21:01:42 <ttx> #startmeeting project
21:01:43 <openstack> Meeting started Tue Mar  4 21:01:42 2014 UTC and is due to finish in 60 minutes.  The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:01:44 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:01:47 <openstack> The meeting name has been set to 'project'
21:01:50 <ttx> #link http://wiki.openstack.org/Meetings/ProjectMeeting
21:02:03 <SergeyLukjanov> o/
21:02:09 <ttx> #topic Feature freeze is upon us (or not)
21:02:20 <ttx> Winter is coming
21:02:29 <ttx> Feature freeze shall hit at the end of this day (i.e. tomorrow morning Europe time)
21:02:40 <ttx> Still plenty of blueprints in-flight
21:02:49 <david-lyle> o/
21:03:04 <ttx> I plan to wait a few hours tomorrow morning to let the gate pipe empty a bit
21:03:18 <ttx> But there will be a number of blueprints that will just need a couple extra hours
21:03:32 <ttx> It's hard to tell how many of those there will be, though
21:04:03 <ttx> Does anyone here think that waiting an extra day would significantly change the numbers ?
21:04:14 <ttx> if it won't change anything, then we should just bite the bullet and cut on time
21:04:35 * dhellmann looks at his open items
21:04:40 <ttx> and then handle coherence and critical features using the FFE mechanism
21:05:05 <ttx> During the 1:1s nobody asked for a delay
21:05:21 <ttx> and the gate has held up (congrats jeblair and crew)
21:05:34 <dhellmann> yeah, kudos to the infra team, very stable
21:05:38 <ttx> the bottleneck was not really in the infra
21:05:42 <stevebaker> ttx: so does FFE mean a blueprint gets until rc1 to land?
21:05:51 <ttx> stevebaker: no
21:06:06 <ttx> stevebaker: it means it gets until an agreed deadline to land
21:06:21 <ttx> ususally, you'll hear things like "as long as it merges this week, I'm fine with it"
21:06:49 <ttx> because disruptive stuff has to land ASAP. Close to RC1 you might just introduce regressions, or new critical bugs
21:07:04 <stevebaker> ok
21:07:40 <jd__> we're pretty good on the Ceilomeer side FWIW, no need for extra delay
21:07:40 <ttx> RC1 is when you don't have release-critical bugs left. Your features have to land before enough for you to be able to catch the bugs in them if any
21:08:19 <ttx> OK, I don't hear anyone asking for an extra day, I suspect it wouldn't make that much of a difference
21:08:49 <markmcclain> I'd rather stick to time and deal with exceptions
21:08:53 <dhellmann> +1
21:08:57 <david-lyle> +1
21:09:13 <ttx> You should set things that you preemptively would like to ask a FFE for to "high". I'll push those to RC1 for further discussion, and defer to Juno anything else
21:09:25 <ttx> we can still adjust later
21:09:29 <ttx> but that would be the base line
21:09:38 <ttx> I'll cut the branches tomorrow morning Eu time
21:09:49 <ttx> and then catch up with you about FFEs during the day
21:09:58 <ttx> another long day ahead :)
21:10:17 <ttx> I would like to go through as many of them as possible, since I'm in vacation next week
21:10:30 <ttx> and I want to facilitate the job of the people who will replace me
21:10:54 <ttx> whom I should probably confirm now
21:11:06 <ttx> err
21:11:19 <ttx> OK. any question on that ?
21:11:20 <russellb> replace or cover for you?  :-)
21:11:25 <ttx> cover*
21:11:28 <russellb> ok good
21:11:47 <ttx> mostly using common sense to discuss FFEs
21:11:53 * russellb nods
21:12:05 <ttx> sdague volunteered some time ago, I hope he is still up for it :)
21:12:10 <sdague> ttx: sure
21:12:13 <ttx> yay
21:12:27 <ttx> sdague: Those weeks sure passed fast
21:12:33 <sdague> yeh, no lie
21:12:55 <ttx> #topic Clean Log enforcement, take 2 (sdague)
21:13:06 <ttx> sdague: floor is yours
21:13:09 <sdague> thanks
21:13:30 <sdague> back during i2 we started enforcing no *new* errors in logs
21:13:43 <sdague> because we spent a month building a whitelist
21:13:51 <sdague> then the giant 2 weeks of gate hell happened
21:14:06 <sdague> and we turned it off, because it was add additional races
21:14:31 <sdague> however, we really do want to figure out a way to make forward progress on this
21:14:43 <sdague> so the new proposed idea is to enforce 1 log file at a time
21:14:56 <sdague> so if we find a log file is clean, we lock it down
21:15:05 <sdague> like n-cond.txt
21:15:08 <sdague> it's clean, lock it down
21:15:19 <dhellmann> I like the incremental approach
21:15:28 <markmcclain> that's sound very manageable
21:15:29 <sdague> so a change which causes an ERROR or TRACE on a successful test run, is marked a failure
21:15:34 <ttx> once clean, always clean
21:15:38 <sdague> then tackle these one at a time
21:15:44 <sdague> instead of the whitelist approach
21:16:03 <sdague> and mostly wanted to get buy in from PTLs on the approach
21:16:10 <russellb> willing to give it a shot
21:16:14 <sdague> before we start enforcing
21:16:20 <russellb> maybe not this week, heh
21:16:27 <russellb> but sounds like a reasonable approach to see how it goes
21:16:34 <stevebaker> I like this plan
21:16:38 <sdague> yeh, probably I'd vote to start it next week
21:16:49 <sdague> and figure out what we can safely lock down
21:17:08 <dhellmann> sdague: we're doing something similar in oslo with python 3 support in modules
21:17:22 <dhellmann> the pattern makes me feel good about the approach :-)
21:17:35 <sdague> dhellmann: cool :)
21:17:45 <sdague> ok, that was it. Mostly open for questions if people had them.
21:17:52 <sdague> or see what objections exist
21:18:19 <ttx> sounds good
21:18:34 <ttx> looks like everyone likes it
21:18:42 <ttx> next topic ?
21:18:57 <ttx> #topic Red Flag District / Blocked blueprints
21:19:03 <ttx> markmcclain: you mentioned Neutron IPv6 patches state potentially adversely affecting Horizon
21:19:26 <markmcclain> yes
21:19:45 <markmcclain> I expect a few the IPv6 patches will land Wed/Thu
21:20:05 <markmcclain> and amotoki who works on both projects said there were a horizon items waiting on them
21:20:10 <ttx> markmcclain: so that's part of a blueprint that would get an FFE ?
21:20:24 <ttx> or just bugfixes?
21:20:38 <markmcclain> yes
21:20:42 <ttx> david-lyle: you have a blocked blueprint on that, right
21:20:56 <david-lyle> yes, neutron-subnet-mode-support
21:21:10 <ttx> so this would likely need an FFE as well
21:21:39 <ttx> ok, it's High already$
21:21:47 <david-lyle> yes
21:22:27 <ttx> markmcclain: you already have 3 blueprints on https://launchpad.net/neutron/+milestone/icehouse-rc1
21:22:45 <ttx> markmcclain: was wondering if those were abusive placement or preemptive FFE strikes
21:22:57 <ttx> Ukraine-style
21:22:59 <markmcclain> pre-emptive
21:23:10 <ttx> ok
21:23:26 <ttx> markmcclain: so we'll review them tomorrow as well
21:23:34 <markmcclain> yes
21:24:04 <ttx> david-lyle: I think you can move neutron-subnet-mode-support to icehouse-rc1 at this point, unlikely to be unblocked in time
21:24:12 <david-lyle> ack
21:24:14 <ttx> Any other inter-project blocked work that this meeting could help unblock ?
21:25:01 <ttx> #topic Incubated projects
21:25:14 <SergeyLukjanov> ttx, o/
21:25:18 <ttx> kgriffs, SergeyLukjanov, devananda: o/
21:25:26 <SergeyLukjanov> ttx, everything done for i3 https://launchpad.net/savanna/+milestone/icehouse-3
21:25:27 <kgriffs> o/
21:25:31 <ttx> https://launchpad.net/savanna/+milestone/icehouse-3
21:25:44 <ttx> SergeyLukjanov: OK, will cut your branch first thing tomorrow
21:26:00 <ttx> https://launchpad.net/marconi/+milestone/icehouse-3
21:26:03 <kgriffs> we have 2-3 patches in flight but they should land shortly
21:26:17 <SergeyLukjanov> ttx, docs/tests are planned for rc1 + several bug fixes and probably several improvements to one of the plugins
21:26:20 <SergeyLukjanov> ttx, thx
21:26:20 <ttx> SergeyLukjanov: planning to enforce feature freeze ?
21:26:35 <ttx> SergeyLukjanov: ok, that answers my question
21:27:02 <ttx> kgriffs: OK, when the page is all green (implemented / Fix committed) I'll cut
21:27:02 <SergeyLukjanov> ttx, yup, I think that we need only one/several FFEs and we're ok with trying to have FF
21:27:10 <kgriffs> ttx sounds like a plan
21:27:11 <SergeyLukjanov> ttx, the main FFE is for renaming :)
21:27:22 <ttx> SergeyLukjanov: you're the sole decider on your FFEs, fwiw
21:27:41 <ttx> I don't really need to be involved, except for friendly advice
21:27:55 <SergeyLukjanov> ttx, ok, great, thx
21:28:08 <SergeyLukjanov> ttx, I think that I'll have some questions :)
21:28:16 <ttx> kgriffs: just move off to -rc1 the stuff that didn't make it
21:28:45 <kgriffs> ttx: does that mean we can still land more features between i-3 and rc1 ?
21:29:08 <SergeyLukjanov> ttx, heh, I've already -2'd some own patches with comment that it'll be better to postpone to Juno
21:29:31 <ttx> kgriffs: yes you can, nobody really depends on you. The earlier you switch to bugfix mode the better the quality of the end release though
21:29:58 <kgriffs> ok, I was planning on just saying the FF for i-3 is FF for rc-1 as well
21:30:11 <SergeyLukjanov> ttx, is it correct that FF starts technically right after the milestone cut?
21:30:19 <devananda> ttx: o/
21:30:20 <kgriffs> and only allowing critical bug fixes into the branch
21:30:24 <ttx> kgriffs: yes
21:30:29 <ttx> SergeyLukjanov: yes
21:31:01 <ttx> devananda: https://launchpad.net/ironic/+milestone/icehouse-3
21:31:18 <devananda> ttx: last bp up there needs to be bumped (will do in a moment)
21:31:28 <ttx> devananda: just move non-completed stuff off i3 to icehouse-rc1 so that the page is all green
21:31:36 <devananda> ttx: ack
21:31:36 <ttx> (implemented / Fix committed)
21:31:47 <ttx> devananda: when it's in that situation I'll cut
21:31:51 <ttx> the MP branch
21:32:03 <devananda> will do right now
21:32:08 <ttx> then you may or may not attach feature-freezing semantics to that milestone
21:32:17 <devananda> would haev done this morning, but been distracted by the tripleo sprint
21:32:19 <ttx> enjoy it while you can :)
21:32:37 <devananda> hehe. we will be feature-freezing as far as major features
21:32:49 <devananda> but continuing to iterate on bugs as we add CI and integrate with tripleo
21:33:12 <ttx> sounds good
21:33:17 <ttx> any question on that ?
21:33:24 <devananda> nope
21:33:32 <SergeyLukjanov> ttx, probably, it'll be clearer to move FF to the March 4 in https://wiki.openstack.org/wiki/Icehouse_Release_Schedule?
21:33:45 <SergeyLukjanov> ttx, I can do it if you ack
21:33:50 <ttx> isn't it on Match 4 already ?
21:33:58 <ttx> FeatureFreeze, StringFreeze (Mar 4)
21:34:30 <ttx> SergeyLukjanov: looks good to me ?
21:34:34 <devananda> ttx: do I need to create the RC1 targets?
21:34:49 <SergeyLukjanov> ttx, heh, see it now, 01:34am makes me inattentive
21:34:53 <ttx> devananda: ah. Thought I did that.
21:35:19 <ttx> devananda: will fix now. Same for Marconi and Savanna
21:35:27 <devananda> thanks
21:35:38 <kgriffs> rock on
21:35:52 <SergeyLukjanov> ttx, I've already created rc1 in savanna
21:36:08 <ttx> SergeyLukjanov: coolthx
21:36:11 <ttx> done
21:36:49 <ttx> #topic Open discussion
21:37:16 <ttx> As discusse last week, design summit session suggestion site shall open on Thursday/Friday
21:38:16 <ttx> If there are no other questions or last-minute topic, let's enjoy 22 extra minutes
21:38:27 <ttx> i can certainly use them
21:38:50 <SergeyLukjanov> ttx, great, thank you
21:38:56 <dhellmann> sounds good
21:39:15 <ttx> #endmeeting