16:00:52 <j^2> #startmeeting openstack-chef
16:00:56 <openstack> Meeting started Mon Jul 13 16:00:52 2015 UTC and is due to finish in 60 minutes.  The chair is j^2. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:58 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:00 <openstack> The meeting name has been set to 'openstack_chef'
16:02:17 <j^2> hey everyone!
16:02:30 <markvan> Howdy
16:02:35 <jklare> hi
16:02:51 <sc`> good $TIME_OF_DAY
16:03:03 <j^2> :D
16:03:11 <j^2> so i’ll give everyone a couple mins to join
16:03:25 <j^2> AGENDA: https://etherpad.openstack.org/p/openstack-chef-meeting-20150713
16:05:20 <j^2> ok, cool
16:05:26 <j^2> so this sohuld be pretty quick…ideally
16:05:33 <j^2> #topic Announcement(s)
16:05:53 <j^2> Toyko Summit ATC passes got released this morning 13-07-2015, check your inboxes for the information.
16:05:53 <j^2> We're providing all ATCs who contributed to the Liberty cycle (priorto July 8, 2015)
16:06:06 <j^2> anything else?
16:07:05 <markvan> quick session review?
16:07:18 <j^2> not sure what you mean about that
16:07:20 <markvan> We have two submitted right now, any others regarding Chef?
16:07:36 <j^2> oohh
16:07:50 <j^2> i have one on the Mailing list one too
16:08:03 <j^2> anyone else have talks they are submitting?
16:08:18 <jklare> nope
16:08:21 <j^2> we should try to help each other out
16:08:29 <j^2> advertise and what not right?
16:09:03 <sc`> i, most likely, won't be at the summit unless i have a remarkable change in luck
16:09:26 <markvan> yup, just making sure folks know there are some Chef sessions scheduled and get word out...
16:09:51 <jklare> sc` broke any mirrors lately?
16:09:51 <j^2> rock on
16:09:54 <j^2> sc`: sorry to hear that
16:10:04 <j^2> i could talk to edgar for you ;)
16:10:41 <sc`> ha
16:10:41 <j^2> any other announcements?
16:12:12 <j^2> cool, lets go on to Previous Business
16:12:24 <j^2> #topic  <sc`> c7 still requires some modifications to common to start to converge. still working on getting a converge
16:12:33 <j^2> sc`: updates?
16:12:46 <sc`> the main obstacle at this point is mariadb
16:12:53 <j^2> as with last week?
16:12:59 <j^2> dude hasnt merged yet eh?
16:13:01 <sc`> yup. no chance from last week
16:13:06 <sc`> s/chance/change.
16:13:07 <j^2> #link https://github.com/sinfomicien/mariadb/pull/73
16:14:15 <sc`> we'd also need to get a new version pushed to supermarket
16:15:21 <j^2> true dat
16:15:35 <j^2> #topic <j^2> specs.openstack.org
16:15:50 <j^2> yeah i still havent done anything with this, i’ll update on _what_ I’ve been doing here in a bit
16:15:57 <j^2> i’ll still own it though
16:16:17 <j^2> #topic <all> How to handle cross project CI testing.
16:16:23 <j^2> so this one is the big one
16:16:51 <j^2> i havent seen any tests yet; but markvan’s update with the wiki has been greate
16:17:21 <j^2> is there any questions or thought on this, or can we take this off the agenda going forward?
16:17:58 <markvan> a side note, we caught our first bug with the new Depends-On support: see https://review.openstack.org/#/c/198997/
16:18:08 <jklare> i think some of this should probably move into a spec
16:18:20 <j^2> jklare: can you explain that?
16:19:09 <markvan> so basic cross project support is there, for cookbooks and repo, now just need to get CI gate going to really see this work
16:19:18 <jklare> i think the first option about leaving in the old function is basically about a good workflow we should adapt
16:19:51 <jklare> the non-voting ci job can be done, if my patch mentioned below gets merged and we port our ci to all cookbooks (in reverse order)
16:20:27 <jklare> the cross cookbook dependencies are a bit harder i think, since i have not found a good way to use zuul-cloner for this so far
16:20:52 <jklare> but since we are nearly done with our own gerrit-zuul-jenkins stack, i will invest some more time into that in the near future i think
16:21:12 <jklare> so maybe we could move this cross cookbook part into a spec
16:21:19 <jklare> or a blueprint
16:22:33 <markvan> how are cross cookbook dependencies diff from Depends-On support?
16:22:59 <j^2> i thought the Depends-On support covered the cross, wait a sec let me reread whats going on
16:24:09 <jklare> maybe i missed something about the depends-on thingy
16:24:27 <j^2> ah so the blueprint ot capture the ideas, but yeah the depends on thingy should do what you’re discribing as your gap jklare
16:24:48 <j^2> it figures out what order it’s supposed to merge for the verifid/gates
16:24:51 <j^2> let me find the link
16:25:07 <j^2> #link http://docs.openstack.org/infra/manual/developers.html#cross-repository-dependencies
16:25:13 <j^2> yeah that whole section
16:25:26 <j^2> it takes a bit to grok, but it’ll make sense
16:25:56 <jklare> so this will also allow the integration testing with multiple patches?
16:26:09 <markvan> yup
16:26:20 <jklare> ah ok, then i missed that part
16:26:39 <markvan> check out the log from the patch I mentioned above, its one of the first to kick this support on
16:27:29 <jklare> kk, will do
16:28:05 <j^2> #topic <markvan> keystone under apache, blueprint patches submitted
16:28:10 <j^2> so this is interesting
16:28:30 <j^2> it seems that this has been resolved
16:29:04 <markvan> yup, for kilo, the new recipe is in, but not activate in any role, that will be a change for liberty
16:29:21 <sc`> are there any plans to backport this?
16:29:58 <j^2> sc`: you’re more then welcome to ;)
16:30:03 <markvan> humm, had not thought about it.   it might be a bit ugly since I needed to bump up the apache cookbook version to get some needed bugfixes
16:30:38 <sc`> j^2: :)
16:31:49 <j^2> markvan: do you mind scoping the backport and publishing what you think would be needed to be done?
16:32:03 <j^2> it might be worth looking into for Juno, being LTS,
16:32:12 <j^2> but depends on the effort involeded
16:32:15 <j^2> sc`: that good for you?
16:32:26 <jklare> btw: i am really not sure about this juno lts thingy
16:32:39 <jklare> i think we should probably not worry too much about that
16:32:45 <j^2> yeah?
16:32:51 <sc`> j^2: yeah. if it's a big hassle, then it's not worth the cycles
16:32:54 <j^2> icehouse EOL’d last week
16:33:22 <j^2> markvan: good on your end?
16:33:38 <markvan> I can take a quick peek and see what would be needed
16:34:27 <j^2> #topic <j^2> AIO-Neutron Update: https://review.openstack.org/#/c/196858/
16:34:35 <j^2> now i’m pretty pround of this guy
16:34:45 <j^2> though i need some help
16:35:27 <j^2> the main issue is OVS needs a second adapter to work to slurp up the NIC and do it’s black magic
16:35:46 <j^2> HP cloud, doesnt have eth1 RS does have eth1
16:35:55 <j^2> so we have to create a dummy interface
16:35:58 <j^2> which is what i did
16:36:32 <j^2> now creating the pools are my problem, i cant seem to relyably get the gateway and create a floating ip pool out of it
16:36:47 <j^2> if anyone is willing to step up and help that would be amazing
16:37:52 <j^2> any high level questions on this?
16:38:35 <jklare> sure, maybe we can talk about this later and i will try to help where i can
16:38:42 <j^2> jklare: sounds good
16:39:15 <j^2> anything else?
16:39:48 <j^2> #topic Manila Cookbook: https://github.com/jjasghar/cookbook-openstack-shared-filesystem
16:39:56 <j^2> so I’ve taken a stab at the Manila cookbook
16:40:02 <j^2> it doesnt converge or anything yet
16:40:25 <j^2> and i don’t want to move it into the openstack namespace till it “works” but i’d love some other eyes on it
16:40:42 <j^2> it’s just a framework right now
16:41:42 <j^2> i just randomely picked one of the other “not-so-huge” projects to say that we can make cookbooks for something else
16:41:59 <j^2> alanmeadows: was supposed to give us Desegnate ;)
16:42:09 <jklare> sadly i do not have too much time this week and will be out of office next week
16:42:26 <j^2> :(
16:42:45 <j^2> also the “orchestration cookbook” works still out of the box
16:42:48 <markvan> I'll give it a quick review when I get a chance...but dont know much about manila
16:42:50 <j^2> i validated that last week
16:43:05 <j^2> markvan: yeah that’s one reason why i picked it, it’s basicly “shared” nfs
16:43:17 <j^2> i’m not sure _why_ you’d want it, but it’s there
16:43:49 <markvan> yeah, some folks here are interested since we have the slick gpfs product..
16:44:10 <alanmeadows> I've been a total slacker on that (been going through home closing this month), I actually have some working code for designate that just needs to be cleaned up since it was just used internally for a small project.
16:44:29 <alanmeadows> Do we have a time line for when we're aiming for introducing a net-new book?
16:45:15 <j^2> alanmeadows: when it’s ready? ;)
16:45:19 <alanmeadows> lol
16:45:53 <alanmeadows> I think I can turn something around thats good to get the ball rolling in about two weeks
16:46:04 <alanmeadows> if that is sufficient you can still count me in
16:46:05 <j^2> alanmeadows: rock on, i’ll start bugging you then ;)
16:46:18 <alanmeadows> please always feel free :)
16:47:29 <j^2> cool, i forgot one announcement
16:47:40 <j^2> #topic  OpenStack Ops Mid-Cycle Meetup
16:47:45 <j^2> I’m gonna be there
16:47:54 <j^2> Tuesday, August 18, 2015 at 9:00 AM - Wednesday, August 19, 2015 at 5:00 PM (PDT)
16:47:54 <j^2> Palo Alto, CA
16:47:54 <j^2> https://etherpad.openstack.org/p/PAO-ops-meetup
16:47:54 <j^2> http://www.eventbrite.com/e/openstack-ops-mid-cycle-meetup-tickets-17703258924
16:48:05 <j^2> if anyone else could that would be amazing
16:48:40 <markvan> yeah, I asking about that one, but odds are not good
16:49:12 <sc`> i'll probably make it to that since i'm local
16:49:18 <alanmeadows> It's across the bridge for me (30m away), so its just about getting some time allocated, so odds are pretty good
16:49:33 <jklare> too far for me
16:49:33 <j^2> it’s useally a good oppertunity to just discuss stuff, anad hell if we get a couple of us there we can powwow on somethings
16:50:13 <j^2> there really is a great productivity boost when we’re all together
16:50:34 <j^2> alanmeadows: ooh, if you’ll be there we can put those finishing touches on desginate, and maybe even get it shipped then
16:50:51 <j^2> i have a cluster we can inject it in too if we want
16:51:17 <j^2> to validate
16:51:22 <j^2> anyway...
16:51:23 <alanmeadows> that will be a good forum to discuss
16:51:27 <j^2> rock on
16:51:33 <j^2> #topic General Discussion
16:51:44 <j^2> open floor? or a second to close the meeting?
16:52:34 <jklare> would be great if you could take a look at this periodic test patch
16:52:45 <j^2> link?
16:52:49 <jklare> or both that are mentioned in the etherpad
16:53:05 <jklare> https://review.openstack.org/#/c/194164/ should build it at 6am UTC every day
16:53:09 <j^2> awesome
16:53:12 <j^2> ah, cool
16:53:14 <jklare> https://review.openstack.org/#/c/194181/ to move it from experimental to non-voting in check
16:53:16 <j^2> yeah i’ll take a look myself
16:53:49 <jklare> these have been out for some time now and before i start kicking some infra ppl it would be good to have some feedback from you
16:54:10 <j^2> i’d love experimental to move to non-voting, it’s annoying having to write check experinmental every time
16:54:25 <markvan> yup, will also review them
16:56:21 <j^2> cool anything else?
16:56:47 <jklare> nothing from me
16:57:42 <j^2> cool, thanks everyone
16:57:45 <j^2> #endmeeting