17:00:48 #startmeeting third-party 17:00:49 Meeting started Tue Aug 4 17:00:48 2015 UTC and is due to finish in 60 minutes. The chair is krtaylor. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:50 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:00:53 The meeting name has been set to 'third_party' 17:01:16 anyone here for Third party CI Working group meeting? 17:01:27 hi 17:01:56 hi asselin_ 17:02:00 hi 17:02:12 o/ 17:02:34 o/ 17:02:42 hi marcusvrn mmedvede rfolco 17:03:05 so here is the agenda for today: 17:03:08 #link https://wiki.openstack.org/wiki/Meetings/ThirdParty#8.2F4.2F15_1700_UTC 17:03:26 basically a carry-over from last meeting 17:03:42 mainly because I was on vacation all last week 17:03:58 * asselin_ was also on cavation last week 17:04:13 ah, good to know 17:04:14 well...not in a cave :) 17:04:22 hehheh 17:04:26 I was on vacation last month....hehehe 17:04:38 looks like we landed some good patches and contributions in the repo 17:04:43 good stuff 17:05:05 so does anyone have any quick announcements? deadlines approaching? 17:05:39 just an fyi, recent upper-constraints update broke f21 gate 17:05:45 both us and upstream 17:05:54 yes, thanks mmedvede 17:06:02 well, upstream was broken already, but it broke it further :) 17:06:26 anything else? 17:06:45 #topic Common CI 17:06:56 downstream puppet continues to move along 17:07:17 anything you'd like to work through here asselin_ mmedvede 17:07:21 yes, nodepool refactoring is next 17:07:31 lots of progress has been made 17:07:53 would be good to focus reviews on those...it's a big change 17:08:37 #link nodepool patches to review are on etherpad: https://etherpad.openstack.org/p/common-ci-sprint 17:09:57 #link https://review.openstack.org/#/q/topic:downstream-puppet,n,z 17:10:56 yes, the topic has a lot more changes 17:11:09 was looking, lots of good work 17:11:42 it would be good to focus on nodepool since it is a big change 17:11:58 yes and mmedvede, yours look to be reviewed and ready to go, just need approval 17:12:23 krtaylor: mine a low priority. But yes, a couple of them are ready 17:13:21 lots of depends-on links 17:13:31 for nodepool, thats good 17:13:43 mmedvede, I have those on my list too, but have been prioritzing nodepool 17:15:07 ok, anything else here, other than do reviews? 17:16:02 ok, we can come back to it if needed in open discussion 17:16:12 i think that's it 17:16:21 #topic Spec for infra hosting a monitoring dashboard 17:16:45 this has got some good reviews, not for content, which is a good sign 17:17:42 jhesketh didn't like the naming, which I can certainly change, but was hoping to just have the other spec moved by now 17:17:45 #link https://review.openstack.org/#/c/194437/ 17:18:42 I did appreciate him not holding it up for naming in the rollcall, that was cool 17:18:52 sweston, are you around? 17:19:11 I didn't get a chance to follow up last week due to vacation 17:20:16 ok, well, I'll ping him later - maybe we just need to move it with a new patchset to third-party-ci-tools 17:20:43 and capture the comments in a txt file 17:21:32 then we can abandon the infra spec 17:22:53 so, also in the agenda and related is the last comment dashboard got merged to third-party-ci-tools 17:23:26 patrickeast isn't in channel 17:23:43 #link last-comment https://git.openstack.org/cgit/stackforge/third-party-ci-tools/tree/monitoring/lastcomment-scoreboard 17:24:00 I took an action to start an email discussion on which way to go - scoreboard or last comment 17:24:09 I didn't get to that before vacation 17:25:13 I personally feel like we should just stay with scoreboard for now, but I don't have a strong preference 17:25:23 I just want to get something working now 17:25:27 yesterday 17:25:27 unrelated, could people with +2 take a look at https://review.openstack.org/#/c/194490/ and approve maybe (scoreboard square stats, instead of piecharts) 17:25:54 mmedvede, will do 17:26:04 thx 17:26:26 also, I feel that waiting on an email discussion would slow this whole process down again 17:26:51 but, we can, if it needs to happen 17:27:18 ok, not much input here so I'll go to the mail list 17:27:38 next then 17:27:41 Do you think it would help if we throw together scoreboard POC puppet deploy? 17:28:11 and have it depends-on the spec? yes I do think that woul dhelp 17:28:15 would help 17:28:16 mmedvede, I think so, would be valuable regardless fo those setting it up 17:28:25 ++ 17:28:57 asselin_: cool. I would look into creating module on github, or inside the third-party tools repo 17:29:28 hm, repo is fine, I can refresh spec with a link 17:29:41 +1 for third-party tools repo 17:29:59 mmedvede, that would be very helpful 17:30:33 asselin_: I am not clear on what would happen if infra would deploy it. Would they split out puppet module, or just use it as is out of third-party repo 17:31:16 that is the only concern for creating a separate repo for the module 17:31:22 that would be a guide template I would think, it would be a new module 17:31:28 mmedvede, yes tha's a good point. 17:31:45 so if it's in github already, you can just import it. 17:31:56 right 17:32:11 ok, I change my vote to new github repo :) 17:32:20 I assume they would want control over it, and would not be ok to deploy from third-party tools :) 17:33:32 mmedvede I guess I was thinking it would be used and moved, just a place to put it while spec was approved 17:33:39 separate github repo needs to happen, it is just a question whether we start in third-party tools, and then cut it out 17:33:49 exactly 17:37:14 ok, so, I guess it can be fixed any way it is done, mmedvede what ever you want to do would be great 17:37:19 #action mmedvede look into creating puppet-scoreboard module 17:37:31 +1 mmedvede decides :) 17:37:37 +1 17:37:39 thanks :) 17:38:17 thank you mmedvede , that will help - this is really important to start improving the trust in CI systems 17:38:25 +1 17:38:47 anything we can do to make this as painless for infra to implement is a plus 17:39:00 anything else on this? 17:39:41 #topic TPCI Repo patches for review 17:39:55 so I added a topic this week, that I like to see continue 17:40:21 that is to remind everyone to go review our patches 17:40:53 #link https://review.openstack.org/#/q/project:stackforge/third-party-ci-tools+status:open,n,z 17:41:13 feel free to add patches that you feel need a broader audience 17:41:32 like the one you brought up mmedvede 17:42:03 if time permits, we can review/approve them in this meeting 17:43:23 so, lets come back to that 17:43:38 #topic Open Discussion 17:43:41 anyone? 17:44:02 else, I am looking at: 17:44:06 #link https://review.openstack.org/#/c/194490/ 17:45:19 unfortunately amazon scoreboard instance is down 17:45:32 yes, mine is not coming up 17:46:15 the code is working. I did not want to ninja-merge it all on my own 17:46:44 mmedvede, appreciated, I agree with your comment 17:46:47 I asked to use bar charts instead of pie charts 17:46:51 but it would be nice to see it 17:47:31 pie is the "legacy" way of doing it, for those that remember radar 17:47:58 I don't have a preference really 17:48:32 mmedvede, did it not look right? 17:48:57 piecharts actually just a bad type of chart to use. 17:49:17 hard to compare 17:49:21 hehheh, ok, fair enough 17:49:31 research is on my side :) 17:49:57 yeah, that is not really important in the end, as long as we have something working 17:50:05 +1000 17:50:34 patrickeast may be on vacation, I'd hate to hold this up 17:50:54 if you have seen it and it is close enough, I'll approve it 17:51:04 I guess we could deploy it internally :) 17:51:56 I don't have it deployed yet, but I will review it still 17:52:26 yes, I saw the patch work. I can switch to patch commit internally and test 17:52:27 mmedvede, how hard is it to update our internal with this patch? 17:52:36 krtaylor: very easy 17:52:39 ++ 17:53:52 ok, we are getting close to time, anything else from anyone? 17:56:28 thanks everyone! 17:56:34 thanks krtaylor 17:56:41 thank you krtaylor 17:56:45 #endmeeting