22:02:10 #startmeeting horizon 22:02:10 Meeting started Tue Sep 10 22:02:10 2013 UTC and is due to finish in 60 minutes. The chair is gabrielhurley. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:02:11 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 22:02:14 The meeting name has been set to 'horizon' 22:02:21 hi folks 22:02:29 Hello 22:02:29 hello 22:02:32 hello 22:02:33 Hello 22:02:33 hey all 22:02:39 Heya 22:02:46 hello 22:02:54 hi 22:03:19 #topic overview 22:03:29 I'm gonna paraphrase ttx from the start of the last meeting here... 22:03:33 The havana-3 milestone has been cut, and we're in feature freeze. The two FFE blueprints that came up are also merged. Thanks everyone for work on that. 22:03:46 The goal this week is to come up with a clear list of release-critical bugs (bugs targeted to the havana-rc1 milestone) 22:03:55 This is usually best done by going through all bugs and targeting the appropriate ones... then monitor the new ones filed 22:04:07 Wwhen we get that list down to 0 we can create the havana release branch and tag our first release candidate 22:04:14 Then master can be open for icehouse business 22:04:33 Note that each project can have their RC1 at a different time (it happens whenever ready). Ideally all RC1s would be out before the end of the month. 22:04:50 ttx also created an Icehouse-1 milestone for us for targeting things forward 22:05:14 (I copied and pasted most of that, just changing some grammar) 22:05:26 Any questions on all that? 22:05:30 Nice 22:05:35 Daisy is planning around Sept 20th-24th to have the Horizon translations ready, btw 22:05:39 When is going to happen some planning for Icehouse? 22:05:40 awesome 22:05:50 jcoufal: aside from Hong Kong? ;-) 22:06:03 there's no formal planning sessions scheduled outside of the summit 22:06:09 we can talk in these meetings, on the ML, etc. 22:06:09 gabrielhurley: yeah, aside from HK :) 22:06:20 if the core team needs to sync up we can schedule a google hangout or something 22:06:30 feel free to spearhead any discussions you want to have 22:06:42 sounds good, thanks 22:06:43 though the focus for the next week or two should be polishing the H release 22:06:54 of course 22:07:42 I think the release stuff is all I have for project-wide business. 22:07:48 #topic blueprints 22:08:05 the two Neutron FFE blueprints are merged as of today 22:08:26 there are a bunch of reviews to continue polishing those, but we nailed the target for getting them in. great job. 22:08:41 amotoki has been diligently posting patch after patch of improvements. ;-) 22:08:43 thanks everyone. 22:09:06 thanks amotoki_ 22:09:19 I'm not aware of any other blueprints that would be considered for FFE, so I think we are *truly* in feature freeze now. 22:09:28 which makes the "blueprints" portion of the meeting really short. 22:09:41 For this phase I'll carry over into a bugs section though 22:09:45 #topic bugs 22:10:08 My goal for this week is to do a complete inventory of the Horizon bug list and do two things: 22:10:35 1. move anything that I think is a regression in Havana or a (fixable) bug in a new feature into the RC1 milestone 22:11:13 2. Close out old tickets which are either fixed, have not been reproduced/updated in a long time, or which are no longer relevant 22:11:43 the current RC1 buglist is pretty short by comparison to previous releases 22:11:46 I expect it to grow 22:12:12 on the plus side, most of them are not exeptionally deterimental issues 22:12:43 As of right now everything seems well in hand 22:12:51 we'll see how it looks next week 22:12:59 does anybody have bugs they particularly want to call attention to? 22:13:38 for the other Core folks, feel free to triage bugs into the RC1 milestone, but if you do I recommend leaving a new comment explaining where the bug is at or why it's important, otherwise it might get triaged back out. 22:14:03 Makes sense 22:14:11 okay then... moving on 22:14:15 #topic open discussion 22:14:22 I have two items for the open discussion this week 22:14:42 1. I wanted to talk for a minute about our blueprint triage and targeting processes 22:15:38 I've noticed (particularly at the end of H3) that blueprints were just "showing up" in the H3 list. And not like when a community member proposes a BP for Havana and the like, but actually targeted and accepted... 22:16:08 I encourage Core folks to have the power to target/untarget blueprints and bugs, but we need to make sure we're in sync about them 22:17:02 My preference is that we discuss the priorities for a release/cycle at the start of the cycle so we set the full list appropriately (as we've done for most of this past Havana release) 22:17:19 and if there are BPs that should come in later we bring them up with the team at this meeting or via email 22:17:35 and anytime you change the targeting of a blueprint please leave a comment in the whiteboard section 22:17:59 I usually do this when I bump blueprints out, but I think it's good to do when targeting to a specific milestone too 22:18:07 gives us a good record and helps set expectations 22:18:17 anybody else have thoughts on that? 22:18:28 makes sense 22:19:03 gabrielhurley: would you expect reasoning in whiteboard also for the initial targeting of bp/bug? 22:19:07 Lemme rephrase... Core folks, sound good? 22:19:12 jcoufal: ideally 22:19:16 yep 22:19:18 Yep 22:19:22 makes sense. 22:19:36 yup 22:19:38 like, I might leave a note that says "targeting to H3 to make sure Nova has time to implement the feature before we do" 22:20:04 or "bumping to Icehouse to make sure we have lots of time to get this right" 22:20:14 which brings me to my second topic... 22:21:02 2. I've got some time this week from my employer to work on interesting projects of my own, and I'm currently spending that working with the realtime stuff (message bus listener and websocket server) 22:21:29 I don't have specific findings so far other than to say that the POC code was of great help to me in experimenting with it and it's definitely on the right track 22:21:52 I think we'll want to do a bunch of architectural work on how we go about transforming and filtering that data to turn it into items Horizon understands and can take action on 22:22:00 some sort of plugin model akin to ceilometer 22:22:12 cool 22:22:21 but I just wanted to report back that I am actively involved with it and it's totally gonna happen in I 22:23:19 the current implementation with tornado and socksJS may or may not be what we want to stick with, btw 22:24:06 we should think about how far back we want to try and add support (e.g. which fallbacks we care about behind websockets) 22:24:22 interesting topics for the future. 22:24:23 :-) 22:24:30 anybody else got stuff they want to talk about? 22:25:49 gabrielhurley: will there be a following-up in the summit about the version api? 22:26:02 ah yes. that. 22:26:08 that would be a good idea 22:26:20 I just got a bit burnt out on it before 22:26:26 the discussion tends to turn into a format war 22:26:54 I think it'd be good to look at what Nova has done in the v3 API and what Neutron's been up to and to try and bring that back into a single vision 22:27:16 I'm not sure if I have the bandwidth to run that this summit, so if someone else wants to own that I'd be in favor of playing a supporting role personally. 22:27:41 I've got another initiative I'm spearheading this time that will be unveiled/proposed soonish 22:29:45 anybody else? 22:29:50 if not, we'll make this a short meeting 22:29:56 I can look into the version api, need to read-up on the old thread to get up-to-speed. 22:30:25 yeah, I'd have to review too 22:30:38 I am interested in what the other projects have done in the interim though 22:31:09 any thought to the keystone v3 multi-backend support 22:31:27 I'll look into Nova and Neutron as suggested. 22:31:51 per domain backends like ldap for one, sql for another? 22:32:42 I think we're fairly insulated other than setup time 22:32:56 yeah, I think I can honestly say I have zero thoughts on that 22:32:56 domain creation that is 22:32:58 thoughts welcome 22:32:59 :-) 22:33:17 it's not a topic I've considered at all thus far 22:33:25 well done for finding one 22:33:34 I'll invest some more time looking into it, but that may have some implications in I 22:33:41 sounds good 22:35:40 okay. let's call it a short meeting and get on with our days/to bed depending on timezone. 22:35:46 have a great week folks 22:35:52 Thanks 22:35:54 #endmeeting