15:00:06 #startmeeting third-party 15:00:07 Meeting started Mon Apr 4 15:00:06 2016 UTC and is due to finish in 60 minutes. The chair is lennyb. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:11 The meeting name has been set to 'third_party' 15:00:20 Hello 15:00:28 hello lennyb 15:00:35 thanks for chairing the meeting 15:00:41 anteaya: hi 15:00:48 I had to cancel my vacation so I'm around 15:00:52 but do proceed 15:00:59 how are you? 15:01:03 good thanks 15:01:14 looking forward to actually taking a vacation the next time 15:01:17 how are you? 15:01:31 hi everyone 15:01:44 I am ok, thanks and good luck next time 15:01:56 thank you 15:01:58 hi mmedvede 15:02:03 does any of you something to discuss? 15:02:39 hi 15:02:39 I do not have anything to discuss today 15:02:43 hi asselin 15:02:47 asselin: hi 15:03:15 I have a question. looks like our CI is posting empty comments on 'rebase' #link https://review.openstack.org/#/c/266540 15:03:29 any suggestion why and how this can be 'removed' 15:04:35 lennyb, which one is your ci?' 15:04:48 mellanox ci, look for the last comment 15:05:10 well rebase and recheck both start wtih re 15:05:21 what is your regex to match recheck? 15:05:31 can you paste your zuul layout.yaml for the pipeline? 15:06:11 #link http://paste.openstack.org/show/492885/ 15:07:49 we had similar issue, when no job was triggered due to filtering. This was 'fixed' in zuul by eantishev commit ( if I recall correctly ) 15:09:08 your system posted Build succeeded. which I don't see in the paste provided 15:09:52 that might be the default when none is provided 15:10:10 I wondered if that was the case 15:12:11 anteaya: I agree with asselin, it looks like default value 15:12:55 so on rebase do you want it to run tests, and it isn't running tests? 15:13:09 that appears to be what you want, yes? 15:13:35 or what do you want to do on rebase? 15:13:36 anteaya: rebase does not change code, so I do not want to run tests. 15:13:56 okay so when it rebases, it shouldn't test and it shouldn't comment for your system 15:14:08 I can add trigger event for rebase and run tests, but I dont see any point to do it 15:14:08 so it isn't running tests, but why is it commenting? 15:14:24 no, a rebase is effectively a new patch set 15:15:13 asselin: if rebase is actually a new patch set, then I should add event to be triggered and run tests. 15:15:59 anteaya: exactly my question - my it's commenting if it was not running 15:16:16 well asselin does offer a good point 15:16:50 would you consider testing rebases? 15:17:07 anteaya: I will update my CI. Issue is closed :) 15:17:21 thank you for bringing up the topic for conversation 15:17:40 Any other topics? 15:17:58 I don't have anything I wish to discuss today 15:18:43 I have another one. 15:19:22 * lennyb pasting code 15:20:13 #link http://paste.openstack.org/show/492889/ 15:21:07 I have 2 pipelines that should trigger 2 jobs. But it looks like the second one posts a 'recheck' comment for CI. 15:21:20 what is a better way to implement it? 15:21:48 ?? why does the 2nd one post a recheck comment? 15:22:27 i dont know yet. 1sec, I will find a comment 15:23:57 lennyb: I wonder why you have 2 pipelines. 1 pipelines should normally be enough 15:24:25 or you have different recheck comments for each 15:24:46 mmedvede: one pipeline works great, I just did not want to trigger both jobs 15:25:10 mmedvede: yeap, different recheck comments 15:25:12 that's good point, 2 pipeline would lead to 2 comments per patch set, which isn't allowed (last I checked) 15:25:41 it isnt' allowed 15:25:44 you are correct 15:25:54 asselin: so it's 'by design'. ok I will use 1 pipeline 15:25:57 thanks 15:26:03 lennyb: yes, makes sense to have different pipelines if you have to have different settings for both. But if you only want conditionally trigger a job depending on project, that can be done on the job level 15:26:17 lennyb, you could leave the 2nd pipeline and only include the recheck comment option 15:26:41 i.e. delete lines 9 & 10 from here: http://paste.openstack.org/show/492885/ 15:26:49 right, so I missed that it would give two comments on the patch :) 15:27:23 asselin, mmedvede: thanks. 15:27:25 and make sure your other pipeline doesn't repond...i.e. make them mutually exclusive 15:27:42 but one pipeline is simpler (by design) 15:28:07 Issue is closed. thanks. 15:28:19 thank you 15:28:23 Any other issues? 15:28:29 lennyb: do share your results after your changes 15:29:27 anteaya: it's working. I moved to 1 pipe line as workaround and I will leave it 15:30:45 If there are no other topics I will close the meeting. Any objections? 15:30:55 I do not object 15:31:00 thanks for chairing lennyb 15:31:29 anteaya, asselin, mmedvede: thanks for being here. See you next time 15:31:38 thanks lennyb 15:32:07 anteaya: do you need me tomorrow? I can chair as well, it's a day time here :) 15:33:48 lennyb: yes, that would be lovely thank you 15:33:54 been up all night for 4 days 15:34:00 if I am awake I will join you 15:34:17 anteaya: sure, np. 15:34:22 #endmeeting