16:02:28 #startmeeting openstack_chef 16:02:28 Meeting started Mon Apr 4 16:02:28 2016 UTC and is due to finish in 60 minutes. The chair is sc`. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:02:30 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:02:32 The meeting name has been set to 'openstack_chef' 16:02:44 hi sc` 16:02:46 :) 16:02:54 howdy! 16:03:28 i have a topic i wanna talk about 16:03:41 #chair jklare 16:03:42 Current chairs: jklare sc` 16:03:43 the 2 working sessions and the fishbowl one we have at the austin summit 16:03:51 #topic sessions in austin 16:04:01 sooo 16:04:20 as stated in the mails from ttx, we got all the sessions we wanted and now have to fill them 16:04:42 woot 16:05:11 i think we need at least one of the working sessions to do the actual release work for all the projects 16:05:14 hi 16:05:18 hi markvan 16:05:22 speaking of, my plans are confirmed for austin. i'll be there 16:05:35 sc thats great 16:06:00 o/ 16:06:07 imo we should use the fishbowl one to try to get some feedback on all the refactored stuff 16:06:19 if we announce it early enough with a great title 16:06:25 we might even get some people in :D 16:06:54 for sure. i've heard a couple of "i've seen something about that" but nobody has really seemed to bang around on it in depth. if they are, they aren't talking 16:07:11 i could put some slides together as a introduction to all the refactoring we have done 16:07:36 so we can use them to explain what we did and ask people for their opinion 16:07:59 sounds good to me 16:08:08 but we need an amazing title 16:08:40 how about "The best OpenStack deployment tool got even better!" 16:08:44 :) 16:09:06 might at least provoke "some" feedback 16:09:16 that is catchy :D 16:10:45 i sort of see the cookbooks as filling one of the needs to provide a nearly production-ready deployment system without *too* much fussing with it. approaching it from that angle may be good, so as to not directly take puppet head on 16:11:05 ;) 16:12:07 for example, in my own deployment, i haven't had to diverge a significant amount from how the branch exists upstream. my team has wrapped a lot of stuff around those cookbooks, but the core is mostly the same as github 16:12:48 how about "The Biggest Loser - Running to production with OpenStack Chef" 16:12:54 yeah, the "wrapping" ability is a strong point to make. It was just a bit harder to wrap/extend this in the past 16:13:35 we are nearly finished with our first beta mitaka deployment on xenial and just have to wrap the original upstream cookbooks 16:13:38 markvan: yeah. we've had to contort the wrappers a bit in some cases (like when i added a hacked up server-apache.rb) 16:14:00 jklare: nice! 16:14:39 i think i had enough coffee for today... 16:14:46 i'm just getting started 16:15:00 * sc` vibrates ever so slightly 16:15:20 so markvan any ideas on that title for the fishbowl session= 16:15:21 ? 16:15:43 should we do something provokative to get more people in? 16:16:19 or rather go with something like "Your opinion on the all new and shiny OpenStack Mitaka chef cookbook" 16:16:25 j^2: you lurking? any ideas? 16:17:14 this has really be a revolution of the older cookbooks, so something like "Revolution in OpenStack deployment, Do you agree?" 16:18:00 less provokative but also catchy :) 16:18:11 indeed. that is one way to approach it 16:18:13 because it really is 16:18:23 and people are going to get blindsided after liberty 16:18:28 and its more obvious that it is a feedback session 16:19:17 i would put "Next Level Deployment" also in it :D 16:20:16 it hurts me to say this, but "the next generation of deploying openstack" 16:20:17 but i agree with marks 16:20:35 hehe 16:21:21 unrelated: the jenkins jobs still "succeed", despite converge failing 16:21:32 bring on the Trek'ers....next generation ;) 16:21:46 because the logs successfully save, the job "succeeds" 16:21:46 ^^ 16:22:03 if we get all the Trek'ers in, we might need a bigger room 16:22:07 but well 16:22:40 gotta nerd it up somehow ;) 16:23:18 how about "Living the refactoring dream and still going to production with OpenStack Chef" 16:23:47 since we had this super big discussion about refactoring cycles and stuff 16:24:06 and instead of talking about it, we actually went for it :) 16:24:18 i mean some other projects did too 16:24:20 but still 16:25:01 the refactoring was indeed a major undertaking, and it seems to be working so far, even if i can't build openstack in ci 16:25:53 sc` did not have the time in the last weeks to look at it, but we are using the upstream cookbooks from master and can build a real multi node openstack setup 16:26:02 so might be related to packaging 16:26:04 however, with my repo and common changes, the logs are properly saved pass or fail. it just always "passes", even if converge fails. i'm thinking that's a jenkins artifact 16:26:08 i'm convinced it's packaging 16:26:22 mysql for trusty seems to be a bug that isn't fixed until xenial 16:26:53 centos is a python dependency. i'm going to bump rdo again to latest and see what happens 16:27:36 so, wanna sleep over all the topic suggestions and decide tomorrow evening(EU)/morning(USA) 16:27:41 ? 16:27:52 again, many, many, many, many apologies for spamming the hell out of the channel with all my patches, but we can get everything we need out of ci with them 16:28:45 i do like the refactoring one, jklare 16:29:40 i would give j^2 the chance to also think about it and push some suggestions 16:29:48 maybe: "Living the dream: refactoring and still going to production with OpenStack & Chef" 16:29:58 he is usually quite good with that stuff :) 16:30:15 yeah. that's why i wanted to solicit his input :) 16:30:19 so imho we can decide tomorrow 16:30:29 +1 16:32:20 cool 16:32:36 nothing else from my side 16:32:38 :) 16:32:50 sc`: as for the ci passing when pass fails, I think we need to save the chef client rc and then return that at the end of the job to keep jenkins happy 16:33:14 markvan: maybe i can put it in the rescue part then 16:34:12 i'll poke at it and see what happens 16:34:31 sc`: yeah, something like that should work 16:36:26 that's pretty much all i have 16:37:01 i probably have one or two more patches in this case. gives me a little more to chew on 16:40:12 alright, with that said. i'll give back 20 minutes unless anyone has anything more 16:40:34 nope, thanks for the 20 16:40:44 see you on the channel! 16:40:48 #endmeeting