15:34:06 <markvan> #startmeeting Openstack-Chef 15:34:07 <openstack> Meeting started Thu Mar 5 15:34:06 2015 UTC and is due to finish in 60 minutes. The chair is markvan. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:34:08 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:34:11 <openstack> The meeting name has been set to 'openstack_chef' 15:34:12 <openstack> j^2: Error: Can't start another meeting, one is in progress. Use #endmeeting first. 15:34:12 <j^2> heh 15:34:13 <j^2> sorry 15:34:20 <j^2> running late this morning 15:34:26 <markvan> np 15:34:31 <j^2> i can’t seem to computer for the life of me 15:34:44 <markvan> ha 15:35:02 <markvan> #topic kilo initial patches 15:35:14 <markvan> down to last 4 patches 15:35:17 <markvan> #link https://review.openstack.org/#/q/status:open++branch:master+topic:bug/1426424,n,z 15:35:40 <markvan> after that, we should be good to go 15:35:49 <j^2> awesome 15:36:20 <markvan> A good discussion point for next Mon would probably be to walk the blueprints and clean those up to see listof what we want for kilo. 15:37:27 <j^2> chef 12 15:37:29 <j^2> ;) 15:37:57 <cmluciano> +1 I think chef 12 should be a high priority 15:38:34 <markvan> yup, after we get the infra patch for using rake, that should be easy bump for Gemfiles. 15:39:26 <markvan> #topic stable juno 15:39:38 <j^2> in theory we might want to remove the Gemfiles, or leverage them like we do for the chef-repo 15:40:33 <markvan> yeah, that another good discussion to have, I like the idea of syncing with the Chef DK levels, but others still like Gemfiles (as we saw for the repo) 15:41:26 <j^2> yep 15:41:48 <markvan> just a heads up for juno stable, we have some folks that are looking to add some simple attributes for keys in conf files that are missing. They will do kilo and then backport to juno. 15:42:28 <markvan> should be straight forward since branches are so close right now. 15:43:33 <markvan> #topic metadata nova api 15:43:36 <markvan> #link https://bugs.launchpad.net/openstack-chef/+bug/1415220 15:43:38 <openstack> Launchpad bug 1415220 in OpenStack + Chef "[chef-repo][testing-stack] nova-api and nova-api-metadata conflict" [High,In progress] - Assigned to Samuel Cassiba (scassiba) 15:44:14 <markvan> please review this bug and see if mylast few comments make any sense, we some agreement, we can fix this 15:45:03 <markvan> #topic ubuntu neutron 15:45:30 <markvan> #link https://bugs.launchpad.net/openstack-chef/+bug/1314751 15:45:31 <openstack> Launchpad bug 1314751 in OpenStack + Chef "[network] openvswitch agent broken, uses ml2 conf on ubuntu" [High,In progress] - Assigned to Mark Vanderwiel (vanderwl) 15:45:40 <j^2> lovely 15:46:07 <markvan> This one has been a thorn... but I'm trying a different approach. patch https://review.openstack.org/#/c/161495/ 15:46:54 <j^2> #link https://review.openstack.org/#/c/161495/2/recipes/default.rb is the meat of it 15:47:13 <markvan> please review. With that patch, I'm trying to standup the aio_neutron on ubuntu, based upon these instructions: https://fosskb.wordpress.com/2014/10/18/openstack-juno-on-ubuntu-14-10/ 15:47:36 <openstackgerrit> Merged stackforge/cookbook-openstack-object-storage: Initial kilo updates https://review.openstack.org/159971 15:47:53 <markvan> yup, pretty minor change, but seems to be working, just need to tweak the network attriubtes to make it work with our test suite. 15:48:01 <j^2> nice 15:48:17 <j^2> 👍 on this from what i can see so far 15:48:19 <markvan> but I'm no neutron expert... 15:48:53 <markvan> that's the topics I had. 15:48:55 <j^2> i don’t think anyone is 15:48:56 <markvan> #general discussion 15:49:27 <j^2> i don’t really have anything honestly 15:49:32 <cmluciano> in checking with the RDO folks, we currently don’t have any valid packages to test with kilo. 15:49:45 <j^2> so we’d have to try to get trunk? 15:49:49 <j^2> if we wanted to test 15:50:30 <cmluciano> They publish some packages, but it was a real “here be dragons” type of package 15:51:19 <cmluciano> from the #rdo channel 15:51:22 <cmluciano> number80> cmluciano: RDO kilo is not released, we only have snapshots build through Delorean 15:51:25 <cmluciano> <apevec> cmluciano, there's current Delorean snapshot but it eats kittens etc. 15:54:05 <markvan> looks like they will need some time. 15:55:11 <j^2> yeah 15:55:21 <markvan> I was thinking about hacking back in to use rdo juno, just for sanity test, but don't have time. The internal ibm folks have a kilo repo and will be using the new kilo cookbooks against that shortly, so that should shake out any obvious bugs 15:55:52 <markvan> /kilo/redhat bsed kilo/ 15:56:02 <j^2> yeah we really do need to get centos7 working with juno too 15:56:31 <j^2> completely, if we can’t get RDO i think if you’re looking for things to focus on this is a great time to context switch back so we can get this resolved 15:58:00 <markvan> would be nice to see others interested in juno/centos step in, I'll probably be full bore on kilo for a while 15:58:53 <j^2> ah cool 15:59:05 * markvan wow zuul looks tame today (for a change) 15:59:14 <j^2> yeah i’m gonna try to circle back on it, but i should have a new release of knife-openstack here soon 16:00:43 <markvan> interesting...that reminds me of another topic. Until we get the CI going (with infra or travis or ?), we will need to decide on what type of gates to run against the repo. The other cookbooks have style, lint, unit. 16:01:35 <markvan> but right now we only have style/rubocop for repo. Would be nice at least to something like spiceweasel was doing where the syntax of the roles and env are checked. 16:02:01 <j^2> chef-provisioning does that iirc 16:02:14 <j^2> when it uploading it to chef-zero 16:02:33 <j^2> break one of the roles and try to spin up the stack that should verify it 16:03:02 <markvan> yeah, but for basic gates in infra, we can't to the full aio_test 16:03:45 <j^2> oh 16:04:21 <j^2> yeah i don’t think we’ll be able to till after the fact of the patch being uploaded and jenkins doing the non-voting +1 -1 thing 16:04:36 <j^2> we’ll have to rely on that and do diligance verifying it that way 16:04:53 <markvan> not a big deal, but I think we can run the same type of knife -z like spwiceweasel did under covers against our roles and env in a simple rake task 16:05:42 <markvan> yeah, for now it does the spiceweasel in the gate, but we will have flexibility with the rake support. 16:07:33 <markvan> any thing else to discuss now? 16:08:32 <wenchma_> #link https://bugs.launchpad.net/openstack-chef/+bug/1428689 16:08:33 <openstack> Launchpad bug 1428689 in OpenStack + Chef "[chef-repo] Should update the version of cookbook build-essential" [Undecided,In progress] - Assigned to Ma Wen Cheng (mars914) 16:08:42 <wenchma_> a little issue 16:09:18 <wenchma_> need minor change , but urgent , because it made jenkins fail 16:09:50 <openstackgerrit> Merged stackforge/cookbook-openstack-compute: Initial kilo updates https://review.openstack.org/159990 16:09:57 <wenchma_> this patch https://review.openstack.org/#/c/161754/ 16:10:16 <wenchma_> please review with it 16:10:55 <j^2> kk 16:13:20 <markvan> wenchma_: humm, we're using 1.4.2 in the chef repo Berks just fine? What's different about the client cookbook? 16:14:04 <wenchma_> markvan: https://github.com/stackforge/cookbook-openstack-client/blob/master/metadata.rb#L13 16:14:22 <markvan> oops, my bad this is in the repo side 16:14:50 <wenchma_> qhttp://logs.openstack.org/42/161142/2/check/gate-openstack-chef-repo/fcc91e4/console.html#_2015-03-05_14_23_14_246 16:15:00 <wenchma_> here is the error log 16:16:20 <markvan> yup, see that now. Ah, the client was pushed after repo, so repo did not fail. 16:17:29 <markvan> last 2 kilo patches are rechecking now, then will be ready 16:18:04 <markvan> and yes wenchma_ will also be needed to close this out: https://review.openstack.org/#/c/161754/ 16:18:26 <markvan> any other topics? 16:18:40 <cmluciano> I’m good 16:19:14 <wenchma_> markvan: ok 16:19:20 <j^2> :) 16:19:27 <markvan> #endmeeting