19:23:56 <juliashapovalov2> #startmeeting refstack
19:23:56 <openstack> Meeting started Mon Sep  8 19:23:56 2014 UTC and is due to finish in 60 minutes.  The chair is juliashapovalov2. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:23:58 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
19:24:01 <openstack> The meeting name has been set to 'refstack'
19:24:35 <rockyg> #meeting refstack
19:25:02 <rockyg> Thankyou juliashpovalov2
19:25:14 <rockyg> Rollcall
19:25:21 <catherine_d> here
19:25:21 <rockyg> o/
19:25:42 <juliashapovalov2> here
19:26:21 <rockyg> #topic status
19:26:48 <rockyg> (juliashapovalov2 needs to do all commands)
19:28:10 <catherine_d> refstack.org site is down ...
19:28:17 <juliashapovalov2> #topic status
19:28:30 <rockyg> So, where are we?  what needs core review?  Please put up the links to the reviews that need core.
19:29:41 <catherine_d> David just join ...
19:30:32 <zehicle> o/
19:30:41 <catherine_d> not sure why this one fail gate .. https://review.openstack.org/#/c/117651/
19:31:50 <rockyg> zehicle: were you on the same flight as davidlenwell?
19:32:03 <davidlenwell> I am here
19:32:10 <rockyg> Any waiting for core review?
19:32:14 <davidlenwell> give me a minute to catch up
19:32:20 <davidlenwell> zehicle: was on a call that ran long
19:32:24 <davidlenwell> I was in the air
19:33:40 <rockyg> nothing to catch up on, yet;-)
19:35:35 <zehicle> I've got a review for the designated sections shema
19:37:59 <rockyg> I need to review that.
19:38:42 <rockyg> zehicle:  question: should the definitive havana defcore json files live under OpenStack governance?
19:39:03 <rockyg> Once settled, it should be owned by the board, right?
19:39:29 <zehicle> yes
19:39:58 <zehicle> there's no other place for files like these that also have gerrit voting
19:40:05 <fcarpenter> (sorry I'm late)
19:40:52 <rockyg> That would mean that all the files in refstack should be generated from the governance files, such as capabilities.json
19:41:11 <catherine_d> zehicle : so the tests for havana will be from tempest-stable-havana and not from the current tempest master right?
19:41:38 <juliashapovalov2> question regarding the havanacore: shouldn't this file correspond to tests in Tempest master? i mean if tests will be changed or deprecated these changes should be reflected in json also
19:43:03 <rockyg> Remember:  tests for Havana are against the Havana *release*, not against the current state of OpenStack projects.
19:43:37 <juliashapovalov2> it would be inconsistent to support one OpenStack version only from separate branch
19:43:59 <zehicle> juliashapovalov2, we do want the tests to be from trunk tempest
19:44:12 <zehicle> davidlenwell, did we make progress on getting UUIDs for tests?
19:44:15 <zehicle> that would be needed
19:44:22 <rockyg> fyi:  #link https://review.openstack.org/#/q/status:open+project:+stackforge/refstack,n,z
19:44:23 <rockyg> Link to open reviews
19:44:33 <zehicle> rockyg, so we need to have a way to know which tests were there for each release
19:44:47 <zehicle> I'm multithreaded, sorry about the lags
19:44:51 <davidlenwell> not yet .. I'll have answers this week
19:45:14 <zehicle> juliashapovalov2, +1.  I think we agreed in ATL that we'd be from trunk for all releases including havana
19:45:57 <catherine_d> davidlenwell and zehicle: woudl adding uuid to test be a good topic for design session on Paris?
19:46:11 <davidlenwell> I Want to solve it long before paris
19:46:17 <rockyg> Zehicle: so we are from trunk, but are we taking tests fro HEAD, or from when the release was cut?
19:47:17 <juliashapovalov2> tests may be released much later then official OpenStack version//
19:47:26 <davidlenwell> the data still doesn't mean anything over time without the uuid's
19:47:28 <zehicle> rockyg, the later - from when the release was cut
19:47:44 <zehicle> we cant have tests that shows up in Icehouse be part of the havana critieria
19:48:06 <zehicle> we're limited to the tests that were there at the release
19:48:13 <zehicle> otherwise it's a moving target
19:48:23 <zehicle> besides, for core, we're talking about stable features only
19:48:35 <catherine_d> so if it is from release the refstack-client need to be able to test at release level ...
19:48:39 <zehicle> so there should not be a lot of new tests that change the functionality
19:49:08 <zehicle> catherine_d, we'd test all the tests and then apply a "which release does this pass filter"
19:49:12 <catherine_d> <zehicle>  the problem is not for the new test ... but for the tests that are relocated
19:49:35 <zehicle> it would be interesting if we tested icehouse and  could say it passed core for havana & icehouse.  frankly, it should be able to do that
19:49:46 <zehicle> catherine_d, yes.  that's why we need IDs
19:50:08 <catherine_d> yes ... so without UUID .. are we stuck now?
19:50:22 <juliashapovalov2> what format of UUID's do you expect?
19:50:35 <zehicle> I
19:50:36 <zehicle> I
19:50:51 <juliashapovalov2> we can mock them for now, i beleive
19:50:51 <zehicle> I'd like to have a workaround, but I do think we'll need UUIDs to complete the cycle
19:51:11 <catherine_d> <zehicle> agree ...
19:51:12 <zehicle> we could also have a lookup table
19:51:32 <catherine_d> I think it will take sometime for uuid feature to be implemented
19:51:58 <rockyg> also, zehicle, does DefCore have a repository in governance yet?
19:52:55 <zehicle> rockyg, no
19:53:16 <zehicle> did not realize that there was a dedicated repo for that - is that what the TC is using?
19:53:24 <zehicle> if so, we can move the /defcore items to that
19:53:33 <rockyg> woa!  uh, what just happened?  Flood o' messages
19:54:17 <zehicle> ah, ok.  #link  https://github.com/openstack/governance
19:54:25 <zehicle> I'd be happy to move the defcore stuff to that
19:54:46 <zehicle> but want to make sure that we've got the right Core reviewers.  Who manages that repo?
19:56:46 <rockyg> That one looks like TC.  Lemme do some more looking, though.
19:59:41 <zehicle> the question to me is, does it help refstack to have those outside of the repo?  I don't really care where they reside
20:00:31 <rockyg> I really think the changes, once official, need to be owned by the foundation, not by the implementers of the certification software
20:01:53 * redrobot is ready to chair barbican meeting
20:01:57 <rockyg> Can we end meeting and move to Refstack?  Barbican has this now?
20:02:19 <catherine_d> sure
20:02:30 <rockyg> I think Juliashapovalov2 needs to 3endmeeting
20:02:36 <rockyg> #endmeeting that is...
20:02:41 <juliashapovalov2> #endmeeting