16:00:22 <david-lyle> #startmeeting Horizon
16:00:23 <openstack> Meeting started Tue Jan 28 16:00:22 2014 UTC and is due to finish in 60 minutes.  The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:27 <openstack> The meeting name has been set to 'horizon'
16:00:37 <david-lyle> Hello Horizon folks
16:00:39 <lblanchard> hey all
16:00:42 <jpich> Hey
16:00:43 <t_mazur> hello o/
16:00:45 <akrivoka> hi
16:00:47 <jcoufal-mobile> _o/
16:00:47 <mrunge> good morning
16:00:54 <jomara_> ahoy
16:00:58 <lsmola_> hello
16:01:42 <jrist> what does one say to greet the 1600 UTC time
16:01:47 <jrist> I suppose evenin'
16:01:53 <matty_dubs> Good day!
16:01:59 <jpich> Hello is a safe one too :-)
16:02:09 <amotoki> hello
16:02:10 <jpich> Greetings
16:02:12 <doug-fish> Greetings!
16:02:14 <doug-fish> aww.
16:02:20 <jpich> hah :P
16:02:21 <rdopieralski> jrist: have a cup
16:02:29 <jrist> cheers!
16:02:43 <matty_dubs> jrist: But hey, it's 5 o'clock somewhere, so 'Good evening' works! ;)
16:02:49 <julim> hi all
16:02:55 <david-lyle> I don't have a lot of general info today since we're proceeding the project meeting now
16:03:07 <rdopieralski> matty_dubs: it's actually 5 here on the continent
16:03:27 <akrivoka> it's 5 pm in most of Europe
16:03:35 <david-lyle> On the plus side, the gate is back in good condition and the stable branches are also open
16:03:41 <t_mazur> Here we have 8pm instead of 2am
16:03:58 <david-lyle> Several items have merged in the last week
16:04:16 <david-lyle> So the frustration level should hopefully be subsiding
16:04:19 <mrunge> it's good to know, stable branches are back to merge things
16:04:42 <mrunge> that was not the case for the previous weeks
16:05:13 <jpich> Many thanks to the people who worked hard to address the gate bugs, on master and stable
16:05:24 <david-lyle> yes, once the milestone mass push went through, the required fixes could trickle in
16:06:14 <david-lyle> jpich: absolutely a lot of people have spent a great amount of time keeping things moving
16:06:19 <amotoki> one of the reason of failures of neutron isolated jobs seems to hit kernel bug :-(
16:06:46 <jpich> amotoki: ouch
16:07:02 <mrunge> amotoki, wow, that must be interesting to debug, though
16:07:35 * mrunge wonders, if that even differs from environment to environment
16:08:04 <david-lyle> #topic Icehouse-3
16:08:11 <amotoki> salvatore and russelb did great jobs. but nobody has identified why.
16:08:23 <david-lyle> https://launchpad.net/horizon/+milestone/icehouse-3
16:08:44 <david-lyle> This is extremely overloaded
16:08:58 <david-lyle> And I need to start paring it back to reality
16:09:30 <david-lyle> RBAC is moving along and should make it
16:09:44 <david-lyle> UX enhancement is also on track
16:10:14 <david-lyle> Django 1.6 needs to get done in icehouse
16:10:36 <mrunge> david-lyle, I don't think I can finish the separation
16:10:44 <jpich> The "Low" priority blueprints are not really being tracked for the release though, aren't they? It seems only a little (!) overloaded if we don't consider them
16:10:56 <amotoki> i have a quesiton on django 1.6 support for havana. it can be discussed later.
16:11:10 <david-lyle> The only thing blocking there is I want to get another release of django-openstack-auth that uses keystone v3 by default for authorization
16:11:35 <david-lyle> Once that's in we can propose the openstack/requirements change
16:11:40 <mrunge> david-lyle, I'll add that to the blueprint; lots of javascript is required to move out for separation.
16:12:31 <david-lyle> volume backups is stalled and needs a refactoring and is at risk at this point
16:12:45 <lsmola_> david-lyle: i will sent you update about ceilometer bps in a week
16:13:06 <david-lyle> Can others give me an idea if the bps assigned to you still make sense for icehouse and are likely
16:13:26 <david-lyle> Most service teams are going to feature freeze on Feb 18
16:13:36 <lsmola_> david-lyle: so far it doesn't seems that many patches in Ceilometer I need will land, or will land to late
16:13:37 <david-lyle> I think this may be a bit premature for us
16:13:57 <david-lyle> lsmola_: I may them all out then
16:14:04 <lsmola_> david-lyle: ok
16:14:09 <mrunge> I like the idea of a common feature freeze
16:14:16 <lsmola_> david-lyle: we can put them back if something changes
16:14:17 <mrunge> to focus on bug fixing
16:14:43 <david-lyle> And just to make sure everyone's clear if something is moved out, that doesn't block it from coming back in if it's ready
16:14:55 <david-lyle> mrunge, I do too, but 3 weeks is not much time
16:15:20 <mrunge> david-lyle, I absolutely agree
16:15:34 <doug-fish> david-lyle:  okay, so I'm new here, but what does it matter if something is moved out or not then?  (trying to get a handle on how we decide what is in or not)
16:15:56 <doug-fish> is "moving out" just a way that we communicate with each other?
16:15:59 <david-lyle> I think it's a good target, but Feb 18-March 5 I will be more likely to make an exception
16:16:15 <david-lyle> doug-fish release planning
16:16:16 <jpich> I'm happy with the progress on the integration tests, I think it'll be ready on time with a first set of tests (it's pretty self-contained and doesn't really affect the rest of the code base anyway so unlikely to be an issue). Cinder v2 support probably should be a blueprint too and is going well so far
16:17:05 <jpich> doug-fish: Icehouse-3 is the last milestone of the cycle so features that don't make it by the feature freeze deadline and get "moved out" of the list won't be included in Icehouse
16:17:27 <david-lyle> doug-fish: icehouse has 3 iterations in the release, we are in the third now, soon to finish.  Getting moved out at this point would mean your feature will not be part of the integrated icehouse release
16:17:53 <david-lyle> For places doing CI, this is not a big deal
16:18:06 <doug-fish> david-lyle, but didn't you suggest that even if its moved out, if its delivered in a timely manner it could still be delivered?
16:18:26 <david-lyle> Anything moved out will be targeted to Juno
16:18:28 <doug-fish> I don't want to derail the meeting too long with my novice questions.  Maybe I'll bring it up later in the regular channel.
16:18:43 <david-lyle> sure, let's take it offline
16:18:53 <doug-fish> great.  thx
16:18:55 <david-lyle> it's really about setting release expectations
16:19:08 <david-lyle> jpich: awesome
16:19:24 <amotoki> doug-fish: UNLESS the feature moved out is VERY important as a community.
16:19:56 <david-lyle> jpich: so that will be a tempest review?
16:20:28 <jpich> david-lyle: We decided at summit to have in the horizon repo to begin with, while we learn to write robust UI-driven tests
16:20:55 * david-lyle suffers short and long term memory loss
16:21:07 <david-lyle> jpich: thanks for the reminder
16:21:16 <jpich> david-lyle: I have a review up in openstack/horizon for running the tests, just chatting with a couple of people who had offered help writing the tests themselves then we'll start posting them in a 2nd review
16:21:28 <jpich> david-lyle: No worries! It's fresh in my mind atm :-)
16:21:55 <david-lyle> jpich: thanks for the hard work here
16:22:10 <jpich> :)
16:22:14 <david-lyle> mrunge: how are we looking on the separation?
16:22:36 <mrunge> david-lyle, well. looots of javascript stuff needs to go away
16:22:51 <mrunge> and I don't think, I can handle that in icehouse-3
16:23:07 <mrunge> all that js should be moved out
16:23:35 <mrunge> good thing is: we can do that step by step
16:23:43 <david-lyle> mrunge: so we may need to bump?  I think the js is a big effort, but a requirement for the split to work properly
16:24:06 <mrunge> yupp
16:24:22 <mrunge> I think, all that js is too much to handle here
16:24:28 <david-lyle> yeah
16:24:35 <david-lyle> summit topic for sure
16:24:40 <mrunge> e.g. we bundle jquery 1.7
16:24:46 <mrunge> which is ancient
16:24:59 <david-lyle> yes
16:25:08 <jpich> mrunge: Would additional hands maybe help, if there are people who can volunteer the time?
16:25:08 <david-lyle> but i-3 is too late to start changing all that
16:25:18 <mrunge> jpich, definitely
16:25:31 <mrunge> and tasks are doable in parallel
16:26:07 <david-lyle> well, lets see how things progress, if anyone wants to lend mrunge some help, please contact him
16:26:21 <david-lyle> jomara_: heat still on target?
16:26:39 <jomara_> david-lyle: yeah! no issues road blocking me at the moment other than my own incompetence!
16:26:40 * jpich rereading the blueprint - I forgot we want to depend on python libraries instead of bundling too
16:27:09 <mrunge> jpich, for what bp?
16:27:09 <jomara_> david-lyle: if any of those bps become risky time wise, ill let you know
16:27:21 <david-lyle> jomara_: thanks
16:27:37 <jpich> mrunge: the separation bp at https://blueprints.launchpad.net/horizon/+spec/separate-horizon-from-dashboard
16:27:43 <mrunge> ah yes.
16:28:28 <david-lyle> ok, any other bp or bug updates for icehouse?
16:28:29 <mrunge> that just moves bundling out of horizon.... jpich^
16:29:07 <david-lyle> mrunge: we can tackle parts for icehouse too, and leave the remainder for juno
16:29:21 <amotoki> I see RBAC for network is proposed as a part of neutron feature (router snat).
16:29:28 <mrunge> sounds like a plan for me david-lyle
16:29:34 <amotoki> I think It should be done as a separate patch.
16:29:51 <jpich> mrunge: I wonder if this should be a separate blueprint?
16:29:57 <david-lyle> amotoki: not sure I follow
16:30:33 <amotoki> I will comment it on the bp. https://review.openstack.org/#/c/54276/ it has no reference to BP
16:30:59 <david-lyle> does neutron not support policy already?  I may have made a bad assumption
16:31:20 <amotoki> neutron itself supports policy already.
16:31:33 <amotoki> the patch is for horizon.
16:31:55 <david-lyle> amotoki correct
16:32:02 <amotoki> david-lyle: so your assumption is correct.
16:32:42 * david-lyle wonders how to spin luck to look like knowledge
16:33:11 <david-lyle> ok, we haven't started digging into the RBAC checks in neutron yet, but hope to this week.
16:33:30 <david-lyle> The initial bp is just to guard the actions in the table view
16:33:41 <david-lyle> more extensive support is to follow in separate bp
16:34:07 <amotoki> agree.
16:34:27 <david-lyle> amotoki: excellent
16:34:53 <david-lyle> #topic Integration of Tuskar UI
16:35:01 <david-lyle> into the Horizon Program
16:35:15 <jrist> new Gerrit workflow?
16:35:41 <david-lyle> launchpad now has a tuskar-ui drivers group that includes existing tuskar-ui core and horizon core
16:36:16 <david-lyle> gerrit is now gated by horizon-core, I will work with infra to see if we can add another group
16:36:41 <david-lyle> I haven't pinged them about it yet because they've been just a little busy holding things together :)
16:37:35 <david-lyle> so right now, if a tuskar-ui patch is reviewed and ready to merge ping me or another horizon core and we will review and approve
16:37:59 <david-lyle> I'll let you know when I find out more from infra on what is possible
16:38:37 <david-lyle> But tuskar-ui is now officially part of the Horizon program as far as OpenStack is concerned
16:38:42 <david-lyle> questions?
16:38:43 <jpich> yay, welcome :D
16:38:46 <jcoufal> alright, thank you david-lyle for all this work
16:38:51 <jcoufal> \o/
16:38:53 <tzumainn> awesome, thanks!
16:39:15 <akrivoka> great, thanks David for making it happen
16:39:26 <lsmola_> yaaaay
16:39:33 <jrist> woot woot
16:39:33 <jtomasek> cool!
16:39:48 <lsmola_> thanks david-lyle
16:40:03 <amotoki> for horizon reviews: the following search pattern in gerrit will help you all:
16:40:06 <amotoki> status:open (project:openstack/horizon OR project:openstack/django_openstack_auth OR project:openstack/tuskar-ui)
16:40:22 <lsmola_> cool
16:40:35 <t_mazur> amotoki: thanks!
16:40:49 <david-lyle> We are also looking at setting up a separate weekly meeting for tuskar-ui, roughly around the same time, but on a different day
16:40:55 * jpich uses the Watched category
16:40:59 <david-lyle> to make sure both get enough time
16:41:09 <akrivoka> amotoki: thanks, that's helpful
16:41:14 * david-lyle is watching way too many projects
16:41:42 <jcoufal> david-lyle: if that's needed, we will try to figure out more details this week
16:42:19 <david-lyle> So we'll work that out and send an email to the mailing list
16:42:28 <jcoufal> yup
16:42:35 <david-lyle> #topic Free for all
16:42:43 <jcoufal> :) nice topic
16:44:05 <amotoki> the next havana stable release will be freeze this week: Jan 31.
16:44:06 <doug-fish> This is one of the most restrained fre for alls I have ever seen
16:44:38 <david-lyle> amotoki: are you handling the strings for this?
16:45:07 <amotoki> i haven't yet. I can take care of it.
16:45:23 <david-lyle> let me know if you want someone else to take it on
16:45:33 <amotoki> thanks.
16:45:53 <amotoki> i see backport patch series of django 1.6 support for stable/havana.
16:46:10 <jpich> amotoki: Thanks, for handling the strings
16:46:19 <amotoki> do we add django 1.6 support for havana?
16:46:48 <jpich> This sounds more like adding a new feature, and not really appropriate for a backport, imo
16:47:28 <david-lyle> I think the need would be based on how likely existing installs will update their django version
16:47:53 <david-lyle> which I wouldn't think is very likely
16:48:40 <david-lyle> mrunge, amotoki: do you have a strong opinion here
16:48:48 <mrunge> oh yes
16:48:59 <david-lyle> just checking :)
16:49:05 <amotoki> https://review.openstack.org/#/q/status:open+project:openstack/horizon+branch:stable/havana,n,z
16:49:07 <mrunge> there is a change, which makes it incompatible with earlier havana
16:49:38 <mrunge> https://review.openstack.org/#/c/66253/
16:50:07 <mrunge> the middleware change changes the cookie
16:50:09 <amotoki> no strong opinion on it, but mrunge's point is important.
16:50:19 <mrunge> which would become a pain for users
16:50:45 <mrunge> if that is improved, I'd be ok with it
16:51:37 <david-lyle> sounds like a questionable enough change to leave out
16:52:04 <david-lyle> the option is there for deployers to cherry-pick if this is important to them
16:52:24 <mrunge> if that part of the change is not applied, the whole backport does not make sense at all
16:52:31 <jpich> Agreed. A backward incompatible change that will cause issues to any user with a session cookie doesn't sound appropriate for backporting
16:52:40 <david-lyle> right
16:52:48 <david-lyle> mrunge, I was referring to all
16:52:50 <mrunge> I think that was MaxV to introduce this
16:52:56 <amotoki> and I am not sure django 1.6 support works without new openstack_auth release. I don't follow the status.
16:53:09 <david-lyle> amotoki: good point
16:53:45 <david-lyle> to much change for a backport 75% of the way to the next release
16:54:19 <MaxV> hi all o/
16:54:51 <mrunge> hey MaxV, could you please elaborate, why you're trying to backport patches for django-1.6 to havana?
16:55:50 <mrunge> it's a rather huge backport, so there must be a good reason for this
16:56:45 <MaxV> mrunge: zigo (a debian packager) tried to install on debian testing and was caught because of the compatibilty of django 1.6
16:57:11 <MaxV> the default package on testing use django 1.6
16:57:18 <mrunge> MaxV, would it be an option for him to wait until icehouse is released?
16:57:44 * mrunge wonders, if horizon is the only package breaking on Django-1.6 ...
16:58:04 <mrunge> we have basically the same situation for Fedora
16:58:55 <zigo> Hi there!
16:59:07 <mrunge> hey
16:59:24 <mrunge> zigo, you were asking for django-1.6 support in hanvan?
16:59:29 <mrunge> havana?
16:59:55 <mrunge> it's a rather big change, and currently still unfinished for icehouse
17:00:05 <zigo> mrunge: I already carry such a patch !
17:00:33 <mrunge> zigo, which version of django_openstack_auth are you using?
17:00:38 <mrunge> ... just curious?
17:01:08 <mrunge> no, please let's move that to #openstack-horizon
17:01:31 <mrunge> zigo, is there an option for you to wait until icehouse is released?
17:01:39 <zigo> http://anonscm.debian.org/gitweb/?p=openstack/horizon.git;a=blob;f=debian/patches/add-support-for-Django-1.6.patch;h=f777e6189b6ac6eb1f91d7ca237e077e41c43680;hb=9bd1026c637a49a765a97df0abc03ab9a4fc2d72 <--- A big fix for the url.defaults which I would like to see in Havana.
17:02:08 <zigo> mrunge: I'm working on Sid, and I would like everything to work there... It's *already* in Sid.
17:02:13 <mrunge> the middleware change is unlikely to be merged, zigo
17:02:22 <david-lyle> Looks like we're over time.  We can continue here or #openstack-horizon
17:02:27 <zigo> I use openstack-auth 1.1.3-2
17:02:33 <zigo> 1.1.3 I mean ...
17:02:33 <david-lyle> #endmeeting