16:00:04 <j^2> #startmeeting openstack-chef
16:00:05 <openstack> Meeting started Mon Jun  8 16:00:04 2015 UTC and is due to finish in 60 minutes.  The chair is j^2. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:06 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:08 <openstack> The meeting name has been set to 'openstack_chef'
16:00:13 <j^2> hey everyone!
16:00:35 <j^2> our agenda is located here:
16:00:37 <jklare> hey
16:00:40 <j^2> #link https://etherpad.openstack.org/p/openstack-chef-meeting-20150608
16:00:41 <sc`> howdy
16:01:30 <j^2> i’ll give everyone a couple mins to join
16:01:54 <j^2> the agenda is pretty packed too, and markvan iirc isn’t going to be here either
16:03:16 <j^2> ok
16:03:24 <j^2> #topic Announcements
16:03:49 <j^2> just to make sure everyone knows
16:04:06 <j^2> in the cross project meeting last week, there is going to be 3 guides written for liberty
16:04:45 <j^2> install guide, cloud admin guide, and ha guide
16:05:13 <j^2> anyone have anything else?
16:05:36 <jklare> nope
16:05:48 <j^2> #topic Previous Business
16:05:58 <j^2> sc`: can you fill us in on the status of c7?
16:06:54 <sc`> i'm able to converge with the addition of the rdo-manager repos, which eliminates the need to modify the roles. there's still a packaging issue ongoing
16:07:54 <sc`> that also uncovered issues with mariadb
16:09:06 <sc`> specifically, the service name on c7 is 'mariadb' instead of 'mysql', and /var/log/mysql doesn't get the correct permissions applied, resulting in an error where mariadb can't write out the binlogs
16:10:08 <sc`> the additional repos that needed to be added to openstack-common are at https://raw.githubusercontent.com/rdo-management/rdo-manager-release/master/rdo-manager-release.repo
16:10:11 <j^2> nice
16:11:06 <j^2> any way we can add a review in to do a ohai check and put in the correct repo?
16:11:49 <sc`> there has been discussions of doing away with the midstream repos, but currently that's where patches yet-to-be-upstreamed land
16:11:59 <j^2> boooo
16:12:02 <j^2> ok, thanks
16:12:18 <j^2> we’ll skip marks update on libvirt
16:12:23 <sc`> we can add them in, but there's no gpg validation on the midstream repos
16:12:27 <j^2> but i did want to aks about 15.04
16:12:29 <j^2> thoughts?
16:13:12 <jklare> we are working on 15.04
16:13:26 <jklare> because it has systemd and gets all the updates
16:13:31 <jklare> but thats basically it
16:13:51 <jklare> i think if we want to fix the libvirt issue, we need to add the -updates repo
16:14:00 <j^2> nice
16:14:24 <jklare> not sure when libvirt 1.2.12 will be ported to trusty
16:14:38 <j^2> cool, but we might work on 15.04, but offically announce it i think is what i was more looking at
16:15:07 <jklare> i think one of the issues with 15.04 would be the integration testing part
16:15:14 <jklare> since we want to use the infra nodes
16:15:18 <j^2> true
16:15:19 <jklare> and those only run 14.04
16:15:31 <jklare> which is fine, since 15.04 is not an lts release
16:15:34 <j^2> yep
16:15:44 <jklare> so i dont think we should necessarily jump on it
16:15:57 <jklare> but rather try to pull in some repos if needed
16:16:42 <j^2> works for me.
16:17:02 <j^2> do we have the steps required or at least something mapped out on how to get 15.04 working?
16:17:35 <jklare> i will try to push the stuff we have upstream
16:17:55 <j^2> rock on
16:18:11 <j^2> #topic fauxhai
16:18:23 <j^2> so sc` i know you have so issues with this
16:18:37 <j^2> i bounced around internal chef asking about getting a new gem pushed
16:18:57 <j^2> everyone came back to me saying that it should just pull the .json from github if it can’t find it locally
16:19:06 <j^2> (for 7.1.json that is)
16:19:17 <j^2> i said it’s not working how we expect, and we had to put that hack in
16:19:49 <sc`> it should and does, but not without the hack that already exists, which is suboptimal
16:20:03 <j^2> :(
16:20:32 <j^2> yeah it seems we cant do anything other than give an extremely compelling reason to convince seth to push another gem
16:20:58 <j^2> if we want to put our case on that issue, he’s willing to listen
16:21:10 <j^2> #link https://github.com/customink/fauxhai/issues/141
16:21:24 <jklare> or we can pull in the gem from a commit we consider stable
16:21:28 <j^2> but chef doenst control that gem, we are at the whims of Seth
16:21:34 <jklare> not sure if thats better than our workaround
16:22:07 <j^2> yeah me neither
16:22:19 <j^2> table this till we have more people?
16:22:54 <sc`> the only benefit i see from pushing a new gem would be to not have to run fauxhai as root
16:23:14 <j^2> which is just wrong anyway
16:24:05 <jklare> chef could also release a new version of chefdk including a fixed fauxhai version
16:24:30 <jklare> this way, at least everybody with chefdk would not need a workaround
16:24:54 <jklare> but that would also mean that chefdk contains a gem version thats not released yet
16:25:11 <sc`> still suboptimal :(
16:25:26 <j^2> yeah chefdk’s versions and keeping things up-to-date are trying to be kinda behind the curve from what I’ve seen
16:25:32 <j^2> it’s more a safe then sorry route
16:25:51 <jklare> if so, i think there is no optimal solution to the problem for now
16:26:01 <j^2> sounds correct
16:26:13 <jklare> so we just have to workaround it until fauxhai is released
16:26:31 <j^2> #topic Eastern Hemisphere standup?
16:26:43 <j^2> jklare: is it going well? is it worth it?
16:27:53 <jklare> j^2 there has been no change since the last monday meeting on this topic, not sure why its on the agenda for this one :)
16:28:15 <j^2> just to make sure that we discuss it. even if it’s something like you just said
16:28:24 <jklare> ok
16:28:29 <j^2> #topic core reviewer process
16:28:42 <jklare> also no change since 2 meetings
16:28:57 <j^2> so at Chef i’m gonna be getting some help with my responsibilites
16:28:58 <jklare> and i think we finished discussing it on the summit
16:29:15 <j^2> should we put the new person through the full process first?
16:29:30 <jklare> the full process?
16:29:50 <j^2> having the time they step up, do the meetings commit and all?
16:29:55 <j^2> everything we’ve discussed
16:30:03 <j^2> then propose them as a core?
16:30:07 <jklare> sure
16:30:09 <j^2> ok
16:30:42 <j^2> i just wanted to see if there was any objection or we should stick to our process because we created it
16:30:46 <j^2> if that makes sense
16:30:58 <jklare> i think the process is fine
16:31:01 <j^2> kk
16:31:14 <j^2> #topic client cookbook
16:31:26 <j^2> markvan and i had a great time pairing on the client cookbook last week
16:31:47 <j^2> i have slated another couple hours this week for it, and i’d love if people joined us
16:32:06 <j^2> we have more or less the foundation that we need to make this happen
16:33:21 <jklare> cool
16:33:37 <j^2> :D
16:33:38 <sc`> nice
16:34:25 <j^2> #topic  chefdk 0.6.0
16:34:28 <j^2> thoughts?
16:34:50 <sc`> i'm with jklare in that we should wait for the chef-provisioning issue
16:34:59 <sc`> to be fixed
16:35:03 <j^2> makes sense
16:35:16 <j^2> i just want to make sure we mention it here :)
16:35:53 <j^2> #topic specs
16:35:59 <j^2> so it seems there is a specs site:
16:36:06 <j^2> #link specs.openstack.org
16:36:19 <j^2> we can move to it if yall want to
16:36:21 <j^2> thoughts?
16:37:04 <jklare> what does moving to it mean
16:37:18 <jklare> ?
16:37:42 <jklare> is it just linking our repo or does it mean to abandon the old repo and use something different?
16:38:18 <j^2> it means we move our specs to the offical site, and our repo gets depricated
16:38:23 <j^2> it’s more “big tent” stuff
16:38:56 <sc`> i think it'd make sense to move toward that as part of the big tent effort
16:39:07 <j^2> cool
16:39:08 <jklare> sounds good
16:39:15 <j^2> they still havent moved us as of yesterday
16:39:27 <j^2> but we are slated for the “next maintance window"
16:39:30 <j^2> whenever the fuck that is
16:39:31 <j^2> :P
16:39:56 <j^2> ill keep my eye on it, updates when i got em
16:40:23 <j^2> #topic cross project ci testing
16:40:37 <j^2> opps, that’s mark
16:40:46 <j^2> #topic Open Discussion
16:40:58 <j^2> anything on yalls end?
16:41:11 <jklare> i just read a bit about the specs thing
16:41:14 <jklare> :)
16:41:42 <jklare> i think we will just keep our repo as is and only link it via some files in the projects-config repo
16:42:00 <jklare> maybe we have to refactore the repo structure a bit
16:42:02 <jklare> not sure yet
16:42:08 <jklare> will have to read a bit more
16:43:24 <j^2> cool
16:43:29 <j^2> sc`: anything on your end?)
16:43:46 <sc`> i'm good
16:44:38 <sc`> still chasing down mariadb inconsistencies
16:48:51 <j^2> nice
16:49:50 <j^2> ok, if there is nothing else
16:50:01 <j^2> i’ll end now and give everyone back 10
16:50:04 <j^2> mins
16:50:12 <jklare> :)
16:50:16 <j^2> #endmeeting