21:02:35 #startmeeting 21:02:36 Meeting started Tue Jun 19 21:02:35 2012 UTC. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:02:38 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:02:47 Agenda @ http://wiki.openstack.org/Meetings/ProjectMeeting 21:03:03 vishy: run 21:03:09 #info Only 2 weeks left until the milestone-proposed cut for Folsom-2 (July 3) 21:03:37 #topic Actions from previous meeting 21:03:42 * chmouel to post about python-swiftclient separation to the ML: DONE 21:03:50 * chmouel to discuss with CI on enabling swift in devstack-gate 21:04:00 ttx: chmouel asked me to comment. he's not here 21:04:00 I think I just saw notmyname discussing it in a previous meeting ? 21:04:20 that hasn't been done, but I discussed it briefly with the ci team in their meeting today 21:04:39 so there is a little more to talk about, but the conversation is definitely started 21:04:45 I can take it as an action item 21:05:00 #action notmyname to pursue discussion on enabling swift in devstack-gate 21:05:08 * jgriffith to update the ML with Cinder progress 21:05:18 I don't think I've seen that 21:05:37 let's carry over 21:05:45 #action jgriffith to update the ML with Cinder progress 21:05:52 * ttx to clarify Horizon 2012.1.1 release process and fix CI to match 21:06:14 ttx: email sent out to cinder core and select others with no feedback 21:06:27 markmc took the lead on that, we now have proposed 2012.1.1 tarballs for all projects that use such versioning, based on current stable/essex branches. 21:06:35 jgriffith: could you send it to the general ML ? 21:06:36 jgriffith: you might want to cc openstack 21:06:46 please 21:06:51 #help Please test proposed 2012.1.1 deliverables: https://lists.launchpad.net/openstack/msg13251.html 21:06:55 Yeah, I can do that, but it's awkward right now anyway because everything is in Draft 21:06:59 Nobody can pull it 21:07:14 But yes, after this meeting I'll send an updated to ML 21:07:17 jgriffith: thx 21:07:21 ty 21:07:31 #topic bp-issues script 21:07:44 #info I finally took the time to write a tool to single out generic issues with blueprints 21:08:10 heckj, notmyname, bcwaldon, vishy, ohnoimdead, danwent: You can use it to find issues about your plan and fix them before I ask you to do so in the meeting 21:08:16 Think of it as a pythonized version of me. 21:08:22 ttx: excellent 21:08:23 #link https://github.com/ttx/bp-issues 21:08:26 import ttx 21:08:27 sounds nice 21:08:30 ttx: please call it ttx.py 21:08:36 It catches unassigned/unprioritized stuff, things that are missing from series goals, bad dependency prioritization, etc. 21:08:38 yeah ^^ 21:08:49 #action ttx to rename script to ttx.py 21:08:50 ttx: or thierrorize.py 21:09:17 It's still a bit early so probably will have a few false negatives 21:09:30 and will be updated to catch more stuff as you get better :) 21:09:42 #topic Keystone status 21:09:53 heckj: hello again :) 21:10:00 #link https://launchpad.net/keystone/+milestone/folsom-2 21:10:03 ola! 21:10:15 Looks good, just a couple of questions: 21:10:23 * https://blueprints.launchpad.net/keystone/+spec/stop-ids-in-uris (Guang Yee) 21:10:34 You said last week that is actually dependent on implement-v3-core-api... 21:10:42 Does it make sense to track it as a separate objective ? 21:10:52 Should it depend on (or be depended on by) implement-v3-core-api ? 21:10:53 yep - meant to shift it back/merge it, didn't get that done 21:11:07 it depends on implement-v3-core-api 21:11:25 #action heckj to repurpose stop-ids-in-uris wrt implement-v3-core-api 21:11:33 ok, thx 21:11:34 yep 21:11:36 * https://blueprints.launchpad.net/keystone/+spec/rbac-keystone (dolphm) 21:11:51 You mentioned last week that this is also depending on implement-v3-core-api and would likely be split ? 21:12:18 split between... a pre-v3 and a post-v3 thing ? 21:13:02 have the split blueprints related to V3 & RBAC up, but not fully assigned to people, milestones, etc 21:13:11 some of that just nailed down this morning in the keystone meeting 21:13:31 OK, that might explain the following... 21:13:36 ttx.py picked up the following issue: 21:13:46 https://blueprints.launchpad.net/keystone/+spec/rbac-keystone-api 21:13:46 * Not in series goal while targeted to a series milestone 21:13:55 yeah, that's it exactly. 21:14:03 kewl, will let you fix 21:14:13 heckj: anything else ? 21:14:33 fix in now - nope- questions? 21:14:38 How is v3-api feedback/implementation going ? Any cut date on the feedback ? 21:15:08 heckj: I've kept the images api v2 spec open through development - I would suggest you do the same 21:15:09 excellent feedback - have some ML responses to make. Hoping to nail down consensus sufficient to begin implementation in another 5-10 days 21:15:23 bcwaldon: damned good idea 21:15:34 heckj: once you start implementing it, you find some pretty dumb stuff 21:15:45 totally believe it 21:15:52 heckj: so you could start implementing it right now ! 21:16:06 (let's see if that trap works) 21:16:14 ttx: patches welcome! 21:16:30 heckj: I noticed the term "tenant" was still in the v3 API draft... so it's not changing to "project"? /troll 21:16:31 touch� 21:17:11 ready to switch to swift? 21:17:20 would you all tar and feather me if I changed it all RIGHT NOW!?!? 21:17:40 s/tenant/project/g 21:17:42 #topic Swift status 21:17:48 notmyname: yo 21:17:51 hi 21:17:53 Next version should be 1.5.1, do you have any plans already ? 21:18:10 I've been out this past week and just catching up today. no plans yet. I 21:18:14 We know it will include the swiftclient split and an important bugfix... 21:18:15 miskey 21:18:24 I'll work on that this week 21:18:26 yes 21:18:35 I'll work on setting a date this week 21:18:56 notmyname: bp-issues picked up two Essential specs listed in your Folsom plans, without any milestone yet: 21:19:03 * https://blueprints.launchpad.net/swift/+spec/extract-client-lib 21:19:04 * https://blueprints.launchpad.net/swift/+spec/keystone-middleware 21:19:15 Should those both be targeted to 1.5.1 ? 21:19:20 the first will be in 1.5.1 (we just talked about it) 21:19:30 the 2nd I'm not sure yet 21:19:40 ok, targeting the first one 21:19:51 notmyname: Anything else ? 21:20:02 chmouel said he was going to pick it up, but no work has been done yet 21:20:15 no, nothing else right now. questions? 21:20:18 notmyname: isnt there already a python-swiftclient? 21:20:45 bcwaldon: the separation hasn't landed in a release yet 21:21:03 notmyname: so python-swiftclient is the product of extract-client-lib? 21:21:11 correct 21:21:13 ok 21:21:20 i am pleased 21:22:02 #topic Glance status 21:22:23 bcwaldon: hello 21:22:28 #link https://launchpad.net/glance/+milestone/folsom-2 21:22:29 ttx: why hello there 21:22:41 * ttx refreshes to avoid last-minute tricks 21:22:48 heh 21:22:49 ttx: I only changed one bug on you this time 21:22:52 bcwaldon: So far, looks like you're still on your way to complete api-2 in folsom-2 ? 21:23:00 ttx: doubt it 21:23:16 ttx: I'm more comfortable with f-3 for all of api-2 21:23:28 Oh ? Which part might not make it ? 21:23:38 api-v2-store-access ? 21:23:47 ttx: possibly 21:23:59 ttx: theres been a lot of refactoring going on recently 21:24:08 Was wondering if api-v2-images-pagination and api-v2-images-sorting were actually not implemented yet. 21:24:13 ttx: I think we're at a place that we can make measurable progress again, but with only 2 weeks left, i'm not 100% confident 21:24:22 ttx: they are 21:24:28 ttx: yes, markwash speaks the truth 21:24:43 ttx: some of the work we're doing could fit under a few different bps 21:24:49 markwash: awesome! set "implementation status" to implemented please 21:24:52 ttx: like pagination-related links 21:25:10 ttx: sure, sorry I missed that 21:25:29 ttx: the tenant-specific swift container storage may slip to f3 21:25:41 ttx: its not as important as previously thought 21:25:46 ttx: and it hasnt been started, afaik 21:25:50 bcwaldon: right 21:26:01 bcwaldon: bp-issues just picked up the following issue for you: 21:26:06 ttx: a lot should land in the next 7 days, though 21:26:07 * https://blueprints.launchpad.net/glance/+spec/kill-registry-server : Targeted to a milestone but has no assignee and unknown status 21:26:19 ttx: actually gonna wait until we get the next and first links finished before. . sorry, don't know where my head is 21:26:22 bcwaldon: ^^ 21:26:32 markwash: ok, sorting is done, right? 21:26:41 nod 21:26:49 ttx: we could argue about it, or just leave it 21:26:56 markwash: ack. matybe just add a comment on the whiteboard for the one that is still pending 21:27:59 bcwaldon: let's argue about it another day. I did my share of arguing today 21:28:03 bcwaldon: Anything else you wanted to mention ? 21:28:15 ttx: no sir 21:28:20 Questions on Glance ? 21:28:59 #topic Quantum status 21:28:59 is anybody else here feeling like a stakeholder in v2? 21:29:04 oops 21:29:12 feels like we're just in a room by ourselves, which is fine too :-) 21:29:28 * ttx lets 20 second for a stakeholder to show up 21:29:34 :-) 21:29:49 danwent: hey 21:29:53 hello 21:29:55 #link https://launchpad.net/quantum/+milestone/folsom-2 21:30:03 markwash: shh, don't let anyone find out we're changing things! 21:30:08 latency should be good b/c we're both in france 21:30:19 heh... Still very busy, looks like you should start postponing stuff that will obviously miss the bus, if any 21:30:31 Would like to review the essential stuff: 21:30:36 already have postponed several things. 21:30:46 https://blueprints.launchpad.net/quantum/+spec/improved-nova-quantum-integration 21:30:58 You mentioned potentially dropping other stuff to get this completed... how is it going so far ? 21:31:17 so Trey will again have cycles to work on this. Yesterday we also pulled in yong to work on it as well. 21:31:29 still may main concern, though tr3buchet expressed confidence 21:31:44 https://blueprints.launchpad.net/quantum/+spec/quantum-dhcp: still on track ? 21:31:57 this was the thing that was really off-track last time. 21:32:23 ah? misunderstood then 21:32:38 we now have a design together and the developer making progress, which is a big improvement. my personal opinion is that the scope of the design is too much work, but he's prioritizing the basic case first, so I think we'll be ok. 21:32:49 too much work for F-2 that is. 21:33:14 ok, we'll see next week how it goes 21:33:28 https://blueprints.launchpad.net/quantum/+spec/new-cli: looks almost there ? 21:33:32 indeed. i've asked for everything to be in for an initial review next week. 21:33:45 ttx: yes, very close. i'm doing some final testing on it. 21:33:58 great. 21:34:00 we'll probably merge the main branch tomorrow, with any outstanding issues being filed as additional bugs 21:34:00 danwent: Anything else ? 21:34:13 still a ton to do, but really happy with progress the team is making 21:34:35 commits and reviews are way up in the past month or so…. i should put a graph together 21:35:00 * ttx senses a blogpost coming 21:35:03 Questions on Quantum ? 21:35:23 yes, but F-2 work before blogpost :) 21:35:43 #topic Nova status 21:35:52 vishy: how was that coffee ? 21:35:58 not good 21:36:10 ttx: I made it back in time, and the coffee did too 21:36:17 vishy: come to my place. I bought an automatic espresso machine with grinder 21:36:21 ttx: but it was not in the right location 21:36:33 ttx: in the cup or in my tummy would have been the right location 21:36:44 ttx: unfortunately it was on the ground and on my shirt 21:36:48 == wrong location 21:36:55 hmm. Suboptimal 21:36:55 :| 21:37:00 agreed 21:37:03 #link https://launchpad.net/nova/+milestone/folsom-2 21:37:22 Looks like we are a bit behind. Should probably start deferring stuff that we already know won't make it 21:37:29 First let's review the status on the essential stuff: 21:37:38 * https://blueprints.launchpad.net/nova/+spec/general-host-aggregates (jog0) 21:37:54 jog0 mentioned last week that this was on track -- but maybe some parts might not land, so it could be split 21:38:10 jog0: yes i was chatting today 21:38:23 russellb: any progress on the no-db stuff? 21:38:26 ttx: after sorting out some details with vishy earlier today the first part is on target. Expect a patch set for review later this week 21:38:35 jog0: cool 21:38:43 * https://blueprints.launchpad.net/nova/+spec/finish-uuid-conversion (mikal) 21:38:51 do we have news on that ? 21:39:13 vishy: very litle ... been distracted by other things this cycle 21:39:26 been looking this week though 21:39:47 still some chance for folsom-2, but more likely folsom-3 21:39:54 russellb: that is what I was thinking 21:40:05 vishy: my understanding is that there is more coming to finish-uuid-conversion ? 21:41:03 ttx: I think the prop in is the last 21:41:14 ok 21:41:14 * https://blueprints.launchpad.net/nova/+spec/volume-decoupling (vishy) 21:41:21 My understanding is that Cinder can now fully be used, so this is complete ? 21:41:36 (or not) 21:41:52 ttx: well there are a few more things on the whiteboard, but those have their own blueprints in the other projects 21:42:10 ttx: so I suppose we could just link those blueprints and mark it implemented? 21:42:15 if nothing else needs to land in nova, should be implemented yes 21:42:26 What's your plan for F2: keep both options and default to nova-volume ? 21:42:44 and use this one: https://blueprints.launchpad.net/nova/+spec/extract-nova-volumes 21:42:57 F2 keep both and default to nova-volume 21:42:57 right 21:43:09 F3 switch the default 21:43:12 agreed 21:43:24 (after the PPB declares Cinder core (or not)) 21:43:33 trusted-messaging (ewindisch) is also marked "not started", I suppose it's unlikely to hit F2 now ? 21:44:13 * russellb hasn't heard anything about itlately 21:44:32 Finally I was wondering if those two were not already completed: 21:44:39 * https://blueprints.launchpad.net/nova/+spec/zeromq-rpc-driver (ewindisch) 21:44:39 * https://blueprints.launchpad.net/nova/+spec/lvm-disk-images (Boris Filippov) 21:44:49 top one yes 21:44:53 zero-mq-rpc-driver is complete 21:44:58 ok, adjusting 21:45:05 yes both implemented 21:45:26 Wanted to discuss a bit about Nova bug triaging 21:45:34 The BugTriage day had a good effect on cutting down the number of untriaged bugs, but the numbers are increasing again: 21:45:40 http://webnumbr.com/untouched-nova-bugs 21:45:53 vishy: how do you suggest we solve that ? 21:45:59 Should we run BugTriage days more often ? 21:46:04 Should we add BugTriaging to the ReviewDays tasks ? 21:46:11 Should we just encourage more devs to do BugTriaging ? 21:46:18 Should we recruit a team of triagers ? 21:46:26 Should we write a thread on the ML about it ? 21:46:42 ttx: yes 21:46:55 we need more help triaging 21:46:56 yes..to which ? 21:47:11 thread to ml to recruit might be a good start 21:47:31 vishy: should we open the Nova bug supervisor team to empower more people ? 21:47:56 (I proposed to do it across the board but some smaller projects complained) 21:48:14 ls 21:48:31 vishy: ML thread : should I, should you ? 21:48:49 lets try recruiting first 21:49:43 vishy: you take the action to do the ML post or should I do it for you ? 21:50:04 ttx: you can do it! 21:50:07 vishy: Anything else ? 21:50:18 #action ttx to raise a new thread about Nova bug triaging 21:50:33 just tracking down some interesting racy network bug that seems to have crept in 21:50:41 otherwise the stability has been pretty good 21:51:01 especially considering the amount of changes that have gone in 21:51:18 Indeed. Questions on Nova ? 21:52:20 #topic Horizon status 21:52:28 ohnoimdead: still around ? 21:52:32 o/ 21:52:36 #link https://launchpad.net/horizon/+milestone/folsom-2 21:52:46 Nice progress overall, don't have any comments! 21:53:02 Just one question about https://blueprints.launchpad.net/horizon/+spec/global-ajax-communication 21:53:16 It's marked "Deferred": should we remove it from Folsom series / folsom-3 milestone ? 21:53:48 yeah, probably. that one sort of turned into a rather large conversation 21:54:12 ok, will un-folsom3-it as a start 21:54:15 ohnoimdead, ttx: given what the quantum guys were saying earlier in the meeting, it sounds like https://blueprints.launchpad.net/horizon/+spec/readd-quantum-support may be in jeopardy. 21:54:39 I guess that should also be directed at danwent 21:54:58 ttx: sounds good 21:55:09 gabrielhurley: arvind was supposed to send me an update today 21:55:14 but i haven't heard from him. 21:55:29 we really narrowed the scope of what we're targeting for F-2, and arvind said he was comfortable with it. 21:55:40 danwent: gotcha 21:55:42 will ping him and include you all 21:55:43 gabrielhurley: ok, when you get the answer from arvind you can set to "slow progress" or "blocked" with e acomment on the whiteboard 21:55:51 gabrielhurley: we can kick to f-3 if necessary 21:56:06 yes, we can 21:56:16 gabrielhurley: (only keep it "good progress" if it's on track 21:56:17 ) 21:56:23 ttx: duly noted 21:56:26 ohnoimdead: Anything else ? 21:56:43 nope, i think we are looking good. we got a couple of interns for the summer helping out as well. ;) 21:56:50 Questions for Horizon ? 21:57:20 #topic Other Team reports 21:57:33 Any other team lead with a status report ? annegentle ? 21:58:04 Well.... yeah, kinda 21:58:21 Cinder should be available in the next couple of days (out of draft) 21:58:30 jgriffith: good good 21:58:42 Four of the five blueprints should be implemented (at least for first pass) 21:58:55 nova volume decouple is what I'm trying to finish up now 21:59:05 that's all 21:59:07 ok 21:59:07 thanks 21:59:15 #topic Open discussion 21:59:18 jgriffith: I'll set up a meeting to talk docs 21:59:28 Anything else, anyone ? 21:59:34 annegentle: yes, I haven't forgotten, just still wasn't ready :( 21:59:42 jgriffith: no worries 21:59:50 Note that during a webinar today the events team announced that our next Design Summit would be held in San Diego on the week of October 15th 22:00:10 Mark your calendars 22:00:22 still working on a "deployment template" document that helps people inform others about their deployment 22:00:33 still invite people to work on an operations manual 22:00:46 annegentle: business as usual, I see 22:00:59 ttx: :) 22:01:21 annegentle: anything else you wanted to mention before we close it ? Looks like we always are the only ones listening in the end :) 22:01:32 ttx: I wonder if adding a "deployers news" section to this meeting would be useful? 22:01:54 annegentle: to inform users of largish changes ? 22:02:05 or to brag about deployments ? 22:02:19 ttx: or for deployers to bring their bugs/track stuff? 22:02:25 everyone wants to track stuff :) 22:02:44 annegentle: hm, they are supposed to ask questions on each project status update 22:02:50 not the only ones... 22:02:58 ttx: ah, ok, it's embedded throughout 22:03:02 * ttx hugs heckj 22:03:29 annegentle: yep 22:03:35 ok, let's close it 22:03:38 #endmeeting