04:00:22 #startmeeting third-party 04:00:22 Meeting started Wed Mar 25 04:00:22 2015 UTC and is due to finish in 60 minutes. The chair is krtaylor. Information about MeetBot at http://wiki.debian.org/MeetBot. 04:00:23 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 04:00:26 The meeting name has been set to 'third_party' 04:00:31 hi 04:00:35 o/ 04:00:37 Hi asselin 04:00:43 hi sweston 04:00:50 hi krtaylor 04:00:52 hi sweston...good to see you again 04:01:06 anyone else here for third party ci working group meeting? 04:01:26 hi asselin .. thanks, I have been buried in switch design goodness :-) 04:06:12 uh oh, looks like our host left 04:06:24 yeah...was wondering what's up 04:07:44 guess we will wait 04:09:50 argh, I dropped connection 04:09:52 what was the last thing I posted? 04:09:54 sweston, asselin ^^^ 04:10:21 we figured .. krtaylor: anyone else here for third party ci working group meeting? 04:10:30 OMG 04:11:00 hehheh, ok, well I said a bunch more :) 04:11:13 cut/paste time 04:11:16 sweston, yes, thanks for joining, I did review the spec 04:11:25 the format for the 0400 meeting is usually a bit more laid back, since there are usually just a few in attendance 04:11:33 sweston I know what you want to discuss :) 04:11:40 so maybe lets start with asselin 04:11:42 ah ok 04:11:56 asselin, do you have anything to bring up? 04:11:58 #link https://review.openstack.org/#/q/status:open+topic:downstream-puppet+owner:%22Ramy+Asselin+%253Cramy.asselin%2540hp.com%253E%22,n,z 04:12:16 got a few patches up for the in tree 3rd party ci. 04:12:34 yes, 4 have merged 04:12:52 started with the log server as it's probably the simplest piece / lowest risk & not changing much 04:13:22 really? 4? I need to take a look and see 04:14:27 I guess the governance changes 04:14:32 oh, actuall more: https://review.openstack.org/#/q/topic:downstream-puppet,n,z 04:14:38 actually too 04:14:43 still working on the last one? all i see is a rakefile and metadata.json 04:15:10 sweston, which one's that? 04:15:30 initial files for the puppet-openstackci repo 04:16:05 there's a README & license. anything else needed for that? 04:16:57 I was expecting a manifest and some modules 04:17:05 not that I can see 04:17:13 no...not yet 04:17:19 that's in the depends on change 04:17:29 gotcha 04:17:37 sorry I mean needed by change 04:18:03 initial populate, I like a smaller patchset 04:18:26 I see now 04:18:28 yeah...I hope I did it right..not a puppet expert 04:18:54 me either, just enough to understand what our system is doing 04:18:57 I still need to test it actually. But that's the starting point. 04:19:21 I will take a look ... I have a new puppet server I have been dying to try out 04:19:35 sweston, great thanks! 04:19:57 asselin, sure :-) 04:20:08 I find it's easier to submit something and then test b/c i want to run it on a clean vm 04:20:14 * krtaylor need to do some more reviews 04:20:41 I guess I could scp from my dev box 04:21:02 I will download your module files and do some testing tomorrow 04:21:29 so that's it...just wanted to say some progress was made 04:21:44 yes, and it has 04:21:52 hopefully this will iron out any issues with the plan forward 04:21:52 good stuff, thanks asselin 04:22:22 speaking of plan, I created an etherpad for topics to discuss 04:22:33 not much there yet, but we have a week or two 04:22:48 #link https://etherpad.openstack.org/p/liberty-third-party-ci-working-group 04:23:22 if you put something there, please put your nick next to it 04:24:17 we'll do like last time I am assuming, load the etherpad up, prioritize and then see if we can get a slot 04:24:51 sweston, asselin, ack? 04:24:57 ack 04:25:00 ack 04:25:07 cool, wondering if I dropped again 04:25:17 sweston, wil you be in vancouver? 04:25:36 I hope so 04:25:37 asselin, I do not know yet if I will make this one .. 04:25:53 krtaylor, asselin, will you be there? 04:26:02 yes 04:26:08 I got approval, so I can say yes for sure 04:26:55 I'll have to talk to infra and see if they have time for us a slot again this summit, with all the discussion on CI, I hope so 04:28:05 ok, so if you all think of topics to discuss, throw them in the etherpad, I'll push out some thoughts there tomorrow 04:28:15 sure 04:28:21 * asselin_ waiting to not be first :) 04:28:30 hehe 04:28:42 asselin_, hehheh, np, I'll jump in 04:29:22 sweston, so I did read the spec several times again and have a few suggestions for a simplified approach for the monitoring dashboard spec 04:29:40 excellent, I'd love to hear them 04:29:50 #link https://review.openstack.org/#/c/135170/ 04:29:58 just for completeness 04:30:10 the dashboard server is back up, btw ... #link http://dashboard.triniplex.com 04:30:26 yes, excellent 04:31:06 sweston, any data showing up? 04:31:08 imo, the accountability throws it off 04:31:36 just a list of servers and operators right now 04:32:08 sweston, so maybe just start by saying that the design is for a way for third party systems to graphically see how their system is running compared to Jenkins 04:32:22 nice, I like that 04:32:28 krtaylor, ++ 04:32:29 that is the criteria that most everyone is looking for 04:32:59 then that removes all controversial topics from the spec, and judgement is not passed 04:33:04 yeah, basically a read-only version 04:33:15 for now 04:33:18 start small 04:33:27 also simplify the implementation 04:33:42 sweston, that's probably what you were going to do anyway 04:33:53 maybe start with describing a gatherer and a default UI 04:34:09 ok, I will modify accordingly 04:34:20 maybe expand to discuss mysql backend for historical data 04:34:34 but you can pro drop the API service for now 04:34:36 asselin, yes was going to start with initial data collection and go from there 04:35:04 cool, I'll be happy to add all of my comments to the spec, but I wanted to discuss them first 04:35:09 so the rest of the discussion, which are good ideas, can go in a new spec, v2 04:35:18 agreed 04:35:32 krtaylor well, I can leave it out of the spec, but that is built into the design .. the api is where all of the other calls go through, it holds the logic for operation 04:36:16 similar to the way which storyboard is built, as per jhasketh's suggestion on one of the early drafts of the spec 04:36:16 ok, that's fine, I don't see that as a show stopper 04:36:27 yes, exactly 04:36:43 ok 04:37:29 I will also remove the data collection from the gerrit stream, it complicates things too much 04:37:58 the vast majority of comments were around the criteria and classifying failures 04:38:04 yes, good idea 04:38:09 anything to simplify 04:38:27 once approved we can add/change 04:38:44 yes, the accountability is really what is complicating this 04:39:15 but I would also like to see mention of infra hosting the dashboard 04:39:37 krtaylor, why? 04:39:59 additionally, I would like to remove the section on identification of a system which may be malfunctioning 04:40:06 sweston, +1 04:40:22 sweston, yes, thats what I meant 04:40:38 that deserves it's own spec 04:40:46 krtaylor, ok, just making sure we are on the same page 04:40:57 I'll add comments, it will make it more clear, hopefully 04:41:01 asselin_: +1, yes, it does 04:41:04 asselin, yes 04:41:14 oops asselin_ 04:41:30 one is your proxy I presume? 04:41:50 * asselin_ is at home 04:41:56 on a different irc client 04:42:04 good to know 04:42:41 I can save my notes to create a second spec somewhere down the line, but for now should we agree to start with a simpler version of this one? or should a second spec be referenced in the first? 04:43:08 also, sweston we should get patrickeast involved here too, he had some good ideas 04:43:35 sweston, either abandon this one and start a more simple version, or just strip this one down first 04:43:51 I don't see a need to reference the second in the first, other than to say, "these topics are out of scope and can be addressed in another spec" 04:43:57 sweston, I vote for the second that way it maintains the history 04:44:07 asselin_, agreed 04:44:12 agreed 04:44:39 ok, I think that gives me plenty to work with, new spec will be out tomorrow for reviews 04:45:04 sweston, if ok with you, I'll defer comments until tomorrow morning, I will make more sense after a nights sleep 04:45:19 krtaylor, sure thing 04:45:56 thats all I had 04:46:04 did you all have anything else? 04:46:13 I am done, as well 04:47:24 ok, well I'll wrap this one up then 04:47:35 unless there is anyone else lurking? 04:47:55 thanks for the input, krtaylor, asselin 04:48:07 you bet, glad to see you back 04:48:20 yes 04:48:38 thanks you all, good meeting! 04:48:45 goodnight 04:48:54 Night! 04:49:00 #endmeeting