17:02:27 #startmeeting murano 17:02:28 Meeting started Tue Oct 6 17:02:27 2015 UTC and is due to finish in 60 minutes. The chair is kzaitsev_mb. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:02:29 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 17:02:31 The meeting name has been set to 'murano' 17:02:35 HI 17:02:46 #topic rollcall 17:02:48 0/ 17:03:00 hi 17:03:22 Hey! You're late =) 17:03:27 hi 17:04:25 not much for agenda today =) serg and stan won't be able to attend. 17:05:17 #topic Action Items Review 17:05:54 1) release new version of the client and propose change to global-requirements 17:06:45 I believe we did, but it's not possible to add clients to global requirements 17:08:13 and 2d) we're still working on app-catalog integration. ativelkov & I are working on that one. 17:09:08 #topic Liberty release 17:10:08 o/ 17:10:16 kzaitsev_mb: good to know 17:10:29 global openstack releses on Oct 15. 17:10:53 Let's review any bugs we still have for L release, shall we 17:11:45 #link https://launchpad.net/murano/+milestone/liberty-rc2 17:12:36 I seee 2 bugs, still up for l2 and L in general 17:12:48 bug 1502589 is clear and I have a solution for it 17:12:48 bug 1502589 in murano mitaka "Deadlock in DB on package deletion" [High,Confirmed] https://launchpad.net/bugs/1502589 - Assigned to Alexander Tivelkov (ativelkov) 17:12:52 will try to commit tomorrow 17:13:01 k, great! 17:13:32 #link https://launchpad.net/bugs/1502589 17:13:46 #info ativelkov is about to fix 1502589 17:13:53 however, bug 1502437 is trickier: I still cannot understand why it happens. If the reason is the amount of environments, then it is actually not a bug 17:13:53 bug 1502437 in murano mitaka "Murano picks cidr for subnet which is already used" [Critical,New] https://launchpad.net/bugs/1502437 17:14:25 we have a limit on number of environments which may co-exist in a tenant (i.e. share the same murano router) 17:14:35 by default that limit is 20 17:14:39 ativelkov: AFAIK, this bug reproduced under heavy load... 17:14:50 freerunner: yup, that was a rally test 17:15:22 however, the test was supposed to be run in several tenants, so it is unclear how many environments it creates per tenant 17:17:18 I have detailed logs, will need to analyze it 17:17:24 Should it be critical then? 17:17:39 kzaitsev_mb: I don't think so 17:17:48 cause, looks like you need specific conditions to reproduce it. 17:18:12 yup: it's actually a race condition for a cidr 17:18:51 wonder what Serg would say, but looks like we can downgrade Importance 17:20:00 kzaitsev_mb: Agreed. 17:20:49 And looks like there is no easy way to eliminate that race condition, cause murano puts cidr to heat, so there is a chance that 2 murano envs would request heat to create networks with same cidrs 17:21:21 and the error would not come up, untill both of the stacks are launched 17:22:42 so looks like there is no easy way for an env to know, that someone requested the same cidr =/ 17:25:34 ok, the bug is definitelly there. I believe, that we should downgrade it to High, and it would not hurt us to fix it after the release. And probably backport it to j/k/l if we find a suitable solution. 17:25:38 huh 17:25:49 Serg was the one, who marked it as Critical 17:25:50 =) 17:26:18 so we'll need to convince him tomorrow anyway, that the bug is not that Critical =) 17:27:04 #action discuss https://bugs.launchpad.net/murano/+bug/1502437 severity with Serg, continue considering fixes. 17:27:05 Launchpad bug 1502437 in murano mitaka "Murano picks cidr for subnet which is already used" [Critical,New] 17:27:40 #link https://launchpad.net/murano-apps/+milestone/liberty-rc2 17:28:42 in the apps we have a couple of k8s bugs, assigned to Stan, I believe I even saw a couple of fixes on review, so hopefully they'll be through. 17:29:13 so yep, looks like nothing to discuss there =) 17:30:47 #link https://launchpad.net/python-muranoclient/+milestone/0.8.0 17:32:20 a couple of bugs in progress and a medium bug about CLI. 17:33:32 doesn't look like there are any blockers anywhere across our repositories =) 17:35:46 thats cool 17:36:30 #topic Open Discussion 17:38:53 Looks like we've done nice job fixing bugs =) 17:43:43 k, so if no one comes up with any topics — I'll end the meeting early =) 17:43:55 in a couple of minutes ) 17:50:01 k, thanks everyone! =) see you in #murano and next week. 17:50:12 #endmeeting