21:01:39 <ttx> #startmeeting
21:01:40 <openstack> Meeting started Tue Apr  5 21:01:39 2011 UTC.  The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:01:41 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic.
21:01:48 <ttx> Welcome everyone to our weekly meeting...
21:02:09 <ttx> Today's agenda:
21:02:12 <ttx> #link http://wiki.openstack.org/Meetings
21:02:27 <ttx> #topic Current release stage: QA
21:02:41 <ttx> We are 2 days away from GammaFreeze, which will happen late on April 7.
21:02:52 <ttx> So we have two more days of unrestricted bugfixing frenzy.
21:03:08 <ttx> #info Once we hit GammaFreeze we are less than one week before release, so we want to avoid introducing last-minute regressions
21:03:23 <ttx> #info We concentrate on fixing a subset of release-critical bugs ("RC bugs"), which will be added at:
21:03:27 <ttx> https://launchpad.net/nova/+milestone/cactus-rc
21:03:35 <ttx> https://launchpad.net/glance/+milestone/cactus-rc
21:03:39 <ttx> https://launchpad.net/swift/+milestone/1.3-rc
21:03:54 <ttx> If after GammaFreeze you want to propose random bugfix branches for merging, you should ask for their inclusion in the RC bugs list.
21:04:17 <ttx> See process at http://wiki.openstack.org/GammaFreeze -- basically you need to explain why the benefit of the bugfix clearly outweighs the risk of regression that it brings
21:04:25 <ttx> Questions ?
21:04:43 <soren> How does that actually work?
21:04:53 <soren> I mean.. How do I "ask for their inclusion tin the RC bugs list"?
21:05:03 <ttx> "See process at http://wiki.openstack.org/GammaFreeze"
21:05:03 <soren> Where do I click? :)
21:05:06 <soren> Ah :)
21:05:45 <ttx> Next topic...
21:05:53 <ttx> #topic Cactus Release status
21:06:07 <ttx> We have 8 "High" bugs for Nova, 1 for Glance and 1 for Swift
21:06:25 <ttx> We are still finding rather high-impact bugs every day, so I fear Cactus is still not the "highly stable" release we hoped for...
21:06:41 <ttx> But probably the best we can do until we get stronger automated smoketests on commits
21:06:52 <ttx> Thoughts on how stable Nova/Glance are ?
21:07:49 <blamar__> Unknown due to lack of sufficient testing?
21:07:53 <jaypipes> ttx: Glance is more stable now that functional tests are run on each commit and migrations are actually being tested properly. At least, that's my impression ;)
21:08:37 <ttx> jaypipes: we are still bitten by some bugs like bug 747855 which I find surprising
21:08:38 <uvirtbot`> Launchpad bug 747855 in nova "Incoherent use of is_public creates havoc when using euca-describe-images" [High,Triaged] https://launchpad.net/bugs/747855
21:08:54 <jaypipes> ttx: not a Glance bug.
21:09:30 <ttx> jaypipes: oh, sure, was meant as an example of "high impact bugs we still find"
21:09:37 <jaypipes> ttx: it's a Nova bug due to the image service abstraction translation of keys going to and from Glance or any other image service.
21:10:19 <ttx> anyway, we'll make it as stable as we can :)
21:10:20 * jaypipes was under the impression we would get Nova smoketests automated through Hudson for Cactus?
21:10:42 <ttx> jaypipes: If we get them, we get them too late.
21:11:12 <westmaas> I think Vek has some tests in a branch, not sure if they are ready to go.
21:11:45 <Vek> I do indeed have some tests in a branch; been working on them with a local QA guy.
21:11:46 <ttx> We have high hopes for Diablo, as we'll get a QA coordinator soon
21:12:01 <ttx> Recent stats on the Nova stabilization effort:
21:12:03 <Vek> they focus on the openstack API, v1.0
21:12:05 <soren> What is a "Vek"?
21:12:10 <soren> Oh, it's a person!
21:12:11 <soren> Hi!
21:12:12 * Vek is a Vek
21:12:16 <westmaas> haha
21:12:26 * glenc is not a Vek
21:12:27 <ttx> Last week we had 65 bugs opened and 27 fixes committed
21:12:33 <ttx> This week we had 32 bugs opened and 44 fixes committed
21:13:30 * Vek is very glad glenc is not a Vek
21:13:47 <cynb> one Vek is enough
21:13:52 <ttx> Looks like we would have needed 2 more weeks of bugfixing and two less weeks of feature development... but that would suppose everyone participates to bugfixing, which is not the case anyway.
21:14:17 * ttx stops his late rant and switches to next subject
21:14:25 <ttx> #topic Welcome to our Diablo PTLs
21:14:38 <ttx> Congrats to vishy, jaypipes and notmyname for their election as PTL !
21:14:50 <ttx> I started annoying them to get the design summit process finalized...
21:14:51 <vishy> \O/
21:14:59 <vishy> ^ note the big head
21:14:59 <uvirtbot`> vishy: Error: "note" is not a valid command.
21:15:07 <johnpur> thanks ttx for jumping on this!
21:15:52 <ttx> Looks like we are in agreement that the process at http://wiki.openstack.org/Summit will be used for this summit, so keep on filing sessions
21:16:52 <westmaas> is there a deadline for filing sessions?  Still looking, but haven't seen it yet.
21:17:12 <ttx> westmaas: no deadline... but might be a good idea to have one, actually
21:17:32 <spectorclan> ttx: I will have room names and sizes by day for you tomorrow after we finalize in the morning my time
21:17:33 <ttx> westmaas: usually people come up with new ideas the day the summit starts
21:17:45 <ttx> spectorclan: cool, that's what I need right now
21:17:56 <ttx> spectorclan: any way we could stay in the same rooms for the 3 days ?
21:18:07 <spectorclan> yes, that is what we are trying to do. I see no reason why we can;t
21:18:23 <ttx> spectorclan: ok, cool
21:18:34 <ttx> ok, moving on to the next topic
21:18:36 <sandywalsh> ttx, is it possible to point an existing BP at summit for inclusion? Or do they need to be new bp's?
21:18:58 <ttx> sandywalsh: it is possible. Just "propose for sprint" ods-d
21:19:04 <sandywalsh> thx
21:19:07 <ttx> #topic Dedicated Doc Day
21:19:16 <ttx> annegentle: Floor is yours...
21:19:26 <annegentle> thanks ttx.
21:19:52 <annegentle> Just wanted to give y'all the encouragement to drop-everything-and-doc this Wednesday.
21:20:17 <annegentle> I'll be on Skype at annegentle, IRC at the same, and either merging into the openstack-manuals project or poking core for merges for RST docs
21:20:33 <annegentle> I'm happy to answer any questions about where a certain bit of doc should land, formats, templates, etc.
21:21:39 <annegentle> Any questions? Comments? Doc to fling?
21:22:32 <ttx> I'll definitely give some wiki pages a lifting
21:22:33 <soren> is there a todo list?
21:22:46 <soren> Or should we just go mad on anything and everything?
21:23:00 <annegentle> soren: http://wiki.openstack.org/Documentation/ToDo <--- great starting point
21:23:19 <annegentle> priorities for this release were docstrings and reference info (flags for example)
21:23:39 <annegentle> and API documentation
21:24:14 <annegentle> I'd also appreciate reviews of the revisions on docs.openstack.org - I plan to have a new top-level page pointing to both Bexar and trunk/Cactus docs by tomorrow.
21:24:29 <annegentle> so it can also be drop-everything-and-review-doc day if you wish
21:24:46 <sandywalsh> annegentle, it was a good tip that you can do RST for the wiki
21:25:12 <annegentle> sandywalsh: yes, it's a great way to write once pub twice
21:26:06 <annegentle> that's it! Thanks all.
21:26:25 <ttx> annegentle: I'll try to update http://wiki.openstack.org/FlagsGrouping as well
21:26:50 <ttx> #topic Open discussion
21:26:56 <ttx> anything, anyone ?
21:27:01 <spectorclan> Is everyone registered for Summit?
21:27:15 <ttx> Any issue I should know about, 9 days before Cactus release ?
21:27:18 <spectorclan> We updated the max from 350 to 390 today so I have more slots
21:28:10 <ttx> spectorclan: One of the issues with the way we handled registration is that we have no clue how many people are attending conference, and how many stay for design summit.
21:28:16 <westmaas> ttx: where is that review priority page again?
21:28:29 <soren> westmaas: http://wiki.openstack.org/reviewslist
21:28:31 <spectorclan> I have that data and can send it to you shortly, just need to run the report for you
21:28:38 <soren> westmaas: Err... No.
21:28:41 <ttx> westmaas: what soren said
21:28:45 <soren> westmaas: http://wiki.openstack.org/reviewslist/
21:28:46 <soren> There.
21:28:50 <soren> (now with a slash at the end)
21:29:49 <vishy> we have over 350 registered?
21:30:01 <spectorclan> vishy: we have more like 400+
21:30:07 <spectorclan> big waiting list
21:30:32 <ttx> vishy: should be fun to try to have a productive technical discussion with ~200 people in the room
21:30:46 <jaypipes> ttx: will be fun with even 20 people in the room.
21:30:54 <ttx> jaypipes: ;)
21:31:01 <spectorclan> vishy: i expect most will be for conference and not deisign summit
21:31:11 <ttx> jaypipes: I could think of one fun discussion involving just 2 people.
21:31:28 <jaypipes> heh
21:33:00 <masumotok_> Can we see the list of registered bp for diablo summit? the result of https://blueprints.launchpad.net/sprints/ods-d/+addspec is same as https://blueprints.launchpad.net/nova/diablo?
21:33:30 <ttx> masumotok_: we still have to review the current proposals and accept them
21:33:49 <ttx> masumotok_: we = PTLs (+ me on the generic release discussions)
21:34:15 <masumotok_> so few days laster, everyone can check! :)
21:34:26 <eday> ttx, jaypipes: I think we shold designate one room those types of discussio for entertainment purposes :)
21:34:35 <ttx> masumotok_: some of them appear on https://blueprints.launchpad.net/sprints/ods-d already
21:34:55 <ttx> eday: with a ring and giant gloves, like justinsb suggested
21:35:39 <masumotok_> tttx: good list, thanks!
21:35:44 <eday> ttx: totally, how many OSS projects say they've made decisions based on hand-to-hand combat?
21:36:16 <ttx> eday: I was kinda expecting you and soren end up in the mud over eventlet vs. twisted
21:36:26 <ttx> eday: I was disappointed.
21:36:31 <Vek> heh.
21:36:39 <annegentle> whiffle bats? Yes I watch Attack of the Show.
21:36:40 <dragondm> I can think of a few linux kernel process scheduler discussions that probably involved blood ...
21:36:43 <eday> ttx: I'm a pacifist
21:36:47 <ttx> eday: your calm is so disarming.
21:37:54 <ttx> ok, let's wrap up
21:38:03 <ttx> and get some more bugs fixed
21:38:10 <ttx> #endmeeting