20:00:59 <david-lyle> #startmeeting Horizon
20:00:59 <openstack> Meeting started Wed Dec 16 20:00:59 2015 UTC and is due to finish in 60 minutes.  The chair is david-lyle. Information about MeetBot at http://wiki.debian.org/MeetBot.
20:01:00 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
20:01:02 <openstack> The meeting name has been set to 'horizon'
20:01:11 <david-lyle> who around ?
20:01:12 <hurgleburgler> ( ಠ‿ಠ)┘
20:01:14 <r1chardj0n3s> o/
20:01:15 <mrunge> o/
20:01:26 <rhagarty_> o/
20:01:42 <tsufiev> o/
20:02:02 <itxaka> \o/
20:02:18 <sqchen> o/
20:03:09 <david-lyle> ok, let's get started
20:03:49 <david-lyle> General announcements, django_openstack_auth 2.1.0 and 1.2.1 have both been released
20:04:22 <david-lyle> 2.1.0 introduced an issue with liberty Horizon and we've merged a fix, but will release 2.1.1 shortly
20:05:06 <david-lyle> second item, gerrit is down for an upgrade, I hope you've all seen the announcements, should be back up in an hour or so, IIRC
20:05:55 * tsufiev wonders if jobs start working better after that
20:06:28 <david-lyle> yes, grenade job for horizon is broken
20:06:33 <david-lyle> I believe
20:06:42 <david-lyle> can't release without gerrit though
20:07:20 <david-lyle> We're in the middle of M-2 which is slated to close Jan 21
20:07:41 <david-lyle> All things seem to be making progress
20:08:10 <david-lyle> are there any items people are worried about that were >= high priority for Mitaka?
20:09:30 <david-lyle> everything is perfect, wonderful
20:09:31 <r1chardj0n3s> xstatic has hit a wall; it's taken a bit of a back seat in the last couple of weeks so I've not been able to agitate for it in #infra
20:09:32 <mrunge> single page navigation?
20:09:49 <mrunge> is marked as high prio
20:09:51 <david-lyle> I was a little to quick
20:09:55 <david-lyle> *too
20:10:22 <r1chardj0n3s> neill focused on the swift ui re-work, which is now done, but he's unlikely to be able to re-roll his single page patch in the short term, but I can ask
20:10:58 <r1chardj0n3s> swift re-work for some values of "done" which include "needs a bunch more polish"
20:11:09 <tsufiev> r1chardj0n3s: isn't that the rework being discussed in Invision?
20:11:14 <mrunge> single page and concerns about performance somehow go hand in hand
20:11:33 <david-lyle> single page would be nice before M-3 so we have time to work through any issues
20:12:06 <david-lyle> before M-3 starts that is
20:12:13 <r1chardj0n3s> tsufiev: there's a few steps in the process; getting a basic implementation in angular is step 1; reviewing the interface is step 2; both of those are happeneing
20:12:25 <r1chardj0n3s> step 3 is work in the improvements from step 2
20:12:32 <tsufiev> r1chardj0n3s: got it
20:13:18 <r1chardj0n3s> I'll push single page up further on the priorities heap
20:13:18 <david-lyle> would really like to have xstatic resolved
20:14:01 <r1chardj0n3s> yep, just ran out of tuits(round) to chase that up with #infra
20:14:01 <david-lyle> or we need to make someone an admin on all the repos and bypass infra
20:14:13 <piet> Sorry I'm late
20:14:30 <david-lyle> s/repos/packages/
20:15:10 <david-lyle> any other items falling off the radar for Mitaka?
20:15:51 <r1chardj0n3s> I was also thinking that given the slow progress getting the infra stuff to work that we should (interim at least) just push out the packages we need to, and I can pypi admin people in to enable that to happen
20:16:11 <hurgleburgler> \o/
20:16:30 <david-lyle> r1chardj0n3s: sadly that sounds like the most expedient and reliable path at this point
20:16:53 <tsufiev> performance/sanity testing on scale - but I haven't yet done my homework david-lyle
20:17:05 <tsufiev> (Speaking of other items)
20:17:19 <r1chardj0n3s> performance testing at scale: I've applied for some OSIC time, but am in a queue
20:17:54 <tsufiev> r1chardj0n3s: good, how much time we have to come up with some scenarios?
20:18:38 * tsufiev hopes for at least 1.5 month :)
20:18:47 <r1chardj0n3s> tsufiev: honestly, no idea, but very likely at least that long :-)
20:19:04 * david-lyle hangs head
20:19:52 <david-lyle> at this point a plan for N would be progress
20:20:03 <david-lyle> rather than the vague notion
20:21:22 <david-lyle> last general item
20:21:46 <ducttape_> r1chardj0n3s - does an entire openstack release make it into that setup?  this seems like testing that is constantly / periodically needed
20:22:56 <david-lyle> I propose cancelling at least the next 2 weeks of meetings due to anticipated low attendance/availability
20:23:11 <hurgleburgler> I second that!
20:23:14 <mrunge> +1 david-lyle
20:23:49 <robcresswell> Yeah, next one should be the 6th
20:23:50 <r1chardj0n3s> ducttape_: as far as I'm aware there's no openstack-managed periodic testing using the OSIC facility
20:23:52 <tsufiev> less talk, more code :)
20:24:04 <r1chardj0n3s> +1 on no meetings as I'm not gonna be around anyway ;-)
20:24:12 <david-lyle> ducttape_: depends on the proposed test usage, but it won't be a stable environment
20:24:32 <ducttape_> thinking like for m release, test all the things, at scale, together
20:24:32 <david-lyle> are people back by the 6th?
20:24:48 <ducttape_> yy for skiping until 2016
20:24:53 <r1chardj0n3s> yup, 6th
20:24:58 <hurgleburgler> yeah
20:25:02 <rhagarty_> yes
20:25:04 <mrunge> yup, back on 6th
20:25:05 <piet> yes
20:25:05 <david-lyle> ok will send email canceling the next two
20:25:41 <david-lyle> I lied, one more general item
20:25:45 <david-lyle> Midcycle
20:26:00 <david-lyle> I am having trouble pinning down dates
20:26:25 <david-lyle> I have permission to have it in Portland (hillsborough) which was the top vote getter
20:26:36 <hurgleburgler> I like the 8th or 22nd, personally
20:26:51 <david-lyle> Mid-Late feb are the highest vote getters
20:27:10 <david-lyle> M-3 is slated for March 3
20:27:24 <david-lyle> so the week of the 29th is too late
20:27:43 <david-lyle> I am blocked the week of the 15th
20:27:57 <david-lyle> if I wasn't hosting, that wouldn't be as big a problem
20:28:11 <david-lyle> so that would leave the 8th or the 22nd
20:28:31 <david-lyle> or another local on the 15th and I will miss
20:28:39 <piet> Just as an FYI, the OpenStack UX mid-cycle will likely be in London
20:28:57 <david-lyle> suggestions?
20:29:01 <david-lyle> opinions
20:29:05 <piet> But hope to attend the Horizon mid-cycle if I get approved
20:29:13 <r1chardj0n3s> can't do 8th, sorry. 22nd would be ok
20:29:20 <mrunge> same here
20:29:22 <tsufiev> 22
20:29:27 <hurgleburgler> \o/
20:29:32 <mrunge> 8th is in parallel with devconf.cz
20:29:45 <mrunge> or unlikely to make it then
20:30:00 <robcresswell> I can do any time bar first week of feb
20:30:15 <hurgleburgler> piet when is the ux midcycle?
20:30:16 * ducttape_ can do any bar
20:30:26 <david-lyle> ok, will make is the 22nd
20:30:33 <r1chardj0n3s> \o/ a date!
20:30:36 <piet> We're still deciding..late Jan/early Feb
20:31:36 <david-lyle> for international travelers, day of the week preferred start?
20:31:47 <david-lyle> Tues-Thurs?
20:31:51 <r1chardj0n3s> Tuesday if possible please
20:31:53 <david-lyle> Mon-Wed?
20:32:05 <r1chardj0n3s> so yeah, Tues-Thurs
20:32:20 <david-lyle> ok, will plan on that
20:32:25 <hurgleburgler> that's easier for domestic peoples too
20:33:04 <mrunge> mon-wed would be cheaper for international
20:33:19 <mrunge> for me at least US$ 100
20:33:35 <mrunge> that probably makes a vacation day ;-)
20:33:40 <tsufiev> Interesting
20:33:41 <mrunge> works for me
20:34:00 <mrunge> flying out sat vs. monday saves US$ 500
20:34:31 <david-lyle> ok will schedule the room and line up some suggested hotels
20:34:44 <mrunge> awesome, thank you david-lyle
20:34:52 <r1chardj0n3s> thanks david-lyle
20:34:53 <david-lyle> thanks everyone for your patience
20:34:55 <tsufiev> david-lyle: thank you
20:35:27 <david-lyle> We have a couple of agenda items for today, found here:
20:35:29 <david-lyle> #link https://wiki.openstack.org/wiki/Meetings/Horizon#Agenda_for_December_16_2000_UTC
20:36:11 <david-lyle> #topic discuss local_settings.d/ patchset (see https://review.openstack.org/#/c/243974/ ) (toabctl)
20:36:18 <david-lyle> might be hard without gerrit
20:36:38 <robcresswell> This is the directory with local settings python snippets patch
20:36:42 <mrunge> esp. I don't see toabctl around
20:37:17 <david-lyle> ok, let's table that.
20:37:31 <david-lyle> when gerrit is back up hopefully people will take a look
20:37:47 <david-lyle> #topic will we keep creating tarballs for stable branches? mrunge
20:38:16 <david-lyle> mrunge: I'm not sure I understand the context
20:38:23 <mrunge> yeah, it was just a question coming from downstream
20:38:35 <david-lyle> point releases are or aren't going away?
20:38:43 <mrunge> at least, there has been discussion no to do any stable releases
20:38:55 <mrunge> so: point releases yes or no
20:39:09 <david-lyle> I'd have to reread, but I thought there was some nuance there
20:39:24 * david-lyle has to find mailing list thread too
20:39:34 <mrunge> at some point, that discussion went nowhere, iirc
20:39:49 <david-lyle> I'm not sure that point releases were killed
20:39:56 <david-lyle> but I could be wrong
20:40:05 <tsufiev> Our downstream is merging head of stable branch on a weekly, mrunge
20:40:20 <tsufiev> *weekly basis
20:40:28 <mrunge> tsufiev, I'm rebasing on each point release
20:40:37 <mrunge> cherry-picking important patches earlier
20:41:00 <mrunge> currently, we just have 14 patches on top of liberty
20:41:14 <david-lyle> mrunge: for horizon?
20:41:17 <mrunge> yes
20:41:26 <david-lyle> that seems light
20:41:30 <mrunge> at least 3 translation patches
20:41:34 <david-lyle> no one on liberty?
20:41:44 <mrunge> all on liberty
20:41:48 <david-lyle> still on kilo?
20:41:54 <david-lyle> I meant deployments
20:42:04 <mrunge> I mean, imports from translation tool
20:42:33 <mrunge> I would still like to cut point releases
20:43:19 <david-lyle> lhcheng isn't around, so I'll track down the current plan for stable
20:43:23 <david-lyle> for openstack
20:43:32 <tsufiev> mrunge: isn't it a thing being done for every tc-approved-project?
20:43:48 <david-lyle> tsufiev: there was some talk about not doing that any more
20:43:57 <david-lyle> just not sure where that ended up
20:44:36 <mrunge> and for django-openstack-auth, we did not do any stable backports in the past
20:44:58 <david-lyle> mrunge: we didn't have stable branches until kilo for d-o-a
20:45:05 <david-lyle> and 1.2.1 just released
20:45:06 <mrunge> yes
20:45:12 <david-lyle> which is kilo
20:45:36 <r1chardj0n3s> gerrit is back \o/
20:45:41 <mrunge> I just wanted to mention, it makes sense to keep horizon and d-o-a doing the same type of releases
20:45:58 <david-lyle> mrunge: sure, even though the destinations are different
20:46:03 <david-lyle> :)
20:46:11 <mrunge> yes, right
20:46:12 <david-lyle> let me track down the current plan
20:46:14 <robcresswell> ahhh its different
20:47:28 <david-lyle> and we can determine our plan from there
20:47:35 <david-lyle> robcresswell: no more old view support
20:47:39 <mrunge> great, thank you david-lyle
20:48:00 <david-lyle> #topic Open Discussion
20:48:12 <robcresswell> So on releasing packages
20:48:24 <robcresswell> Can we just use gerrit for voting, then manually release?
20:48:40 <robcresswell> Until we have a real solution
20:48:43 <r1chardj0n3s> sounds fair, yes
20:48:53 <david-lyle> robcresswell: that was the discussed short term work around, yes
20:48:54 <r1chardj0n3s> we'd have to do the gerrit voting anyway
20:49:15 <mrunge> david-lyle, as a pointer: it started here: http://lists.openstack.org/pipermail/openstack-dev/2015-May/065144.html
20:49:17 <r1chardj0n3s> just instead of magic tag-and-release we tag-and-manual-release
20:49:36 <r1chardj0n3s> and ping me if you need to be able to upload and can't
20:49:46 <robcresswell> Great
20:50:53 <robcresswell> I'd really like to see angular updated, and I remember diana had some bootstrap-related issues
20:51:28 <robcresswell> hurgleburgler: ^
20:51:45 <hurgleburgler> I'm gonna be starting on the pie charts today or tomorrow
20:51:48 <ducttape_> meaning that angular stuff is styled differently than the other stuff ?
20:52:10 <hurgleburgler> oh, we need to update the bootstrap to bring in some fixed … I've put in css debt to fix it int he meantime :(
20:54:04 <robcresswell> I meant the angular version we're using. It's 1.3.x... 1.5 is out soon iirc
20:54:39 <tsufiev> robcresswell: I thought you were going to say ng 2.0 :)
20:55:00 <hurgleburgler> :-O
20:55:34 <robcresswell> I looked at the 2.0 syntax earlier and screamed internally.
20:56:02 <david-lyle> funny I had that reaction with the 1.* series syntax
20:56:07 * david-lyle ducks
20:56:10 <r1chardj0n3s> :-)
20:56:39 <rajatvig> Actually, Angular Bootstrap in use (0.11) is no longer documented and possibly not supported.
20:56:54 <david-lyle> just how we like it
20:56:57 <hurgleburgler> :(
20:56:58 <robcresswell> So, upgrades
20:57:11 <robcresswell> r1chardj0n3s: Post-christmas lets get releasing some packages.
20:57:11 <david-lyle> might be nice to get back to the common era on the libs
20:57:17 <r1chardj0n3s> robcresswell: yep
20:57:22 <robcresswell> Following gerrit reviews of course.
20:57:38 <r1chardj0n3s> *cough* sure
20:58:26 <robcresswell> :o our gerrit version is actually reasonably modern now. I wonder if there are some nice APIs.
20:58:48 <piet> Hi Folks, were starting a meeting with Bruno from Catalyst in two minutes to discuss some concerns around network topology and the Launch Instance workflow
20:58:52 <hurgleburgler> We should just check them all in :-D
20:59:42 <piet> If you want to attend (first come/first serve) https://plus.google.com/hangouts/_/event/c8j8km27k3retmn7jgplp94g61c
21:00:07 <robcresswell> I'm on PTO now btw. So happy holidays everyone :D
21:00:08 <david-lyle> ok, times up. Thanks everyone and no meetings for two weeks. Enjoy the time.
21:00:15 <david-lyle> #endmeeting