14:00:28 <nikhil_k> #startmeeting glance_artifacts_sub_team
14:00:28 <openstack> Meeting started Mon Aug 24 14:00:28 2015 UTC and is due to finish in 60 minutes.  The chair is nikhil_k. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
14:00:30 <ativelkov> o/
14:00:32 <openstack> The meeting name has been set to 'glance_artifacts_sub_team'
14:00:33 <dshakhray> o/
14:00:44 <nikhil_k> We are all open discussion today :-)
14:00:50 <nikhil_k> #topic Open Discussion
14:01:12 <ativelkov> OK
14:01:19 <ativelkov> so, we have a number of reviews
14:01:32 <ativelkov> which await for some kind of a fast track mergement
14:01:37 <mfedosin> I'm here
14:01:41 <mfedosin> o/
14:01:53 <ativelkov> #link https://review.openstack.org/#/c/207374/
14:02:04 <ativelkov> #link https://review.openstack.org/#/c/214185/
14:02:17 <jokke_> o/
14:02:17 <ativelkov> #link https://review.openstack.org/#/c/214600/
14:02:32 <ativelkov> #link https://review.openstack.org/#/c/216180/
14:02:45 <ativelkov> These are mostly the bugs discovered when adopting V3 in Murano
14:03:00 <mfedosin> And there will another one for ranged vesion filtration today
14:03:35 <jokke_> ativelkov: May I raise one concern I added to the comments of 214600 as well
14:04:25 <mfedosin> folks, I want to discuss one thing - disable v3 by default in Liberty
14:04:47 <jokke_> ativelkov: I don't think the agreement with FastTrack was that you FT approve your own changes without review, no? I think the point was that the changes are reviewed within the Artifacts group and then FT'd
14:05:08 <mfedosin> I think it's reasonable, because no one except Murano won't use it in near future
14:05:39 <nikhil_k> mfedosin: makes sense
14:06:05 <jokke_> mfedosin: if the V3 is going to experimental, I would strongly agree with you
14:06:09 <mfedosin> I made a commit, but several tests failed there
14:06:36 <mfedosin> I'll fix it today, so can I ask you for review then?
14:06:58 <mfedosin> https://review.openstack.org/#/c/215709/
14:07:09 <jokke_> mfedosin: you probably need to hax some functional tests as the tests are ran once for default values, once for thew values specified in the configs iirc
14:07:37 <mfedosin> yeah, I've already made it
14:07:49 <jokke_> ah cool
14:07:54 <ativelkov> jokke_: sure
14:07:55 <ativelkov> I'll ask mfedosin to do that
14:07:55 <ativelkov> My +2 was that I am ok with the tests written to it by dshakhray :)
14:07:55 <ativelkov> (we have some kind of work breakdown between those who write code and those who write tests for it)
14:08:23 <ativelkov> oops, there was some IRC glitch
14:08:43 <jokke_> ativelkov: ah, fair enough, it was just that the commit was owned, authored, committed and +2'd by you ;)
14:09:01 <ativelkov> jokke_: sure, I got you. I'll change it to +1 for clarity )
14:09:16 <mfedosin> there is no approve by him :)
14:09:26 <nikhil_k> so, I think we all pretty much agree on keeping the default off for v3 in Liberty
14:09:31 <jokke_> ativelkov: thanks ... and as you can see, I keep eye for your FTs ;)
14:09:48 <nikhil_k> keep it EXPERIMENTAL and indicate that in the release notes and
14:09:52 <ativelkov> jokke_: thanks :) That's appriciated :)
14:10:24 <ativelkov> I have one more important question to discuss: how do we release EXPERIMENTAL python-glanceclient?
14:10:37 <nikhil_k> one email to tbe sent to ML that indicates the adoption by Murano
14:10:53 <jokke_> ativelkov: we have the feature branch, right?
14:11:07 <ativelkov> nikhil_k: ok, will do that as soon as all patches are submitted for review
14:11:10 <ativelkov> jokke_: yes, we do
14:11:15 <nikhil_k> someone was supposed to create that branch
14:11:22 <nikhil_k> ah cool
14:11:35 <ativelkov> https://github.com/openstack/python-glanceclient/tree/feature/artifacts
14:11:47 <ativelkov> Thanks to jokke_ who aske dhellmann :)
14:11:49 <jokke_> we just need to dedicate version number for it and request release when we need. Just making sure that it's clear enough being experimental due the artifacts. IMO
14:12:13 <nikhil_k> it's a feature branch so the client versioning won't apply on it
14:12:31 <jokke_> ativelkov: I just haven't been following that branch so I have no idea what you guys have been merging in there
14:12:33 <nikhil_k> ativelkov: I think it's best to indicate the rel-mgrs about this particular aspects
14:12:45 <nikhil_k> ativelkov: as they are handling release of the clients/libs
14:12:47 <ativelkov> jokke_: nothing has been merged yet
14:12:55 <jokke_> nikhil_k: if we want  to release from it, we need to have some version due to pbr
14:13:15 <nikhil_k> jokke_: :)
14:13:28 <nikhil_k> I think we should not release anything on the pip from it
14:13:38 <nikhil_k> rather encourage tarball installs
14:13:38 <ativelkov> #link https://review.openstack.org/#/q/status:open+project:openstack/python-glanceclient+branch:master+topic:bp/artifact-repository,n,z
14:14:29 <jokke_> ativelkov: those are all against master not against artifacts! :o
14:14:43 <ativelkov> jokke_: ouch!
14:14:46 <ativelkov> mfedosin: ^
14:14:49 <nikhil_k> -2 :P
14:14:52 <jokke_> yup
14:15:14 <mfedosin> no-no-no :D
14:15:32 <mfedosin> I'll fix it :)
14:15:35 <ativelkov> mfedosin: thanks
14:15:49 <jokke_> nikhil_k: you start from the top of thew list, I start from the bottom. Which one is first on https://review.openstack.org/#/c/197970/ :P
14:16:11 <nikhil_k> haha
14:16:17 <ativelkov> #action mfedosin to change client commit to feature/artifacts branch
14:16:31 <nikhil_k> for some reason gerrit is slow to get to today
14:16:41 <mfedosin> 'cherry-pick to' is okay?
14:16:42 <jokke_> mfedosin: just a hint, you probably want to to rebase of that artifacts branch against master first
14:16:53 <jokke_> might have some problems othewise
14:16:56 <nikhil_k> yep
14:17:04 <nikhil_k> cherry-pick ie
14:17:51 <nikhil_k> & remove the already PRS so that reviewers will be aware of such gotchas
14:17:56 <nikhil_k> PRs*
14:18:16 <mfedosin> we have no code intersection between master and artifacts, so I don't think there will be problems
14:18:39 <mfedosin> but anyway rebase is good
14:18:45 <nikhil_k> most likely not
14:18:54 <nikhil_k> but you could change common file
14:19:19 <nikhil_k> anyways, I think a fresh review shouldn't hurt. keeping a clean slate :)
14:19:31 <bunting> This review has been sitting a while, just thought I would mention it: https://review.openstack.org/#/c/200057/
14:19:50 <jokke_> nikhil_k: mind to add task to document how to build artifacts client for the liberty for those who wishes to try the experimental
14:20:00 <mfedosin> bunting, yes, I remember
14:20:21 <bunting> mfedosin: ah, thanks :)
14:20:40 <mfedosin> yup, I'll talk with Olena tomorrow
14:20:41 <nikhil_k> jokke_: good point, I think we need to chat with rel-mgrs on how they want to take it
14:21:00 <jokke_> nikhil_k: ++
14:21:01 <ativelkov> cannot we release it as version 0.x.1?
14:21:13 <mfedosin> they have holiday in Ukraine today, so they don't work
14:21:14 <nikhil_k> ativelkov: let's see. if need be
14:22:14 <ativelkov> The thing is that murano client uses glance (regular v2 images api) for its own needs. So, we have to have python-glance client in murano's client requirements.txt
14:22:43 <ativelkov> if we put tarball reference there, it will break the requirements job in gate
14:22:48 <nikhil_k> ativelkov: I wassuggesting if the can do a tarball install for liberty?
14:22:53 <nikhil_k> i see
14:23:13 <ativelkov> Technicallu, artifacts support it experimental in Murano in L
14:23:15 <jokke_> ativelkov: what if you use the git reference there?
14:23:49 <ativelkov> i.e. we have "legacy mode" which uses old murano's package repo instead of glance
14:24:24 <ativelkov> jokke_: didn't try, but I believe that requirements job fails in this case as that git reference is not part of global requirements
14:24:26 <nikhil_k> ativelkov: I think let's all hop on to the rel-mgr channel after this meeting
14:24:35 <nikhil_k> we have a freeze on libs soon
14:24:40 <ativelkov> nikhil_k: yup
14:25:06 <nikhil_k> anything else?
14:25:49 <ativelkov> not from my side
14:26:05 <mfedosin> I don't have any artifacts related questions too :)
14:26:20 <nikhil_k> ok, see you all on #openstack-relmgr-office after this.
14:26:22 <ativelkov> ok, then thanks all
14:26:23 <nikhil_k> Thanks all!
14:26:26 <nikhil_k> nice meeting
14:26:31 <mfedosin> thanks!
14:26:32 <nikhil_k> #endmeeting