08:14:50 #startmeeting ptl_sync 08:14:51 Meeting started Tue Aug 26 08:14:50 2014 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:14:52 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 08:14:54 The meeting name has been set to 'ptl_sync' 08:14:57 #topic Nova 08:15:10 #link https://launchpad.net/nova/+milestone/juno-3 08:16:03 #info 8 implemented, 50 in review 08:16:46 It's not totally bad -- but yes, you need to seriously increase the conversion rate here 08:17:03 go from 5 to 25. 08:17:06 :) 08:17:43 johnthetubaguy: how close are the "High" ones ? 08:19:02 ttx: I am told very close 08:19:28 so ironic is risky, but we really want it 08:19:29 It would be... safer to merge them this week basically 08:19:40 (or at least most of them) 08:20:04 right, I think these are ones that get exceptions if they don't make it, but I need to dig a big further, I might demote the scheduler libriary 08:20:23 yeah, ironic will be to the wire I think, but lets see 08:20:58 At least now the list is finite 08:21:06 true :) 08:21:21 and easily prioritized 08:21:38 yeah, we have lots of mediums, I should probably check up on those, and try to demote the ones that are slipping 08:21:47 This is really review money time (now) 08:22:06 next week is the milestone week, I wouldn't expect that much work to get through 08:23:22 Any specific event (like a review day) planned? 08:23:31 not at this point 08:23:41 its a good idea though 08:24:08 It could help in getting a few that are at 99% landed 08:24:18 yeah, maybe try for Thursday or Friday as a review day 08:24:29 Or just a review-specific meeting 08:24:51 single out the ones that are almost there, and give them the final push 08:25:05 yeah, we can try do that informally on IRC 08:25:32 I guess 08:25:37 ok 08:25:56 We'll track further progress during the week 08:25:59 honestly, I feel like I need more insight into how close all these blueprints are 08:26:11 its a bit opaque right now 08:26:29 yeah, would be good to catch up later in the week 08:26:29 I agree that the status is too granular 08:27:08 my plan is, I need to do some more reviews, so I will dig in and just see how its all going 08:27:14 it works well only for single-review things and there aren't so many of those 08:27:23 yes, thats true 08:27:59 anything else you wanted to mention? 08:28:07 Any topic for the meeting today? 08:28:19 I don't think so, I guess I will try demote the priority of a few blueprints where they don't seem close 08:28:27 nothing really to filter up 08:28:36 well, I guess, the slots ideas are still cooking 08:28:52 we kinda quietly opened for K specs 08:28:58 sssh. 08:29:03 yeah, sorry 08:29:04 don't speak too loud 08:29:05 :) 08:29:14 thats exactly my plan... 08:29:33 we have no intention of reviewing them right now but anyways 08:29:38 johnthetubaguy: ok then. I'll be in touch, probably Thursday, to see how well we progressed 08:29:44 awesome, thanks 08:30:00 Thank you! 12:02:08 SergeyLukjanov: ready when you are 12:09:14 ttx, I'm here 12:09:39 ttx, sorry for being a bit later 12:10:09 #topic Sahara 12:10:19 #link https://launchpad.net/sahara/+milestone/juno-3 12:11:35 #info 7 implemented, 4 under review, 5 inprogress 12:11:50 slow progress compared to last week 12:12:15 but still on track 12:12:35 yup, there were some blockers 12:12:48 but I think it's going to be merged till j3 12:13:32 don't forget gate is likely to explode next week 12:13:43 so don't count on Tuesday-Thursday next week that much 12:13:55 I would get everything in this week if I were you ;) 12:13:55 ttx, yeah, our sahara-ci has been exploaded this weekend :) 12:14:25 How is https://blueprints.launchpad.net/sahara/+spec/edp-swift-trust-authentication doing? 12:14:44 is there a lot of code missing? 12:14:58 not a lot, but some code is still missing 12:15:18 I think that all sahara parts of this blueprint will be ready in j3 12:15:29 and will not use FFE for it 12:15:38 OK cool. 12:15:52 anything you wanted to talk about ? Any topic for the meeting later? 12:16:11 I think no topics 12:16:26 I'd like to note that sahara@horizon is going good 12:16:35 and mostly all of the changes has been merged 12:16:45 and we're not going to release dashboard's j3 12:16:56 awesome 12:17:16 we need to keep repo fot some more time because of stable/icehouse branch in it 12:17:29 SergeyLukjanov: right 12:17:42 SergeyLukjanov: ok then, talk to you later 12:17:46 dhellmann: around? 12:17:48 thank you 12:17:52 ttx: hi! 12:17:57 #topic Oslo 12:18:02 dhellmann: hola! 12:18:17 dhellmann: I think we have a way forward for the LP tracking now 12:18:28 the project group? 12:18:33 dhellmann: yes. I'll proceed and ask LP admins to move things 12:18:46 ok, let me know if there's anything you need me to do, too 12:18:48 oslo->oslo-incubator and new oslo projectgroup 12:19:12 the only thing that I could see malfunctioning is that Lp has a (handy) redirection feature for renamed projects 12:19:36 so if you go to lp/svanna you get the lp/sahara page 12:19:47 ah, and we wouldn't want that turned on 12:19:53 but here we want lp/oslo to land on the projectgroup :) 12:20:10 we could call the group openstack-oslo 12:20:13 right, so that's the only left unknown (together with the will of LP admins to accept to create a projectgroup) 12:20:30 I guess we could. 12:20:44 but oslo would be better I think, if we can have it 12:20:48 if they can avoid the bad redirect, oslo is fine, sure 12:21:34 hm. 12:21:48 ctrl-C is too close to ctrl-X apparently 12:21:53 heh 12:21:56 https://launchpad.net/oslo/+milestone/juno-3 12:22:16 I think we established in the past that most of the work here is not FF-sensitive 12:22:37 but still, is it time to defer a few things ? 12:22:41 yeah, at this point I don't expect active adoption of new libraries 12:23:06 dims started working on the logging stuff yesterday, so that may be done even though it was assigned to me :-) 12:23:24 hmm, rootwrap daemon mode should be considered implemented, fixing 12:23:24 we're behind on some reviews, so we'll be shifting gears to focus on finishing things this week 12:23:37 has that been released? 12:23:47 merged, not released 12:23:57 waiting for a few more tests, but we'll release this week 12:24:11 one question is shoudl we do 1.3 or 2.0 12:24:15 ok, I have a short list of other releases for this week but I'll let you handle that when you're ready 12:24:24 is the old API still present? 12:24:29 yes 12:24:43 as long as it's still compatible with existing users, you can stay in the 1.x series 12:24:51 well, its' not strictly an API, but the old way of using it is preserved 12:25:08 ok, makes sense 12:25:10 well, yeah, that's what I meant 12:25:13 1.3 it is 12:25:24 1.3 ships a client API though for the daemon mode 12:25:38 that's fine, that's a new feature not a breaking change 12:25:41 so we'll have to preserve that starting from now 12:26:08 do you foresee removing the old command line API? 12:26:23 Not really. It's still "more secure" 12:26:38 as in simpler. 12:26:38 ok 12:27:36 I do have something to bring up with you, to be thinking about for kilo. 12:27:43 so do you still plan to cover all the j3 targets ? 12:27:56 with some being deferred to rc1 probably ? 12:28:21 I'm going to review that list today and talk to josh and a couple of other people about their status. I'll have it up to date by tomorrow morning. 12:28:25 is there an rc1 target yet? 12:28:31 * ttx tries to wrap his haed around this list but without the incub/lib distinction it's kinda hard 12:28:46 yeah, I wish I could tag bps and filter this list 12:29:06 in the meantime, i just trust you have things under control :) 12:29:06 soon enough 12:29:16 * dhellmann tries not to laugh out loud 12:29:44 dhellmann: so.. kilo 12:29:51 seriously, I think most of this is going to land at some point in juno, but since we're not rushing to make the ff deadline some is likely to be rc1 or later 12:29:59 right 12:30:23 ok, so, we have recently had some trouble with graduating modules that are dependencies of modules that have not graduated 12:30:42 for instance, oslo.utils took strutils out of the incubator, but lots of things use strutils and are not graduating 12:31:22 our original plan was to leave the incubated code alone, but in some cases that makes it harder to adopt because tests in the projects are mocking out the incubated version of a function and some code is calling the non-incubated version 12:31:37 I think that was actually timeutils, but it doesn't really matter 12:31:46 why is it a problem for non-gaduated libs to use graduated libs ? 12:32:16 well, we didn't want to update the incubated code to use the graduated libs because that meant if anyone needed a patch to the non-graduated code they would *have* to adopt the library at the same time 12:32:32 but that's turning into too much trouble, so we want to change that 12:32:51 for kilo, we want to branch early, and then delete things as soon as their graduated library is available for use 12:33:10 the early branch still gives us a way to backport changes for projects that haven't adopted a library 12:33:22 although it occurs to me that the branch might just be stable/juno 12:33:48 stable/juno shall work for you if you branch early in Kilo, yes 12:34:09 when would that stable branch normally be cut? 12:34:28 your "if you branch early" comment makes me think not immediately? 12:34:50 we usually did it sometimes between RC1 and release 12:35:02 ok, that seems like good timing for us 12:35:18 i.e. we would tag the "final" juno oslo-incubator when you want it opened for kilo 12:35:49 and once the final tag is in, we can branch 12:35:52 right, and I think earlier is probably better 12:36:32 I'll need to make sure we have the necessary changes landed first, but I don't think we have much targeted so it shouldn't be an issue 12:36:37 although we can do even earlier if needed, nothing "consumes" that anyway 12:36:48 testing-wise 12:36:59 and I will go through the juno planning stuff we put together and make sure the change to deleting modules isn't going to introduce any other weirdness 12:37:13 ack 12:37:19 ok, we might do it when our last planned graduation happens this cycle, then 12:37:31 that way if we do have juno bug fixes, we have a place to put them 12:37:41 Sounds good. anything else you wanted to mention ? Topics for the release meeting ? 12:37:55 no, that was it 12:38:08 dhellmann: thx! 12:38:11 gordc: o/ 12:38:19 ttx: thanks, ttyl! 12:38:28 o/ 12:38:31 #topic Ceilometer 12:38:41 #link https://launchpad.net/ceilometer/+milestone/juno-3 12:39:20 so we have code for all our bps up and ready for review 12:39:22 #info 3 implemented, 6 in review, 1 in progress, 1 blocked 12:39:42 i think we're making good progress on high priority items... 12:39:51 we've gone through several rounds of reviews on those items 12:40:00 Is https://blueprints.launchpad.net/ceilometer/+spec/paas-event-format-for-ceilometer code already in review ? 12:40:15 it's only marked "good progress" 12:40:35 that bp has a patch up as well but i don't believe it's passed jenkins yet so i've left it as good progress for now 12:40:44 ok 12:40:55 it's been reviewed a few times i believe so i could mark it as review if that's easier to track 12:41:36 https://blueprints.launchpad.net/ceilometer/+spec/grenade-resource-survivability still blocked on https://review.openstack.org/#/c/102354 ? 12:42:08 yes, that's the one issue we have... i switched the item to blocked recently since we made no progress 12:42:10 gordc: if it doesn't need a complete rewrite of the patch, making it "needs code reveiw" will not raise any alarm 12:42:37 ok i'll confirm with owner and switch to needs review 12:42:38 gordc: could you summarize the issue on grenade-resource-survivability, see if I can help unblock? 12:42:59 sure. i'll try my best to summarise 12:43:47 from my understanding there is a debate on what we should be testing in grenade test between joe gordon and chris dent. 12:44:18 to my knowledge chris is pretty much unsure what to do... just that what he put up for review isn't it. 12:44:47 we reached out to joe gordon on thursday at our meeting for clarification and he said he'd contact chris later. 12:45:11 might be worth raising that issue at the meeting today 12:45:17 i just checked with chris today and he hasn't heard from joe yet so i was probably going to reach out to him today to see if we can get some action 12:45:19 we could collect more opinions 12:45:24 yeah. that sounds good. 12:45:40 i think chris is open to anything. just that we're not sure what 'anything' is at this point 12:46:05 #info grenade-resource-survivability blocked pending discussion between jog0 and Chris Dent 12:46:54 aside from grenade bp i think we're ok. hoping to get more reviews once eglynn and llu return in next few days 12:47:30 I would definitely advise to merge as much as you can this week 12:47:35 next week should be a bit busy 12:48:09 yeah. i was hoping to get a bit more merged this past week but we were blocked by a gating issue when tox was upgraded in the gates. 12:48:22 we have a workaround now so we should be able to make some progress 12:48:28 ok, cool 12:48:37 gordc: that's all I had 12:48:45 gordc: anything else for the meeting tonight ? 12:49:03 i think thats all on my end 12:49:08 i'll just track the grenade item for onw 12:49:10 now* 12:49:21 gordc: great, and thanks a lot for stepping up while eglynn was away 12:49:48 np. he did a good job doing a brain dump before he left. 14:01:19 jgriffith: around? 14:07:50 dolphm: you can go now if you're around 14:16:19 ttx: o/ 14:16:39 would have helped if i had irc open 14:18:03 indeed 14:18:06 #topic Keystone 14:18:21 #link https://launchpad.net/keystone/+milestone/juno-3 14:18:46 #info 3 implemented, 5 in review, 2 in progress, 1 blocked 14:19:14 About the blocked one, I raised a thread and will mention it at the cross-project meeting again 14:19:18 bp non-persistent-tokens (currently good progress) has been a series of roadblocks and hurdles throughout juno - we made a ton of progress, but it doesn't look like it's going to make juno-3 14:19:24 ttx: thanks 14:19:38 I'd like to understand if it('s just "annoying" or "blocking" 14:19:48 (the dep thing) 14:20:58 our approach for federating across multiple keystones is to use SAMLv2 (which allows us to federate with non-keystones as well). our approach for generating SAMLv2 docs is with pysaml2 14:21:29 so it's either re-invent an XML + crypto library or use one off the shelf 14:21:50 steve documented a couple alternatives in the requirements review, but it doesn't sound like any others are a better option 14:22:11 and switching libs would probably mean missing juno 14:22:14 yeah, I was wondering if the saml one (which was the less worse alternative) was an option 14:22:40 could be good to explore that scenario, just in case we hit a wall 14:22:53 but my understanding was that it was a pain generator, not a blocker 14:23:03 that one describes itself as an alpha; but at least has a release this year 14:23:07 we could even have a plan to migrate off to saml to remove the pain generator 14:23:29 that's what I want us to clear up tonight 14:24:07 in the mean time, you need to continue landing stuff, nice progress over the last week 14:24:57 44 patches to go at this point :( 14:25:23 I'd recommend you get as much merged this week as you can :) 14:25:33 of course! 14:25:51 We'll review in-week progress in ~2 days 14:26:18 alrighty 14:26:20 anything else you wanted to mention ? 14:26:35 Anything for the meetign (in addition to the saml discussion) ? 14:27:04 no sir 14:27:24 alright, thx! 14:28:16 jgriffith: still not around? 14:29:56 david-lyle: we can start now if you're available 14:30:13 ttx: ready 14:30:15 #topic Horizon 14:30:26 #link https://launchpad.net/horizon/+milestone/juno-3 14:30:38 #info 9 implemented, 38 under code review 14:30:56 Still leaves a lot to process :) 14:31:11 yeah, there are a lot in flight 14:31:34 quite a few are in good shape just need another review 14:31:55 a few are still a little less reviewed and may not make the cut 14:32:03 david-lyle: ok, you should really push early this week, rather than next week 14:32:12 definitely 14:32:30 I like that all the High targets are in 14:32:49 but we want to do our part to add to the semi-annual gate jam 14:32:57 :) 14:33:10 You might want to call a specific meeting to get those last reviews the final +2 14:33:41 Sounds like a good idea 14:33:59 I don't think I can suggest much more from where i stand.. review review review... we'll lok at progress in two days 14:34:17 see how fast the wall is approaching 14:34:25 always too quickly 14:34:50 anything you wanted to discuss at the meeting ? You don't seem to have anything blocked 14:35:10 no, I think reviews are our only hold-up now 14:36:00 david-lyle: then I'll let you concentrate on those :) 14:36:02 david-lyle: thx! 14:36:10 thanks 14:37:06 mestery: ready when you are 14:37:11 \o/ 14:37:16 #topic Neutron 14:37:28 #link https://launchpad.net/neutron/+milestone/juno-3 14:37:46 #info 10 implemented, 54 in review 14:38:00 nice cleanup 14:38:08 Thanks, more to come, but I'm working on it everyday at this point. 14:38:24 but 1/day velocity won't get you very far when the door closes 14:38:29 Agreed. 14:38:38 I've already started booting some out which I don't think have a chance at landing. 14:39:01 anything planned to give the medium/high a final push? 14:39:08 some review-centric meeting? 14:39:21 your meetings have had other issues to discuss recently 14:39:29 I've been thinking of that, perhaps Thursday this week. 14:39:31 so you can't really prioritize reviews there 14:39:35 Right 14:39:47 Speaking of meetings: I plan to completely revamp the Neutron meeting after Feature Freeze 14:39:49 It's become stale 14:39:50 yes, thursday this week is about as much you can wait 14:39:53 Needs a breath of fresh air 14:39:58 ttx: ++ 14:40:15 after that, it will get difficult, unless you work on Labor day 14:40:34 Totally agree, I'll see if I can set it up for Wednesday actually. 14:40:38 That gives us one more day 14:40:46 ++ 14:41:14 I don't think there is anything I can help unblock, apart from reviews you seem in good shape 14:41:15 The good news is we're making progress on many High/Medium community BPs 14:41:21 Agreed 14:41:33 yes, I would focus the review meeting on medium/high tbh 14:41:33 L3 HA, ipset, and security groups RPC refactor are all in good shape. 14:41:39 Absolutely 14:41:45 Low are "best effort" 14:41:52 if time left, you can address low-hanging fruit in Low 14:41:57 yes 14:41:59 but there won't be (time left) 14:42:03 :) 14:43:11 Anything you'd like to discuss at meeting today? 14:43:30 Is the "PTL vs. czar" thing on the agenda? 14:43:34 Or is that a ML thread only thing. 14:43:40 Seems project meeting may be a good place to continue that discussion 14:44:49 we can discuss that, sure 14:45:01 Could you add it to the agenda ? 14:45:01 Well, only if others want as well, maybe the ML thread is dieing down now :) 14:45:10 Will do 14:45:13 thx! 14:45:18 np! 14:45:51 ok then, talk to you later! 14:45:57 jgriffith: still not around? 14:46:18 ttx: later and thanks! 15:22:36 ttx: hey... still up? 15:22:47 jgriffith: sure, just a sec 15:23:42 jgriffith: hi! 15:23:46 #topic Cinder 15:23:48 ttx: howdy 15:24:04 #link https://launchpad.net/cinder/+milestone/juno-3 15:24:16 #info 3 implemented, 16 in review 15:24:21 :( 15:24:35 that's still a lot of review work to complete in 8 days :) 15:24:42 agreed 15:24:50 some are pretty far along 15:24:50 how are the two medium ones going? 15:25:01 those are close 15:25:14 I expect them to get cleaned up and finished in the next day or so 15:25:14 would be great to land them this week to remove a bit of pressure 15:25:18 cool 15:25:23 yeah, totally agree 15:25:43 I've already let it be known drivers are my lowest priority 15:25:56 that must make you a very popular person 15:25:57 There's a couple in there I need to manually update 15:26:03 ttx: Yeah, I'm loved! 15:26:05 :) 15:26:47 do you think you could benefit from a review meeting where you would give a few targeted blueprints a final +2 push? 15:26:50 ttx: I'm not sure about the bugs, but at least I know I can work on those after 3 closes 15:27:02 j3 that is 15:27:16 re: bugs; that's what rc1 is for, unless you have critical ones or regressions 15:27:26 yeah 15:27:29 that's my plan 15:27:40 but I'm certain there will be *new* ones once we cut 15:27:49 based on history at any rate 15:28:07 everyones's pretty distracted right now, between last minute features/reviews and third prty ci 15:28:29 I'll let you go back to it. This week is money time 15:28:37 I'll have an updated turn on the bp list tomorrow 15:28:48 well... one today, and another after the meeting tomorrow 15:28:53 anything you'd like to discuss at meeting today? 15:29:07 i'll look at progress in a couple of days, see how badly we are f*ed 15:29:11 we could continue the summit talks :) 15:29:20 ttx: think positive! 15:29:31 see how amazing the recovery is in 24 hours :) 15:29:38 we have merged 15% of the j-3 targets in 80% of the time 15:29:50 damn you and your statistics 15:29:53 how positive is hat 15:29:54 :) 15:29:55 +t 15:30:17 alright... I'll catch up with you this afternoon 15:30:18 thanks 15:30:23 jgriffith: thx! 15:30:38 notmyname: ready when you are 15:30:47 unless you want to opsmeetup-skip 15:30:54 ttx: I'm here. I'm in the ops meetup right now 15:31:05 nah, I can space a few minutes 15:31:06 notmyname: we can make it quick and painless 15:31:17 thanks for doing the rc tag 15:31:17 like a stab in a kidney 15:31:21 heh 15:31:27 #topic Swift 15:31:36 I'm reminded that I need to send an email to the ML about the RC 15:31:40 #link https://launchpad.net/swift/+milestone/2.1.0 15:31:47 I think I saw one 15:31:55 oh yeah. maybe I didn that 15:32:00 *did 15:32:15 yup. just checked 15:32:22 [openstack-dev] [Swift] 2.1.0-rc tagged 15:32:30 I'll be getting LP filled out this week 15:32:34 looking good so far? 15:33:03 yes, AFAIK. there wasn't much in it that's concerning. work continues on erasure code work 15:33:31 do you think the erasure code stuff would land in the juno final? 15:33:53 or you'd rather introduce it early in kilo when ready 15:34:38 I think it will be very hard to get it in juno. early kilo is more likely. (although there are many people who want it in juno) 15:34:55 #link https://trello.com/b/LlvIFIQs/swift-erasure-codes 15:35:02 that's the living "what's going on" tool 15:35:07 I'll be around on monday for your sign-off on the 2.1.0 tag. If you need backports, contact me asap so that we set up the branch 15:35:20 ttx: thanks. will do 15:35:44 notmyname: anything you'd like to discuss at meeting today? Or will you be enjoying post-meetup activities? 15:36:19 I'll be in the middle of the ops meetup (facilitating a storage discussion) this afternoon and won't be able to make it) 15:36:53 ok, fair enough! 15:36:57 talk to you later then 15:37:08 thanks 15:37:30 zaneb: ready when you are 15:37:36