09:04:45 #startmeeting ptl_sync 09:04:46 Meeting started Tue Feb 24 09:04:45 2015 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:04:47 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 09:04:49 The meeting name has been set to 'ptl_sync' 09:04:51 #topic Heat 09:04:54 sorry, a bit late 09:05:19 #link https://launchpad.net/heat/+milestone/kilo-3 09:05:41 What's the status of those "Unknown" ? Just not started, or really unknown ? 09:06:06 the convergence ones are been worked on 09:06:12 i'll update those 09:06:35 the barbican one i might push to L 09:06:57 OK, and keep "Unknown" for when you have no idea -- use "not started" for those that are just not started 09:07:08 ok, makes sense 09:07:27 Remember Feature Proposal Freeze is Thursday next week, and ISTR you said you would follow it* 09:07:43 that means that everything should be "needs code review" in 10 days :) 09:07:58 wow, ok 09:08:04 so yeah, early defrring of stuff not started that won't make it sounds like a good idea 09:08:21 I mean, you can also decide to not enforce FPF 09:08:41 but generally that doesn't really result in better situations at the end of the line 09:08:48 ok 09:09:00 especially if you piled up enough reviews for the last 2 weeks already 09:09:01 i'll chase the status of these up 09:09:35 i am hopefully most of those will land 09:09:42 they are small bp's 09:09:54 just broken up to spread the load 09:09:54 ok 09:10:00 On the design summit space request side, you indicated that you would confirm the proposed allocation in a meeting 09:10:14 when do you need that info? 09:10:27 well, I don't need a final request, just an idea 09:10:42 we have a meeting in 1.5days 09:10:43 But I need it asap 09:10:54 i ways hoping to canvas there 09:11:03 ok, that works 09:11:07 cool 09:11:48 That's about all I had. i'll add proposed changes to release tracking to discuss at the cross-project meeting tomorrow morning for you 09:12:09 ok 09:12:14 If you can't make it --> https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking 09:12:26 thanks 09:12:39 asalkeld: anything else ? 09:12:46 all ok from my side 09:12:57 zz_johnthetubagu: awake? 09:13:04 asalkeld: ok, have a great day 09:13:08 err..; evening 09:13:15 ttx: you too 09:16:59 ttx: hi, just going to head down to get a coffee, will only be 2 mins hopefully 09:17:26 standing by 09:23:35 ttx: hi 09:24:02 #topic Nova 09:24:05 johnthetubaguy: o/ 09:24:27 #link https://launchpad.net/nova/+milestone/kilo-3 09:24:43 Is that ~current with the Goals + non-goal FFEs list ? 09:24:57 yes 09:25:18 although the non implemented FFEs will get bumped today/tomorrow, in theory 09:25:50 We have feature proposal freeze next week on Thursday 09:25:54 so I need to chase some folks down to see how much they will have by feature proposal freeze 09:26:02 does it look like your goals will make it ? 09:26:29 honestly, a few are touch and go, but looks like they should all get something useful up 09:26:41 I will try push people for answers on some of those 09:26:43 ok 09:27:19 Separate topic, I asked mikal for an idea of the mix between fishbowl and work room sessions for the next summit 09:27:35 did you get an answer from him? 09:27:48 (I expect nova to use 18 sessions as always, but I need an idea of what the mix should look like 09:27:50 ) 09:27:53 No he didn't yet 09:28:01 Was wondering if you discussed it in any Nova meeting 09:28:16 so I think going with the assumption of the same as last time, but I haven't brought that up I am afraid 09:28:54 when is the drop dead time for an answer? 09:29:01 johnthetubaguy: the new thing is that for normal sessions, you can pick between fishbowl where you want large attendance, and work rooms where you would rather keep it for the nova team 09:29:13 ah got you 09:29:37 and that's a team discussion on how you best want to use your time there 09:29:57 other teams generally pickes a 1/3 2/3 mix (1/3 fishbowl) 09:30:01 yeah, for sure, we had lots of good comments about the mix last time 09:30:19 anyway, I'll reach out to mikal 09:30:33 cool, I will put it on the nova meeting agenda before I forget 09:30:55 I'll also discuss at the cross-project meeting tonight evolution in release tracking for Liberty, as previously discussed here 09:31:04 If you can't make it --> https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking 09:31:11 ah, cool, thank you 09:31:21 That's all I had 09:31:29 Anythign on your side ? 09:31:37 cool, thanks for the updates, I don't think there is much with us 09:31:46 ok, ttyl then 09:32:05 thanks, talk to you soon 09:37:54 ttx: FWIW I am totally +1 your ideas in that liberty tracking wiki page, a really nice writeup 09:38:08 cool, thx 13:00:10 ttx: knock, knock, ready when you are 13:01:20 o/ 13:01:26 #topic Ceilometer 13:01:44 #link https://launchpad.net/ceilometer/+milestone/kilo-3 13:01:50 Looks all good 13:02:26 Had a question on design sumit space requests -- you mention on the sheet that you still need to confirm your proposed allocation 13:02:35 shall I consider the guess confirmed now ? 13:02:43 It's not a final request anyway, just an overall idea 13:02:52 so we can work on room layout 13:03:07 yeah, does the initial request look out of line to you? 13:03:21 no no, was just wondering how final that guess was 13:03:30 much less crazy than others iirc 13:03:47 cool, otherwise, I'd be happy go with that 13:03:49 ... with the understanding that there's more horse trading to come 13:03:55 definitely 13:03:58 coolness 13:04:24 In other news we'll discuss liberty release tracking at the cross-project meeting tonight 13:04:44 mostly the idea of switching to tracking what landed rather than predicting what will likely not land 13:04:44 just thinking BTW, does danpb's "modest proposal" warrent a cross project meeting agenda item? 13:04:52 a-ha OK 13:04:55 #link http://lists.openstack.org/pipermail/openstack-dev/2015-February/057614.html 13:05:15 ... or let that discuss play out a bit more first 13:05:19 *discussion 13:05:19 Missed the call for today's meeting, and discussion is only starting 13:05:27 yeap, fair enough 13:05:29 i expect it to bleed onto that release trackign discussion though 13:05:38 yeah, I expect so too 13:05:39 and definitely a topic for next week 13:05:42 agreed 13:06:20 OK, that is all from me 13:06:26 yep, that's all I had also 13:06:30 thanks for your time! 13:06:37 remember FPF is Thursday next week, if you want to enforce it 13:06:53 yeap, I intend to 13:10:17 SergeyLukjanov: o/ 13:11:00 o/ 13:11:04 #topic Sahara 13:11:04 ttx, hi 13:11:13 #link https://launchpad.net/sahara/+milestone/kilo-3 13:11:13 #link https://launchpad.net/sahara/+milestone/kilo-3 13:11:17 oops :) 13:11:39 Looks relatively good, lots of not started 10 days from feature proposal freze 13:12:21 FPF is March 5? 13:12:36 last time I looked yes 13:12:41 But then you can ignore it 13:12:58 since two weeks befroe FF might be a bit heavy for you 13:14:48 yeah 13:14:57 (was checking the release page) 13:15:08 anyway, those "High" not started should probably get goin,g 13:15:18 I'm not expecting new big proposals 13:15:38 ttx, it's a good chance that we'll postpone several of them :( 13:16:14 In other news we'll discuss the possibility to simplify release tracking for liberty at the Cross-Project meeting today 13:16:26 If you can't make it --> https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking 13:17:05 yeah, I saw this item, I'll be on the meetings 13:17:06 That is all I had. Questions on your side ? 13:17:25 (I'm just mostly always very quiet on this meetings :) ) 13:17:34 nope, I think nothing more from my side 13:17:35 heh 13:17:42 ok then talk to you later 13:17:47 dhellmann: around? 13:18:08 ttx, thx 13:19:18 ttx: I'm here, but also just diving into a nova issue 13:19:31 dhellmann: want to postpone ? 13:20:43 yeah, can I have 10-15 min? 13:21:00 I have all the people I need together, so we should be able to resolve the path forward shortly 13:21:12 sure 13:21:15 ping me when ready 13:21:20 i'm not moving anywhere 13:35:32 ttx: ok, that took longer than expected but I think we're clear 13:36:38 ok 13:36:43 #topic Oslo 13:36:50 #link https://launchpad.net/oslo/+milestone/kilo-3 13:37:14 dhellmann: I think you'll want to cull that list early rather than late 13:37:35 ttx: yeah, I need to weed 13:37:41 there is little point in releasing a new lib if nobody has time to switch to using it 13:38:03 do you have a deadline / set date for new graduations ? 13:38:04 well, I don't agree with that, but we haven't seen progress on a couple of those 13:38:20 no, we just work constantly with the idea that adoption can start during the next cycle if it needs to 13:38:47 dhellmann: ok, so we maintain the code on the incubator stable branch ? 13:39:09 right 13:39:10 i.e. if adoption happens over a release boundary it's not such a big deal ? 13:39:13 ok 13:39:14 hang on, we've written this down... 13:39:23 yes, taht rings a bell :) 13:39:51 #link http://specs.openstack.org/openstack/oslo-specs/specs/policy/incubator.html#graduation 13:40:49 hmm, that's actually a bit unclear 13:40:59 "After the first release of the new library, the module(s) should be removed from the master branch of the incubator. During this phase, only critical bug fixes will be allowed to be back-ported to the prior stable branches." 13:41:41 so... new lib is released, code is removed from master, stable branch is cut from master 13:42:00 there is nowhere to backport to for projects that have not switched to the lib yet 13:43:04 we expect them to use the previous stable branch -- so things we delete in master are maintained in stable/juno right now, for instance 13:43:04 issue detected in nova kilo copy of the code -- fix will only be on that copy because there is no incubator stable branch to sync from ? 13:43:35 nope, the library, then stable/juno are fixed and nova syncs just that module from stable/juno 13:44:06 hmm, wait let me illustrate 13:44:26 nova has a copy of incubator foobar module 13:44:53 nova syncs new feature added on oslo-incubator master branch 13:45:11 oslo.foorbar is released 13:45:28 foobar code is removed from oslo-incubator master branch 13:45:37 nova releases kilo 13:45:52 issue detected in nova's foobar code copy 13:46:23 that issue is only present in nova's code copy, not on any of the remaining oslo-incubator branches 13:47:03 oslo-incubator master doesn't have it, stable/kilo doesn't have it (being cut from master), stable/juno doesn't have it (since the feature was added in kilo) 13:47:30 so there is nowhere for nova to sync from, fix has to land in each code copy 13:47:59 ugh, my internet just bounced so I'm not sure if my last reply came through 13:48:04 (only happens if graduation happens on release boundary) 13:48:07 (replay) we have not yet had to change modules in the incubator once we identify them as candidates for graduation, so it hasn't been an issue, yet 13:48:14 right, that step "nova syncs new feature added on oslo-incubator master branch" -- when we identify something for graduation we would try to avoid that step 13:48:21 if it's a bug fix, we sync it to the stable branch, so that's ok 13:48:29 the rule is features are not added to the code in nova before they go into the incubator, so if someone violates that rule it's on them to deal with it 13:48:43 graduation candidates are frozen at the start of the cycle, basically 13:48:46 ok 13:49:03 works if people follow the rule. 13:49:04 so even if we don't get to it right away, we're not tracking differences that would cause issues 13:49:14 thx! 13:49:17 most of this code is pretty stable anyway by the time we're ready for graduation 13:49:35 so, as I said, it hasn't come up, but if it turns into an issue we'll adjust the policy 13:50:02 we used to allow changes and then try to do backports in master, but that made it harder to do things like allow the master branch to have the import statements updated to use graduated modules 13:50:16 #link https://launchpad.net/oslo/+milestone/next-kilo 13:50:45 You target the week before FF for the "likely kilo-final" versions of the libraries ? 13:50:53 that's right 13:51:13 OK, that's all the questions I had 13:51:28 we're working on finalizing that policy: https://review.openstack.org/#/c/153642/ 13:51:44 You might be interested in the discussion on evolution of release tracking we'll have at the meeting today 13:52:06 i.e. switch to tracking what landed rather than track what we think will land 13:52:14 I haven't caught up with email yet, is there an agenda with a link? 13:52:22 https://wiki.openstack.org/wiki/Meetings/CrossProjectMeeting 13:52:28 oh, well, yeah :-) 13:52:30 Topic summary @ https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking 13:52:45 does this relate at all to the product management working group? 13:53:18 dhellmann: not really. i reached out to some of them though to check how valid my assumtions were 13:53:28 ok 13:54:02 I'll read through that before the meeting, thanks for the heads-up 13:54:05 basically they prefer an accurate list of things being worked on, rather than a failed prediction at what will land next milestone 13:54:25 that doesn't mean you shouldn't have key goals and try to get those completed 13:54:40 right, I expect we'll still set goals for ourselves 13:54:41 but it wouldn't be release tracking job to push for those to get completed 13:54:49 * dhellmann nods 13:55:10 me beating the drum to get the goals completed is no longer really efficient 13:55:54 anyway, ttyl 13:56:08 does that mean no more 1:1 meetings next cycle? 13:56:50 the wiki page suggests to switch to on-demand, then syncs at each milestone/intermediary release 13:57:15 that makes more sense -- I'll read the rest so you don't have to type it all out here ;-) 13:57:18 thanks, ttyl 13:57:21 we could still sync weekly, since we are the only ones having a productive discussion there 13:57:41 (and that's because you actually do release management of Oslo* stuff 13:57:52 so we need to share tools/processes 13:58:03 dhellmann: ^ 13:58:25 but between milestones, I found the recent 1:1s to be pretty useless 13:58:37 (for non-oslo) 13:58:57 and I would be glad to have that time back 15:02:32 ttx: Locked and loaded and ready when you are :) 15:03:22 #topic Neutron 15:03:28 mestery: o/ 15:03:46 #link https://launchpad.net/neutron/+milestone/kilo-3 15:04:18 mestery: are all those unknowns real unknowns ? 15:04:27 ttx: I've been updating some of them the last few minutes :) 15:04:38 Some of them are unknowns, and I'm going to start removing them 15:04:38 ah :) 15:05:12 Other question that jumps to mind is how far are those essential things from being code-proposed ? 15:05:32 FPF is Thursday next week (if you still plan to enforce it) 15:05:44 ttx: I need to talk to kevinbenton on the first one, but he's out sick today, I'm hoping he will have something. 15:05:53 The one assigned to salv-orlando will hve some code land but not complete. 15:06:00 Some of the work armax is doign is in refernce to that BP 15:06:27 ok, so you expect a late landing or a ffe on that one ? 15:06:48 If anything, a ffe exception on kevinbenton's, salv-orlando's won't complete in Kilo. 15:07:02 I'll know more once I sync with kevinbenton when's feeling better :) 15:07:18 hmm, ok 15:08:07 Other topic is the design summit space allocation. You actually asked for a total of 20 sessions where there is only time for 18 (assuming you have a slot on every time on Wed and Thu) 15:08:21 so would be good to come up with a total of <=18 15:08:38 OK, I'll do that! Assuming that is total of fish bowl + working sessions? 15:08:42 yes 15:08:50 OK, I'll update that once our meeting is done to be <= 18 :) 15:09:06 Are we likely to get 18 if I move it to that? Or shoudl I go lower? 15:09:17 Whatever works for the broader community I'm good with 15:09:45 well, asking for slightly less would allow you to have sessions in common with Nova 15:10:17 so maybe aiming for 16 will give you some flexibility 15:10:30 ++, good idea! Thanks ttx! 15:10:40 I'd really like some overlap with nova and possibly QA/infra as well 15:10:51 also given that Neutron is generally a magnet for random people, not sure how well the working sessions will work for you 15:11:03 Yeah, we always face that problem as well 15:11:11 Fish bowels don't work well either to be honest 15:11:12 the idea being to not advertise those and make them unattractive, but those peolpe are persistent 15:11:21 having 200 people shouting about network protocols isn't anyone's idea of fun ;) 15:11:39 having 200 people standing up around a 25-people boardroom is even less fun 15:11:40 Stealth mode, I like it ;) 15:11:45 lol 15:11:59 The team is really focusing on making hte working sessions working sessions 15:12:08 Hopefully they are boring enough to keep the marketing people away ;) 15:12:28 8+8 sounds like a good tradeoff 15:12:40 + the full day friday 15:12:53 Ideally we just want a morning session Friday. 15:12:56 or 7+9 and the half day 15:12:59 yes 15:13:36 in other news... will discuss evolution in release tracking for Liberty at the cross-project meeting 15:13:44 I think I already discussed that with you 15:13:54 Yes! But I'm looking forward to that discussion in Vancouver 15:14:23 we'll discuss it today 15:14:34 so that things are ready when the cycle starts 15:14:44 Ah, ok, good! 15:14:46 Perfect 15:15:02 As I've said before, OpenStack is not afraid of change ;) 15:15:04 we migth end up discussing shorter cycles/releases in Vancouver instead :) 15:15:29 ok, that's all I had 15:15:38 That's a good discussion to have. 15:15:40 OK. that's all from me too. 15:15:42 Thanks ttx! 15:16:10 nikhil_k: ready when you are 15:16:18 ttx: o/ 15:16:24 #topic Glance 15:16:38 #link https://launchpad.net/glance/+milestone/kilo-3 15:16:59 Looks good, you might want to set a prio for https://blueprints.launchpad.net/glance/+spec/metadefs-notifications 15:17:33 Also Feature Proposal Freeze (the deadlien for getting code under review) is fast approaching (Thursday next week) in case you intend to enforce it 15:17:36 done 15:18:02 ttx: I was thinking March 12th for Glance 15:18:14 ok, makes sense 15:18:27 thanks 15:18:29 #info Glance may do FPF on Mar 12 instead of Mar 9 15:18:56 That's all I had, anything on your side ? 15:19:10 ttx: I've replied to the google doc for the session schedule 15:19:24 I was curious when would that be published? 15:19:24 nikhil_k: saw that, thx. That will help us work on room layout 15:19:44 this was just a poll to see how we should lay out the rooms 15:19:47 People want me to send email saying that Friday would have a sprint session to help plan travel early 15:20:09 ttx: ah ok, so sending such email would be ok then? 15:20:12 we are still expecting new project teams to appear and need space, so final allocation will happen much later 15:20:31 I think you can bet on at least a half-day Friday 15:20:47 oh, we just have 1 halfday so that seems doeable 15:21:16 yes 15:21:36 anything else ? 15:21:48 that was it, thanks! 15:21:57 nikhil_k: thx, talk to you later 15:22:11 cya, have a nice one 15:26:14 thingee: o/ 15:26:17 ready when you are 15:26:47 ttx: hi! 15:28:21 #topic Cinder 15:28:27 sorry, parallelizing :) 15:28:39 #link https://launchpad.net/cinder/+milestone/kilo-3 15:28:47 Looks pretty godo overall 15:28:50 good* 15:29:14 yes, a lot of these are very close. Some are not going to make the march 1st, code needs to be passing jenkins and ready. 15:29:38 I've been reaching out to people to contributors about march 1st. oh well 15:29:51 March 1st ? 15:30:09 http://lists.openstack.org/pipermail/openstack-dev/2015-February/056964.html 15:30:37 I will untarget bps not ready for review by march 1st. 15:30:44 ok, so FPF 4 days in advance 15:30:59 #info Cinder will do FPF on March 1st 15:31:30 "to people to contributors"...heh sorry still waking up 15:31:57 ok, and then.. review review review 15:32:11 ttx: I'm on it! Hopefully I'll have time with my own bps :( 15:33:13 ok, what else... we'll discuss evolution in release tracking at the cross-project meeting today 15:33:44 the idea that central release management should stop caring about the milestone pages and their bad predictions 15:33:53 feel free to join the fun 15:34:14 heh 15:34:55 that's about all I had in store for you 15:35:01 Anything on your side ? 15:35:07 nope 15:36:07 thingee: ok then, ttyl 15:36:15 david-lyle_afk: ping me when around 15:47:58 ttx: ping 15:48:45 david-lyle: o/ 15:48:49 #topic Horizon 15:49:30 #link https://launchpad.net/horizon/+milestone/kilo-3 15:49:48 Looks generally far from goal 15:49:57 https://blueprints.launchpad.net/horizon/+spec/static-file-bower has no assignee 15:50:22 How much of the feature proposal freeze on March 9 do you plan to enforce ? 15:50:24 I just need to assign it 15:50:55 FPF is code? 15:51:07 trying to recall 15:51:40 code for the blueprint must all be up for review 15:51:54 so that you don't lose time reviewing stuff that won't make it to release anyway 15:52:25 I suspect a couple of high priority items may come late, but I'll generally enforce it 15:52:55 so likely a couple of exceptions, otherwise I will enforce 15:53:12 right -- you might want to remind everyone of the deadline at your next meeting 15:53:28 will do, thanks for the reminder 15:53:35 what else... we'll discuss a lighter form of release tracking to implement in Liberty 15:53:39 at the cross-project meeting today 15:53:52 excellent 15:54:05 I look forward to ito 15:54:07 gist at https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking 15:54:08 *it 15:54:17 anything on your side ? 15:54:44 no I think we just need to get stuff landed 15:55:13 been sprinting to try help big ticket items make progress toward that 15:55:27 which has been fairly successful 15:55:32 but a lot still at risk 15:56:40 ack. I'll let you return to work then :) 15:56:44 ttyl 15:57:40 thanks, later 16:51:32 Ttx: we moved to 1650 right? /me might have put the wrong time on the calendar. 16:52:24 yes! 16:52:27 #topic Keystone 16:52:37 morganfainberg: sorry, too many fires to address at the same time 16:52:55 No worries! I was worried I picked the wrong time for my calendar :) 16:52:58 #link https://launchpad.net/keystone/+milestone/kilo-3 16:53:36 so so so... a bit late to hit FPF on Mar 5 (10 days from now) 16:53:41 So today we are scrubbing that list for anything that doesn't have a chance of landing / significant code done 16:53:50 ack 16:54:00 also one of those has "unknown" status 16:54:17 otherwise looks good 16:54:21 That one is a new Bp, but good work on it. 16:54:30 Iirc. Just was registered this morning. 16:54:46 what else... we'll discuss simplifying release tracking at the cross-project meeting later 16:54:46 Spec was a spec proposal exception but had POC code ready to 16:54:50 Go. 16:55:02 https://wiki.openstack.org/wiki/Release_Cycle_Management/Liberty_Tracking if you miss it 16:55:03 L1 specs will open after k3 is cut. 16:55:44 That looks good at a glance btw. The lighter version. 16:55:45 that's all I had. Anything on your side before you jump in another funny meeting ? 16:56:03 Nope that's it. 16:56:17 I'm off to get coffee :) and review code. 16:56:33 alright then, have a good day 16:57:44 notmyname: ready when you are 17:00:37 ttx: here 17:00:44 #topic Swift 17:01:15 notmyname: so.. sprinting, feature branches... 17:01:29 oh yeah, we didn't meet last week, did we? 17:01:30 notmyname: a vague idea of a window for the next kilo releases ? 17:01:37