15:30:03 <j^2> #startmeeting Openstack-Chef
15:30:04 <openstack> Meeting started Thu Feb 26 15:30:03 2015 UTC and is due to finish in 60 minutes.  The chair is j^2. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:30:05 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:30:07 <openstack> The meeting name has been set to 'openstack_chef'
15:30:18 <j^2> Hey everyone!
15:32:25 <markvan> howdy
15:32:29 <krtyyy> 'sup
15:32:55 <j^2> I'd like to go on record thanking everyone for their flexibility these last couple of weeks/meetings. I've had some personal life excitement and scares and appreciate everyone pitching in
15:33:51 <j^2> And with that statement let's take it to what we all want the discuss...
15:34:10 <j^2> #topic final mile to stable/juno
15:34:27 <markvan> Here's the list of patches that are left for Juno stable:
15:34:28 <markvan> https://review.openstack.org/159254
15:34:28 <markvan> https://review.openstack.org/159252
15:34:28 <markvan> https://review.openstack.org/159265
15:34:28 <markvan> https://review.openstack.org/156416
15:34:29 <markvan> https://review.openstack.org/158845
15:34:47 <markvan> and with that we also need to switch the repo over to the new testing suite.
15:34:57 <cmluciano> Currently testing the metadata ones on centos
15:35:51 <j^2> Ok, so the repo should be able happen ASAP
15:36:18 <j^2> Any of those patches gated by anything we don't control?
15:36:27 <markvan> yup, one pull request I put up yesterday to start to break down the doc
15:36:40 <markvan> no, these patches are straight forward
15:36:47 <j^2> Rock on
15:37:21 <markvan> It's the mariadb stuff that is gated by that cookbook being updated in community, but we really don't need that until kilo
15:38:17 <markvan> The fix for the floating ip was questioned by Jan, so I think that one will be going away as the code works as is, just needs better doc.  I'm getting that doc ready now.
15:39:12 <j^2> Cool this all sounds extremely positive
15:39:24 <markvan> So, I would like to get the repo branched and updated today.  Then get a few confirmations that is still works, and we're ready for the juno branch.
15:40:16 <j^2> Awesome I can take ownership of that assuming I get my laptop
15:41:14 <markvan> ok, I get that last PR looked at and then start creating a patch, that can be done in parallel with the repo branch work.
15:42:47 <j^2> Yep the repo should be independent of anything else, minus the Berksfile.lock but that's after the stamp happens and we have the SHAs
15:45:50 <j^2> And we never got multi nova with CentOS 7 correct?
15:46:10 <cmluciano> j^2: No, I’m getting weird mysql issues
15:46:14 <cmluciano> it just hangs for me
15:46:16 <markvan> j^2: do you have that link to the wiki for the branck work,
15:47:14 <markvan> yeah, I tried it but had the issue with old host-only dhcp entries, have not gone back yet
15:48:39 <j^2> markvan: one sec let me see if I can get you the link
15:49:19 <markvan> nm got it, I'm going to update it a bit more and add a link to it from our releases topic
15:49:56 <markvan> I was going to start adding the list of items to look on the flip side, initial kilo master work items
15:49:58 <j^2> cmluciano: and markvan cool no worries on the CentOS 7 multi nova I just wanted verify I had the information
15:50:11 <j^2> Correct*
15:51:07 <markvan> So, speaking of centos/redhat, I think the basic cookbook are looking good now, so this table should be update to include it now, agreed?  https://wiki.openstack.org/wiki/Chef/GettingStarted
15:51:21 <markvan> (see Releases table near bottom)
15:52:39 <j^2> #agreed markvan on updating the wiki
15:52:49 <cmluciano> #agreed
15:53:19 <markvan> k, I'll do that shortly, and then update it again once the stable juno branch is in place.
15:53:27 <j^2> Kk
15:55:11 <j^2> jklare: any concerns on your end?
15:55:14 <markvan> so, to net for Juno work left:  a few patches (and I'll update the float one to a doc update),  branch repo, patch repo, then branch the cookbooks, cleanup patches
15:55:51 <j^2> markvan: yep
16:02:46 <j^2> Question
16:03:54 <j^2> Has everyone whose committed to our code since Paris summit gotten their ATC for Vancouver?
16:04:32 <cmluciano> I probably have not
16:04:39 <j^2> #topic general discussions
16:04:57 <markvan> humm, our code is in stackforge and that does not count for ATC, only base openstack does
16:05:18 <j^2> Infra patches do though
16:05:28 <j^2> Right?
16:05:47 <sc`> I saw something earlier this week or so about >juno only targeting centos 7. should us folks on 6 be accelerating our testing of centos 7?
16:05:53 <markvan> yup, infra is part of it
16:06:39 <j^2> sc`: yeah that's a very good idea
16:06:40 <markvan> sc`: yes, I saw that as well, base kilo+ want nothing to do with <7 centos/redhat
16:08:24 <markvan> j^2: I usually help out in the openstack security docs to get my ATC
16:08:29 <j^2> so that makes me sad about stackforge then, its like we're no longer part of the OpenStack world. They demoted us even more
16:09:39 <sc`> how about the nova-api and nova-api-metadata conflict? I saw another related bug to it in lp
16:10:02 <markvan> BTW, the ATC email should come from: communitymngr@openstack.org   so check your spam
16:10:27 <sc`> https://bugs.launchpad.net/openstack-chef/+bug/1418797
16:10:28 <openstack> Launchpad bug 1418797 in OpenStack + Chef "[compute] Can not set metadata_listen for nova-metadata-api" [Undecided,In progress] - Assigned to Wei Hu (huwei-xtu)
16:10:46 <j^2> I thought the override rejigger was the way to work around that api compute thingy
16:10:56 <markvan> sc`: yeah, there are a couple of bugs, but the patches are not ready.  For the testing repo we have done the workaround, just remove the metadata from the enabled apis attribute.
16:10:59 <cmluciano> ,
16:11:22 <j^2> sc`: that ^^^
16:12:02 <sc`> ok, cool. I'm glad to help test a longer term solution wherever possible when it emerges
16:12:05 <markvan> As I chatted in the bug, I think we should decide 1 path for metadata, and not try to support 2.  My thought is to go with using the recipe and disable support for the enabled api attribute with metadata.
16:12:41 <markvan> Since we have a workaround now, I did not put this on the list for must have in Juno.
16:12:42 <sc`> I've been trying to get it to disable in the recipe, but haven't had much luck
16:12:48 <markvan> but, open to other thoughts here
16:13:51 <markvan> yeah, I don't like trying to mess with a good recipe, so that's why I suggest we go the other route, just disallow metadata from the enabled_apis and force the use of the recipe path.
16:14:47 <sc`> yep. makes sense
16:16:08 <j^2> Anything else outstanding?
16:17:06 <sc`> I'm still trying to figure out why nova-cert and nova-novncproxy don't get their init scripts on occasion
16:17:20 <sc`> so far, I see it on centos 6/icehouse and centos 7/juno
16:17:23 <j^2> CentOS?
16:17:29 <markvan> I'm good, I'll get the float and repo patches up soon.
16:17:56 <markvan> sc`: yup I saw it once on centos7, but 2nd time it worked, not sure what's going on
16:18:21 <sc`> yeah. not sure if it's just chef being chef, or if there's actually an issue in the cookbooks
16:18:49 <j^2> Maybe a template is clashing with a pkg install of one?
16:23:37 <j^2> Cool if nothing left I'll close the official meeting
16:24:00 <j^2> #endmeeting