11:48:21 #startmeeting ptl_sync 11:48:22 Meeting started Tue Jul 29 11:48:21 2014 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 11:48:23 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 11:48:25 The meeting name has been set to 'ptl_sync' 11:48:27 #topic Ceilometer 11:48:38 eglynn: how did j2 go ? 11:48:45 cool, d'ya wanna a do a quick juno-2 retrospective first? 11:48:56 well, gating for juno-2 was severly restricted by a couple of oslo-messaging bugs causing timeouts in the ceilo units 11:49:05 we didn't get the o-m 1.3.1 release until Friday 11:49:13 (and the test timeouts haven't occurred since) 11:49:19 \o/ :) 11:49:29 but in the end, only two j2 BPs needed last minute bumps to juno-3 11:49:41 dedicated-alarm-database was pretty much approved by Monday of last week 11:49:46 OK, you managed to get 7 bps in 11:50:17 ... but the gating Gods decided not to smile upon the many rechecks of dedicated-alarm-database 11:50:25 Looking at j3 now 11:50:29 https://launchpad.net/ceilometer/+milestone/juno-3 11:50:51 yeah so that, that's been fleshed out a bit 11:50:52 You have 8 blueprints targeted, which sounds pretty realistic, especially with one already completed 11:51:02 do you expect a lot more to show up ? 11:51:15 yeah, the whole SAD freeze turned out to be unenforcable 11:51:26 (with vacations, relocations, job changes etc.) 11:51:28 Remember you can now use "Blocked" if you want to show that a blueprint should be there but didn't have specs approved yet 11:51:38 a-ha, yeah 11:51:40 good call 11:51:50 so I've 3 more that I expect to land 11:51:52 that should allow us to present a more precise roadmap 11:52:02 I'll add blocked entries in LP for those 11:52:03 eglynn: ok, you can add them to the mix as Blocked 11:52:06 cool 11:52:10 will do 11:52:43 otherwise what can I say... merge as early as possible, I expect a busy end of August 11:53:01 yeap, I'm really banging the drum on feature proposal freeze 11:53:20 you plan to enforce FPF at the common date ? 11:53:21 ... despite being the worst offender myself in previous cycles for last minute patches ;) 11:53:33 august 21 iirc 11:53:33 that's the message I've been putting out, yes 11:53:48 ok great 11:53:56 eglynn: anything to add to meeting agenda for today? 11:54:26 I was gonna to raise the per-project functional test suite move 11:54:42 however after discussing with matt, turnout he'll be travelling 11:54:53 so I think it best to punt that to next week's agenda 11:54:53 yeah, and sdague is at the Nova sprint 11:55:02 ok 11:55:04 cool, that settles it so, next week it is 11:55:21 that's it from me 11:55:22 eglynn: ok, that's all I had 11:55:24 thx! 11:55:29 cool, thanks for your time! 11:55:30 * ttx finishes rant 11:55:38 LOL :) 12:04:08 ttx, hey! 12:04:30 ttx, are you ok to talk re sahara? 12:04:48 SergeyLukjanov: yep, rant finished 12:04:58 #topic Sahara 12:05:08 #link https://launchpad.net/sahara/+milestone/juno-3 12:05:09 SergeyLukjanov: how did j2 go for you? 12:05:27 j2 was fully completed 12:05:46 and we have good news - sahara-dashboard was merged to horizon 12:05:59 only some small fixes/features backports on review 12:06:23 so, we're going to close sahara-dashboard repo soon 12:06:58 SergeyLukjanov: this was completed by j2 ? 12:07:01 cool 12:07:09 ttx, several days after jw 12:07:13 j2* 12:07:22 ok 12:07:40 so... j3 12:07:49 https://blueprints.launchpad.net/sahara/+spec/edp-move-examples needs an assignee 12:08:34 You have 11 blueprints targeted, compared to 8 implemented in j3, sounds reasonable 12:08:34 ttx, yup, I'm looking for volunteers for it now 12:08:44 do you expect a lot more to creep in ? 12:09:16 #info looking for a volunteer to implement edp-move-examples 12:09:24 ttx, I think one bp re adding Apache Storm plugin will be added to j3 12:09:29 #info sahara-dashboard was merged to horizon 12:09:47 SergeyLukjanov: is it blocked waiting for spec final approval ? 12:10:00 ttx, oh, I'm wrong, it's already in j3 :) 12:10:31 ok, so the landscape here should be pretty complete ? 12:10:38 ttx, I forget that I've approved spec and targeted it to the j3 a week ago 12:10:43 ttx, I hope so 12:11:01 ttx, I'm not expecting any more huge stuff for j3 12:11:02 #info J3 plan supposed to be complete 12:11:28 SergeyLukjanov: do you plan to enforce Feature Proposal Freeze ? 12:11:40 Aug 21, feature code needs to be proposed by then 12:12:05 ttx, I think it 12:12:24 ttx, I think it's not very important for sahara atm 12:12:29 SergeyLukjanov: OK cool. Anything to add to agenda for meeting today? 12:12:46 ttx, nope, nothing from my side for meeting this time 12:12:58 SergeyLukjanov: ok thx 12:13:09 ttx, thank you! 12:13:56 dhellmann: ready when you are 12:14:34 ttx, just added one more bp to the j3 that was approved in specs 12:14:48 ttx: the toaster just went off, give me a sec to collect breakfast, brb 12:14:52 ok that makes 12 12:15:05 dhellmann: ping me when ready, no hurry 12:18:50 ttx: ready 12:19:31 argh phone call 12:19:40 * dhellmann finishes his breakfast 12:20:57 phone call ended 12:21:00 #topic Oslo 12:21:07 dhellmann: let me know when you're back 12:21:10 here 12:21:31 ttx: thanks for fixing the bugs :) 12:21:32 dhellmann: so.. juno-2 was a bit bumpy with new libs 12:21:58 dhellmann: we'll have to adjust the tools so that they workaround more of LP weirdness 12:22:12 dhellmann: unless you think we should go for one tracker per lib 12:22:26 yeah, I wasn't too worried about it because I think it only affected oslo.db and I expected a release shortly after 12:22:36 which will make this 1:1 a bit uinfun 12:22:37 although I've asked them to wait until the API change for exceptions is ironed out 12:22:43 yeah, I do *not* want that 12:23:13 dhellmann: is your lib tagging sufficiently consistent that we could rely on that in our scripts ? 12:23:15 I would almost rather manage the bps by hand 12:23:27 I suspect not 12:23:30 (the tagging) 12:23:49 I could work more on that, though 12:24:04 Not sure it would solve all issues tbh 12:24:08 yeah 12:24:18 oslo.messaging gets FixReleased at coimmon milestones as well 12:24:25 I'd actually like to be able to use the tool to mark bugs as released when we do cut a library release between milestones 12:24:56 dhellmann: you probably already can -- just need to have milestones there corresponding to versions rather than integrated openstack release milestones 12:25:04 ah, ok 12:25:22 The big problem here is that oslo.messaging for example still has juno-X milestones in 12:25:39 so you don't "release" in LP when you release in pip 12:25:42 yeah, and we don't know the alpha milestones for libraries in advance, since we just increment the value by one 12:26:19 I would have to create a milestone, move all the tickets into it, then use the tool to mark them as released 12:26:19 dhellmann: you can have a "next" milestone and rename it when you release, that's what we do with swift 12:26:23 ah, ok 12:26:35 next-juno 12:26:59 but that would not solve the other libs 12:27:06 we would need next-juno-db, next-juno-config, etc. 12:27:21 well... it would be pretty ugly in the timeline 12:27:23 yeah 12:27:35 how's storyboard coming along? :-) 12:27:35 also not make sense in milestone bug dropdown 12:27:48 It's awesome, you should try it 12:27:56 can't wait! 12:28:01 although projectgroup is not ready yet 12:28:06 which is what you want here 12:28:19 dhellmann: what we COULD do though 12:28:20 seriously, I'm OK with things being a little manual if it means energy goes into that instead of the "legacy" tool 12:28:47 is to create oslo libs projects in LP AND move them all under an oslo parent project 12:29:01 we don't use the openstack parent group that much anymore 12:29:04 I didn't know we could have nested projects 12:29:11 You can, it's just one level 12:29:23 and the project group can't contain bugs/BPs 12:29:34 and you need to beg LP admins to get one 12:29:51 how much longer do you realistically think we'll be using lp? 12:30:01 then you can still see bugs and blueprints at "projectgroup" level 12:30:08 is it worth moving all of the bugs and blueprints around? 12:30:21 So I think infra will migrate to SB for Kilo 12:30:36 then we'll iron out a number of issues 12:30:45 and during Kilo some adventurous projects will join 12:31:00 * dhellmann tries to decide if he's feeling adventurous 12:31:02 ideally so that we can have a strong proposal to move all by L 12:31:43 I think I might like to have oslo move early, rather than kludge something together for lp 12:31:50 dhellmann: with more libs being created, it might be worth to implement the oslo projectgroup 12:31:57 in LP 12:32:24 can we convert the existing project, or would we make a new one and leave that one for the incubator? 12:33:09 we would rename oslo oslo-incubator and ask for an oslo projectgroup 12:33:25 ok, that makes sense 12:33:45 dhellmann: yeah, I think it's realatively easy, and less weird thean trying to use tag to automate stuff 12:33:55 yeah 12:34:15 that said, we still need to convince that juno-X doesn't make sense for libs.. you would just have the same series (juno) not the same milestones as the integrated release 12:34:18 will spec2bp still work if the blueprints aren't in the oslo project? 12:34:32 I know markmc was pretty attached to having juno-x milestones in oslo.messaging 12:34:49 to share the common rhythm I think 12:35:01 yeah, he wants us to act like an integrated project 12:35:19 I'm torn on that. I want us to, but I also see some benefits to not. 12:35:47 spec2bp... hmm... trick is all specs fall into solo-specs, so that was working with oslo-the-LP-project quite well 12:35:55 yeah 12:36:07 ok, let's think about it a bit more 12:36:12 off-sync 12:36:14 we might need an oslospec2bp that lets me specify the project name 12:36:19 and look at j3 now 12:36:45 yes, some option to force project name 12:36:52 we'll need it for edge cases anyway 12:37:04 we've had to find some new drivers for a couple of the graduations for j-3, but did have some people volunteer at the last meeting so I think we'll get those back on track 12:37:10 true 12:37:10 https://launchpad.net/oslo/+milestone/juno-3 12:37:41 are the "unknown" there "not started", or really "unknown" ? 12:37:58 I think those are mostly not started; I'll fix that 12:38:18 we seem to be having more trouble than usual with bp statuses being updated (I thought that happened automatically?) 12:38:25 So that makes 14 BPs, compared to 6 in j2 12:38:45 I think that's doable if most of the ones under review are completed soon 12:38:53 yeah 12:39:08 Also I suspect the graduation ones are not that complex to complete 12:39:16 I'm not sure the cross-project testing one is going to make it. I've had some people object to the idea, so I need to talk to jeblair again 12:39:30 no, they're not hard, but time consuming because of all of the steps 12:39:46 sometimes keeping two on the list prevents to complete any, while keeping one you can complete it 12:39:51 I'm also not counting on anyone adopting those libraries this cycle, so we're going to push right through the rc period with that work 12:40:00 oh, ok 12:40:15 so you may graduate during the RC period, that sounds good 12:40:16 I'd like to have the projects thinking about them for k-1 12:40:33 https://launchpad.net/oslo.messaging/+milestone/juno-3 12:40:52 the amqp driver is blocked on having the dependencies packaged for the distros 12:41:12 gordon is working on that 12:41:14 but otherwise it's reasy? 12:41:16 +d 12:41:34 well, I'm not sure "easy" but there is some code up for review 12:41:48 no I meant ready 12:41:58 ok 12:42:00 ah 12:42:28 so both roadmaps are complete, or you expect new specs to make it in j3 ? 12:43:07 If yes, you can use "Blocked" to add them to the milestone plan until they are accepted 12:43:22 ok 12:43:23 * ttx converts the existing Blocked onto a "Needs Infrastructure" for clarity 12:43:27 I do have 2 new specs that were just proposed 12:43:30 ah, cool, thanks 12:44:10 there's also the one on switching mysql connection libraries, but that's a bigger cross-project concern still 12:44:12 dhellmann: that page is always a best shot at approximating what we think will be in juno... so if you think it will make it, you can add it as Blocked 12:44:39 ok, I'll go through and make sure it is up to date today 12:45:05 ok that's all I had -- anything to add to meeting agenda for today? 12:45:29 #action ttx to think a bit more how to solve the oslo/libraries/fixrelease/milestones conendrum 12:46:02 I don't know if it's meeting worthy, but today is my last day at Dreamhost. I start with HP on Monday, and am taking the rest of the week off. 12:46:48 Cool. But then the rumour already reached me 12:46:58 I've already started using a new email address on the ML, so it's out there if people need to reach me directly. 12:46:59 heh 12:47:16 dhellmann: congrats, i suppose :) 12:47:22 thanks :-) 12:47:30 ttyl 12:47:37 k 14:04:01 dolphm: jgriffith doesn't seem to be around, you can go now if you're around 14:04:16 ttx: o/ 14:04:18 #topic Keystone 14:04:31 dolphm: how did j2 go for you? 14:05:06 specs are still a slow process... which resulted in fewer blueprints and more bumped blueprints than i would have preferred 14:05:19 https://launchpad.net/keystone/+milestone/juno-3 14:05:30 7 blueprints there, compared to 3 landed in j2 14:05:34 on the upside, we seem to have a broader understanding and agreement on the things we landed much earlier 14:06:05 https://blueprints.launchpad.net/keystone/+spec/keystone-to-keystone-federation is marked "unknown" status -- does that mean not started, or really "unknown" ? 14:06:37 it should be just-barely-started 14:06:45 ok will mark it started 14:06:55 dolphm: do you expect more specs to appear on the j3 board ? 14:07:27 one in particular... i expect hierarchical-multitenancy to land, but it should end up in a feature branch, and not be merged with master until K 14:08:31 dolphm: ok, that should be tracked in "next" milestone then 14:08:36 future/next 14:08:41 okay 14:09:08 dolphm: otherwise the plan is thought to be complete ? 14:09:18 openid-connect will be j3 soon 14:10:20 ttx: IIRC, spec2bp was going to have an option to target unmerged specs? 14:10:45 is that still in review, or am i crazy? 14:10:59 still in review 14:11:09 I'm a bit behind 14:11:14 feedback welcome though 14:11:42 https://review.openstack.org/#/c/108041/ 14:12:06 you can already use the "Blocked" status to show up specs that WILL get approved and make j3 14:12:09 let me try it now... 14:12:19 that will make a more comprehensive view 14:12:32 dolphm: do you plan to do a feature proposal freeze? 14:12:35 Aug 21? 14:12:56 yes 14:13:44 ok 14:13:54 #info Keystone planning to enforce FPF 14:14:22 dolphm: alrighty then, anything to add to meeting agenda for today? 14:14:32 looks like the script worked: https://blueprints.launchpad.net/keystone/+spec/openid-connect 14:14:42 not from me 14:16:05 ok 14:16:21 awesomesauce 14:16:25 dolphm: thx! 14:16:29 griff: around? 14:16:39 ttx: hi ya 14:16:40 #topic Cinder 14:16:53 Sorry for not having noticed your new nick around before 14:17:01 nahh... my fault 14:17:05 I should've said something 14:17:11 how was j2 for you? 14:17:22 great, I left it up to Duncan and Russell :) 14:17:26 just kidding 14:17:39 Missed more BP's than I would've liked again 14:17:47 but got the ones I was most interested in landed 14:19:14 7 landed 14:19:19 13 in j3 plan 14:19:26 Lots under review though 14:19:29 Yeah 14:19:34 most of those are carried over 14:19:35 so if you get them merged fast that MAY work 14:19:43 I'm being pretty picky about new things at this point 14:19:56 yeah, my bigger concern is new things coming in 14:20:03 https://blueprints.launchpad.net/cinder/+spec/cinder-storwize-driver-qos is "unknown" status -- is that really unknown, or just not started? 14:20:20 it's really unknown from my view 14:20:36 they submitted the bp but I've had no comittment or updates 14:20:37 ok, great! 14:20:45 just doublechecking:) 14:20:49 :) 14:21:14 I'll get updates on "unknowns" at tomorrows Cinder meeting 14:21:31 griff: do you expect a lot more specs to creep in j3? 14:21:48 ttx: history says they'll try 14:21:59 ttx: but I made it clear that this time around I was going to be picky 14:22:06 griff: but nothing expected from your side ? 14:22:34 once you agree to consider one (and the spec is not approved yet) you can just add it to j3 as "blocked" 14:23:06 so that it apperas on the plan 14:23:10 cool 14:23:12 rather than flies below radar 14:23:23 ok, that's all I had for you 14:23:32 anything you'd like to discuss at meeting today? 14:24:00 mikal_: want to quickly steal 10min now? 14:24:29 nothing for me thanks :) 14:24:36 griff: ok then ttyl 14:31:47 david-lyle: o/ 14:33:14 ttx: o/ 14:33:28 #topic Horizon 14:33:35 david-lyle: how did j2 go? 14:34:08 got most of Sahara in, the rest went in right after j2 tag 14:34:22 yep, heard the good news 14:34:23 I think 12 bps total merged 14:34:34 https://launchpad.net/horizon/+milestone/juno-3 14:34:41 So you have 72 blueprints there 14:34:42 with several more trickling in now 14:34:59 ttx: going to have a fire sale later today, everything must go 14:35:12 including 27 that are already up for review 14:35:21 honestly haven't taken an attempt at cleaning that up yet 14:35:25 if you only merged those that would already be quite a success 14:35:35 agreed 14:35:51 ok, so maybe review it and try to come down to 30-40 14:36:10 anythign above that would just create noise 14:36:14 any that have a status other than started or greater will get cut first 14:36:23 ack 14:36:25 yes, too much review load 14:36:43 you can also adopt an early feature proposal freeze 14:36:54 so that you can easily reject stuff that's not under review yet 14:37:00 the common one is on Aug 21 14:37:06 that's a good idea 14:37:23 even with that I don't expect magic though :) 14:37:38 I think a couple of the need review are stale and may go 14:38:05 70 BPs in 5 weeks means 14/week 14:38:20 is that unreasonable?? ;) 14:38:40 well, we'll see if we have 14 more implemented next week :) 14:39:03 that's about half of those that are already "needs code review" 14:39:24 :) 14:39:48 david-lyle_: note that we'll have a gap analysis progress review at the TC today, you might consider joining us 14:39:53 more likely I expect 14 more random ones to show up 14:39:56 since we'll look at Horizon gap plan 14:40:07 if you can't come, don't hesitate to update the plan 14:40:18 to reflect recent (j2) progress 14:41:01 ttx: ok, not much progress, there are 4 items 14:41:24 I'll likely attend 14:41:30 david-lyle_: anything you wanted to add to the meeting agenda? 14:42:00 no, I may have to miss part of that meeting due to a conflict 14:42:13 ack -- so far nothing is on the agenda anyway 14:42:17 david-lyle_: ttyl 14:42:17 ok 14:42:22 mestery: around? 14:42:25 ttx: o/ 14:42:27 #topic Neutron 14:42:29 ttx: thanks 14:42:33 mestery: how did j2 go? 14:42:48 ttx: It went well, thanks to russellb's help. 14:42:58 cool! 14:43:00 ttx: We merged 11 BPs I think, not bad overall 14:43:20 pales in comparison to your 84 BPs in j3 though 14:43:27 It's a mountain for sure 14:43:48 so that's about 17 to merge every week 14:43:57 That's pretty high 14:44:16 Realistically, if we land 25-30 I'll be happy. 14:44:20 with only 23 being ready to review right now 14:44:37 right, the trick is, how to predict which ones won't make it and only contribute noise 14:44:58 I was going to start pulling things which are in Unknown soon 14:45:22 mestery: you started to make use of the "Blocked" status, does that mean those have specs almost approved? 14:45:42 ttx: Actually, that means that there is a -2 on the patches in most cases. 14:45:52 ok 14:46:12 you can use "slow progress" fo that case if you want to distinguish between the two 14:46:22 ttx: Good idea! I think I'll move to using that. 14:46:36 it's like, you want it in but it's not going in the right direction right now 14:46:52 ack 14:46:59 also lots of BPs have "unknown" status... is that just not started, or really "unknown" ? 14:47:13 Those are really no started 14:47:16 I'll move them to that state 14:47:22 And those will be ones to prune in the coming weeks 14:47:23 ok, cool 14:48:00 mestery: we'll have post-j2 gap coverage plan progress review at the TC today 14:48:13 you mighrt want to be present, or at least update your page 14:48:28 Perfect! 14:48:34 I think markmcclain and I will be ready for this 14:49:08 mestery: also, just to confirm, you're planning to enforce Feature Proposal Freeze on Aug 21? 14:49:18 ttx: Correct 14:49:19 #info Neutron will have FPF on Aug 21 14:49:44 OK, that's all I had -- anything you want to add to meeting agenda for later today? 14:49:55 Nothing from me at this point, thanks ttx! 14:49:58 mestery: cool. ttyl! 14:50:02 ttx: later! 15:29:48 ttx: ready when you are 15:33:20 notmyname: o/ 15:33:24 #topic Swift 15:33:31 notmyname: what news? 15:33:40 ttx: last week we release a new swiftlcient 15:33:48 python-swiftclient 2.2.0 15:33:55 it has support for storage policies 15:33:59 #info python-swiftclient 2.2.0 was released last week 15:34:00 and some other nice things 15:34:33 other than that, we're working on the EC stuff 15:35:24 ok, still planning to try a release around the end of the month? 15:35:43 ttx: as in 2 days? or end of august? 15:35:48 arh 15:35:55 * ttx is already in August for some reason 15:35:57 :-) 15:36:07 ok, end of august seems reasonable :-) 15:36:12 heh, ok 15:36:41 when is the swift gap analysis TC meeting? 15:36:48 planned next week 15:36:53 as discussed 15:37:03 ok 15:37:15 just needed to make sure it was on my calendar :-) 15:37:25 anything you want to discuss at cross-project meeting today? 15:38:00 nothing specifically. I'm interested-ly following the defcore and in-tree gate testing threads 15:38:38 yes, looks like we are converging towards something more swifty 15:38:54 (on the in-tree gate testing) 15:39:16