c/107552 https://review.openstack.org/#/c/107562 17:22:42 <mtreinish> andreaf: ok, yeah I'll take a look at those soon 17:22:51 <andreaf> mtreinish: thanks 17:23:09 <mtreinish> also, for the test-accounts bp we're making progress 17:23:16 <mtreinish> the alternate credential provider has landed 17:23:22 <andreaf> regarding test accounts, your change is merged, and I just changed mine to use deprecated option for tenant isolation 17:23:25 <mtreinish> so now it's a matter of aactually using it 17:23:33 <mtreinish> andreaf: heh 17:24:05 <mtreinish> ok are there any other bps to discuss? 17:24:16 <andreaf> and there are two other changes in progress by psedlak for a tool to generate credentials and its use in devstack 17:24:49 <mkoderer> andreaf: did you checked the -1 from yair on https://review.openstack.org/#/c/107562 17:25:09 <asselin> i haven't had a chance to rework the stree test BP. It's next after I get 3rd party ci working for my team. 17:25:23 <mtreinish> asselin: ok nice 17:25:59 <andreaf> mkoderer: oh yes thanks I almost forgot I'll make a new patch later today for that 17:26:02 <mtreinish> I was curious what the status was on that one 17:26:33 <mtreinish> ok are there any other bps to discuss? Otherwise we can move on 17:27:30 <mtreinish> #topic Grenade 17:27:51 <mtreinish> so we merged the patch to start using javelin2 17:28:06 <mtreinish> which is awesome 17:28:29 <mtreinish> so now we can start at iterate on feature improvements there 17:28:54 <mkoderer> btw I realized that we have too few grenade cores IMHO 17:29:04 <mtreinish> aside from that I'm not aware of any big updates on grenade 17:29:13 <mtreinish> mkoderer: yeah, I agree 17:29:28 <mtreinish> jogo was just added 17:29:37 <mtreinish> but 3 is still kind of few 17:29:40 <mkoderer> I think we have three 17:29:52 <mtreinish> mkoderer: yeah 17:30:11 <mtreinish> the issue is not too many people have been actively reviewing things there 17:30:50 <mtreinish> I would like to grow that 17:31:02 <mkoderer> yep.. I think there aren't many ppl aware of this project 17:31:45 <mtreinish> heh, that might be, but I'm not sure 17:32:12 <psedlak> also imho it's not directly usefull/usable to/by anyone directly 17:32:40 <mtreinish> psedlak: well it's mostly for gating and testing an upgrade locally for dev 17:32:47 <mtreinish> not for doing it in production 17:33:01 <psedlak> mtreinish: right, though i'm not sure if devs are going to use it much localy act. 17:33:09 <dkranz> mtreinish: I am not that familiar with grenade but should that be revisited? 17:33:32 <mtreinish> dkranz: probably not, because it's based on devstack 17:33:38 <dkranz> mtreinish: I think the moves to make tempest not have any devstack dependency helped its adoption a lot 17:33:54 <psedlak> mtreinish: so at the end i think it's "just" kind of thing like "gating glue" 17:34:11 <mtreinish> dkranz: it's not meant to be general purpose tool 17:34:29 <dkranz> mtreinish: I get that 17:34:35 <andreaf> there were some discussions at the infra mid-cycle about making grenade general purpose, but it would be very little left 17:34:49 <andreaf> most of grenade content is devstack specific probably 17:34:55 <dkranz> mtreinish: I was just wondering how much abstraction/pluggability it would need to serve more general purpose 17:35:04 <mtreinish> dkranz: almost everything 17:35:13 <dkranz> mtreinish: ok :) 17:35:19 <mtreinish> it's basically devstack++ 17:35:20 <mkoderer> so I think what be a good idea if some of us have a closer look to grenade? ;) 17:35:36 <mtreinish> mkoderer: heh, yes I agree :) 17:35:42 <andreaf> dkranz mtreinish so it would be good to have a grenade kind of tool for tripleo environment, but how much could we reuse really? 17:36:11 <mtreinish> andreaf: it's more the methodology you can use 17:36:19 <andreaf> yep 17:36:19 <mtreinish> and javelin2 would also be useful 17:36:36 <mtreinish> to test things stay up through an upgrade 17:37:07 <mtreinish> ok is there anything else on grenade? 17:37:54 <mtreinish> ok then lets move on 17:38:04 <mtreinish> #topic Neutron testing 17:38:20 <mtreinish> so salv-orlando is really close to merging parallel full neutron everwhere 17:38:28 <mtreinish> which is exciting 17:38:33 <mkoderer> cool 17:39:18 <mtreinish> mlavalle, salv-orlando: are there any other updates on neutron testing? 17:39:48 <mlavalle> mtreinish: I've been investing time developing the client and api test for the new LBaaS api 17:39:50 <andreaf> mtreinish: so just a note to keep an eye on the full job even if not voting 17:40:05 <mlavalle> mtreinish: I've been sending updates to the ML 17:40:09 <mtreinish> andreaf: oh yes, good point 17:40:14 <mtreinish> mlavalle: yes I've been seeing those 17:40:38 <mlavalle> mtreinish: I have also been helping the group based policy to develop their tests 17:40:55 <mlavalle> they are developing them themelves but I've been training and guiding them 17:41:15 <mtreinish> ok, cool 17:41:57 <mlavalle> and I intend to start developing an scenario test for the new LBaaS api. I will follow the work that andreaf has been doing with test_minmum_basic to convert it to the tempest client 17:42:15 <mlavalle> that's all I have this week 17:42:32 <mtreinish> mlavalle: how does that differ from the existing LBaaS scenario test? 17:42:45 <mtreinish> just covers more of the api? 17:43:03 <mlavalle> mtreinish: the absatractions offered by the new api are different 17:43:08 <mtreinish> ahh, ok 17:43:23 <mlavalle> I really intend to reuse the current scenario test and convert it to the new abstractions 17:43:31 <mtreinish> ok that makes sense 17:43:37 <mlavalle> and make the move to tempest client 17:44:11 <mtreinish> good, that'll make everyone happy :) 17:44:13 <mlavalle> my belief is that by statying close to the new Neutron subprojects we will keep the api and scenario tests fresh 17:44:26 <mlavalle> instead of always catching up 17:44:52 <mtreinish> yeah that'll probably help 17:45:01 <mtreinish> ok is there anything else to discuss on neutron testing? 17:45:01 <mlavalle> ok, that's all I have 17:45:08 <mlavalle> :-) 17:45:40 <mtreinish> ok then let's move on 17:45:47 <mtreinish> #topic bugs 17:46:13 <mtreinish> so I haven't been watching the bug list too closely the past few weeks 17:46:23 <mtreinish> #link https://bugs.launchpad.net/tempest 17:46:41 <mtreinish> but we've got 259 open bugs right now 17:46:42 <andreaf> mtreinish: I didn't really look at bugs either - how do we keep them under control usually? 17:46:57 <mtreinish> andreaf: not very well honestly :) 17:47:16 <mtreinish> I'm thinking we might need to do a bug day soon 17:47:33 <mtreinish> it'll also be helpful if someone volunteered to keep on top of them after the bug day 17:47:50 <mkoderer> yep a bug day is a good idea 17:47:57 <mtreinish> so we don't grow to this level again 17:48:17 <mtreinish> are there any volunteers to organize a bug day? 17:48:36 <mtreinish> maurosr: I believe you did the last one :) 17:49:11 <dkranz> mtreinish: Does that involve more than picking a day and sending out emails? 17:49:27 <mtreinish> dkranz: not much, just tracking the results 17:49:34 <mtreinish> and pestering people to look at bugs 17:49:44 <andreaf> :D 17:49:51 <mtreinish> oh and changing the topic in the -qa channel :) 17:50:26 <mtreinish> well don't everyone volunteer at once 17:50:27 <dkranz> mtreinish: ok, I will find some one or do it 17:50:33 <mtreinish> dkranz: ok awesome 17:50:42 <mtreinish> #action dkranz to find a volunteer to organize a bug day 17:50:49 <mtreinish> ok then lets move on 17:50:49 <dkranz> mtreinish: I is during August too early? 17:50:57 <mtreinish> no I think that'll be fine 17:51:03 <andreaf> mtreinish: does one need special status to triage bugs in lp or how does that work? 17:51:04 <mtreinish> as long as there's a weeks notice 17:51:14 <mtreinish> andreaf: I think there is a tempest bug team 17:51:18 <mtreinish> in lp 17:51:22 <mtreinish> but membership is open 17:51:32 <andreaf> mtreinish, oh ok 17:51:34 <dkranz> For folks here is there any day of week that is bad. Just asking... 17:52:08 <mtreinish> dkranz: well I like to have it before thurs so we can do a summary during the meeting 17:52:15 <mtreinish> but that's not the biggest deal 17:52:33 <dkranz> mtreinish: ok, then it shall be Mon,Tue,or Wed 17:52:42 <dkranz> mtreinish: we can move on 17:52:45 <mtreinish> ok 17:52:53 <mtreinish> #topic Critical Review 17:53:01 <mkoderer> I am worried about https://review.openstack.org/#/c/113900/ 17:53:05 <mtreinish> ok in the time remaining does anyone have any reviews that need more eyes 17:53:15 <mkoderer> I thought flake8 checks the indentation 17:53:26 <mtreinish> mkoderer: it does, that's from pylint 17:53:36 <mtreinish> I'm actually -1 there, because if we don't automate it 17:53:44 <mtreinish> I don't see a point in correcting it periodically 17:53:50 <mtreinish> it'll just cause rebase churn 17:54:00 <dkranz> mtreinish: I agree 17:54:19 <mkoderer> mtreinish: but having over indented lines are ugly 17:54:24 <mkoderer> but anyway :) 17:54:40 <mtreinish> mkoderer: yeah, so we need to check that in the pep8 jobs then :) 17:54:48 <andreaf> mkoderer I think what mtreinish is saying is that we need an hacking rule for that? 17:54:52 <mtreinish> I don't want to fix a style issue unless it's enforcing 17:54:52 <mkoderer> danny is working on that currently 17:55:08 <mkoderer> the pylint jobs in cinder didn't cover it too 17:55:23 <mtreinish> mkoderer: ok, well I'm going to -2 it with that comment until we run the check in the gate we can't merge it 17:55:58 <mkoderer> mtreinish: ok I will talk to danny that he is not sad :) 17:56:09 <mtreinish> I'll try to be more verbose to make it clear to try and not upset him :) 17:56:14 <mtreinish> but on this note: 17:56:22 <mtreinish> #link https://review.openstack.org/98909 17:56:31 <mtreinish> #link https://review.openstack.org/98918 17:56:52 <mtreinish> if we can get those merged it'll turn back on all the rules we had to disable with the last hacking update 17:56:58 <mtreinish> they just need +As 17:57:12 <mtreinish> asssuming they don't need another rebase... 17:57:18 <mkoderer> mtreinish: ok I will have a look 17:57:57 <mtreinish> ok are there any other reviews that people would like to bring up? 17:59:08 <mtreinish> ok then I guess we can end ~1min early 17:59:12 <mtreinish> thanks everyone 17:59:29 <dpaterson> cheers 17:59:31 <mtreinish> #endmeeting