17:01:03 <mtreinish> #startmeeting qa
17:01:04 <openstack> Meeting started Thu May  7 17:01:03 2015 UTC and is due to finish in 60 minutes.  The chair is mtreinish. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:01:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:01:09 <openstack> The meeting name has been set to 'qa'
17:01:13 <mtreinish> hi, who's here today?
17:01:30 <samueldmq> I am
17:01:32 <jlanoux_> hi
17:01:36 <andreaf> o/
17:02:12 <mtreinish> #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Proposed_Agenda_for_May_7th_2015_.281700_UTC.29
17:02:16 <mtreinish> ^^^ Today's agenda
17:02:45 <mtreinish> dkranz, sdague, jogo, mkoderer: around?
17:03:06 <sdague> yep
17:03:57 <mtreinish> well, let's get started, more people might trickle in later
17:04:12 <mtreinish> #topic Summit Session Planning
17:04:18 <mtreinish> #link https://etherpad.openstack.org/p/liberty-qa-summit-topics
17:04:41 <mtreinish> so at last weeks meeting we idenitified a few of the topics to discuss for the slots we have
17:05:16 <mtreinish> I wanted to finalize the list at this meeting so I can take care of submitting the schedule today or tomorrow
17:06:03 <mtreinish> we've got 2 fishbowl slots and 2 work room slots left on the etherpad
17:06:34 <mtreinish> before we go through each item individually was there anything in the ideas section that stuck out to anyone as being necessary for a session?
17:06:42 <sdague> well, given that I can't read the etherpad due to mtreinish's blinding colors, I'll refrain from commenting
17:07:06 <dkranz> o/
17:07:10 <mtreinish> sdague: heh, the red did get a bit more obnoxious after the etherpad upgrade :)
17:07:14 <mtreinish> I'll dial it back a bit
17:07:19 <sdague> it was always terrible :P
17:07:26 <sdague> not a good way to invite collaboration
17:07:56 <samueldmq> sdague, yes I was afraid to
17:08:00 <samueldmq> hehe
17:08:36 <sdague> looking at the list ...
17:08:51 <samueldmq> mtreinish, can I ask a question regarding 'Tempest Scope Revisited' ?
17:09:02 <samueldmq> or should I do it post-meeting in  #qa
17:09:03 <mtreinish> samueldmq: sure, hopefully I'll be able to answer it
17:09:38 <samueldmq> mtreinish, if I understood correctly, we may be able to migrate all the functional tests to services
17:09:51 <samueldmq> and tempest will only remains with integration ones
17:09:59 <samueldmq> is that right ?
17:10:22 <mtreinish> samueldmq: maybe, it's really not so clear cut, which is part of the problem
17:10:36 <mtreinish> that's why we're going to have a session on it
17:10:45 <mtreinish> samueldmq: all we have right now is a process for doing the migration
17:11:05 <mtreinish> samueldmq: https://wiki.openstack.org/wiki/QA/Tempest-test-removal
17:11:11 <samueldmq> mtreinish, k, if we go for this, we will need to establish a process to migrate functional tests, and make sure tempest tests vs service tests correspond
17:11:56 <samueldmq> mtreinish, nice, thanks
17:12:42 <mtreinish> ok, getting back to the list, looking at the sessions it looks like that both UUID topic and dkranz's testing outside the gate have more interest
17:12:54 <mtreinish> so what do people think about a work room for the uuid topic
17:13:05 <sdague> I think a work room for the uuid topic would be fine
17:13:07 <mtreinish> and a fishbowl for testing outside of the gate
17:13:52 <sdague> I guess the other question is about the post gate testing which has some overlap with infra, but I think realistically is going to end up being more qa folks doing it if it happens
17:14:18 <dkranz> mtreinish: ++
17:14:30 <mtreinish> sdague: I mean we did have a QA session on it in Paris and had some work items come out of it
17:14:47 <mtreinish> it just never came to fruition
17:14:56 <dkranz> mtreinish: I thought that was at the thing in Germany which I could not attend?
17:15:16 <mtreinish> dkranz: we had a follow up session in Paris, andreaf led it
17:15:18 <dkranz> mtreinish: anyway we need to figure out some way to get it done
17:15:43 <sdague> mtreinish: if we have the space, that seems like a good place to spend it
17:15:56 <andreaf> dkranz: we talked about that both in Darmstadt as well as in Paris
17:15:57 <mtreinish> yeah, that's what I thought
17:16:43 <mtreinish> ok, I've marked both of those
17:16:53 <sdague> ok, I guess part of this is what do you believe the top 5 things are we need to get done in the next cycle?
17:17:48 <mtreinish> sdague: yeah I was just thinking that, I was thinking we could do a liberty priorities session as the last fishbowl slot
17:18:30 <dkranz> mtreinish: That's a good idea
17:19:07 <mtreinish> sdague: but I see what you're getting at with 5 priorities and 5 fishbowl topics
17:19:59 <dkranz> mtreinish: the problem is we have four people at this meeting so how can we decide that now?
17:20:38 <dkranz> Priorities are meaningless unless we have folks signed up to do stuff
17:20:57 <sdague> mtreinish: yeh, however, a priorities session is also a good idea
17:21:30 <mtreinish> ok, then I'll put the priorities session as the last fishbowl
17:21:33 <sdague> I was more thinking making sure everything that's we all think is important is getting time, or if there is something that's important but unspoken at this point
17:21:51 <mtreinish> sdague: yeah the big but unspoken is what I'm worried about
17:22:59 <mtreinish> but if we are overlooking something there is the meetup day on Friday
17:23:45 <mtreinish> ok, so we have one work room slot left, does anyone have a topic on the list which stands out for it?
17:25:06 <dkranz> mtreinish: how about sending the agenda to the ml with a "speak now or ...". That might flush out a big thing if some one has it but is not a regular at our meetings.
17:25:28 <samueldmq> dkranz, I think it's a good idea
17:26:10 <mtreinish> dkranz: hmm, well it might be a bit too late for that, ttx said the schedules should be done this week
17:27:42 <mtreinish> but I can still do it I guess
17:27:55 <dkranz> mtreinish: well if it goes out now, it is just a choice to use something that is responded in 24h or fill from what we have. I am ok for you to decide.
17:28:37 <dkranz> mtreinish: I think you already moved things that folks said was a "must have"
17:29:42 <andreaf> mtreinish, dkranz: I think something like stages of a test run (e.g. install tests, setup resources, verify config, run, cleanup) fit in the tempest CLI session
17:29:57 <mtreinish> dkranz: yeah, and looking at the list of ideas left there nothing really stands out as needing a session anymore
17:30:11 <andreaf> mtreinish, dkranz: so I didn't propose any session about this specifically
17:30:22 <mtreinish> andreaf: oh, we could do a work room on the cli, dpaterson has a WIP up but the spec has never been finalized
17:30:35 <mtreinish> andreaf: sounds like you just volunteered :)
17:30:44 <andreaf> mtreinish: sure, fine
17:31:04 <mtreinish> hehe
17:32:44 <mtreinish> I mean there is an overlap with sdague's tempest as a system program topic if we do a separate cli one
17:33:32 <sdague> don't call it my session, I've no idea what I'll be able to attend :)
17:33:33 <mtreinish> andreaf: do you think there is enough to talk about just on the cli?
17:33:42 <andreaf> mtreinish: yes - but perhaps this leaves more space in sdague session about the other topics, and in this session we can discuss more in details what kind of CLI we want to have
17:33:47 <mtreinish> sdague: heh, I'll schedule to make sure you can :)
17:34:22 <mtreinish> (assuming that it's possible to do that)
17:34:40 <mtreinish> andreaf: ok, wfm
17:35:00 <andreaf> mtreinish: but still it might be a relatively light session
17:36:09 <mtreinish> andreaf: hmm ok, well I'll use it for now, we can always double up with another topic too
17:36:41 <mtreinish> I also should talk to jeblair because I was thinking of doing an infra subunit2sql topic, but we could use that qa slot for it too if their schedule was tighter
17:37:14 <mtreinish> ok, well I think we've got at least a good grasp on the list, I'll work on it and put it into the tool today or tomorrow
17:37:25 <andreaf> mtreinish: ok - sounds good to me
17:37:36 <mtreinish> does anyone have anything else on this topic?
17:38:44 <mtreinish> ok, then let's move on
17:38:53 <mtreinish> #topic Devstack
17:39:04 <mtreinish> dtroyer: did you survive the lightning?
17:39:25 <mtreinish> sdague: did you want to discuss the wheel thing?
17:41:05 <sdague> the explosion?
17:41:18 <mtreinish> yeah, I think that's what dtroyer was alluding to before
17:41:34 <sdague> beyond the fact that cffi is ... challenging
17:42:07 <sdague> yeh, turns out because there are multiple ways to install things in python, that have different rules, our wheel building actually ruinned the world the other day
17:42:34 <sdague> I think - https://review.openstack.org/#/c/180549/ actually is a more appropriate fix
17:42:45 <sdague> which is just makes sure that pip controls the version in the wheel build venv
17:43:02 <sdague> would be good to get clarkb and dstuft to also ack that
17:43:09 <clarkb> looking
17:43:32 <sdague> there were a number of upstream bugs filed during the process
17:43:56 <clarkb> ya that fix looks good
17:44:07 <sdague> I don't know if anyone is going to stay on top of those to keep this from ruinning the world when cffi 1.0 actually ships
17:44:37 <mtreinish> sdague: hmm, do you have a list of them somewhere?
17:46:17 <sdague> I don't, I was drinking a lot of margaritas
17:46:25 <sdague> clarkb or lifeless might know
17:46:25 <mtreinish> sdague: heh, ok
17:46:38 <sdague> it was tuesday night :)
17:46:52 <clarkb> the one I was aware of was against cffi to fallback on writeable tmp for the file if it needs to
17:47:32 <clarkb> and anyone with a wheel mirror they curate should know to rebuild it once cffi upgrades
17:47:55 <clarkb> this ^ is the real problem and the bug proposes a work around that cffi could ship but does not yet currently
17:49:07 <mtreinish> hmm, ok. Well, I'll follow up after the meeting see if there's anything we need to keep our eyes on pending the next cffi release
17:49:55 <mtreinish> ok is there anything else on devstack?
17:49:59 <sdague> nope
17:50:07 <mtreinish> ok, then let's move on
17:50:22 <mtreinish> #topic Critical Reviews
17:50:40 <mtreinish> does anyone have any reviews they'd like to get extra eyes on?
17:51:31 <mtreinish> #link https://review.openstack.org/178341
17:51:36 <mtreinish> #link https://review.openstack.org/178343
17:51:41 <samueldmq> mtreinish, identity v3 only flag ? can I ask reviews for that ?
17:51:44 <mtreinish> 2 pretty simple tempest-lib cleanups
17:51:45 <dkranz> mtreinish: https://review.openstack.org/179398
17:51:46 <mtreinish> samueldmq: sure
17:51:53 <mtreinish> #link https://review.openstack.org/#/c/179398/
17:51:54 <dkranz> https://review.openstack.org/179401
17:52:03 <dkranz> https://review.openstack.org/179511
17:52:09 <samueldmq> #link https://review.openstack.org/#/c/179663/
17:52:26 <andreaf> mtreinish: #link https://review.openstack.org/#/c/168367/ - not so critical but it has been around for a while
17:52:29 <dkranz> sdague: https://review.openstack.org/179511 removes a bunch of smoke from the object tests
17:52:44 <mtreinish> #link https://review.openstack.org/#/c/168367/
17:52:57 <sdague> dkranz: +2, thanks
17:53:00 <andreaf> mtreinish: #link and this one https://review.openstack.org/#/c/149229/ for the ssh-auth bp
17:53:22 <mtreinish> #link https://review.openstack.org/#/c/149229/
17:53:53 <andreaf> thx, I added some text after the link command :P
17:54:20 <mtreinish> heh, yeah I didn't think meetbot would parse that all correctly :)
17:54:49 <mtreinish> ok are there any other reviews to bring up?
17:56:18 <mtreinish> oh, I have one more link
17:56:20 <mtreinish> #link https://review.openstack.org/#/q/status:open+project:openstack/tempest+branch:master+topic:python3-support,n,z
17:56:26 <mtreinish> those should all be simple to review
17:56:34 <mtreinish> mostly just switch to six in a bunch of places
17:56:55 <mtreinish> not really critical, just high rebase likelihood
17:57:18 <dkranz> I'll do a bunch of reviews this afternoon.
17:57:23 <mtreinish> dkranz: cool, thanks
17:57:38 <mtreinish> ok, I guess with <4 min left, let's end here for today
17:57:44 <mtreinish> thanks, everyone
17:57:50 <dkranz> mtreinish: the first two links I pasted were the ones to complete running the non-admin job (once cli is removed)
17:58:13 <mtreinish> dkranz: oh, yeah feel free to +2 the cli removal patch
17:58:18 <mtreinish> we're at the end of the week
17:58:26 <mtreinish> #endmeeting