15:33:24 #startmeeting OpenStack Chef 15:33:24 Meeting started Thu Feb 19 15:33:24 2015 UTC and is due to finish in 60 minutes. The chair is markvan. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:33:26 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:33:28 The meeting name has been set to 'openstack_chef' 15:33:55 topics for today: juno update, notes from JJ, open discussion 15:34:07 other big topics? 15:35:03 interesting that testing-stack's berksfile is happy, while openstack-chef-repo is not, even though they appear mostly identical 15:36:24 yup, somethings wrong. I have not had time to dig much yet. and now we have this new openssl/postgresql issue. 15:37:27 at least chefdk is back to being happy after purging my system of rvm 15:38:12 ah good. adding a topic for Berks dependencies 15:38:22 #topic Juno update 15:38:41 #link https://groups.google.com/forum/#!topic/opscode-chef-openstack/VGOoOOEOpwg 15:39:05 I updated the list of whereI think were at for Juno. It's getting close. 15:39:47 Just a couple of patches and possibly looking at the mariadb support. 15:40:26 Big step will be changing the repo to the new testing suite. 15:40:50 Any comments on remaining must have Juno work? 15:40:54 does openstack-chef-repo officially have an expiration date? 15:41:01 ^yet 15:41:49 The idea was to branch off the existing repo and replace with testing suite. Timeframe was "any day now" 15:41:58 gotcha 15:42:20 so folks could still use that branched version and make changes to it while getting used to the new suite. 15:43:58 It seems like MySQL is working on centos with jj's patches, wondering if we are still a go on implementing the Mariadb support also. 15:45:04 I know mariadb is being specifically targeted for rhel7, but how about percona? 15:45:25 probably a "nice to have" more than anything 15:45:30 cmluciano: good question. From our teams view, we want to use the platform native database, and that would be mariadb for centos/redhat7. So, I for one would vote to get that support in as the default/native 15:45:30 path 15:47:42 to clarify, for redhat 7, I would like to see mariadb support as the default. for centos, whatever works well is good for me. 15:48:01 I don't know much about percona 15:48:35 this is new. aio_nova fails in the vagrant initialization. you guys using vagrant 1.6 or 1.7 with testing-stack by chance? 15:49:12 1.7.2 15:49:52 When I get a chance today, I'll try out JJ patches for centos 7. 15:50:04 #topic JJ notes on the testing repo 15:50:19 ok, so long story short, AIO-nova works with CentOS7 15:50:19 testing stack is back in a good state 15:50:19 mysql v6 is working with Ubuntu 14.04 15:50:19 both multi and AIO 15:50:19 I’ve started investigating a ceph testing stack 15:50:21 in turn having to learn ceph... 15:50:23 markvan: yeah. trashed berksfile.lock, ./cookbooks and ~/.berkshelf/cookbooks. gist is here: https://gist.github.com/scassiba/e314ae40eaa4494d1d0c 15:50:26 shits cool, but i need to get in the correct mind set 15:50:28 the reviews need a little love, we have some stale code 15:50:30 i have and am ready for the stable/juno branch, i started writing some notes on it 15:50:32 https://wiki.openstack.org/wiki/OpenstackChefStablebranchCreateNotes 15:50:34 as soon as we decied what SHA is stable/juno we can easily create it 15:50:37 i’d like some more +1s and confirmations that https://github.com/jjasghar/chef-openstack-testing-stack/pull/33 is good 15:50:39 and also like someone to spot check the multi-nova issue and see if you can recreate: https://gist.github.com/jjasghar/4d2d184fa23b2575815a 15:50:45 quotes from JJ ^^^ to log in meeting minutes 15:51:24 I'll update my vagrant to latest and see what's good 15:52:26 If we get good testing results with U14 with multi and aio, and use his patch to also get aio wth centos 7, I think with that level we should go ahead and replace the current repo with this. 15:53:05 and objections? 15:53:29 +1, but users will need to be patently aware of how the changes affect them. openstack-chef-repo doesn't spin up vms and such 15:54:15 ideally they'll less README.md, but still :) 15:54:22 sc`: good point. That's why branching off the current repo is important. I guess we can add some addition intro words in the readme 15:54:24 +1, I think a note about deprecation in the readme is a good start 15:55:24 #action add some words to repo Readme about new vs old - markvan 15:56:15 #topic Berks issues 15:57:06 There is a new one based upon a the new 3.x of the openssl cookbook released yesterday. Name of library was changed and breaks the postgresql cookbook. 15:57:28 I doc the issue and PR here: https://github.com/hw-cookbooks/postgresql/issues/225 15:58:08 In addition, looks like we still have some other issue, maybe with the database cookbook v2 vs v4, but still digging into that one. 15:58:30 (and I take blame for some of this, we all my cookbook version bumps) 15:58:38 /we/with/ 16:04:01 I do see that common has apt at >= 2.6.1 and block has apt at >= 2.3.8 16:05:49 smells like a lead 16:06:25 and database shows 4.x in common, but only 2.x is coming up with block. seems like one of those two is probaly the issue 16:06:39 I know it's still not for several more weeks, but anyone attending chefconf this year? 16:06:55 #topic chefconf 16:07:05 as of right now, I'm not 16:08:06 Any recommendation if it's worth the time? I see it's just down the road from me. 16:09:16 I have never been, but heard is was good 16:10:06 it's local to me, too. I've heard good things about it, but never had a chance to attend previously 16:11:20 I would like to go 16:11:31 I'll ping my boss and see what we have in the budget. I can probably justify it since the rest of my group went to Paris for the OS conference. I can bike to this one. ;-) 16:11:33 I am def going to be at the operators meetup in march 16:13:54 Any other topics, if not I'll end the meeting 16:14:40 I got nothin'. go $local_and_or_regional_sports_team! 16:16:44 #endmeeting