19:09:40 <zehicle> #startmeeting Refstack 19:09:41 <openstack> Meeting started Mon Sep 15 19:09:40 2014 UTC and is due to finish in 60 minutes. The chair is zehicle. Information about MeetBot at http://wiki.debian.org/MeetBot. 19:09:43 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 19:09:45 <openstack> The meeting name has been set to 'refstack' 19:09:54 <zehicle> roll call 19:09:55 <zehicle> o/ 19:10:01 <catherine_d> o/ 19:10:10 <juliashapovalov2> here 19:11:39 <rockyg> anyone ping davidlenwell? 19:11:47 <catherine_d> I did 19:12:09 <catherine_d> via IRC ... any other way to ping him? 19:12:38 <rockyg> someone who may have his phone number ;-) 19:19:31 <catherine_d> should we start? It is 20 after .. we can check back with David later .. 19:20:20 <zehicle> +1 19:20:40 <zehicle> I'd like to talk about juliashapovalov2 's recommendation. 19:20:49 <zehicle> I did a pull that reflects her option #1 19:20:53 <davidlenwell> Im here ..s orry .. problem with wirless etwork 19:21:01 <catherine_d> great ... 19:21:24 <juliashapovalov2> hi 19:21:38 <davidlenwell> hi 19:21:52 <juliashapovalov2> seems that uuid issue blocks us a lot and nobody even knows the state of issue discussion 19:22:14 <rockyg> agreed. 19:23:06 <davidlenwell> I can make the discussion more public if that helps 19:23:07 <catherine_d> For the UUID, we need to consider both core list and refstack-client tests 19:23:15 <juliashapovalov2> do we need them for capabilities list only& 19:23:26 <davidlenwell> everything 19:23:40 <davidlenwell> because refstack isn't just the core tester 19:23:50 <davidlenwell> refstack is an interop test tool 19:24:06 <davidlenwell> capabilities extend beyond core by definition 19:24:30 <juliashapovalov2> how client suppose to use them? 19:24:30 <catherine_d> a quick soulution now is to settle on one refstack (which can be update periodically) and use it to generate the core list and for refstack tests 19:24:44 <catherine_d> that can get us miving ? 19:24:49 <catherine_d> moving 19:25:26 <davidlenwell> sure .. if we make a map for a specific sha and used that exclusively 19:25:34 <catherine_d> yes ... 19:25:37 <davidlenwell> but thats also not that useful moving forward 19:25:40 <zehicle> I wanted to make sure that we have a way to track tests for capabilities that can withstand the tests moving 19:25:41 <catherine_d> at least for now 19:26:21 <davidlenwell> zehicle: +1 19:26:32 <rockyg> Would a release label give you what you need? Anything in the test directories with that lable will be what was there at the time of the label. 19:26:35 <juliashapovalov2> failures will let us know) 19:26:36 <davidlenwell> tomorrow at openstack sv .. i will corner some people on this and try to push forward a solution 19:26:53 <rockyg> Kewl. I'll be there, too. 19:27:07 <davidlenwell> rockyg: no .. a release label won't solve the problem 19:27:12 <juliashapovalov2> what solitions are discussed? 19:27:20 <catherine_d> <zehicle> +1 ... but to get us moving let's selltle on using a specific sha for now 19:27:22 <zehicle> per release, we have independent files 19:27:41 <zehicle> catherine_d, yes, I did not make any assumptions about how we'd generate the ID 19:27:47 <zehicle> was expecting us to talk about that 19:27:59 <zehicle> would like them to be <11 chars 19:29:02 <zehicle> my suggestion is to follow juliashapovalov2 option #1. a single master crossref file and that we can use for the capabilities files for each release 19:29:21 <catherine_d> <davidlenwell>what is the differencc between per release vs specific sha? 19:29:37 <davidlenwell> releases can still change paths 19:29:44 <davidlenwell> updates still happen to releases over time 19:29:51 <davidlenwell> its not static 19:29:59 <zehicle> I'm confused. from our perspective the trunk test is always the right test 19:30:03 <catherine_d> ok what we want is sha then ... 19:30:15 <zehicle> we just need a way to track a test even if it's name or path changes 19:30:15 <davidlenwell> zehicle: I can agree to that solution if after tomorrow I still don' thave a solid response from the q/a team 19:30:34 <zehicle> davidlenwell, works for me. I have a pull to review 19:30:49 <davidlenwell> I saw it .. 19:30:55 <catherine_d> <zehicle> the reason is tests sometime are moved from sha to sha 19:32:52 <juliashapovalov2> davidlenwell, please let us know the result once you get it or don't get 19:33:14 <davidlenwell> I am meeting with catherine_d tomorrow after the event .. 19:33:18 <davidlenwell> face to face 19:33:29 <davidlenwell> it will be the primary topic of discussion 19:33:34 <catherine_d> and rocky too .. Rob will you be there? 19:33:36 <rockyg> I'll be at the conf if there's anything you need me for 19:33:46 <zehicle> no, I cannot be there 19:33:54 <davidlenwell> rob isn't in town for this one 19:34:00 <catherine_d> ic 19:34:23 <juliashapovalov2> you can put a note in irc chat 19:34:32 <davidlenwell> I will 19:34:38 <juliashapovalov2> tnx 19:36:18 <davidlenwell> okay .. so given that catherine rocky and I will meet tomorrow .. Do we have anything else to discuss before then? 19:36:35 <catherine_d> could we go through code review? 19:37:07 <rockyg> Critical reviews that need eyes, votes.... 19:37:17 <rockyg> put the links up if there are any 19:37:39 <davidlenwell> sure 19:37:42 <catherine_d> For refstack-client, need decision on this one https://review.openstack.org/#/c/114791/ 19:38:13 <davidlenwell> l thought a face to face about this one would be more productive 19:38:14 <catherine_d> Unit test from Sergey ... 19:38:23 <catherine_d> ok 19:39:02 <zehicle> I'd also like to talk about ubuntu vs centos 19:39:25 <davidlenwell> oh good point 19:39:43 <davidlenwell> catherine_d: we'd like to insure centos/rhel compatibility in the client 19:39:52 <rockyg> ++ 19:40:36 <catherine_d> ya so far we only have setup for Ubuntu .. That would be a work item ... 19:40:45 <davidlenwell> agreed 19:40:54 <davidlenwell> I'll add a story to storyboard 19:41:08 <catherine_d> <davidlenwell> Let put that in storyboard ? 19:42:13 <catherine_d> back to review .. how about https://review.openstack.org/#/c/119826/ and https://review.openstack.org/#/c/120926/? 19:43:18 <davidlenwell> the git ignore one didn't seem critical 19:43:28 <davidlenwell> the other one says not found 19:44:40 <juliashapovalov2> https://review.openstack.org/#/c/120926/ 19:45:18 <catherine_d> thx 19:46:11 <davidlenwell> ahh .. I see 19:46:17 <davidlenwell> I will review it today 19:47:10 <catherine_d> <juliashapovalov2> we will revisit this one after tomorrow's decision? https://review.openstack.org/#/c/115579/ 19:49:13 <davidlenwell> anything else require discussion ? 19:49:38 <catherine_d> how about this one .. should it be merged? https://review.openstack.org/#/q/project:stackforge/refstack,n,z 19:51:10 <davidlenwell> if you are talking about the first thing in that list .. no .. it should be reviewed 19:51:37 <juliashapovalov2> from my side.. guess https://review.openstack.org/#/c/117470/ has higher priority then https://review.openstack.org/#/c/115579/ 19:54:38 <davidlenwell> zehicle and I have an agreement that I'll ignore changes to that file until he or someone form defcore have reviewed them .. which is why I have ignored it 19:54:47 <davidlenwell> get him or joshua to review it 19:54:59 <juliashapovalov2> got it 19:55:14 <davidlenwell> alright folks .. we are basically out of time 19:55:20 <zehicle> yy 19:55:34 <catherine_d> I have a question about https://review.openstack.org/#/c/117470/ . Should there still be a capabilities,json or should the information is now incliuded in havanacore.json? 19:55:44 <zehicle> juliashapovalov2, if you use the existing ID pattern then we can remap them when we have IDs 19:55:44 <rockyg> yah. 19:56:01 <zehicle> havanacore.json is the right onw 19:56:07 <juliashapovalov2> i do 19:56:08 <davidlenwell> catherine_d / rockyg I'll see you both tomorrow 19:56:23 <catherine_d> sure sure 19:56:31 <zehicle> I also added a status field so that we can accept patches to the draft versions 19:56:34 <rockyg> Yeah. Bright and early. I'm not saying I'm gonna be bright, though. 19:56:56 <zehicle> good luck tomorrow all - sorry I'm not there for support! 19:57:17 <zehicle> please let me know if there are discussions about DefCore - we're moving to a board vote Thurs 19:57:36 <catherine_d> <zehicle> that is great .. 19:58:20 <rockyg> See most of you tomorrow. 19:58:32 <zehicle> catherine_d, thanks 19:58:50 * zehicle will miss seeing everyone 20:00:01 <catherine_d> I think that is it for today .... good discussion ... 20:00:27 <openstack> redrobot: Error: Can't start another meeting, one is in progress. Use #endmeeting first. 20:00:45 <catherine_d> we should end our meeting 20:00:45 <zehicle> #endmeeting