14:00:16 #startmeeting tripleo 14:00:17 Meeting started Tue Mar 15 14:00:16 2016 UTC and is due to finish in 60 minutes. The chair is derekh. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:22 The meeting name has been set to 'tripleo' 14:00:25 o/ 14:00:27 o/ 14:00:30 o/ 14:00:30 o/ 14:00:35 Hi 14:01:02 hi 14:01:21 \o 14:01:23 o/ 14:01:32 Hi all, dprince isn't around today so I'm gonna try and not mess this up ;-) 14:02:03 derekh: you have all our support ! 14:02:03 hi 14:02:13 #topic agenda 14:02:14 o/ 14:02:20 * bugs 14:02:20 * Projects releases or stable backports 14:02:20 * CI 14:02:20 * Specs 14:02:20 * open discussion 14:02:30 That is our agenda this week. Anything else to add/subtract? 14:02:35 shardy isn't able to make it bug want us to discuss the timeline re mitaka branching, so we'll discuss that in "Projects releases or stable backports" 14:02:55 anything else ? 2 minutes 14:03:12 o/ 14:03:48 #topic bugs 14:03:57 * dprince is in another meeting, but lurks if needed 14:04:09 ok, any bugs need to be getting attention ? 14:05:54 ok, looks like nothing major to be brought up here 14:06:01 * Projects releases or stable backports 14:06:08 #topic Projects releases or stable backports 14:06:15 Let discuass projects needing releases first 14:06:15 then liberty stable branhces 14:06:15 and finally Mitaka branches 14:06:33 So do any projects need releases ? 14:06:59 is this mic on ? 14:07:00 i planned on doing liberty releases today 14:07:59 slagle: ok, this that of a particular project/s ? or a bunch of them ? 14:08:07 we will need to do master releases after we branch mitaka 14:08:11 derekh: anything needing it 14:08:31 so that we bump the major version on master 14:08:31 #action slagle to do liberty releases where needed 14:08:46 slagle: make sense 14:09:03 #info master releases needed once we branch mitaka 14:09:27 i think we can branch mitaka this week, if there are no objections 14:09:47 we still need to land aodh/gnocchi on master first though 14:09:59 so befor moving onto the timing of the mitaka branch anything else need to be brought up about liberty ? 14:10:17 branching mitaka this week sounds very early 14:10:52 Ok, moving on to mitaka so 14:11:01 This was the list of blockers we had for creating the Mitaka branches https://etherpad.openstack.org/p/tripleo-mitaka-rc-blockers 14:11:33 hows it looking now, somebody(slagle?) mentioned yesterday aodh was the only outstanding thing, if so is it close to merging so we can branch? 14:11:55 aodh is close to merging 14:11:55 I'm wondering if getting IPV6 testing in CI is a blocker to branch Mitaka 14:12:08 the dependent patch has passed CI a few times 14:12:28 #link https://review.openstack.org/#/c/289435/ 14:12:39 the depends-on aren't passing though 14:13:11 i think there are some circular deps, so we might need to merge it all at once 14:13:22 We should also merge https://review.openstack.org/290570 before branching. 14:13:29 EmilienM: the IPv6 gate passed the overcloud creation, now it has a weird error in the after deploy test with image upload. 14:13:39 EmilienM: so I'd say it's not too far from working. 14:14:02 slagle: ack, if the patches are sane, I'm happy to merge together 14:14:37 bnemec: +2 14:14:40 are we going to manage keystone resource with Puppet in stable/mitaka ? https://review.openstack.org/#/c/244162/ 14:14:58 I don't think CI needs to block branching. We don't branch tripleo-ci so we can keep working on CI jobs after the branch happens. 14:15:00 it's to me, a big change and we need to whether decide if we wait for it or not 14:15:08 bnemec: right. 14:15:11 Unless lack of a CI job is blocking a functional change merging. 14:15:37 o/ 14:15:55 EmilienM: that is a good question. someone would need to drive it forward getting it landed 14:16:06 AFIK the blocker to land it was upgrade 14:16:11 Yeah, the patch is broken right now. 14:16:20 I can work on it (again) if you want 14:16:22 EmilienM: RE "branching mitaka this week sounds very early" , I disagree here now that RDO has a mitaka repository we should us it, if we don't we'll be developing mitaka tripleo against master openstack 14:16:43 derekh: I was just comparing with other OpenStack projects 14:17:17 bnemec: I can try (again) to take over this one 14:17:31 personally, i think we could branch without https://review.openstack.org/#/c/244162/, and consider allowing an exception to backport that if we get it working 14:17:41 it's not really a feature 14:17:45 not really a bugfix either 14:17:55 it's a "move" 14:18:05 it will help a lot in composable roles 14:18:18 well that won't be in mitaka 14:18:54 ok, so we don't wait for it 14:19:24 #action EmilienM to work on Manage keystone initialization directly in t-h-t manifests https://review.openstack.org/#/c/244162/ but don't expect to have it in Mitaka 14:19:27 slagle: ok ^? 14:19:30 wfm 14:19:44 Then we definitely need https://review.openstack.org/#/c/290570/ before we branch 14:19:52 trown|ooobootcam: any update on getting an updated consistent repo? 14:20:03 fwiw, rdo mitaka was cut from the consistent repo 14:20:08 which doesnt work with tripleo right now 14:20:32 we do however work with current! b/c i tested it yesterday 14:20:58 so we are in a weird state right now, the periodic ha job will never pass right now as is 14:21:18 slagle: ooo lemme check why we got no consistent repo 14:21:41 yesterday trown|ooobootcam said the delorean builder was down 14:21:51 an in fact, even current hasn't been updated since the 13th 14:22:06 anyway, we should sync with rdo folks on this, we can move on 14:22:28 slagle: yup, its still down, I'll ask on rdo after the meeting 14:23:18 so is that an agreement, we can branch once aodh and the deprecated warning are merged ? 14:23:27 +1 from me 14:23:45 +1 14:23:55 we're in agreement! :) 14:24:09 +1 too 14:24:19 * derekh wonders what quorum is 14:25:13 o/ 14:25:36 +1 from me too 14:26:29 #info we're free to branch mitaka once aodh and and deprecated message have been merged, then patches merged into mitaka should be fixes only 14:26:49 accurate ^ ? 14:27:00 nice, sorry a bit distracted 14:27:02 +1 14:27:03 yes 14:27:44 ok, once that happens we'll need to figure out who does the branching but I think we can coordinate that on our usual channel 14:28:24 #topic CI 14:28:33 We've bumped up the RAM in CI on both the undercloud and overcloud nodes by 1G last week 14:29:09 Some of us also spent a bit of time trying to dubug some of the errors we were seeing, looks like the addition of extra nics may have tickled a bug in our test env hosts 14:29:51 I believe this will fix it, I'll deploy the fix either tonight or tomorrow night https://review.openstack.org/#/c/292480/ 14:30:06 Besides that I think overall our pass rate has increased to something more acceptable (although still not perfect) 14:30:38 the ceph job has been re purposed to be the new upgrades job and has been renamed, merge this so it appears on the cistatus page https://review.openstack.org/#/c/292359/ 14:30:54 Also the patch to start using a mirror server is still hanging out there, https://review.openstack.org/#/q/topic:mirror-server 14:31:02 would be nice to get some eyeballs on it, it should increase reliability and speed. 14:31:15 that all I got on CI anybody got anything else ? 14:31:17 merged https://review.openstack.org/#/c/292359 14:31:27 slagle: thanks 14:32:14 ok, moving on 14:32:16 #topic specs 14:32:56 Would anybody like to draw attention to any specs ? 14:34:38 nope, 14:34:49 #topic open discussion 14:35:13 just an FYI, michchap and I are working on composable roles 14:35:29 we have a PoC : https://review.openstack.org/#/c/292712 14:35:38 EmilienM: cool 14:35:42 and https://review.openstack.org/289459 / https://review.openstack.org/289466 14:35:50 (just the puppet side for now) 14:35:56 we are working on a first iteration where we move out Puppet code from THT to profiles in puppet-tripleo 14:36:19 #info PoC for composable roles https://review.openstack.org/#/c/292712 https://review.openstack.org/289459 / https://review.openstack.org/289466 14:36:35 another FYI - tripleo ui is about to get included into upstream tripleo, we are very excited about that: https://review.openstack.org/#/c/291246/ 14:36:37 we are doing a "service by service" iteration 14:38:34 akrivoka: great job, I wonder if there is anything we need to do to include it in one of our CI jobs, probably worth discussing at some stage, 14:38:54 EmilienM: sounds like a good way to do it 14:39:07 derekh: ultimately yes, but I think we're not there yet 14:39:37 akrivoka: yup, fair enough we can visit the idea once its at the point 14:39:47 *that point 14:39:49 derekh: definitely 14:40:11 derekh: when we do, it sounds like akrivoka is becoming the expert :) 14:40:21 #info FYI - tripleo ui is about to get included into upstream tripleo https://review.openstack.org/#/c/291246/ 14:40:28 ;-) 14:40:59 nice work here! 14:41:20 ok, if there is nothing else, I'll close this early, 2 minutes to get in 14:43:25 ok, thanks all 14:43:27 #endmeeting