15:00:02 <EmilienM> #startmeeting puppet-openstack
15:00:06 <openstack> Meeting started Tue Jun 14 15:00:02 2016 UTC and is due to finish in 60 minutes.  The chair is EmilienM. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:07 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:10 <openstack> The meeting name has been set to 'puppet_openstack'
15:00:14 <EmilienM> #topic agenda https://etherpad.openstack.org/p/puppet-openstack-weekly-meeting-20160614
15:00:14 <chem``> o/
15:00:15 <iberezovskiy> hi
15:00:16 <mwhahaha> hi
15:00:16 <iurygregory> o/
15:00:19 <_ody> o/
15:00:21 <degorenko> hi
15:00:22 <EmilienM> hello folks !
15:00:32 <iurygregory> hello :D
15:01:15 <EmilienM> #topic midcycle
15:01:24 <EmilienM> #link https://etherpad.openstack.org/p/newton-puppet-midcycle-meetup
15:01:36 <EmilienM> so 2 people, iurygregory and mfish are interested by a midcycle
15:02:11 <EmilienM> it's a good sign to tell we don't really need it :)
15:02:18 <iurygregory> yep
15:02:31 <bkero> o/
15:02:32 <EmilienM> though we can finish bug triage that we started at Summit with mfish
15:02:47 <EmilienM> #link https://etherpad.openstack.org/p/newton-puppet-bug-triage
15:03:06 <EmilienM> _ody and I started this triage, maybe mfish (not here) can help to finish it.
15:03:30 <EmilienM> any objection to say "no midcycle during Newton" ?
15:04:05 <EmilienM> ok so no midcycle
15:04:47 <EmilienM> #topic msync
15:04:52 <EmilienM> degorenko: what's up
15:05:00 <degorenko> hey
15:05:13 <degorenko> we have two commits for tacker and congreess
15:05:13 <degorenko> https://review.openstack.org/#/q/topic:cookie_cutter
15:05:34 <degorenko> so, my question is: how we should use cookie cutter and msync for already prepared modules?
15:05:43 <EmilienM> that's a good question
15:05:44 <_ody> EmilienM: I'll do another round today.
15:06:17 <EmilienM> I would 1/ generate a new module with latest bits from cookiecutter and add the bits from old module to the new module (new classes, or new params)
15:06:37 <degorenko> that's also and my suggestion
15:07:16 <degorenko> we can generate stuff for module with our cookie&msync and then update that stuff with already prepared
15:07:25 <degorenko> but in this case - we will have big commit
15:07:29 <EmilienM> radez: does it make sense?
15:07:48 <radez> so we don't need to commit a vanilla cc/msync patch just base future patches on cc/msync structure is what I'm understanding
15:08:14 <EmilienM> well, like any project, iteration is always best. It's not only in our Puppet modules
15:08:38 <degorenko> my main concern in this situation - that we need remove all prepared stuff and then back it again
15:08:39 <iurygregory> to me make sense see a commit with only cookiecutter&mysnc and a commit in chain with the modifications...
15:08:43 <degorenko> that's not good to me
15:08:45 <EmilienM> really, I don't review the latest patches in ec2api/watcher/... because patches are too big
15:08:51 <EmilienM> we can't effectively review this code
15:09:15 <EmilienM> iurygregory: yes
15:09:40 <iurygregory> one way to me is, locally you can have 2 branchs ( old_code and inital_commit)
15:09:58 <iurygregory> old_code will have all code you have send to your repo before
15:10:27 <iurygregory> inital_commit you can 'remove' and just add the structure for cookiecutter&mysnc
15:10:58 <iurygregory> after have the patch upstream you can rebase old_code with the upstream code..
15:11:08 <EmilienM> well, we're trying to solve a problem that should not exist
15:11:12 <degorenko> well, in such case we can break already prepared (and may be used by someone? ) module - if we will remove old stuff
15:11:14 <EmilienM> people should initiate modules in our forge
15:12:19 <EmilienM> iurygregory, degorenko, radez: can you guys sort this out on #puppet-openstack?
15:12:30 <iurygregory> EmilienM, yep
15:12:34 <degorenko> ok
15:12:49 <EmilienM> thx
15:13:11 <EmilienM> #topic ci status
15:13:26 <EmilienM> beaker jobs are moving from Trusty to Xenial this week
15:13:31 <EmilienM> only puppet-swift is blocking
15:13:45 <EmilienM> if someone can have a look
15:13:59 <EmilienM> it is related to the swift init thing that vinsh wrote
15:14:22 <EmilienM> vinsh: I was saying
15:14:35 <EmilienM> puppet-swift is currently blocking us to switch our beaker jobs to xenial
15:14:44 <EmilienM> can you please look at it?
15:14:55 <vinsh> Your talking about the hashsum mis-match issue?
15:14:56 <EmilienM> my infra patch is going to land so puppet-swift gate will break
15:14:59 <EmilienM> yes
15:15:07 <EmilienM> thanks for helping!
15:15:28 <vinsh> Yeah, I am looking at it.. but might not have an easy fix.  Worst case we might have to disable some of the acceptance tests sections on that module till a fix is found.
15:15:30 <EmilienM> also, iberezovskiy and I are working on promoting xenial jobs from mitaka to newton
15:15:41 <EmilienM> we have a blocker in ubuntu packaging that should be solved maybe today
15:15:49 <EmilienM> iberezovskiy: can you follow up on this at your tomorrow morning?
15:15:51 <vinsh> I am reading through the core-puppet "file" provider code to see where their puppet bug is at... its tough.
15:16:01 <iberezovskiy> yep, of course
15:16:11 <EmilienM> vinsh: maube ask _ody for some help, he's a puppet4 superstar
15:16:32 <_ody> Yeah that if fine.
15:16:41 <EmilienM> and a last thing about CI, I'm working on RDO trunk promotion (last promotion was Friday) and right now everything is broken but under control. Will keep you updated on our channel
15:16:42 <vinsh> Right on.. this one is ugly.
15:16:49 <_ody> I have time today too.
15:16:56 <EmilienM> _ody: thanks
15:17:01 <EmilienM> do we have anything else about CI?
15:17:46 <EmilienM> #topic release status
15:17:57 <EmilienM> less than one month before newton-2
15:18:38 <EmilienM> I would like to see some WIP on acceptance testing for new modules, ie: ec2api, watcher, tacker and congress (I'm currently writting puppet-octavia)
15:18:50 <EmilienM> can we have a least one person per module as a liaison?
15:19:12 <EmilienM> radez: you ok to be liaison for congres & tacker?
15:19:31 <EmilienM> basically, we need to make sure we have some packaging in RDO so we can deploy the services in our beaker jobs and actually test the module
15:19:51 <bkero> I can take one
15:19:55 <radez> EmilienM: sure
15:20:05 <degorenko> i guess danpawlik can be liason for watcher ? and mflobo for ec2api ? thoughts?
15:20:17 <EmilienM> thanks
15:20:19 <radez> bkero: which one?
15:20:21 <EmilienM> yes +1
15:20:22 <iurygregory> mflobo have time for ec2api?
15:20:33 <degorenko> don't know - that's why asking here
15:20:35 <bkero> radez: I'll handle congress if you'd like to give it up
15:20:46 <iurygregory> he is not here XD
15:20:54 <radez> bkero: yea that works for me, I'll take tacker
15:20:56 <EmilienM> basically, /join #rdo, and ask for some help if needed
15:21:05 <EmilienM> it can take time but we'll do out best
15:21:08 <iurygregory> we should ask him degorenko ,+1 =)
15:21:16 <EmilienM> excellent
15:21:27 <EmilienM> #topic open discussion
15:21:43 <EmilienM> do we have outstanding bugs or patches / or any questions this week?
15:22:41 <EmilienM> seems like no
15:22:52 <EmilienM> /me afk all afternoon
15:22:56 <EmilienM> have a good week folks!
15:22:59 <EmilienM> #endmeeting