17:01:06 <mtreinish> #startmeeting qa
17:01:07 <openstack> Meeting started Thu Nov 20 17:01:06 2014 UTC and is due to finish in 60 minutes.  The chair is mtreinish. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:01:08 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:01:10 <openstack> The meeting name has been set to 'qa'
17:01:18 <mtreinish> hi who's here today
17:01:37 <jun_xie_1> i am ~ hello~
17:02:00 <mtreinish> #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Proposed_Agenda_for_November_20th_2014_.281700_UTC.29
17:02:05 <mtreinish> ^^^ Today's agenda
17:02:28 <sdague> o/
17:03:25 <mtreinish> dkranz, mkoderer: around?
17:03:51 <mtreinish> well, I guess we can get started, and maybe more people will trickle in :)
17:04:04 <sdague> maybe they forgot the tz shift :)
17:04:14 <mtreinish> ah, yeah probably
17:04:30 <aalekov> tz ok. I
17:04:44 <mtreinish> dtroyer: around?
17:04:54 <mtreinish> #topic XML Testing (mtreinish)
17:05:11 <mtreinish> so I just wanted to make a quick mention of xml in tempest
17:05:25 <mtreinish> a lot of patches have been coming through to disable xml testing on master
17:05:38 <mtreinish> and moving forward the openstack rest apis are just going to be using json
17:05:55 <mtreinish> so at this point I just wanted to make it clear that we should probably not accept new xml tests in tempest
17:06:00 <sdague> +1
17:06:20 <aalekov> +1
17:06:23 <jun_xie_1> +1 so how r we going to handle the existing ones?
17:06:28 <dtroyer> o/
17:06:31 <jun_xie_1> should we try to remove them>?
17:06:42 <mtreinish> jun_xie_1: they need to stay around until juno is eol
17:06:53 <mtreinish> because we're running them on the stable branches
17:06:53 <sdague> yeh, mostly, how soon do we think we can remove them, because it simplifies the code quite a bit to not have it in there
17:07:19 <mtreinish> it's the same branchless tempest quagmire for removal of anything
17:07:28 <sdague> mtreinish: well, I kind of wonder in this case
17:07:43 <sdague> the TC didn't say XML is part of the API, we specifically did not say that
17:07:51 <sdague> so these are extra tests for project specific things
17:07:57 <sdague> which aren't part of the openstack interface
17:07:58 <mtreinish> sdague: well at least for icehouse it was before that I think
17:08:19 <sdague> the TC resolution about the API was pre icehouse
17:09:06 <mtreinish> sdague: do you want to take it to the ml then? I'm fine with ripping it out, but I want to make sure people aren't expecting the stable branch coverage
17:09:20 <sdague> sure, can do
17:09:36 <mtreinish> #action sdague to take removing xml test question to ML
17:09:51 <mtreinish> ok does anyone else have anything on this topic?
17:10:30 <sdague> nope
17:10:31 <mtreinish> #topic Spec Reviews
17:10:49 <mtreinish> ok does anyone have a pending spec review to bring up
17:11:03 <mtreinish> this is your chance to raise it
17:12:08 <mtreinish> ok I guess not
17:12:14 * mtreinish thinks this will be a quick meeting
17:12:49 <mtreinish> #topic Blueprints
17:13:08 <mtreinish> ok are there any open blueprints we need to discuss today?
17:13:48 <mtreinish> sdague: on the javelin2 stuff did you want to respin that spec to outline the new plan
17:13:59 <mtreinish> err I guess jogo volunteered for that didn't he
17:14:09 <sdague> yes, jogo volunteered on that
17:14:20 <mtreinish> ok I'll bug him to update that then
17:14:54 <mtreinish> looking at the open tempest bp list the only thing I see is https://blueprints.launchpad.net/tempest/+spec/branchless-tempest-extensions
17:15:06 <mtreinish> which I think salv-orlando has patches pending for
17:15:31 <mtreinish> so if we could get eyes on those it would be good, this is something I'm actually constantly being asked about
17:16:01 <mtreinish> #link https://review.openstack.org/126422
17:16:17 <mtreinish> #link https://review.openstack.org/116129
17:16:54 <mtreinish> I guess the only thing on bps (assuming no one else has one to bring up) is how we're going to handle devstack bps (if they exist)
17:17:06 <mtreinish> but we can use that as a segway into the next topic
17:17:38 <dtroyer> mtreinish: I have one or twp BPs on my post-summit list, so there will be some
17:18:11 <mtreinish> #topic Devstack
17:18:31 <sdague> yeh, I should honestly probably write down the LIBS_FROM_GIT one as it turned into slightly more effort than I expected
17:18:42 <sdague> and testing turned out to be kind of critical
17:18:44 <mtreinish> dtroyer: ok, well I think we decided to try using specs for devstack too, but I wasn't sure I've ever seen a bp on devstack before
17:19:06 <mtreinish> so I wasn't sure what criteria should be used for what's a bp, etc
17:19:08 <sdague> I think we've not done it before, but it might be worth while for some of these
17:19:25 <dtroyer> mtreinish: I think they exist but we've never actually used it.
17:19:40 <sdague> like the venv discussion on the list probably should be written down before implemented, because of the edge cases.
17:19:50 <dtroyer> We should formalize the bigger projects a bit as it is getting hard to know where some of this stuff is coming from
17:19:58 <jogo> mtreinish sdague:  I did
17:20:31 <dtroyer> jogo: thanks, that's one I was thinking about.
17:20:41 <sdague> dtroyer: agreed
17:21:12 <dtroyer> I don' tthink these are BPs, but I want a way to track things like the sudden appearance of Gluster support
17:21:33 <dtroyer> which specifically is what I knew was only a matter of time after we added Ceph
17:21:38 <mtreinish> dtroyer: well you could do something like what we do in tempest for new tests
17:21:56 <mtreinish> have a lighter spec template for new feature support
17:22:17 <sdague> dtroyer: it seems like what we probably want is a more clear in / out / as an internal plugin / as an external plugin line
17:22:17 <mtreinish> and make it more procedural so there is a single place to track it all
17:22:18 <dtroyer> mtreinish: sounds good…common process FTW
17:22:47 <sdague> the tuskar review kind of brought up that idea for me
17:22:49 <mtreinish> dtroyer: do you want to draft up a patch to expand the qa-specs readme for the devstack process? and maybe a second template
17:23:08 <dtroyer> sdague: exactly. we've crossed the philosophical line we had a year or so ago
17:23:19 <dtroyer> mtreinish: sure
17:23:51 <sdague> dtroyer: at least we have this nice new docs tree to document it better now :)
17:24:06 <sdague> speaking of which, can we also bring back the devstack logo into the docs?
17:24:07 <mtreinish> #action dtroyer to write up devstack specific sections for qa-specs repo
17:24:10 <sdague> I miss it
17:24:26 <mtreinish> sdague: it should be doable, sphinx supports images
17:24:42 <dtroyer> sdague: I'll look in to it.  I don't have the source but can probably track down jake to get it...
17:24:46 <mtreinish> I guess it's more a matter of do you put it in the git repo vs hosting somewhere
17:24:53 <sdague> dtroyer: ok, cool
17:24:59 <dtroyer> in the repo, it was before
17:24:59 <mtreinish> I'm not sure if sphinx lets you do remote images
17:25:09 <sdague> yeh, in the repo is fine
17:25:15 <mtreinish> ok cool
17:25:42 <mtreinish> ok is there anything to discuss on devstack?
17:25:45 <dtroyer> That's all I had on my mind…
17:26:09 <sdague> just fyi, LIBS_FROM_GIT looks fairly stable now
17:26:20 <mtreinish> sdague: oh, cool
17:26:29 <sdague> the glance_store fix is inbound, and there are some tests to cross check thing sanely
17:26:30 <mtreinish> are you going to attempt to backport it?
17:26:42 <mtreinish> or is it too large a change?
17:26:51 <sdague> I'm going to see how bad it is
17:27:09 <mtreinish> ok
17:27:50 <mtreinish> ok let's move on
17:27:56 <mtreinish> #topic Grenade
17:28:13 <mtreinish> is there anything to discuss about grenade this week?
17:28:18 <mtreinish> jogo, sdague: ?
17:28:41 <jogo> no big changes coming through that I know of
17:29:15 <sdague> just jogo needs to do the ansible stuff :)
17:29:17 <mtreinish> jogo: ok, I'm looking forward to the new javelin2 (or will it be javelin3) spec
17:29:26 <sdague> neutron is voting now
17:29:28 <sdague> that's new
17:29:31 <jogo> mtreinish: should I just append the original spec?
17:30:16 <mtreinish> jogo: whatever you prefer, you can just respin the existing spec
17:30:29 <mtreinish> or push a new one to replace it, the net effect is the same
17:30:39 <sdague> jogo: and circle with mordred he has a bunch of pieces here you can probably reuse
17:31:18 <jogo> sdague: yeah
17:31:30 <jogo> mtreinish sdague: will start working on this once I get live migration testing working for nova
17:31:31 <mtreinish> sdague: oh awesome, so the recent changes from jakub landed
17:31:34 <jogo> (getting super close)
17:31:39 <mtreinish> jogo: cool
17:31:53 <jogo> (to getting the testing working not live migrate itself)
17:32:30 <mtreinish> ok, let's move on
17:32:38 <mtreinish> #topic Bugs
17:32:54 <mtreinish> #link https://etherpad.openstack.org/p/Tempest-bug-report
17:33:02 <mtreinish> #link http://keystone-weekly-bug-report.tempusfrangit.org/weekly-bug-reports/tempest-weekly-bug-report.html
17:33:10 <mtreinish> we've got a bunch of new bugs this week
17:33:18 <mtreinish> and it looks like we're falling behind on the triage again
17:33:45 <mtreinish> it looks like it's dkranz's rotation for triage
17:34:06 <mtreinish> I'll circle back with him on it
17:34:21 <mtreinish> does anyone have any critical bugs they'd like to raise?
17:35:30 <mtreinish> ok I guess not
17:35:34 <mtreinish> let's move on then
17:35:53 <mtreinish> #topic Critical Reviews
17:36:07 <mtreinish> does anyone have any reviews that they'd like to get extra eyes on?
17:36:36 <mtreinish> now's your chance to bring them up
17:36:50 <jun_xie_1> speaking of reviews, so what is up with jenkins? i just noticed that many failed
17:37:23 <mtreinish> jun_xie_1: do you have a specific example?
17:37:25 <sdague> jun_xie_1: if you are more specific about what's failing, that would be good
17:37:44 <jun_xie_1> https://review.openstack.org/#/c/135860/
17:37:45 <mtreinish> I suspect you're seeing the latest icehouse reqs issue though
17:37:53 <jun_xie_1> https://review.openstack.org/#/c/134889/
17:38:01 <mtreinish> yeah that's https://bugs.launchpad.net/taskflow/+bug/1394647
17:38:18 <jun_xie_1> ok got it thx
17:38:24 <sdague> mtreinish: oh... also - https://review.openstack.org/#/c/125095/ is a possible fix for that
17:38:29 <dtroyer> when should we expect that to propogate through?
17:38:31 <sdague> but it needs to be fixed
17:38:50 <mtreinish> sdague: ok I'll take a look
17:38:55 <mtreinish> dtroyer: it's in the gate now
17:39:10 <mtreinish> dtroyer: we can ask clarkb, fungi, or jeblair to promote it
17:39:16 <mtreinish> I think it's still towards the bottom
17:39:28 <dtroyer> OK.  I must have been watching the wrong thing last night…
17:39:51 <mtreinish> dtroyer: there was another almost identical failure yesterday with oslo.vmware
17:40:07 <dtroyer> ah, right, that's what I was watching
17:40:29 <mtreinish> it's been 2 days in a row that I've woken up to an oslo lib breaking icehouse...
17:40:55 <mtreinish> we're totally going to last the extended support window
17:41:15 <mtreinish> ok well I guess no one has any reviews to bring up
17:41:22 <mtreinish> I guess I'll open the floor then
17:41:26 <mtreinish> #topic Open Discussion
17:41:50 <mtreinish> does anyone have a topic they'd like to bring up that wasn't on the agenda?
17:42:40 <kirshil> yes, who and when will do network func tests migration to neutron?
17:42:43 <mkoderer> hi sry, I am late
17:43:13 <mtreinish> mkoderer: heh, no worries
17:43:42 <mtreinish> kirshil: as of now there are no concrete plans to remove anything from tempest, neutron has to spin up functional testing first
17:44:14 <clarkb> mtreinish: they have functional testing fwiw
17:44:17 <mtreinish> after that, there is a process for removing things from tempest
17:44:47 <mtreinish> clarkb: yeah, but I was refering to duplicate coverage to warrant removing something from tempest
17:44:56 <mtreinish> from what I remember the suite was still pretty small
17:45:00 <clarkb> rgr
17:45:11 <kirshil> does it mean that new tests to tempest are not allowed?
17:45:40 <mtreinish> no, new tests are still allowed
17:47:36 <mtreinish> ok is there anything else? otherwise we can end here
17:47:46 <kirshil> put the question in this way: if i wanna suggest new network func test- to which project then?
17:48:13 <kirshil> tempest or neutron?
17:48:29 <mtreinish> kirshil: honestly without any context on what your trying to add I can't say
17:48:50 <mtreinish> you can try tempest and if it's outside the scope in tempest that should be caught in review
17:49:12 <kirshil> f.e. some generic ipv6 related test not dealing with any vendor plugin?
17:49:18 <mtreinish> I'm not sure what the scope of neutron functional testing is so you;d have to ask in -neutron
17:49:30 <kirshil> ok thk got it
17:50:16 <mtreinish> sure, np
17:50:34 <mtreinish> ok, if there isn't anything else I guess we'll call it
17:50:36 <mtreinish> thanks everyone
17:51:02 <mtreinish> #endmeeting