21:01:08 #startmeeting project 21:01:09 Meeting started Tue Jun 25 21:01:08 2013 UTC. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:10 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 21:01:11 o/ 21:01:13 The meeting name has been set to 'project' 21:01:19 #link http://wiki.openstack.org/Meetings/ProjectMeeting 21:01:31 3 weeks left to H2. Most work should now be started, and a good half should be merged or under review already 21:01:41 We'll spend a bit extra time on Swift (1.9.0) and Neutron (renaming plan). 21:01:52 and Swift will go right after the general stuff 21:01:55 thanks 21:01:58 #topic General stuff 21:02:17 Nothing on my side 21:02:21 sdague, annegentle, jeblair/mordred: news from QA/Docs/Infra teams ? 21:02:29 yes 21:02:31 soon-to-be-programs ? 21:02:34 #link http://lists.openstack.org/pipermail/openstack-docs/2013-June/002014.html 21:02:43 shoot 21:03:00 we're planning on doing the quantum->neutron rename on saturday july 6 at 1600utc 21:03:01 that's the latest from doc land -- Shaun McCance started as a contractor for Cisco last week to sort through install docs. 21:03:20 we would like to have core reviewers from as many openstack projects as possible on hand during that time 21:03:44 seriously. this one _could_ break in weird places 21:03:53 in case we find that there is something in some project that needs fixing in order for gating to work 21:03:55 annegentle, awesome, shaun rocks 21:04:10 jeblair: do we have a wiki page to track the transition details ? 21:04:14 also - infra team is all going to be offsite in nyc thursday or friday of this week - so if people could try to not break everything, that would be neat 21:04:31 jeblair, ah, interesting - ok 21:04:39 * markmc wonders how he's going to remember that 21:04:55 ttx: no, but that's a good idea... perhaps if there was one with a sign-up sheet for each affected project 21:05:10 we'd know what kind of representation we can expect 21:05:16 ttx: I'm working on a wiki for the transition 21:05:17 i will set that up and link to it in the email announcement 21:05:23 jeblair: +1 21:05:28 i will add that to markmcclain's wiki page 21:05:45 ttx: [end] 21:05:59 annegentle, jeblair: thx 21:06:36 #topic Swift status 21:06:40 notmyname: o/ 21:06:40 hi 21:06:43 #link https://launchpad.net/swift/+milestone/1.9.0 21:06:50 Got your email about cutting a candidate today for release on July 2 21:07:00 yes, we are very close 21:07:02 How is that plan going so far ? 21:07:05 with your conference and stuff 21:07:12 I see bug 1180680 on the 1.9.0-critical list, is that the blocker ? 21:07:22 is that the dispersion report? 21:07:25 oh, you fixed it 21:07:32 who broke the bot 21:07:42 yes dispersion report 21:07:44 if so, yes. it's got one +2. just needs one more. 21:07:55 and we need one more test run for the write affinity 21:08:03 NothingDone: merged one hour ago 21:08:05 ow 21:08:11 notmyname: merged one hour ago 21:08:16 ah, cool 21:08:21 OK, well, send me an email or an IRC ping with the SHA to cut milestone-proposed from, when you have it. 21:08:37 with changelog in 21:08:37 will do. I've also got a little LP work to finish with it 21:08:46 and bump to next version on master 21:08:47 ya, authors/changelog/affinity 21:08:54 the normal dance 21:08:59 ack 21:09:11 On the page linked above, "Multi region replication" and "Proxy affinity (writes)" are still marked in progress. Should they be considered "Implemented" ? 21:09:15 I've quite excited about this release. lots of good features, not the least of which is full global cluster support 21:09:26 ttx: when the write affinity lands 21:09:35 heh, fair enough 21:10:08 missed your mention of it above 21:10:12 notmyname: anything else you wanted to raise ? 21:10:22 ttx: I'm good 21:11:26 #topic Oslo status 21:11:35 markmc: hi! 21:11:36 yo 21:11:38 #link https://launchpad.net/oslo/+milestone/havana-2 21:11:55 so, I knocked a few bps out to havana-3 21:12:01 still seems optimistic 21:12:15 like a few of those probably won't make it, but I can't quite predict which yet 21:12:28 * ttx refreshes 21:12:39 Looks on track to me... 21:12:47 review review review 21:12:50 yeah 21:12:53 too good to be true 21:12:55 http://russellbryant.net/openstack-stats/oslo-openreviews.txt 21:13:03 we've got plenty of reviewing to be done 21:13:03 you can't really defer more though 21:13:05 https://review.openstack.org/#/c/32445/ looks weird btw 21:13:13 ttx, can't really defer more? 21:13:49 jd__, yeah, I can't remember the details, but danpb did persuade me this would be useful in all projects 21:13:51 well, you already have 9 blueprints in h3 so given your velocity I'd say that deferring them means deferring again at the end of H3 21:14:05 ttx, ah, got it 21:14:17 it's not as if you had an empty h3 still :) 21:14:23 markmc: anything you wanted to raise ? 21:14:26 markmc: I actually meant the status of the review, it's like the commiter didn't notice the test failed and he's waiting on something 21:14:33 nothing huge, but ... 21:14:42 I'll do oslo-config-1.2.0a3 snapshot this week 21:14:55 main change is ordering of priority of CLI args 21:15:05 also, monty is putting integration testing in place for pbr 21:15:14 and we'll have a slew of changes landing once that's in place 21:15:20 so expect another pbr release 21:15:24 ack. Questions about Oslo ? 21:15:30 and we should be doing hacking 0.5.6 soon too 21:16:08 * ttx needs to work on streamlining library release announcements 21:16:10 jd__, ah, thanks 21:16:18 ttx, with release notes 21:16:27 * markmc started doing release notes for oslo.config 21:16:46 yeah, implement what we discussed at the summit about automatically building those announces 21:17:03 because currently it's a bit of an unpredictable mess 21:17:07 from git log? 21:17:15 human editing is preferable IMHO 21:17:27 anyway, digression 21:17:29 markmc: ok, will discuss off-meeting 21:17:35 #topic Keystone status 21:17:50 ayoung: representing keystone ? 21:17:59 Here 21:18:02 or is jcannava proxying here as well ? 21:18:07 #link https://launchpad.net/keystone/+milestone/havana-2 21:18:08 Keystone H2 Blueprints: 21:18:21 For the most part looking good. A couple questions 21:18:26 jaypipes still planning on doing Region support? 21:18:57 Looks like you're a bit behind, yes 21:19:01 inherited-domain-goals in particular looks more likely to hit by havana-3 at that point ? 21:19:05 ayoung: last I checked, termie still has the review blocked. 21:19:20 jaypipes, and I think we need to override that 21:19:42 jaypipes: so that work is started ? 21:19:58 jaypipes: shoudl I set "Needs code review" as status ? 21:20:11 Allow project roles to be inherited from owning domain probably underway, but Henry Nash off this week. 21:20:25 ayoung: so it could still make h2 ? 21:20:30 ttx it was just the api doc that had gotten posted 21:20:35 ttx, unlikely 21:20:35 ttx: dolphm asked me to make a proposed change to the API spec first, which I did. termie poo-poo'd it and -2 the review. 21:20:43 jaypipes, link? 21:21:03 jaypipes: so I'll mark it "started". 21:21:18 (or "Blocked" 21:21:19 ) 21:21:46 ayoung: please look into it and unblock it if you can 21:21:50 shardy: any progress on https://blueprints.launchpad.net/python-keystoneclient/+spec/delegation-impersonation-support ? 21:21:54 ttx, We'll make it happen 21:21:56 ayoung: https://review.openstack.org/#/c/27563/ 21:22:11 adding review link and marking it blocked 21:22:19 ttx: not yet, been gated on too many heat task 21:22:41 tasks, planning to look at it later this week 21:22:48 ok, we'll discuss the impact on heat-trusts later in the meeting 21:22:55 ayoung: Anything more about Keystone ? 21:22:59 other H2 tasks seem to be on target 21:23:00 yes 21:23:08 Horizon wants updated client for V3 Auth. according to lcheng 21:23:10 Is anyone else able to cut a python-keystoneclient release while Dolph is gone? Horizon needs a new release now that the v3 auth code is finally done and merged. It's currently a blocker for us. 21:23:13 lol 21:23:39 mordred: I'm looking at you... 21:23:41 gabrielhurley: I /could/. But dolphm should be back very soon, so maybe it can wait tomorrow 21:23:52 is he back that soon? 21:23:52 I did not do it 21:23:55 also 21:23:59 if it's one day that's fine 21:24:03 Unified auth. Projects are all over the place. Would liketo make keystone client the libary of choice there, but sort of competes with the common cli effort. 21:24:08 if it's a week it'd be good to have someone do it sooner 21:24:24 gabrielhurley: if he is not back tomorrow ping me and I'll tag 21:24:32 ayoung: does it? 21:24:32 He's gone this week and next, IIRC 21:24:33 can anyone confirm when he's back? 21:24:54 ayoung: uh 21:25:00 sorry, I am confusing the issue. Lets finish the cut new client discussion first 21:25:04 ayoung: Ithought unified CLI was still using underlying libs - so it would also consume keystoneclient for auth, no? 21:25:09 * mordred shuts up 21:25:11 mordred: i hope so. 21:25:26 so, yeah, we should cut a new client. 21:25:31 +1 21:25:49 gabrielhurley: fire an email, cc dolph, ayoung and me. If he is not around, I'll push the tag for you 21:25:53 k 21:25:56 doing now 21:25:56 Sounds good 21:25:59 just want to give him an oppty to complain about the idea 21:25:59 ttx: can't ayoung do it? 21:26:00 OK, back to the auth thing 21:26:06 ttx: it's anyone in core isn't it? 21:26:10 right now, not all of the other CLIs use keystoneclient 21:26:17 in -drivers I think. 21:26:20 ah 21:26:37 ayoung: I just discovered that in troveclient, in fact. I was angry 21:26:43 jamielennox just sent me an email with a list of the status 21:26:54 So for example... 21:26:56 ayoung: could you push that to -dev ? 21:27:02 (status) 21:27:10 Cinder: 21:27:10 - Uses requests 21:27:10 - Does not use keystoneclient 21:27:10 - Uses sleep f 21:27:18 Yes, will do so 21:27:30 then we can come up with a cleanup plan 21:27:36 wanted to make sure there was nothing of a security nature in it first 21:27:38 ayoung: I am strongly in favor of fixing that 21:27:49 ayoung: sure 21:27:54 * mordred will gladly sit on the sidelines and cheer on the effort 21:27:59 ok, we need to move on 21:28:04 but are we OK with saying that all of the current clients change over to using keystoneclient as a dependency for auth? 21:28:20 ayoung: I think that would be a sane statement 21:28:25 ++ 21:28:42 [end] 21:28:46 #topic Ceilometer status 21:28:50 * ttx runs 21:28:51 o/ 21:28:55 jd__: hey 21:28:56 oh I had a keystone Q 21:29:01 o/ 21:29:09 ayoung: I'd also like to know when it's ok to publish a v3 WADL to api.openstack.org/api-ref.html? I didn't realize Dolph was out that long. 21:29:10 annegentle: ah! 21:29:12 #undo 21:29:12 (sorry) 21:29:13 Removing item from minutes: 21:29:14 rollback; 21:29:20 jd__: hee 21:29:20 annegentle: shot 21:29:35 annegentle: ask your question :) 21:29:54 ayoung: mostly a timing thing, when is it "correct" to document v3 21:30:09 ayoung: I think now's ok but wanted to confirm 21:30:19 man maybe I missed him... I'll email 21:30:22 annegentle, I haven't thought about it. I will have to get back to you. There is a V3 API doc, of course, so the answer seems to be "now" 21:30:33 ayoung: ok, sure. 21:30:44 #topic Ceilometer status 21:30:51 #link https://launchpad.net/ceilometer/+milestone/havana-2 21:31:02 Slightly behind schedule, I'd say 21:31:15 eglynn and dhellmann in particular look like they are late 21:31:22 and you can't defer that much to havana-3 21:31:24 I agree 21:31:46 did they give you any signs that they would catch up by working 24-h shifts ? 21:31:47 ttx: I'm catching up, nose to grindstone ... 21:31:51 I will pull their ears 21:32:06 ttx: dhellmann is supposed to start this week 21:32:13 jd__: consider mine pulled :) 21:32:20 I'm fine with stuff not making it... but I prefer to have those roadmaps reflecting reality as much as they can 21:32:40 eglynn: if you think you can still make it, then I'm fine 21:32:48 jd__: anything you wanted to mention ? 21:32:50 ttx: cool 21:32:55 ttx: nop 21:33:00 Questions on Ceilometer ? 21:33:18 #topic Glance status 21:33:24 markwash: o/ 21:33:25 #undo 21:33:30 nice try. 21:33:30 darnit 21:33:32 #link https://launchpad.net/glance/+milestone/havana-2 21:33:38 refresh please. . . 21:33:49 refreshed. 21:34:14 Look slightly behind, but not too bad 21:34:24 yeah, we had a bit of a culling 21:34:28 How is async-glance-workers going so far ? 21:34:46 ttx: I have had to push that off somewhat, and need to reflect that change in the assignee 21:34:57 but it has active work 21:35:06 markwash: still likely to hit h2 ? 21:35:13 ttx: as far as we can tell now, yes 21:35:25 Iccha didn't start working on api-v2-property-protection yet ? 21:35:34 looks like it will be deferred again 21:35:40 not that I've seen, but it is scheduled for h2 timeframe 21:35:44 There is not a lot of room left for deferring to h3. 21:36:04 things in h3 might get pushed to I, to make room for higher priority h2 misses 21:36:08 will start working on it week hopefully once the design is approved 21:36:16 *next 21:36:18 * markwash approves design 21:36:22 :-) 21:36:36 markwash: shall set up the next/ongoing stuff as soon as I can beat LP into submission 21:36:47 * markwash cheers 21:36:50 markwash: anything you wanted to mention ? 21:36:55 yes 21:37:01 (for the record, the API does not expose series goal at all) 21:37:01 "Add sheepdog store" 21:37:14 this is all ready, except we're having a bear of a time getting it to pass py26 tests in jenkins 21:37:24 zhi yan has narrowed it down to an eventlet issue 21:37:48 basically I need some help from infra folks to get me oriented on jenkins boxes 21:37:50 #help :-) 21:38:01 that is all 21:38:10 mordred/jeblair ^ 21:38:16 Questions on Glance ? 21:38:28 #topic Neutron status 21:38:32 markmcclain: hi! 21:38:37 #link https://launchpad.net/quantum/+milestone/havana-2 21:38:37 hi 21:38:57 Slightly behind, but nothing too big 21:39:02 And you have plenty of room to defer to h3 21:39:12 yeah.. I expect a few will defer 21:39:28 also the rename will be disruptive 21:39:32 do you have a URL set up for the renaming plan ? 21:39:47 which we could add to the minutes of the meeting ? 21:40:05 not yet.. I'll push to -dev list 21:40:40 ok, everything else seems to be under control 21:40:42 markmcclain: anything you wanted to raise ? 21:40:59 #link https://wiki.openstack.org/wiki/Network/neutron-renaming 21:41:08 nothing else to raise 21:41:11 cool, thx 21:41:18 Questions on Neutron ? 21:41:33 #topic Cinder status 21:41:38 jgriffith: hola! 21:41:42 #link https://launchpad.net/cinder/+milestone/havana-2 21:41:42 Howdy!! 21:42:00 Still a bit slow progress overall... but then you have room in h3 to defer 21:42:04 I've got requests out for Huaweii fellows and one of the DB guys 21:42:11 Tons o'room :) 21:42:14 markwash: can you drop by #openstack-infra tomorrow? i'm about to head out to dinner. 21:42:23 Should https://blueprints.launchpad.net/cinder/+spec/netapp-unified-driver be considered 'implemented' since https://review.openstack.org/#/c/33168/ was merged ? 21:42:26 I think most of these are going to make it 21:42:39 Umm...yup 21:42:46 will fix 21:43:03 Still have 4 blueprints proposed @ https://blueprints.launchpad.net/cinder/havana/+setgoals 21:43:12 can't wait until we don't have to care for that anymore 21:43:17 jgriffith: anything on your mind ? 21:43:30 you and I both.. everybody keeps sneaking stuff in there 21:43:38 Nope, just the note about pecan to the ML 21:43:43 other than that we're rollign along 21:43:48 Questions on Cinder ? 21:44:08 jeblair: sure, thanks! 21:44:11 #topic Nova status 21:44:14 russellb: hey 21:44:17 #link https://launchpad.net/nova/+milestone/havana-2 21:44:30 hey 21:44:44 A slight bit behind, I'd say, from a 10,000 feet viewpoint 21:44:45 mountain of code reviews to do 21:44:48 yeah. 21:44:57 i'm starting to move stuff "Not Started" to havana-3 21:45:00 Critical point being Christopher Yeoh's 10 High blueprints. Could you give us an update on that ? 21:45:26 Yeah, so those are all subsets of the v3 API work, which he broke up. More than just him are working on them 21:45:30 and they've made lots of good progress 21:45:44 russellb: so that's still on track 21:45:47 yeah 21:46:00 russellb/vishy: about deprecate-nova-network, should that work start early so that it's completed by h3 ? 21:46:02 reviewers probably need to do a better job at giving their reviews attention 21:46:20 yeah, i haven't heard anything about that 21:46:28 vishy and garyk were going to work on it 21:46:35 russellb: I haven't made much progress on that side 21:46:36 * ttx doesn't like Essential/h3 stuff 21:46:41 vishy: figured 21:46:53 another thing was full tempest support, which i'm not sure we're even tracking 21:47:13 vishy: what's your plan ? Start soon ? Doesn't sound like a no-brainer to me ;) 21:47:26 ttx: well we don't have one obv! 21:47:28 or do it all within h3 ? 21:47:39 obv? 21:47:42 obviously? 21:47:51 ttx: garyk was investigating trying to move from nova network to quantum via live migration 21:47:51 obviously. 21:48:00 and i was going to look at moving without live migration 21:48:22 I have managed to replicate a few different forms of nova-network in quantum 21:48:31 one of them is pretty much a no-go 21:48:45 vishy: "essential" says we really need to have a plan though. whereas "high" says we really want to have a plan :) 21:48:46 but i haven't tested the actuall unplug and plug 21:48:46 ah interesting ... would be good to have a write-up on those 21:49:03 russellb: i will put something together 21:49:03 ttx: probably should downgrade to High at this point. 21:49:06 vishy: awesome 21:49:16 * markmcclain is interested in what you've found 21:49:24 russellb: tat would mean go for another release with nova-network not deprecated, right ? 21:49:33 (if we miss it) 21:49:36 ttx: it would ... and i guess i'd really rather not 21:49:38 russellb: the multi-host network (not merged yet) stuff is required to get really close 21:50:06 russellb: I will write up some stuff and try to the unplug plug stufff 21:50:12 ok cool 21:50:12 russellb, vishy: ok 21:50:27 we'll track progress on that on future meetings 21:50:37 sounds good 21:50:46 and maybe downgrade to High if it starts to smell funny 21:50:49 russellb: anything else you wanted to mention ? 21:51:15 don't think so 21:51:17 Any question on Nova ? 21:51:22 we're running behind on reviews, hoping we can improve tht 21:51:30 #topic Heat status 21:51:35 shardy: o/ 21:51:37 o/ 21:51:40 #link https://launchpad.net/heat/+milestone/havana-2 21:51:56 Good progress, still slightly behind, I'd say 21:52:16 moved a couple of BPs to h3 that were not started or blocked 21:52:18 we shoudl have a better view next week 21:52:28 yes, h3 is filling up fast 21:52:34 otherwise, think we should be OK with what is there 21:52:44 you moved heat-trusts to h3, right ? 21:52:45 the problem recently is review overhead 21:53:16 Was wondering if https://blueprints.launchpad.net/heat/+spec/provider-upload was not completed already with https://review.openstack.org/#/c/33008/ 21:53:28 ttx: yes, it's taken, way, way longer than I expected to get the suspend/resume code merged, so the keystoneclient and heat-trusts stuff has slipped 21:54:04 ttx: Possibly, I need to check with asalkeld at our meeting tomorrow if there are more patches coming 21:54:11 shardy: anything else you want to raise ? 21:54:32 ttx: don't think so, no, thanks 21:54:32 Questions about Heat ? 21:54:48 #topic Horizon status 21:54:51 gabrielhurley: o/ 21:54:54 #link https://launchpad.net/horizon/+milestone/havana-2 21:55:10 A bit late at this point, and not so much room to defer to h3 21:55:53 \o 21:56:08 yeah, some things are gonna fall off 21:56:11 The "igh" in particular are not in a very good shape 21:56:15 high 21:56:36 I'll rearrange whatever's not started by next week (including knocking a few off of H3 most likely) 21:56:46 gabrielhurley: +1 21:56:53 Two blueprints proposed @ https://blueprints.launchpad.net/horizon/havana/+setgoals fwiw 21:56:59 gabrielhurley: anything you wanted to mention ? 21:57:24 not particularly. you can have the last minutes for the other projects 21:57:54 Questions on Horizon ? 21:58:07 #topic Incubated projects 21:58:18 Hi! 21:58:23 devananda: o/ 21:58:30 just one thing for today 21:58:31 Looks like we don't have hub_cap around 21:58:37 A key feature for Ironic being feature-complete (and thus reaching an RC) is porting the PXE driver, which is currently blocked on porting image tools into glanceclient. Specifically, these two patches: 21:58:41 https://review.openstack.org/#/c/34320/1 and https://review.openstack.org/#/c/33327/ 21:58:54 so I'd like to raise awareness and hope some reviewers can come look at those :) 21:59:10 it's not the only thing on our plates -- but it's a big one 21:59:30 #help markwash: devananda needs glance-core help on https://review.openstack.org/#/c/34320/1 and https://review.openstack.org/#/c/33327/ 21:59:46 thanks much :) 21:59:52 devananda: are you likely to have something ready by the h2 milestone ? 21:59:57 nope 22:00:00 h3 is also unlikely 22:00:38 devananda: note that it affects the chnaces of proper graduation 22:00:49 indeed. 22:01:04 we'll cover that in a future meeting 22:01:20 i.e. progress or not progress :) 22:01:26 our time is up 22:01:30 #endmeeting