20:03:56 #startmeeting tripleo 20:03:57 Meeting started Mon Aug 19 20:03:56 2013 UTC and is due to finish in 60 minutes. The chair is lifeless. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:03:58 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:03:59 #topic agenda 20:04:01 The meeting name has been set to 'tripleo' 20:04:07 bugs 20:04:08 Grizzly test rack status 20:04:08 CI virtualized testing progress 20:04:08 open discussion 20:04:12 #topic bugs 20:04:17 https://bugs.launchpad.net/tripleo/ 20:04:17 https://bugs.launchpad.net/diskimage-builder/ 20:04:17 https://bugs.launchpad.net/os-refresh-config 20:04:17 https://bugs.launchpad.net/os-apply-config 20:04:18 https://bugs.launchpad.net/os-collect-config 20:05:19 lifeless: I'm begging debugging bug 1182249 and bug 1183223 today 20:05:21 Launchpad bug 1182249 in tripleo "quantum configuration is overly hardcoded" [Critical,In progress] https://launchpad.net/bugs/1182249 20:05:22 Launchpad bug 1183223 in tripleo "nova-compute.yaml missing parameters" [Critical,In progress] https://launchpad.net/bugs/1183223 20:05:24 beginning 20:05:46 jog0: any news on bug 1184484 ? 20:05:48 Launchpad bug 1184484 in neutron "Quantum default settings will cause deadlocks due to overflow of sqlalchemy_pool" [High,Fix committed] https://launchpad.net/bugs/1184484 20:06:14 I'm fairly sure no-one is on the POC changes yet; but I think we're right on the cusp of that being a tackleable thing 20:06:42 bug 1213967 was filed overnight, I want to attract jog0's attention and confirm it's nova, then submit a revert for that patch 20:06:43 Launchpad bug 1213967 in tripleo "Seed VM filtering out compute resources" [Critical,Triaged] https://launchpad.net/bugs/1213967 20:06:58 cause you know, being able to boot baremetal is a thing 20:07:25 I'll ping markmcclain about bug 1189385 later today 20:07:26 Launchpad bug 1189385 in tripleo "quantum-server hung up it's listening port" [Critical,Triaged] https://launchpad.net/bugs/1189385 20:08:06 Any other bug discussion? 20:08:24 lifeless: re POC bugs, I have one critical Heat bug I need to handle today/tomorrow, and then I think I can tackle them (the other two criticals I'm on are I think mostly "rewrite the bug and demote to High or Medium". 20:08:38 SpamapS: cool 20:08:48 ok 20:08:52 #topic Grizzly Rack status 20:09:04 still ticking along. And see ^^ 20:09:45 anything else on the racks? 20:09:50 erm, rack? 20:10:23 #topic CI virtualised test progress 20:10:44 so last week I had some slowness issues with lxc, but they've disappeared this week, can't replicate 20:11:16 right now I'm working on what we'll need to modify in boot-seed-vm to support lxc 20:11:54 libvirtd support for lxc exists and I got it working, just need to work through making a .xml file that does everything we want in lxc (will be similar to kvm, but needs some modifications) 20:12:19 cool 20:12:24 it doesn't support everything though, no shutdown, just destroy, but that shouldn't be a big deal since it's just the seed vm 20:12:33 do you need any assistance ? 20:12:59 not at the moment, do you use lxc with libvirtd at all? 20:13:14 I have not, though I am aware of the existence of it 20:13:28 others may have though ;) 20:13:37 ok, converting an lxc .conf file into an .xml file is an interesting task, I'll shout if I get stuck 20:13:46 ok 20:13:55 anything else, or next topic ? 20:13:57 oh, and tracking notes here: https://etherpad.openstack.org/tripleobaremetallxc2013 20:14:00 that's it :) 20:14:07 #link https://etherpad.openstack.org/tripleobaremetallxc2013 20:14:12 #topic open discussion 20:14:45 I'd like to get more folk promoted to reviewer 20:14:55 we have ng/spamaps/devananda/my/arata 20:15:03 s/my/me/ 20:15:25 I figure if we have 2 folk in each broad tz - us, asia, eu - we could go to two x +2 to land 20:15:40 without adversely impacting velocity 20:16:06 and that would also help with folk in those broad areas who currently often have to wait for reviews 20:19:35 crickets? 20:19:39 that would require people to start +1'ing/-1'ing who are not already reviewers 20:19:41 Am I still online? 20:19:42 I have not seen much of that 20:20:00 * derekh is interested and will try to spend more time on reviews to help make it happen 20:20:07 yes lifeless u are 20:21:03 at least u were 20:21:12 I am :) 20:21:38 Ok 20:21:44 Are folk generally happy with reviews? 20:22:01 no 20:22:04 Are they prompt enough? Are they useful? Are we catching defects, not [just] making the codebase pretty? 20:22:54 my only gripe is exactly what you brought up.. we need more reviewers 20:23:25 i'm generally happy with the reviews. agree that more reviewers would be nice. 20:24:01 in the neutron area, they are anythin but prompt, they can be helpful, but in the final stages they seem to be a bit more code my way items. 20:24:17 i can devote some time to doing reviews. you don't have to be a "core" reviewer to just +1/-1 right? 20:24:25 slagle: that is correct. 20:24:31 slagle: anyone with a gerrit account can +1/-1 20:24:37 We need a unicode character that represents a bike shed. 20:24:38 k. i'll start looking at some stuff 20:24:48 dkehn: :( I feel for you 20:25:09 lifeless, I know, you are my sounding board 20:25:16 dkehn: how is the neutron + neutronclient + nova patchset looking? 20:25:28 dkehn: last week it looked like nova was ready to land? 20:26:38 lifeless, went through the last round on the neutron and client waiting for someone to review it, not sure where that will go, markmcclain indicated that we are there, but ....., on the nova side, I've got to file a bug 20:26:52 dkehn: cool! 20:27:21 lifeless, gto hit with a nasty alembic process change that cost me some time 20:27:47 SpamapS: you mentioned the other day that you'd like a bit more connectivity with folk; I was wondering - perhaps we should setup 2/3 hangouts a week - one NZ-US friendly, one US-EU friendly, and perhaps one EU-NZ friendly 20:28:01 lifeless, will take to you offline about the russellb comments 20:28:08 dkehn: ok 20:28:18 lifeless, https://review.openstack.org/#/c/31061/ 20:28:32 SpamapS: something explicitly social and chatty vs formal progress like this meeting 20:28:48 lifeless: indeed, I feel that bandwidth is too low between participants and the summit is way too far apart to make up for the loss in cohesion. 20:29:02 SpamapS: I can fedex you some superglue 20:29:30 no thanks, that always ends in something being glued to something it shouldn't be glued to. 20:29:40 SpamapS: lots of cohesion :) 20:29:46 like a pencil glued to the bottom of a desk. :) 20:30:39 so, who would attend such a hangout(s) ? 20:30:47 o/ 20:32:40 tripleo hangout? I'd join in if tz allowed it 20:33:25 I wouldn't mind a larger scale but less frequent too.. 20:33:32 like once a month 20:33:57 Since we seem to have low interest in the meeting, I'll think about some ideas for how to get people more connected and present next week. 20:34:25 well we haven't got ng's attention right now, nor GheRivero, I suspect. 20:34:41 and devananda will be heading off to burning man ~ now, so also likely not tuned in. 20:34:55 I suspect a little socialisation via the dev list might get more interest 20:35:07 I'd be happy to attend some hangouts 20:35:46 ok will revisit the idea next week and/or on dev list 20:37:02 cool 20:37:03 last call for discussion... 20:37:13 In toci land, I had a jenkins jobs triggering on various projects for a week or so a couple of weeks ago, but then it broke because toci fell behind various tripleo changes, would people be interested in getting this reporting on gerrit? 20:37:19 cue crickets 20:37:33 derekh: -yes- 20:37:54 Unfortunatly toci breaks often because its mostly me keeping it uptodate, but if it was visibly reporting that might help out that situation and help catch regresions. 20:37:57 I believe though, that pleia2's work is exactly focused on that 20:38:00 derekh: hell yes 20:38:26 derekh: I'd really quite like toci to be gating on incubator/tie/tht set 20:38:29 derekh: but 20:38:54 derekh: I'm not super happy about the redundancy we have between incubator and toci 20:39:10 ok, I'll get it set up again and we can see how it goes 20:39:13 derekh: [but tools for deployers aren't themselves CI, so making toci the source of truth feels wrong] 20:39:36 lifeless: ya, as much as possible I'm using the scipts in incubator 20:39:51 lifeless: but the devtest instructions themselves are another story 20:40:18 derekh: yeah; I'm nearly finished on the arc of removing API calls from within the image 20:41:08 derekh: at which point I think I'm going to go drink some bubbly :) 20:41:14 my work is sort of a basic subset of toci, I'll certainly be referencing it once I get to the automation phase 20:41:37 (right now just working to get it running manually) 20:41:48 derekh: pleia2: Once BM is gate-tested, I'd surely like a zuul owned cloud hosted tripleo full test story. 20:41:56 that would be very nice 20:42:18 lifeless: would be cool 20:42:30 mordred is also working on some takeovernode stuff for d-g, I have a review to do 20:43:05 the job I was doing was booting seed + undercloud 20:43:22 I'm just doing seed + bmnodes 20:44:08 pleia2: I don't think takeovernode will help [much] in the cloud test story as we need isolated networks 20:44:21 anyways I'll be out most of this week so probably wont get to it until next week. 20:44:28 pleia2: unless/until there are two CI test clouds with full private SDN 20:44:40 derekh: kk 20:44:46 lifeless: yeah, there is that :\ 20:48:54 ok 20:48:56 #endmeeting