16:00:50 <sc`> #startmeeting openstack_chef
16:00:51 <openstack> Meeting started Mon Jun 27 16:00:50 2016 UTC and is due to finish in 60 minutes.  The chair is sc`. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:52 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:56 <openstack> The meeting name has been set to 'openstack_chef'
16:01:03 <jklare> o/
16:01:09 <sc`> ohai!
16:02:26 <sc`> judging from previous meetings, i think it's just the two of us
16:02:47 <jklare> yeah, i guess most of the others went away to focus on other projects
16:03:38 <sc`> alright. i don't think we need an agenda for the time being
16:03:51 <jklare> :D
16:04:17 <jklare> i am quite happy that we are done with stable/mitaka and from my side everything looks good so far
16:04:35 <sc`> yeah, for sure. my team is starting to gnaw on it and see what they can do
16:04:49 <jklare> sounds good
16:05:15 <jklare> any plans on how we want to continue with the openstack-chef project during the newton cycle?
16:05:32 <sc`> in order to land newton, we need openstackclient
16:05:50 <jklare> sounds reasonable
16:05:54 <sc`> that's before we can even try converging
16:06:31 <jklare> so you would go for replacing the service pythonclients by the openstackclient or rather try to abstract stuff into a wrapper with some ruby libraries
16:06:33 <jklare> like fog
16:07:20 <sc`> we do have the client cookbook that abstracts to fog, but it might need some love before we can use it for newton
16:08:11 <jklare> yeah, looked at it a while ago
16:08:16 <jklare> there is not too much there
16:08:23 <sc`> yeah. not much to it
16:08:46 <sc`> i'm looking at it, going to see what can be used, or if python-openstackclient is the way to go
16:10:33 <sc`> the good news is that in order to do 'rake unit' on keystone, not much needs to change
16:11:18 <sc`> so we should be able to get newer chefdk in, even if integration can't converge at the moment
16:11:52 <sc`> only because integration still isn't voting :)
16:12:46 <jklare> :)
16:14:25 <sc`> i'll be spending some time on making 'rake unit' work with chefdk 0.15.15 with the master branch, and proposing those changes
16:14:34 <jklare> cool
16:14:46 <sc`> so we can get movement on at least one of the things we discussed at the summit :)
16:15:01 <jklare> i think its a good first step
16:15:20 <jklare> i will try to look at the apt cookbook update so we can make xenial work
16:15:30 <sc`> awesome
16:22:41 <sc`> i was looking at lp, and i'm going to have to figure out how to 'release' mitaka so we can move forward there
16:23:47 <jklare> what do you mean by "release" ?
16:24:17 <sc`> according to https://launchpad.net/openstack-chef we're still actively focused on mitaka
16:25:00 <sc`> okay, looks pretty straightforward
16:25:25 <jklare> should be released now :D
16:25:39 <sc`> yep, i see it
16:26:23 <sc`> thanks :D
16:27:31 <jklare> np, i forgot that i wanted to do that :D
16:27:54 <sc`> rock on. i'll register newton today
16:28:01 <jklare> great
16:28:12 <jklare> any thoughts about the "discontinued" cookbooks?
16:29:09 <sc`> swift and sahara? i'd love to get them current, but if nobody actively wants to use either one, it may be a bunch of work for nothing
16:29:47 <jklare> +1
16:30:39 <sc`> i haven't seen any patches on swift since... kilo i think
16:30:50 <jklare> i think everybody is using ceph
16:30:52 <sc`> beyond the "let's get this released" patches
16:32:38 <jklare> ^^
16:33:35 <sc`> i'm fine with letting swift and sahara be as-is for now. if someone wants to step up in the meantime, they're quite welcome
16:36:14 <jklare> agreed
16:36:37 <sc`> back to the client stuff, i think we will be pretty much forced down the path of using python-openstackclient or making fog work
16:36:59 <sc`> everything i'm seeing suggests openstackclient is where to be
16:38:04 <jklare> yeah, but writing a wrapper around a pythonclient seems less than ideal
16:38:15 <jklare> i think using a ruby library would be much more comfortable
16:38:29 <jklare> although it would be more work initially
16:38:50 <sc`> i think the main question there is who would be doing it
16:39:32 <jklare> if nobody maintains the cookbook, we could fork back from suse :D https://github.com/crowbar/crowbar-openstack/tree/master/chef/cookbooks/swift
16:40:10 <sc`> :D
16:40:47 <jklare> i mean the funny part is, that they are also maintaining these cookbooks and most of the time doing exactly the same fixes and patches we do :D
16:41:38 <sc`> didn't suse contribute to the openstack cookbooks when they were still in the stackforge namespace?
16:41:46 <jklare> yeah
16:41:52 <sc`> that is amusing
16:41:53 <jklare> thats why this is so funny
16:42:46 <jklare> but thats off topic :D
16:43:04 <jklare> so yeah, we should focus on the client cookbook
16:43:48 <sc`> looks like the crowbar cookbooks don't approach the client either. just relies on python-*client bindings. doesn't work for newton i imagine
16:45:52 <sc`> that's about all i have, since we seem to be diverging from topic
16:46:07 <jklare> ^^
16:46:38 <sc`> btw, i fixed liberty-rc1 on lp. that typo was bugging the shit out of me
16:46:48 <jklare> :D
16:46:57 <jklare> these things are horrible
16:47:35 <sc`> unless we have anything else for today, i think we'll call it a few minute early
16:47:41 <jklare> +1
16:47:58 <sc`> sounds good. see you on the regular channel o/
16:48:13 <sc`> #endmeeting