22:00:25 <gabrielhurley> #startmeeting horizon
22:00:26 <openstack> Meeting started Tue Mar 19 22:00:25 2013 UTC.  The chair is gabrielhurley. Information about MeetBot at http://wiki.debian.org/MeetBot.
22:00:27 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
22:00:29 <openstack> The meeting name has been set to 'horizon'
22:00:48 <gabrielhurley> hi everybody
22:00:56 <stevebaker> hi
22:00:59 <david-lyle> hi
22:01:06 <gabrielhurley> #topic Overview
22:01:19 <gabrielhurley> So, we've got ourselves what looks like an RC!
22:01:25 <gabrielhurley> https://launchpad.net/horizon/+milestone/grizzly-rc1
22:01:52 <gabrielhurley> There are a couple bugs I'd like to point out in the bugs part of this meeting, but none that I consider blockers at this stage
22:01:57 <ttrifonov> aloha
22:02:07 <vkmc> Hey all o/
22:02:09 <gabrielhurley> I also went in and added release notes and translations this morning
22:02:35 <gabrielhurley> the review to bump the version (and signal the cut of the RC) is in gerrit currently.
22:02:54 <gabrielhurley> unless we identify something that should be a blocker in this meeting I'll merge that and trunk will be open for Havana
22:03:26 <gabrielhurley> any fixes after that have to be explicitly ported to the Grizzly release  branch
22:03:30 <gabrielhurley> so let's try not to have to do that ;-)
22:04:03 <gabrielhurley> things are looking good though
22:04:22 <gabrielhurley> Writing the release notes is always a great exercise and I feel really positive about what we've accomplsihed in Grizzly
22:04:49 <gabrielhurley> it's a big stride forward in solving the most common UX complaints. it just feels better. plus the new features that made it in are very cool.
22:04:54 <lcheng> hello
22:04:55 <gabrielhurley> so great job to everyone.
22:05:06 <jpich> Go team
22:05:12 <gabrielhurley> any general/process questions before I move to bugs?
22:05:43 <gabrielhurley> going once
22:05:47 <gabrielhurley> going twice
22:05:47 <kspear> well done all
22:05:48 <lcheng> just  a quick question, does the 'switchable' class supposed to work on workflow as well?
22:06:07 <gabrielhurley> lcheng: good question, and off the top of my head I'd say yes, but I've never tried it.
22:06:52 <gabrielhurley> #topic bugs and blueprints
22:06:53 <lcheng> I was trying this out, couldn't get it to work. Might log a bug for this one. ;-)
22:07:11 <gabrielhurley> go for it
22:07:32 <gabrielhurley> just so people are aware, this late breaking bug was pretty serious: https://bugs.launchpad.net/horizon/+bug/1155876 and https://bugs.launchpad.net/horizon/+bug/1156198
22:07:33 <uvirtbot> Launchpad bug 1155876 in horizon "Quota updating failed with Http 400 Error" [Critical,Fix committed]
22:07:50 <gabrielhurley> it was caused by Nova failing to properly deprecate the inputs their API accepted
22:07:55 <gabrielhurley> we got a fix in, but that was a close one
22:08:34 <gabrielhurley> it's a reminder that we on the Horizon team have to be extra-diligent in being aware of other projects' changes
22:08:47 <gabrielhurley> and if they do things which aren't backwards compatible we need to call them on it
22:09:56 <gabrielhurley> looking forward to bugs that didn't make the RC blocker list
22:10:09 <gabrielhurley> this one sucks, but isn't the end of the world and I even release noted it: https://bugs.launchpad.net/horizon/+bug/1148372
22:10:10 <uvirtbot> Launchpad bug 1148372 in horizon "edit flavor deletes the flavor if special char is used" [Medium,Confirmed]
22:10:26 <gabrielhurley> we need to do more careful validation on the input there on our side before sending it to Nova
22:10:43 <gabrielhurley> I don't see it as a blocker
22:10:56 <gabrielhurley> if we roll an RC2 I might consider adding it in to be fixed
22:11:34 <jpich> I'm wondering about https://bugs.launchpad.net/horizon/+bug/1156817 and its impact on distros after seeing mrunge's comment (#3). Any thought?
22:11:35 <uvirtbot> Launchpad bug 1156817 in horizon "Regression due "rm invalid args when updating quotas with nova api "" [High,Fix committed]
22:12:25 <gabrielhurley> I'm not sold on that argument, for this reason:
22:13:06 <gabrielhurley> the argument that Fedora 18 includes an older cinderclient is moot because it doesn't include this version of Horizon. If you're gonna run an updated package for horizon why can't you update your copy of cinderclient?
22:13:26 <gabrielhurley> I don't like adding special-cased code paths for the clients in general.
22:13:50 <kspear> I agree. Fedora 19 is their target for Grizzly
22:14:08 <kspear> There are also existing incompatibilities
22:14:17 <kspear> the openstack_auth version for one
22:14:36 <stevebaker> fwiw, heat assumes a recent cinderclient
22:14:39 <gabrielhurley> If you wanna run latest, then run latest. Updating a package without updating its dependencies is a bad idea in general.
22:14:40 <jpich> That sounds like a fair point. I'm not super familiar with packaging but remember reading about the evils of pinning so it sounded worthwhile addressing that comment
22:15:00 <gabrielhurley> the only problem with pinning for distros is the distro needs to have that version or later
22:15:19 <gabrielhurley> I try to pin to wide swaths of versions where possible to avoid making it harder for people
22:15:56 <gabrielhurley> OpenStack as a whole could adopt a policy of supporting X versions back in the clients, and then I'd feel bound to do something about it
22:16:05 <gabrielhurley> but the clients are frustratingly wild-west style
22:16:13 <gabrielhurley> so I don't like supporting incompatible versions of them
22:16:43 <jpich> Cool, makes sense :)
22:17:22 <gabrielhurley> k
22:17:30 <gabrielhurley> that should probably be noted on the ticket
22:18:06 <gabrielhurley> anyhow, the other big bug area
22:18:24 <gabrielhurley> there's a slew of Quantum-related bugs, but given how late those features landed it's an unaviodable consequence. A sampling here:
22:18:25 <gabrielhurley> https://bugs.launchpad.net/horizon/+bug/1153525
22:18:26 <uvirtbot> Launchpad bug 1153525 in horizon "Some Lbaas fields require ints but declared as CharFields" [Low,Confirmed]
22:18:29 <gabrielhurley> https://bugs.launchpad.net/horizon/+bug/1151101
22:18:30 <uvirtbot> Launchpad bug 1151101 in horizon "The displayed network, router, topology info of admin user is confusion" [Medium,Confirmed]
22:18:36 <gabrielhurley> https://bugs.launchpad.net/horizon/+bug/1156433
22:18:37 <uvirtbot> Launchpad bug 1156433 in horizon "network topology can not show up in some version of firefox browser" [Medium,Triaged]
22:18:40 <gabrielhurley> https://bugs.launchpad.net/horizon/+bug/1146926
22:18:41 <uvirtbot> Launchpad bug 1146926 in horizon "Cannot create a network topology other than very simple" [Medium,In progress]
22:19:26 <gabrielhurley> I release noted some of those, and generally made it clear that these are late-breaking features and should be treated somewhat like a "beta" in Horizon.
22:20:24 <gabrielhurley> If anyone wants to argue for blockers out of any of that, now's the time
22:21:01 <gabrielhurley> okay
22:21:16 <gabrielhurley> those are the only bugs that really concerned me that I didn't mark as blockers for the RC
22:21:25 <gabrielhurley> anybody got any others they're deeply worried about?
22:23:25 <gabrielhurley> great!
22:23:32 <ttrifonov> https://bugs.launchpad.net/horizon/+bug/1154391
22:23:33 <uvirtbot> Launchpad bug 1154391 in horizon "Cannot launch Dashboard due to Import error " [Medium,Confirmed]
22:23:49 <ttrifonov> This is not a blocker
22:24:10 <gabrielhurley> the PR to openstack_auth is merged, and I'm gonna cut a new release for it to PyPI today
22:24:29 <ttrifonov> coool
22:24:31 <gabrielhurley> that only affects people (devs) who had an old version of django installed but are running newer versions of the code
22:24:42 <gabrielhurley> I have to take off in about 5 minutes, but before then I'll open it up for general discussion
22:24:50 <gabrielhurley> #topic open discussion
22:25:02 <kspear> one comment on the release notes
22:25:06 <stevebaker> can we talk Heat integration?
22:25:15 <kspear> the theme support bp became a no-op i think
22:25:32 <kspear> the review was abandoned
22:26:15 <gabrielhurley> kspear: I thought it got merged
22:26:45 <gabrielhurley> stevebaker: is there something specific you want to talk about for it prior to the summit in a couple weeks where it can be discussed in depth?
22:26:51 <stevebaker> I'm doing some development on the Heat UI at the moment. I won't be at Summit, and I hope to have something worth pushing well before that
22:27:15 <gabrielhurley> stevebaker: gotcha. I haven't had time to focus on what's there so far so I may not be of immediate use.
22:27:18 <stevebaker> assuming that fits with release/feature roadmap
22:27:23 <kspear> gabrielhurley: https://review.openstack.org/#/c/20804/
22:27:24 <gabrielhurley> oh yeah, absolutely
22:27:42 <kspear> gabrielhurley: unless there was another review i missed?
22:27:57 <gabrielhurley> kspear: the blueprint got marked implemented... https://blueprints.launchpad.net/horizon/+spec/theme-support
22:28:00 <gabrielhurley> now I'm not sure
22:28:14 <gabrielhurley> looks like that was inappopriate
22:28:25 <stevebaker> I'm paring back Dan's work to the bare essentials, I'll probably have a few newbie questions for #openstack-horizon
22:28:26 <kspear> the conclusion was it was already possible to do via django
22:28:47 <gabrielhurley> kspear: okay. that needs to be fixed in the release notes. I'll put up a review to axe it in a moment.
22:29:09 <kspear> gabrielhurley: great
22:29:13 <gabrielhurley> stevebaker: sounds good. this meeting is a good time to ask questions in general. today it's just a lot of RC business and I have to take off ASAP
22:29:31 <stevebaker> ok
22:30:03 <jpich> stevebaker: You're welcome to, the channel is quiet but there are a few of us keeping an eye on it and answering questions in there
22:30:42 <stevebaker> when I've got something ready to review I'll push to gerrit, just to discover what you're expectations are for merging
22:30:45 <gabrielhurley> I'm bad about it, sadly. I have too many chat windows open already :-(
22:30:58 <stevebaker> jpich: cool, thanks
22:31:16 <gabrielhurley> release notes fix: https://review.openstack.org/#/c/24853/
22:31:23 <gabrielhurley> okay. I gotta go.
22:31:31 <gabrielhurley> I'm gonna close the meeting but feel free to keep talking
22:31:34 <gabrielhurley> thanks everyone!
22:31:36 <gabrielhurley> #endmeeting