17:02:28 <dkranz> #startmeeting qa
17:02:29 <openstack> Meeting started Thu Jun  4 17:02:28 2015 UTC and is due to finish in 60 minutes.  The chair is dkranz. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:02:31 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:02:33 <openstack> The meeting name has been set to 'qa'
17:02:38 <cdent> o/
17:02:44 <dkranz> Who is here today?
17:03:00 <riwinters> o/
17:03:10 <dkranz> #link https://wiki.openstack.org/wiki/Meetings/QATeamMeeting#Proposed_Agenda_for_June_4th_2015_.281700_UTC.29
17:03:57 <dkranz> ok, not too many folks today, so let's get started
17:04:02 <hogepodge> o/
17:04:15 <dkranz> #topic specs reviews
17:04:35 <dkranz> I put in a link to the tempest plugin spec
17:04:53 <dkranz> This is one of the big issues out of the summit for dealing with "big tent"
17:05:03 <dkranz> Does any one have any comments about that?
17:05:24 <dkranz> #link https://review.openstack.org/#/c/184992/
17:06:02 <dkranz> ok mtreinish I think will get to that when he gets back
17:06:11 <dkranz> Any other specs to bring up?
17:06:52 <dkranz> ok, any blueprints to discuss?
17:07:41 <cdent> :)
17:07:47 <dkranz> sdague, dtroyer : Anything to say about devstack or grenade today?
17:08:52 <sdague> only a little
17:09:04 <dkranz> sdague: ok
17:09:07 <sdague> I've got some grenade out of tree plugin support up in a review, it's really rought
17:09:09 <sdague> rough
17:09:20 <sdague> the sahara, heat, and ironic folks are looking at using it
17:09:41 <sdague> but my plate was full with other things this week, so I haven't made much progress on moving that forward yet
17:09:50 <sdague> goal is L1 to get it all sorted though
17:10:06 <dkranz> sdague: sure. Have you heard anything from manila? They were interested in the tempest plugin
17:10:19 <cdent> I'm hoping to join ceilometer to the grenade plugin party soon, but still warming up post-summit
17:10:28 <dkranz> cdent: great
17:10:59 <sdague> dkranz: I have not
17:11:05 <cdent> related to that I'm officially the ceilo qa liaison now which should change nothing as I was effectively it already
17:11:15 <dkranz> cdent: ok, good to know
17:11:24 <dkranz> sdague: I will check with mkoderer
17:11:45 <dkranz> sdague: while you are here, any suggestions on how we can get the tempest periodic jobs working?
17:11:54 <dkranz> sdague: they have been busted for a month
17:12:04 <sdague> dkranz: I don't know, I haven't looked at why they are broken
17:12:14 <sdague> and am unlikely to dive into it any time soon
17:12:15 <dkranz> sdague: I asked in the infra channel but no one really responded and Matt said he had inquired as well
17:12:23 <dkranz> sdague: I know, but it is an infra issue
17:12:26 <sdague> do you know what's wrong with them?
17:12:43 <dkranz> sdague: They all fail with a missing directory.
17:12:58 <dkranz> sdague: But the jobs seem to run on the experimental queue ok
17:13:17 <dkranz> sdague: there is something about periodic and I think tempest may be the only  project with such jobs
17:13:32 <dkranz> sdague: I'll provide more details and send something to the infra list
17:14:08 <dkranz> sdague: thanks
17:14:18 <dkranz> Does any one have anything else to discuss today?
17:14:26 <dkranz> hogepodge: anything on the defcore side?
17:14:45 <hogepodge> We have a new dev who's going to look at the flagged tests
17:15:02 <hogepodge> Particularly glance images right now.
17:15:22 <dkranz> hogepodge: yeah, have you looked at the boot-from-volume issue?
17:15:41 <dkranz> hogepodge: there seems to be some lack of clarity about whether booting from glance is required
17:15:48 <hogepodge> dkranz: I haven't looked at much of anything since Vancouver.
17:15:54 <dkranz> hogepodge: ok :)
17:16:02 <hogepodge> dkranz: ok. I'll look into it.
17:16:11 <dkranz> hogepodge: seems long ago already
17:16:20 <hogepodge> If you have any links, I can get up to speed.
17:16:45 <dkranz> https://bugs.launchpad.net/tempest/+bug/1436314
17:16:45 <openstack> Launchpad bug 1436314 in tempest "Option to boot VM only from volume is not available" [Medium,Confirmed] - Assigned to Soumit (soumit-mishra)
17:17:14 <dkranz> hogepodge: the question is whether tempest should be able to run against a cloud that has boot-from-volume only
17:17:38 <dkranz> hogepodge: right now there is a test that does boot from volume, but the image tests assume you can boot from an image
17:17:51 <hogepodge> dkranz: ok.
17:18:16 <dkranz> #topic Open Discussion
17:18:24 <dkranz> Anything else some one wants to discuss?
17:19:14 <hogepodge> I'm going to write up a post proposing what it means to be an interop test.
17:19:18 <sdague> no, I'm good, just digging out and figuring all the things I signed up for
17:19:30 <dkranz> sdague: :)
17:19:44 <dkranz> hogepodge: that would be helpful
17:19:57 <hogepodge> Just my point of view to try and be strategic about how we can create better api consistency across clouds.
17:20:11 <dkranz> hogepodge: I still have this idea that we should be able to say that all of tempest is interoperable except for some tests we flag as not
17:20:37 <hogepodge> dkranz: there are issues with assumptions, though.
17:20:55 <dkranz> hogepodge: and then defcore could specify what tempest config should say rather than listing tests
17:21:05 <hogepodge> dkranz: If I want to test cloud x for interop, is it reasonable to force me to have access to two tenants?
17:21:57 <dkranz> hogepodge: I would say that if there is an api that requires two tenants to test it, then yes.
17:22:23 <dkranz> hogepodge: and there are such apis
17:23:14 <dkranz> hogepodge: I think is we have the idea that interoperability is to be verified, we have to do that. What is the alternative?
17:24:07 <hogepodge> dkranz: depends on what you define as interoperability.
17:24:34 <dkranz> hogepodge: I assumed it was "this api can be used on any cloud that is OpenStack(TM)"
17:24:44 <hogepodge> dkranz: all apis will lose in practice. A subset, with expected behaviors, but with possible implementation differences is better.
17:25:01 <hogepodge> dkranz: I just want to know if an application will run.
17:25:16 <dkranz> hogepodge: an application will run if it uses apis that are interoperable
17:26:10 <dkranz> hogepodge: I really don't understand. If an api is not working, then it is something that we declare to be non-interop, or there is a bug
17:26:38 <hogepodge> dkranz: that's not the issue. The issue is some tests assume you have more access to resources than would be reasonable to expect.
17:26:41 <dkranz> hogepodge: anyway, I'll take a look at your doc when it is ready
17:27:04 <hogepodge> dkranz: I want to limit the max knowledge you need to have. Ideally, endpoint and credentials are all you should need.
17:27:26 <hogepodge> dkranz: we don't live anywhere in that world, though.
17:27:45 <ekarlso> oh
17:27:49 <dkranz> hogepodge: I agree, but there may need to be two. There can certainly be real applications involving vms from more than one tenant
17:28:00 <ekarlso> while it's a meeting not to be rude but can I ask for a review +A and a release ?
17:28:04 <ekarlso> of tempest lib
17:28:13 <dkranz> ekarlso: which one?
17:28:35 <ekarlso> https://review.openstack.org/#/c/183999/ < been nagging about it for days and it's blocking designate's QA patches
17:28:41 <ekarlso> s/QA/tempest
17:29:31 <dkranz> ekarlso: ok, I'll look right after the meeting
17:29:46 <ekarlso> dkranz: we need a release also to use it..
17:29:50 <ekarlso> pretty please !
17:29:51 <dkranz> ekarlso: right
17:30:19 <dkranz> ekarlso: I am not sure I can do a release. mtreinish has been doing them I think
17:30:36 <dkranz> ekarlso: he is on Japan time this week
17:30:52 <dkranz> Anything else to discuss, or reviews needed?
17:31:24 <dkranz> hogepodge: please let me know when your document is available. I'd like to see your thoughts
17:31:45 <dkranz> ok, so we will end early today
17:32:37 <dkranz> thanks every one
17:32:44 <dkranz> #endmeeting