16:00:57 #startmeeting Mistral 16:00:58 Meeting started Mon Mar 2 16:00:57 2015 UTC and is due to finish in 60 minutes. The chair is rakhmerov. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:01:00 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:01:03 The meeting name has been set to 'mistral' 16:01:09 hi 16:02:23 hi ! 16:02:47 let's wait for a couple of mins 16:05:47 Hi all, bit late, apps. 16:06:15 hi 16:06:25 ok, let's skip action items review 16:06:40 because we haven't had meetings for a while 16:06:51 #topic Current status (by team members) 16:07:40 my status: keep working on engine refactoring, a few patches are waiting to be reviewed and we need to wait for Dmitri when he says "ok" to merge them in 16:08:02 also worked with Nikolay on finding a solution for the bug with 'join' 16:09:30 status: worked on new YAQL related changes, added YAQL in policies and tried to fix 'join' bug. Also almost fixed bug with incorrectly finishing workflow when we have delayed task 16:11:10 yep, pls see a minor comment from me in 30 sec 16:11:27 my status: along with w4dcode and lakshmi, worked on deploying mistral on munti-node environment and doing stress tests. 16:11:57 :) 16:12:08 ok 16:12:17 NikolayM, rakhmerov thank you very much for quick turnaround on those irritating bugs we found. 16:12:30 no problem 16:12:49 we need to do it even faster 16:13:14 ok, I wanted to bring up a couple of things up to discussion 16:13:16 as for loadtests: we deployed Mistral on ngnix and apache and run it at the load of 100 complex parallel workflows. 16:13:34 yes 16:13:45 cool 16:13:59 At ~500 parallel workflows, it begins to break down both on a single vm and on distr env. 16:14:06 btw, dzimine, we started seeing that issue with oslo.messaging again after Winson's patch 16:14:21 More details on breaking points to come. 16:14:28 ok, good 16:14:47 hm... often or intermittent? in tests? 16:15:57 intermittent 16:16:12 yes, in test 16:16:36 so looks like placing monkey patching into tests/__init__.py doesn't help 16:17:10 ok let's find a place for it so that it doesn't mess apache deployment. a bug? 16:17:13 to track? 16:17:15 it's not a critical urgent issue though but anyway we need to get more details from Winson about what was wrong before 16:17:37 I guess we don't have a bug yet 16:17:55 #action rakhmerov, create a bug for oslo.messaging & monkey patching issue 16:18:13 ok 16:18:21 let's move on 16:18:36 #topic A bug with 'join' 16:18:54 so now we agreed to merge in a temporarily solution 16:19:01 I already +2ed it 16:19:23 so, dzimine, you can take a look and approve it if you and Winson find it ok 16:19:39 Nikolay also added a test that's skipped for now 16:19:45 it's already in. 16:19:49 ooh, ok 16:20:17 so, my question is: do we need to think about it more or we agreed on the solution suggested today? 16:20:23 with versioning 16:20:39 if yes, then I think we could knock it down this week 16:21:30 dzimine, maybe we you could just discuss it with Winson 16:21:35 that's what comes to my mind 16:21:45 because he may have some other ideas 16:22:24 rakhmerov: can you just repost your comment to mailing list so it's easier do dig out... 16:22:34 ok, I will 16:22:57 #action rakhmerov: repost our discussion about how to fix 'join' bug into ML 16:23:06 ok 16:23:23 #topic Changing team meeting time 16:23:29 thanks! it was convincing to me last night but I was a bit brain-tired. 16:23:36 yes! 16:23:57 that's why I think it makes sense to think about it a little bit longer 16:24:01 :) 16:24:30 not because I don't trust you but because it seems like too important decision to make it in a rush 16:24:46 ok 16:25:05 another thing that I wanted to bring up is changing our team meeting time 16:25:39 from now on it'll be difficult for me to attend meetings on Mondays at 16 UTC 16:26:20 so I just want you guys maybe to propose time frames convenient for you 16:26:36 1 hour later will be fantastic. 16:26:46 for me as well 16:27:01 ah, but it WILL become 1 hour later with daylight saving time. 16:27:04 but as far as I understand for Nikolay it's not so good, is it? 16:27:22 so for me 1 hour earlier will work, too (although not so convinient :)) 16:27:25 but I am ok. 16:27:41 yes, for me a hour later is not very good 16:28:17 regardless of day of week, right? 16:28:20 let's move meeting for other day 16:28:32 Tuesday? 16:28:35 yes, the best day for me is Tuesday 16:28:37 yes 16:29:06 same time would be good for me but the problem is we'll have to find another channel 16:29:22 because looks like this channel is occupied on Tuesdays 16:29:50 or, the other alternative is just to have shorter meetings on Mondays 16:30:10 say, from 16.20 till 17.00 UTC 16:30:13 40 mins 16:30:22 if we don't find more convenient time 16:30:47 ok, let me search for available slots and suggest some options 16:31:10 #action rakhmerov: suggest some time slots for a new time of team meetings 16:31:22 #topic Open Discussion 16:31:39 any more topics/questions? 16:31:50 just fyi, for me and Nikolay time after16 UTC is not good 16:32:00 yes, I got it 16:32:14 thanks 16:32:38 it's a challenge because we are all in different time zones 16:32:39 :) 16:32:55 ok 16:33:33 so, I don't have any more topics for today 16:33:47 I'm too deep in coding ) 16:34:05 from your side? 16:35:05 none from my 16:35:25 dzimine? 16:36:56 ok, let's finish then 16:37:00 thanks for coming 16:37:06 bye 16:37:36 #endmeeting