08:03:06 #startmeeting ptl_sync 08:03:07 Meeting started Tue Aug 12 08:03:06 2014 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 08:03:08 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 08:03:10 The meeting name has been set to 'ptl_sync' 08:04:07 Hi 08:04:10 #topic Nova 08:04:25 #link https://launchpad.net/nova/+milestone/juno-3 08:04:53 Status looks coherent, there are 5 "unknown" states that could use clarification 08:05:03 Yep 08:05:11 And the usual "we could do more code review" comment applies 08:05:15 As far as progress goes, last week we had 1 implemented, 29 in review 08:05:20 This week: 08:05:24 We're trying to identify things to review in the meetings, but its early days 08:05:38 #info 2 Implemented, 33 under review, 27 in progress and 11 Not started/Unknown 08:06:27 So I'd say you need to convert to "Implemented" faster, and yes, singling out "blueprints of the week" could allow faster iterations and more conversion 08:06:38 Agreed 08:06:46 So yeah, we're trying it but we could be more active 08:06:53 I haven't had a lot of code review time in the last week 08:07:07 As I find things which are important and under review I can ping cores and ask them to take a look 08:07:42 We are 9 days away from FPF, so I expect a lot of those 38 in progress/notstarted to not make it 08:07:59 Agreed 08:08:02 We can bump them then though 08:08:07 Maybe starting to aggressively defer stuff where assignees are fine to drop it to Kilo 08:08:25 Ahhh, as in ping people and ask them if they're going to make it or not? 08:08:39 Like Dan Berrange has 5, and only one in review 08:08:39 Do you plan on sending a general FPF reminder to the mailing list? 08:08:44 Or does each project do that? 08:09:06 It's more project-specific, since some projects do not follow the FPF 08:09:13 Ok, fair enough 08:09:15 I'll send reminders for the feature freeze itself 08:09:22 which is cross-project 08:09:35 but FPF I think is only followed by 3 projects 08:09:54 Ok, I will send an email 08:09:56 jaypipes has 3 blueprints to, none under review yet 08:09:59 too* 08:10:14 In those cases they could already pick one to drop 08:10:22 Agreed 08:11:16 so my advice would be to review the top, which ones are 99% there and just need a bit more attention to pass 08:11:31 and the bottom, which ones are very unlikely to make it and should be proactively dropped 08:11:35 As in those in "needs code review" 08:11:37 that way you reduce the slate 08:12:13 otherwise the sheer number of targets make it very difficult to make any kind of significant progress 08:12:46 Quick look at the High/Essential prio stuff 08:12:56 There are 6 High 08:13:07 4 already under code review 08:13:22 The top two are ongoing efforts 08:13:30 They're unlikely to be complete in Juno 08:13:46 Which you could argue means they're mistargetted 08:13:48 ok, it's like never actually "complete", just merge as much as you can ? 08:13:53 But we also really want to land the code 08:13:57 Yep 08:14:01 ok, that's fine 08:14:02 Well, they'll be compelte one day 08:14:06 Just not instantly 08:14:15 So high actually looks ok to me on that basis 08:14:25 I would prioritize up the stuff that affects other projects 08:14:36 i.e. ironic and cinder clientv2 08:14:38 Like the ironic thing? 08:14:55 Yep, ironic is already getting pretty good attention -- its currenlty waiting for review comments to be addressed 08:15:01 ok 08:15:09 Cinder I'd have to check 08:15:12 I think that's all I had 08:15:21 not sure how dependent Cinder is on that feature though 08:15:30 i suspect you could keep using v1 08:15:42 Yeah, I think they want to remove v1 IIRC 08:15:47 Anything you'd like to discuss at cross-project meeting today? 08:15:53 Just looking at it, the code might have been abandoned? 08:16:12 So, we're starting to get requests for kilo specs to open 08:16:18 I'd be interested in how other projects are handling that 08:16:30 I personally feel its a distraction from Juno, but I think I am going to lose that argument 08:16:39 hmm, ok. mestery had a thread about that on the ML a few weeks ago 08:16:49 The counter argument being that the people wanting to write those specs aren't interested in fixing Juno bugs anyways 08:17:00 Hmmm, I shall go looking for that. I missed it. 08:17:00 People should know they can ignore nova-specs in review for now 08:17:10 Oh, I think the reviewers do 08:17:12 so posting changes there doesn't hurt THAT much 08:17:19 Its more we haven't given any way for someone to propose a kilo spec 08:17:28 but yeah, just the fact that people ask is a bit worrying 08:17:42 Its people who have a feature to add and then want to run away 08:17:54 You know the type 08:18:40 I can add a topic for the meeting today about "When to open Kilo specs" 08:18:54 That would be good 08:18:54 Thanks 08:19:06 ok done 08:19:17 mikal: all set? 08:19:24 Yep, done 11:44:41 ttx: knock, knock, ready when you are ... 11:45:15 eglynn: o/ 11:45:19 #topic Ceilometer 11:45:56 #link https://launchpad.net/ceilometer/+milestone/juno-3 11:46:16 so I think that's our full slate of BPs for juno-3 11:46:30 #info 2 implemented, 4 under review, 6 in progress 11:46:43 good progress on most things with priority > Low 11:47:01 with the exception of the gnocchi dispatcher, but that slightly decoupled in any case 11:47:12 i.e. forward-looking to kilo 11:47:47 so I'm not overly-concerned but will keep a close eye on it 11:47:52 still a long way to go 11:48:34 yeap, so I've an additional resource (cdent) to help out with two of the higher priority BPs 11:48:41 ok, cool 11:48:42 (in addition to the main authors) 11:48:49 so that should expedite progress 11:49:13 You don't plan to enfore a feature proposal freeze anyway, so for all of those you have 3 weeks left 11:49:34 (that's cdetn helping out on central-agent-partitioning & bigger-data-sql specifically) 11:50:04 yes, I'm planning to push the feature proposal freeze idea as hard as possible 11:50:15 OK, if there is little progress by next week we may want to cut down a number of things that will obviously not make it 11:50:28 but let's give them all a chance for the time being 11:50:43 Anything you want to discuss at meeting today? 11:50:51 yeap, I'll pass that mess onto to gordc, and will do another pass on Friday before I finish up 11:50:59 yeap couple of things 11:51:06 major snafu in Fedora-land has killed the ceilo py26 gate for the last day 11:51:17 mongodb packages were retired from EPEL6 by mistake 11:51:21 (couchdb was the intended victim) 11:51:22 heh 11:51:32 see https://www.redhat.com/archives/pulp-list/2014-August/msg00002.html & https://www.redhat.com/archives/pulp-list/2014-August/msg00016.html 11:51:36 ?o??db 11:51:48 yeap :) 11:51:56 the bad commit has been reverted and the version re-tagged, pbrady is actively chasing +1s on restoring 11:52:07 but may take a while for the restore to propogate thru' 11:52:16 in the meantime, ceilo gating is blocked :( 11:52:22 ok 11:52:22 also possibly marconi & trove? 11:52:42 #info Ceilo py26 fgate blocked on Fedora snfu removing mongodb packages 11:52:46 as a stop-gap, we may need to install the RPM direct from the Fedora package build system on the centos-6 jenkins slaves 11:52:55 * eglynn will talk to the infra folks about this 11:53:15 Did you discuss the best time for 1:1 syncs with Gordon starting next week? 11:53:46 I mailed him with those alternate times as he was offline, no response yet 11:53:53 ok 11:53:54 I'll chase on IRC this afternoon 11:54:12 #action ttx to see best time for Gordon for upcoming 1:1 sync with Ceilometer 11:54:22 not that we're going to solve this now, but ... 11:54:25 otherwise will discuss it with him directly 11:54:43 the other thing I'm concerned about is the demotivational/distraction effect of that future of integration release discussion 11:54:53 from a timing perspective re. the push for juno-3 11:55:07 but as I said, we're not going to solve that now 11:55:20 trick is we need to review several incubation/graduation requests during August 11:55:43 all I can say is that this discussion doesn't affect Juno 11:55:52 it's all about Kilo 11:56:02 BUT some kilo-affecting decisions need to be taken asap 11:56:13 especialy which projects to graduate or not 11:56:14 yeah, I was assuming that re the Kilo orientation 11:56:39 eglynn: btw we should have time to discuss gnocchi today 11:57:07 yeah, cool, there's been some activity on that ML thread 11:57:16 * ttx is a bit late with email 11:57:21 ETOOMANYFIRES 11:57:39 I'll be around during the TC meeting when you need to call on me 11:57:51 anything specific you want to discuss at the cross-project meeting? 11:58:22 it might be worth clarifying the documentation contribution workflow 11:58:41 specifically the heavyweight XML docbook markup that we've been struggling with 11:59:04 versus a new lightweight RST-based workflow that the doc team are piloting with heat 11:59:14 eglynn: sounds good. Could you edit https://wiki.openstack.org/wiki/Meetings/ProjectMeeting#Agenda_for_next_meeting ? 11:59:19 will do 11:59:22 thx! 11:59:35 talk to you later, then 11:59:41 SergeyLukjanov: around? 11:59:46 OK 12:11:13 ttx, I'm here 12:11:44 ttx, sorry, just returned from the lunch 12:13:20 #topic Sahara 12:13:24 SergeyLukjanov: o/ 12:13:43 #link https://launchpad.net/sahara/+milestone/juno-3 12:14:17 ttx, hey 12:14:21 #info 2 implemented, 4 udner review, 8 in progress, 1 not started 12:14:44 That's 4 more blueprints since last time we looked at it 12:15:01 this is also progressing slowly 12:15:17 Do you plan to use a feature proposal freeze next week? 12:15:28 i.e. drop everything that is not under code review yet? 12:15:40 ttx, I think so 12:17:10 #info Sahara might use FPF 12:17:12 ttx, I'll talk with folks on the meeting this week about slow progress blueprints and probably move some of them to the next 12:17:35 Looking at the high-prio items... 12:17:47 https://blueprints.launchpad.net/sahara/+spec/edp-swift-trust-authentication is marked "slow progress" 12:18:12 although 3 merged out of 4... is it blocked on something ? 12:18:24 ttx, it's extremely complicated 12:18:36 ttx, and it looks like we'll move it to next 12:19:01 ok, do it early rather than kate, will free up reviewers time 12:19:08 ttx, it's about changing auth and credentials distribution strategies for data sources 12:20:12 ttx, all current CRs are backward compat and adds support for moving to the new strategy, and useful by itself 12:20:22 I'm just not sure that it'll be fully completed in J 12:21:33 I prefer that you get the others merged rater than try this one and fail the others as a result 12:21:51 so maybe move it to next at the next meeting 12:22:10 OK, anything you wanted to discuss at cross-project meeting today? 12:22:22 ttx, I think everything is going ok 12:22:31 dashboard stuff going ok too 12:22:35 SergeyLukjanov: ok cool, talk to you later then 12:22:39 dhellmann: around? 12:22:44 hi, ttx 12:22:47 #topic Oslo 12:22:49 ttx, thx 12:23:08 #topic https://launchpad.net/oslo/+milestone/juno-3 12:23:18 #topic https://launchpad.net/oslo.messaging/+milestone/juno-3 12:23:25 #topic https://launchpad.net/oslo.vmware/+milestone/juno-3 12:23:26 ... 12:23:43 dhellmann: I've been giving some thoughts on that mess ^ 12:24:30 I think we should be using an oslo project group in LP, and use library versions as milestones 12:24:32 hmm, there aren't milestones set up. I thought the vmware guys were going to use the oslo project so we wouldn't keep making this worse :-/ 12:25:14 ttx: that seems like it would help 12:25:16 dhellmann: i'll post something to that effect 12:25:27 we did talk about that last time, iirc, but I didn't follow up on making it happen 12:25:37 Still need to check how blueprints would show up 12:25:48 for bugs, it's pretty nice 12:25:59 yeah, I'm worried if we switch to versions for milestones it will be confusing when we talk about the cycle schedule 12:26:38 we'd keep the same "series" 12:27:34 but yeah, still on my todo list to explore all the consequences of the proposal :) 12:27:43 in the meantime... 12:27:47 ok 12:28:28 Most of the oslo-incubator BPs are about graduation, which is not really a FF thing 12:28:36 right 12:29:05 and the remaining ones are more "good if they go in, not a big deal if they don't" 12:29:31 Although I guess I would like us to wrap up the daemon mode thing 12:29:46 but no big deal if we warp it up first thing in Juno 12:29:49 in Kilo 12:30:13 So just drop stuff you know won't make it, to avoid settign wrong expectations 12:30:49 dhellmann: anythign you want to discuss at meeting today? 12:31:05 yeah, I think we're OK now that everyone is back from vacation and we can start doing more reviews 12:31:10 nothing from me this week 12:31:24 OK then, thanks for your time 12:31:29 ttyl 14:01:06 jgriffith: ready when you are 14:02:40 ttx: https://bugs.launchpad.net/oslo/+bug/1321080/comments/34 14:04:23 dolphm: commented 14:13:04 dolphm: I guess we can talk now, if you're around 14:15:08 ttx: o/ 14:15:32 #topic Keystone 14:16:13 #link https://launchpad.net/keystone/+milestone/juno-3 14:16:52 #info 0 Implemented, 4 Under review, 2 in progress, 6 Blocked/Not started 14:17:04 That looks a bit late to me 14:17:31 agree 14:17:58 on the upside, openid connect is supposedly complete somewhere outside of gerrit 14:18:15 i don't think we'll have anything to deprecate this cycle, so that bp will just go away 14:18:27 and audit support for federation is a small change 14:18:32 that's our three "Not Started" 14:18:46 The blocked ones are those waiting for spec approval? 14:19:00 two of the Blocked ones are mostly blocked on "Do we really need this?" but the work is easy if we agree 14:19:02 yes 14:19:10 hmm, ok 14:19:15 But still, time is running out 14:19:21 so you have to make up your mind now 14:19:50 because you need to start finalizing some of those 14:19:53 the scary ones in terms of amount of work and complexity of reviews are keystone-to-keystone federation and non-persistent pki tokens 14:21:04 right, I'm a bit concerned with keystone-to-keystone federation 14:21:09 what's the current state ? 14:21:21 I think that would be really good if that landed in juno 14:21:38 so maybe moving non-persistent pki tokens to next makes sense 14:21:54 there were a lot of radically different options to approach it, and we didn't decide on a solution until our hackathon in early july. it's made good progress since then, but we didn't exactly choose the lightest weight solution. it's going to be close if it makes it 14:23:03 do you think moving stuff out of the way would help ? 14:23:15 or is it completely separate 14:23:38 openid connect looks like the only one that might directly interfere 14:24:02 i'll follow up with steve on that one 14:24:35 dolphm: do you plan to enforce FPF next week? 14:24:44 yes, aug 21? 14:25:52 yep 14:26:06 OK, anything for the meeting agenda today? 14:26:17 nope 14:26:23 #info Keystone plans to enforce FPF 14:26:30 ok then ttyl! 14:27:12 o/ 14:33:21 david-lyle: o/ 14:33:32 ttx: o/ 14:33:33 #topic Horizon 14:34:21 down to 63 blueprints 14:34:23 https://launchpad.net/horizon/+milestone/juno-3 14:34:34 heh 14:34:57 #info 7 implemented, 28 under review, 24 in progress, 4 unknown 14:35:13 yeah still way too much load, working on pruning the list, and trying to keep up with cleaning out those added everyday 14:36:19 yep, 3 weeks left 14:36:29 several are very mature and should merge soon 14:37:03 ok, hopefully we'll see a clear progress next week 14:37:11 also we have FPF on August 21 14:37:27 so you can leverage that to remove stuff that is just not there yet 14:37:29 I set Horizon's at Aug 12 14:37:37 or 14 14:37:43 that would be Thursday 14:37:47 cool 14:38:01 Looking at High prio stuff... 14:38:04 yeah, already too much, can't handle more 14:38:21 https://blueprints.launchpad.net/horizon/+spec/remove-javascript-bundling is under review -- in good shape? 14:39:12 it's close, had a couple of bug fixes that set it back a little, should be ready soon 14:39:12 i see two more changes needing review/merge 14:39:17 ok 14:39:19 https://blueprints.launchpad.net/horizon/+spec/launch-instance-ux-enhancement is in progress -- how far is it? 14:39:39 That one still very much WIP 14:39:51 this is stalled and I may have to move out of Juno 14:40:24 unfortunately 14:40:39 ok, drop it early if you do 14:40:45 so that we don't communicate wrong expectations 14:41:08 sure, I'll bring it up in the Horizon team meeting 14:41:22 make sure there's not work I'm missing 14:41:26 ok 14:41:33 anything you want to discuss at meeting today? 14:41:51 no, I don't think so 14:42:02 alrighty then, talk to you later 14:42:08 mestery: ready when you are 14:42:09 ttx: thanks 14:42:11 ttx: o/ 14:42:14 #topic Neutron 14:42:35 been following your mailing-list adventures 14:42:41 #link https://launchpad.net/neutron/+milestone/juno-3 14:42:45 ttx: It's been fun :) 14:43:16 #info 3 implemented, 39 under review, 11 in progress, 32 unknown/blocked 14:43:34 so those unknown are a bit of a concern I guess 14:43:39 Agreed 14:43:44 It's likely I'll cull most of those next week. 14:43:51 For the most part, they are all low priority, best effort. 14:43:53 also feels like you could do with 32 less blueprints in your plate 14:43:58 Agreed 14:44:22 mestery: if all core reviewers really ignored those and prioritized their reviews properly, I would say we don't mind them 14:44:41 but just having such a busy slate is discouraging 14:44:51 so the numbers just hurt 14:45:08 Yes, in the future, we are going to be better about saying no earlier on. 14:45:17 It's something we as a team acknowledge we need to do better at 14:45:20 Looking at the essential/high stuff 14:45:32 https://blueprints.launchpad.net/neutron/+spec/neutron-dvr-fwaas has unknown status 14:45:44 does that mean you lst contact with the people working on it ? 14:45:46 Yes, that one concerns me, they owner says "it's coming", but I have my doubts. 14:46:03 I plan to follow up again on that one today to see if it has a chance. 14:46:06 or that you have no idea so it's as bad as not started as far as you're concerned 14:46:14 I think the latter. 14:46:18 ok 14:46:27 There was a lot of activity on it a few weeks back on an email htread 14:46:30 and then it's died on the vine :( 14:46:57 How is https://blueprints.launchpad.net/neutron/+spec/neutron-ovs-dvr going ? 14:47:18 That one is "done", they have left it open while they fix a few bugs which have fallen out. 14:47:25 But the functionality has merged a few weeks back 14:47:30 So, that's a HUGE win for us 14:47:45 frankly if we ignore the Low/Unknowns it doesn't look that bad 14:47:52 Agreed 14:48:09 Like almost all the >=Medium are proposed 14:48:10 L3 HA is also very important, and it's coming along nicely at this point as well. 14:48:22 https://blueprints.launchpad.net/neutron/+spec/l3-high-availability 14:48:42 so maybe you can just bluntly leverage FPF to remove all the extra stuff 14:48:51 That is my plan :) 14:49:04 * mestery will be in Chicago next week for Linuxcon and plans to spend Friday morning cleaning this up. 14:49:06 That's post FPF 14:49:12 Which is next Thursday 14:49:13 But then you should also finalize as many as you can while it's not too difficult to do so 14:49:21 Yes 14:49:52 Theer are two things that concern me on the recent threads 14:50:17 The first one is the disagreement on the nova-net->neutron migration. It looks like a dev vs. ops thing 14:50:32 where devs say, better to have cold migration than nothing 14:50:47 and (some) ops say we don't like tat as a solution 14:50:50 Yes, that one is concerning, as we thoughyt at the nova mid-cycle we had agreement 14:51:00 russellb summarized it well with the pony 14:51:03 Consider migration a "WIP" I guess right now, which I know isn't good 14:51:07 Yes :) 14:51:33 The other is the whole "where goes GBP" 14:51:47 Did you make progrss on that at the recent meeting? 14:52:08 i want a pony. 14:52:09 I think so, the bottom line is there is not enough consensus, and the incubator appears like a good place for things like this. 14:52:19 The incubator proposal has some issues, which we're working through. 14:52:25 russellb: I heard that two times today from my daughters already 14:52:30 nice! 14:52:39 russellb: Or a unicorn. Unicorns are nice too. :) 14:52:46 russellb: the problem being, I actually have enough room for two ponies or more 14:53:03 how about time to take care of them? 14:53:15 i want to visit your town :) 14:53:21 sounds peaceful. 14:53:22 russellb: that's the plan. 14:53:43 get me cloned so that I can actually take care of all the ponies. OpenStack and otherwise. 14:54:13 mestery: ok, so it's not stalled 14:54:41 ttx: The incubator appears to be the way forward here. 14:54:57 mestery: ok cool 14:55:07 mestery: anything you want to discuss at meeting today? 14:55:20 I don't see any obvious next step on the pny migration though 14:55:38 ttx: Maybe we can discuss when projects are opening Kilo specs? 14:55:42 I'd like to wait as long as possible 14:55:45 Given where we are in Juno. 14:55:51 I think mikal and I are in agreement there at least. 14:55:56 It's actually on the agenda already 14:56:00 ttx: Perfect. :) 14:56:06 mikal added it this morning 14:56:09 cool 14:56:14 ok, so let's discuss that today 14:56:21 Thanks! 14:56:24 mestery: talk to you later then 14:56:30 ttx: ack, later! 15:30:21 ttx: hello 15:30:39 notmyname: o/ 15:30:53 #topic Swift 15:31:37 Getting close to that "end of august" deadline 15:31:45 hmm...I just sent an email to openstack-dev and it was help for moderation 15:32:02 notmyname: yeah, it hits a too many recipioent limit, I moderated it through 15:32:08 ah ok 15:32:14 not your fault, the original mail had the same issue 15:32:20 oh ok 15:32:32 it's lie he copied 20 people on it 15:32:38 like* 15:33:10 so, for an end-of-august release. I'll be looking at that this week 15:34:33 ok, so we should know nect week? 15:34:38 next* 15:35:17 yes. well, I think it's pretty easy to say we can have one. there's several nice little things. but by next week I can give a better idea of a date and what's in it 15:35:38 #info more info next week on potential end-of-august intermediary swift release 15:36:16 notmyname: anything you want to discuss at meeting today? 15:36:47 ttx: no. I'm working on the gap stuff still. 15:37:11 ack 15:37:37 OK, well I guess we'll have more to talk about next week once you know what you want to do for the intermediary release 15:38:03 Anything else you wanted to mention here? 15:38:14 ..thinking... 15:38:33 no 15:38:44 ok then, talk to you later 15:47:39 zaneb: o/ 15:47:48 o/ 15:47:50 #topic Heat 15:47:54 is it Tuesday already? 15:48:08