17:08:29 #startmeeting Rally 17:08:30 Meeting started Tue Jul 1 17:08:29 2014 UTC and is due to finish in 60 minutes. The chair is boris-42. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:08:30 hi 17:08:31 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:08:33 The meeting name has been set to 'rally' 17:08:36 olkonami 17:08:37 hi 17:08:40 rediskin 17:08:43 hi 17:08:46 hi 17:10:17 Okay let's just start discusssion 17:10:23 #topic stress runner 17:10:28 olkonami pls go on 17:11:10 I'll send new patchset with fixed comments in a few minutes 17:11:51 olkonami anything to discuss? 17:12:13 yes, my questions from rally chat 17:12:28 I think it's better to specify "max_failure_rate" in percents. It will be more clear and it is done so in sla patch. Are you agree? Also about this parameter name, it is better to indicate that it is associated with a stop and that it is different from sla "max_failure_percent". May be something like "stop_failure_percent", what do you think? 17:14:21 olkonami stop_on_failure_percent 17:14:56 olkonami probably ) 17:15:24 olkonami btw 17:15:34 olkonami ah no 17:16:04 olkonami so how user will setup? 17:16:09 olkonami 0.10 ? 17:16:13 olkonami or "10" 17:16:19 olkonami or "10%" ? 17:16:34 I think 10 17:16:42 will be more clear 17:17:08 olkonami I don't have strong opinion on that 17:17:14 olkonami so I'll be ok with both 17:17:33 that it is percents shold be specified in parameter name 17:17:48 or what do you mean? 17:17:49 olkonami ok 17:18:02 olkonami hm no I am ok with 10 17:18:19 olkonami and stop_on_failure_precent: x 17:18:53 Okay so let's move 17:18:58 #topic RPS runner 17:19:03 oanufriev hey there 17:19:09 oanufriev any update? 17:19:11 yo! 17:19:21 \o 17:19:41 i sent the patch for review and got -1. 17:20:02 have a question about one comment 17:20:17 https://review.openstack.org/#/c/102363/6/rally/benchmark/runners/rps.py,cm 17:20:28 here it is - the second-one 17:21:40 should i reuse implementation from glance? 17:22:16 oanufriev ? 17:22:37 oanufriev what implementation? 17:23:17 implementation of rate limitting of requests 17:23:24 oanufriev I am not sure 17:23:33 oanufriev actually I am sure that we shouldn't use this 17:23:43 oanufriev we don't need speed limit 17:24:01 ok. got it 17:24:19 oanufriev olkonami I will try to find time to review your patches 17:24:31 oanufriev olkonami sorry currently I am quite busy with osprofiler 17:24:36 #topic SLA 17:24:43 rediskin so is it finish finally? 17:25:12 boris-42: this patch got +2, so i think it quite ready 17:25:19 https://review.openstack.org/#/c/98158/ 17:25:21 rediskin okay I'll take a look 17:25:27 rediskin what about functional tests ? 17:25:33 #topic func tests 17:25:36 but there is a lots of criteria to be added 17:26:08 rediskin for example? 17:26:46 only two criteria done, and i believe it is not enough 17:27:03 people will want more criterio =) 17:27:04 rediskin any examples? 17:27:17 maximum time per specific atomic action 17:27:33 rediskin ouh ok 17:27:35 rediskin yep 17:27:45 rediskin zoo okay we will work on that 17:27:59 rediskin so what about function tests in rally-cli job? 17:28:27 it finally works 17:28:46 rediskin so we can result script that made k4no 17:28:52 you can see results in comments for this patch https://review.openstack.org/#/c/102899/ 17:28:52 rediskin to display results btw 17:29:02 job called check-rally-dsvm-cli 17:29:19 rediskin where is HTML?? 17:29:24 =) 17:29:38 no html yet 17:30:08 rediskin so you can rest from tempest in future 17:30:11 i focused around helper methods for tests 17:30:15 and tests itself 17:30:19 rediskin okay I will review it 17:35:23 #topic open discussion 17:35:31 so do we have anything to discssu? 17:35:36 discuss? ) 17:36:50 so seems like we can end it=) 17:36:53 woohoo=) 17:40:42 oh 17:40:55 https://review.openstack.org/#/c/103706/ 17:41:56 -3k lines =) 17:52:11 #endmeeting