17:07:30 <davidlenwell> #startmeeting refstack
17:07:31 <openstack> Meeting started Mon Jul 28 17:07:30 2014 UTC and is due to finish in 60 minutes.  The chair is davidlenwell. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:07:33 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:07:35 <openstack> The meeting name has been set to 'refstack'
17:08:08 <rockyg> o/
17:08:23 <davidlenwell> roll call?
17:08:27 <catherine_d> o/
17:08:41 <fcarpenter> o/
17:10:47 <juliashapovalova> hello
17:10:48 <davidlenwell> agenda: refstack-client status, api status, general ui and feature discussion, summit talks.
17:12:36 <davidlenwell> #topic refstack-client status
17:13:17 <davidlenwell> I've just finished reviewing the latest patch.. turned my -1 to  +1 .. I'd like some other eyes on the review before merging it.  https://review.openstack.org/#/c/108218/
17:13:32 <davidlenwell> catherine_d: Do you want to give us any more information than that?
17:14:19 <catherine_d> Once we merge this code .. People can start testing with refstack-client ...
17:14:34 <davidlenwell> catherine_d:  well after the api is hosted
17:14:44 <catherine_d> The next step is to process the data and send only passing data back to refstack ...
17:16:22 <catherine_d> I encourage the refstack team to start testing with this code .. .  Perhaps Alex/Rocky can start to use it? ...
17:16:57 <juliashapovalova> >>(20:14:44) catherine_d: The next step is to process the data and send only passing data back to refstack ...   sergey s did such a feature some a week ago? probably the same code could be used
17:17:16 <davidlenwell> juliashapovalova: there is an open review for this ..
17:17:28 <davidlenwell> well its -2'd .. the same code could be used. yet
17:17:29 <davidlenwell> yes
17:17:41 <catherine_d> Sergey's code is merging the sunit format results ...
17:18:13 <davidlenwell> true .. the decided format is in the spec is simply an array
17:18:16 <rockyg> alex has been fighting with RDO
17:18:26 <juliashapovalova> is it changed?
17:18:34 <catherine_d> wha we need is to pass a JSON as defined by the spec https://review.openstack.org/#/c/105901/
17:18:37 <zehicle_at_dell> o/
17:19:03 <fcarpenter> ( https://review.openstack.org/#/c/108218/ has an additional +1 now )
17:19:16 <catherine_d> We are not sending back results with subunit data format as we did earler
17:19:19 <davidlenwell> https://github.com/stackforge/refstack/blob/master/specs/approved/api-v1.md
17:19:28 <juliashapovalova> oh, ok
17:19:39 <juliashapovalova> is the appropriate story&
17:19:44 <juliashapovalova> ?
17:20:00 <davidlenwell> juliashapovalova:  that link clearly shows the format for submitting results
17:20:17 <juliashapovalova> thanks
17:20:21 <davidlenwell> thanks fcarpenter
17:20:50 <davidlenwell> lets move on to the next topic... because the client can't complete its mission until the new api is hosted.
17:21:13 <davidlenwell> #topic api status
17:22:53 <davidlenwell> I've been dragging my feet a little bit on this. But I am going to attempt to have a review in at the end of my day for the api code.. Then tomorrow I will focus my energy on getting the api hosted. sadly we can't just jump right into infra hosting us.. they don't move that fast.. but I will get that ball rolling tomorrow as well.
17:24:06 <catherine_d> For testing, could we stand up a local refstack and use the API to post data in local refstack?
17:24:45 <davidlenwell> after I get todays review in..
17:24:56 <davidlenwell> you could check out my review and stand up an api server
17:25:12 <catherine_d> great
17:25:32 <davidlenwell> right now I have most of the code written but need to debug a little and get an api server start script working.
17:25:55 <davidlenwell> okay any other questions about the api?
17:26:14 <davidlenwell> moving right along..
17:26:16 <zehicle_at_dell> I have a small board update about it
17:26:29 <davidlenwell> zehicle_at_dell: about the api?
17:26:35 <zehicle_at_dell> sort of
17:26:42 <davidlenwell> okay .. lets hear it
17:26:49 <zehicle_at_dell> we talked about the "pass only" decision from ATL
17:27:03 <zehicle_at_dell> and hiow it made it easier to companies to post results w/o negative results
17:27:20 <zehicle_at_dell> and they agreed that it would be OK if companies want to claim their results as long as it was OPTIONAL
17:27:41 <davidlenwell> you mean publicly claim them ?
17:27:47 <zehicle_at_dell> so that supports the direction that we've been going
17:28:04 <zehicle_at_dell> the thinking right now is that they would post them on the marketplace
17:28:07 <zehicle_at_dell> not through refstack
17:28:14 <zehicle_at_dell> they would upload to refstack
17:28:33 <zehicle_at_dell> but we don't have to have an attenstation process (company Foo stamps results Bar as valid)
17:28:48 <davidlenwell> zehicle_at_dell: that brings me actually cleanly into the next topic.
17:28:55 <zehicle_at_dell> cool
17:29:14 <davidlenwell> if we're done with triggering tests and collecting the results.. its time to start getting deep into what we are doing with the data.
17:29:46 <davidlenwell> a. we need an attenstation process for vendors.
17:30:46 <davidlenwell> or is what you are saying devoid that need and place it on the market place?
17:31:17 <zehicle_at_dell> yes
17:31:26 <zehicle_at_dell> for this cycle, we could skip it
17:31:40 <zehicle_at_dell> and allow vendors to self post in the marketplace
17:31:58 <zehicle_at_dell> would not be "official" like we want to eventially drive
17:32:15 <zehicle_at_dell> but it would be easy enough to verify so there's little risk of fabrication
17:32:27 <davidlenwell> okay.. well I'd like to increase our value to the general community
17:32:45 <zehicle_at_dell> I think that if they are using our results and re-posting then that's a big win
17:32:59 <zehicle_at_dell> right now, they just say "YES, we comply" so it would be an improvement
17:33:18 <davidlenwell> we have stuff for defcore.. we have stuff for vendors to say they are compliant..
17:33:42 <davidlenwell> I want to use the varified results data to build an interop map
17:33:49 <davidlenwell> verified
17:33:53 <zehicle_at_dell> +1
17:34:03 <davidlenwell> I think we have the man power to pull it off right now
17:34:19 <zehicle_at_dell> we'er still working off the original stories that we drafted last year
17:34:35 <zehicle_at_dell> it would help to have a populated schema
17:34:57 <davidlenwell> agreed
17:35:24 <juliashapovalova> and wiki page should reflect the latest reqs change as well i beleive
17:35:59 <davidlenwell> juliashapovalova: that and the home page of refstack.org  which will probably soon be something like interop.openstack.org or something.
17:36:01 <catherine_d> so when they post their results to the market place .... How is the check to core test list done?
17:36:21 <davidlenwell> catherine_d: I think they will link to a score card that we host
17:36:29 <rockyg> I just need the final decision to update the wiki pages
17:36:44 <davidlenwell> rockyg: decision about what?
17:37:02 <rockyg> names, etc.
17:37:12 <rockyg> like interop.openstack.org.
17:37:27 <davidlenwell> that decision is a ways off
17:37:32 <zehicle_at_dell> rockyg, +1
17:37:43 <rockyg> And pointers to ref changes
17:38:04 <rockyg> ^ref^req
17:38:25 <davidlenwell> rockyg: I think what juliashapovalova was reffering to is an update in message on the wiki about what the heck we are doing and how folks can use it.
17:38:36 <davidlenwell> a lot has changed.
17:39:08 <davidlenwell> we simplified our mission.. and we should update our public facing text to reflect that
17:39:25 <juliashapovalova> yes?  a lot of stuff is moved out of scope
17:39:35 <juliashapovalova> *,
17:39:40 <davidlenwell> rockyg: perhaps you and I should have a sit down about that and we can write up new content.
17:40:05 <catherine_d> should we have a face-2-face?
17:40:29 <rockyg> Cool.
17:40:29 <davidlenwell> catherine_d: one that is not engineering .. but documentation / message / content driven
17:40:33 <catherine_d> a face-2-face meeting sometime next week?
17:40:47 <rockyg> Mybe RWC?
17:40:53 <rockyg> or SC
17:40:57 <davidlenwell> I'm open to that.. and I'd even travel to the south bay for it
17:41:06 <zehicle_at_dell> We'd talked about having one the 14th
17:41:27 <davidlenwell> thats a ways off .. we can do an engineering one on the 14th
17:41:40 <davidlenwell> but what I would like to do is just spend time working on website/wiki copy
17:41:50 <davidlenwell> which wouldn't require all parties
17:42:00 <zehicle_at_dell> I may be in SJC on 8/8
17:42:08 <zehicle_at_dell> can't commit yet
17:42:19 <zehicle_at_dell> will be in SFO on 8/14
17:42:28 <davidlenwell> zehicle_at_dell: let me know asap .. otherwise we can skype you in to this
17:42:53 <zehicle_at_dell> I have tx to be there 8.8
17:43:02 <zehicle_at_dell> the issue if if I'm available - I'm booking meetings
17:43:16 <zehicle_at_dell> the 14th is already blocked for DefCore / Refstack
17:43:37 <davidlenwell> we'll keep that blocked for an engineering meetup
17:43:58 <davidlenwell> I want you joshua and I to work out graphing for defcore
17:44:05 <davidlenwell> we should be collecting data by then
17:44:14 <catherine_d> yes agree ..
17:44:35 <zehicle_at_dell> ok
17:44:43 <davidlenwell> so 8/8 rockyg .. can you commit to a f2f in the south bay near a cal train station?
17:44:56 <davidlenwell> documentation / wiki / website copy meeting
17:44:58 <rockyg> Yup.
17:45:25 <davidlenwell> hopfully we can get aligned with zehicle_at_dell.. we'll finalize the details next meeting.
17:45:31 <rockyg> Lots of choices.  RWC, Mtn Vw, Sunnyvale
17:45:48 <davidlenwell> I'll plan on spending the bulk of the day in the south bay
17:46:03 <rockyg> Cool.
17:46:25 <davidlenwell> okay moving on because we are running short on time.
17:46:34 <davidlenwell> #topic Summit talks
17:46:54 <davidlenwell> catherine_d: I heard a rumor you might want to give a talk?
17:47:02 <davidlenwell> the deadline to submit is this week.
17:47:18 <rockyg> Today for the genreal conf.
17:47:26 <davidlenwell> yes.. not design sessions
17:47:31 <rockyg> I assume midnight UTC
17:47:32 <davidlenwell> I am definatly planning a design session or two
17:47:35 <catherine_d> I think we can give a talk about our experience in using Tempest to test non-devstack env?
17:47:48 <davidlenwell> catherine_d: you should submit that
17:48:00 <zehicle_at_dell> deadline is today
17:48:04 <rockyg> Sounds good.  Just need an abstract and a short bio
17:48:21 <catherine_d> which name should I put in along? David, Rocky, Rob?
17:48:38 <davidlenwell> I'm happy to colaborate if you like
17:48:43 <davidlenwell> but I didn't want to step on your toes
17:48:51 <davidlenwell> I'm planning two design sessions
17:49:00 <catherine_d> no .. we are the refstack  team
17:49:20 <catherine_d> I think it is a good lessons learned session ,,,
17:49:26 <zehicle_at_dell> catherine_d, we've got plenty of session - go ahead and run that one yourself
17:49:37 <catherine_d> ok
17:49:38 <zehicle_at_dell> I could recommend a Dell person if you want
17:49:44 <catherine_d> sure
17:49:45 <rockyg> catherine_d, I could do it, but i submitted two for new contribs..so go for it.  we'll be there to back you up
17:49:51 <zehicle_at_dell> David Paterson (I think you've been copied on emails w/ him)
17:50:11 <zehicle_at_dell> he's got a lot of experience about this (just not directly w/ Refstack)
17:50:17 <catherine_d> ok Do I need to contact him? Or you will let him know?
17:50:45 <davidlenwell> catherine_d:  yes .. like I said .. we all have your back if you want other people on stage.. but I think you should be the focus.
17:51:18 <zehicle_at_dell> catherine_d, I'll reach out and do a re-intro
17:51:25 <catherine_d> yes I am the focus... but I would like the talk to not an IBM talk ... it is a refstack talk ..
17:51:37 <davidlenwell> sounds good to me :)
17:51:45 <davidlenwell> we need a platform to tell the masses.. USE REFSTACK
17:51:51 <catherine_d> yes  .. this is a community effort
17:52:00 <davidlenwell> so the more places at the summit that is talked about .. the more results we'll get in our database
17:52:21 <davidlenwell> catherine_d: can you get the submission in today?
17:52:22 <rockyg> maybe we could support catherine_d by supplying some data?
17:52:27 <catherine_d> yes I will
17:52:33 <davidlenwell> you have like 6 hours
17:52:40 <catherine_d> :-)
17:52:54 <davidlenwell> 6 hours and 7 minutes
17:53:09 <davidlenwell> awesome!!
17:53:19 <davidlenwell> anything anyone else needs to talk about?
17:53:23 <zehicle_at_dell> yy
17:53:32 <zehicle_at_dell> board meeting & tests json file
17:53:53 <catherine_d> zehicle_at_dell:  will we have a list of core tests for branchless Tempest soon?
17:53:58 <zehicle_at_dell> yy
17:54:09 <zehicle_at_dell> I've nearly got the complete json ready
17:54:20 <zehicle_at_dell> I've got all the capabilities in a json
17:54:27 <catherine_d> I check RDO against Havana cores.. rght now it is about 60% pass
17:54:28 <zehicle_at_dell> need to load in the tests next
17:54:43 <catherine_d> need to see how the new core list is ..
17:54:50 <zehicle_at_dell> davidlenwell, can you update the topics?
17:55:09 <catherine_d> lots of test are no longer there .. Let me give a new test list
17:55:10 <davidlenwell> #topic tests json
17:55:11 <zehicle_at_dell> #link https://wiki.openstack.org/wiki/File:DefCore_Capabilities_Scoring.pdf
17:55:27 <zehicle_at_dell> last week, the board approved the Havana capabilities
17:55:34 <davidlenwell> yay
17:55:35 <zehicle_at_dell> (designated sections is pending until September)
17:56:00 <zehicle_at_dell> there was some last minute movement on the capabilities as we got scores from the TC and I zero'd out the other 0.5s
17:56:25 <zehicle_at_dell> but basically, no real movement.  some of the capabilities are closer to the "line" than before
17:56:40 <davidlenwell> :)
17:56:50 <zehicle_at_dell> I've been working to turn that into a unifited JSON file that matches the DefCore meeting schema
17:57:01 <zehicle_at_dell> I'm nearly done with it and will have a patch soon
17:57:04 <davidlenwell> So you'll have that submitted for review some time this week?
17:57:11 * zehicle_at_dell knows I've been saying that for a while
17:57:12 <zehicle_at_dell> yes
17:57:18 <zehicle_at_dell> before Wedensday
17:57:18 <davidlenwell> +1
17:57:47 <catherine_d> +1
17:58:02 <davidlenwell> with fully qualified test names correct?
17:58:21 <zehicle_at_dell> yes
17:58:32 <zehicle_at_dell> had a question about that
17:58:41 <zehicle_at_dell> the current schema did not account for paths
17:59:04 <davidlenwell> json scmema or are you talking about the storage db?
17:59:09 <zehicle_at_dell> so I've got some names where the path is includes area.parent/child.file.test
17:59:20 <zehicle_at_dell> json schema
17:59:23 <davidlenwell> zehicle_at_dell: can we offline this in #refstack?
17:59:27 <zehicle_at_dell> yy
17:59:28 <davidlenwell> we're just about out of time
17:59:34 <zehicle_at_dell> need a small break first
17:59:49 <davidlenwell> zehicle_at_dell: that works.. I am communiting to the office after this
17:59:55 <catherine_d> we need to .. I can work with zehicle_at_dell to get the fully qualified name ..
18:00:09 <davidlenwell> awesome!!
18:00:15 <davidlenwell> okay folks.. until next time
18:00:19 <davidlenwell> #endmeeting