21:01:34 <ttx> #startmeeting 21:01:35 <openstack> Meeting started Tue Jan 17 21:01:34 2012 UTC. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 21:01:36 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic. 21:01:47 <ttx> Today's agenda: http://wiki.openstack.org/Meetings/ProjectMeeting 21:02:13 <ttx> #topic Actions from previous meeting 21:02:20 <ttx> * gyee to talk to zns about service-endpoint-location milestone targeting 21:03:01 <ttx> zns, gyee: ? 21:03:19 <zns> I have to search back on the outcome of that. I asked and got acknowledgement, but not sure about the status. I'd lean towards saying it won't make it in E3. 21:03:44 <ttx> nor in Essex ? 21:04:01 <zns> If it won't be in E3, then it won't be in Essex. Refer to last meeting :-) 21:04:25 <ttx> that was a loaded question :) 21:04:29 <ttx> * vishy to ping HP on openstack-api-ssl status 21:04:50 <vishy> done 21:04:51 <zns> jsavak and I have talked about starting an F branch for all new features. WHat's your take on that? 21:04:55 <vishy> result was undetermined 21:04:59 <jaypipes> hehe 21:05:14 <vishy> they said that they are under a lot of time pressure but they would try to update it this week 21:05:25 <ttx> zns: an official F branch sounds like a good way for developers not to work on bugfixes. 21:05:31 <jaypipes> vishy: please feel free to use me to get answers on that. happy to help. 21:05:43 <vishy> it is just supporting SSL for endpoints, so I actually think that one is ok post e-3 if need be 21:05:48 <zns> ttx: maybe we open an F branch after E4? E5? 21:05:53 <vishy> I'm still hoping they get it fixed though 21:05:58 <ttx> after E4 yes. 21:06:03 <zns> ttx: deal 21:06:05 <ttx> #topic Keystone status 21:06:16 <ttx> #link https://launchpad.net/keystone/+milestone/essex-3 21:06:21 <vishy> jaypipes: ok, I will include you in communication in the future 21:06:22 <ttx> zns: Is the status on this page accurate ? 21:06:28 <zns> yes 21:06:34 <jaypipes> vishy: cheers 21:06:41 <ttx> What's the status on: 21:06:45 <ttx> * https://blueprints.launchpad.net/keystone/+spec/keystone-client (anotherjesse) 21:06:57 <mtaylor> ttx, zns: re: opening f branch 21:06:58 <ttx> (on time for next week ? 21:07:21 <mtaylor> ttx, zns: we should talk, possibly offline, about that 21:07:36 <zns> ttx: that's the one my team is most concerned about since we have not put resources on it. Need input from anotherjesse. 21:07:39 <ttx> mtaylor: sure 21:07:48 <ttx> * https://blueprints.launchpad.net/keystone/+spec/keystone-configuration (dabo) 21:08:15 <zns> That one is almost ready for merge. SHould be ready. 21:08:21 <ttx> * https://blueprints.launchpad.net/keystone/+spec/keystone-logging (jheck) 21:08:38 <ttx> heckj, rather 21:09:01 <heckj> ttx: huh? 21:09:05 <zns> I've put some work into that. keystone-configuration includes improvements to it. Not sure about status from jheck, but I think we've made enough progress through other blueprints to call it good. 21:09:08 <mtaylor> ttx, zns: (you guys move too fast ...) python-keystoneclient is in all the systems and producing tarballs and whatnot now, in case that's helpful for keystone-client status 21:09:52 <ttx> zns: if any of those 3 miss next week, are you considering an exception for them, or will it be F ? 21:10:14 <ttx> heckj: status on keystone-logging ? 21:10:24 <zns> mtaylor: I'm not sure it is being used by all projects, though (maybe only horizon?). Keystone middleware, for example, doesn't use it yet. 21:10:28 <ttx> anotherjesse1: status on keystone-client blueprint ? 21:10:33 <mtaylor> zns: gotcha 21:11:10 <mtaylor> well, fwiw, it's hooked in to all the things needed for other projects to start depending on it in pip-requires and whatnot 21:11:11 <heckj> ttx: good progress, should be able to wrap up shortly 21:11:22 <ttx> heckj: cool, thx 21:11:25 <zns> ttx: I'm not considering any exceptions as of now. If anything, I've considered getting the feature in asap and stabilizing after E3, but I hear resistance to E3 being less unstable. 21:11:33 <heckj> ttx: expecting that to be resolved fully by E4 - hoping to close by E3 21:11:44 <ttx> #info In other news, we should have a python-keystoneclient tarball deliverable in Keystone E3 21:12:13 <ttx> #action anotherjesse to sync up with zns on keystone-client blueprint status 21:12:18 <ttx> zns: Anything else ? 21:12:24 <zns> nope 21:12:26 <ttx> Questions for Keystone ? 21:12:27 <zns> tx 21:12:28 * jaypipes still thinks the keystone client should be called keystone-cops 21:12:40 <jk0> jaypipes: +1 21:12:41 <zns> lol! 21:12:57 * mtaylor starts the gerrit renaming process... 21:13:00 <ttx> #topic Swift status 21:13:04 <ttx> notmyname: o/ 21:13:07 <notmyname> hi 21:13:11 <dragondm> +1 21:13:30 <anotherjesse1> ttx: the library stuff is done for the client - the CLI is being worked on as we speak - about 1day of work 21:13:32 <notmyname> not much to report this week. working on new features and bug fixes. I expect the next release to be fairly soon 21:13:40 <ttx> anotherjesse1: thx 21:13:48 <notmyname> to include for posting and some bug fixes 21:13:57 <notmyname> s/for/form/ 21:14:02 <ttx> notmyname: will it include object versioning ? 21:14:15 <notmyname> probably not yet. 21:14:23 <ttx> notmyname: so 1.4.6 it will be ? 21:14:31 <notmyname> versioning is probably still a few weeks out 21:14:36 <notmyname> ya. the next release is 1.4.6 21:14:51 <ttx> notmyname: let me know as soon as you have a target date 21:14:54 <notmyname> will do 21:14:58 <ttx> notmyname: Anything else ? 21:15:12 <notmyname> not unless there are questions 21:15:16 <ttx> Questions on Swift ? 21:15:40 <jaypipes> I did, but now I can't remember... doh. 21:15:51 <jaypipes> oh, that was it.. 21:16:01 * ttx waits 21:16:02 <notmyname> ask away 21:16:29 <jaypipes> notmyname: someone on #openstack this morning was asking about whether Swift has a stable/diablo-type branch that backported fixes are going to? 21:16:59 <notmyname> jaypipes: only major security issues (of which there have been none) 21:17:11 <notmyname> it hasn't happened, so it hasn't been done 21:17:18 <jaypipes> notmyname: ah, gotcha. 21:17:24 <ttx> #topic Glance status 21:17:34 <ttx> #link https://launchpad.net/glance/+milestone/essex-3 21:17:38 <jaypipes> notmyname: but in the case there is, y'all will create a stable/1.4.x or whatever branch? 21:17:53 <notmyname> jaypipes: I'm sure we'll figure something out 21:18:07 <jaypipes> notmyname: ok. 21:18:21 <ttx> jaypipes: a few questions on status... 21:18:24 <jaypipes> notmyname: just wanted to make sure I give correct information to people that were asking about it. 21:18:29 <notmyname> thanks 21:18:30 <ttx> * https://blueprints.launchpad.net/glance/+spec/refactor-internal-api (jaypipes) 21:18:39 <jaypipes> ttx: refresh your page. ;) 21:18:50 <ttx> jaypipes: ah! 21:19:15 <ttx> jaypipes: it's gone! 21:19:30 <jaypipes> ttx: all blueprints and bugs on the E3 status page are in proper order. The only remaining blueprint is one that bcwaldon is working on (adding policy support to Glance, same as Nova) 21:19:42 <bcwaldon> jaypipes: will be MP'd today 21:19:50 <ttx> bcwaldon: so on track for completion ? 21:19:56 <bcwaldon> yep 21:20:05 <jaypipes> ttx: we're in code review on the Swift-style multiprocess/multigreenthreadpool servers, but that should be good to go by E3 21:20:21 <jaypipes> ttx: and the remaining unassigned bugs will be assigned by EOD tomorrow 21:20:54 <ttx> as a sidenote, Essex was a good success in containing the number of new features. Now if we can convert that into buggfixing frenzy... 21:21:00 <jaypipes> ttx: a couple of the Low priority bugs may get pushed out to E4 or so, but the 1 High and 4 Mediums should be done by E3 21:21:24 <ttx> jaypipes: OK 21:21:37 <ttx> jaypipes: Anything else ? 21:21:44 <jaypipes> ttx: I've been kind of splitting my time recently between Glance and Tempest (about 40/60%), so sorry about the lag in status updates on Glance. 21:22:02 <ttx> you prefer to bust me with last minute updates 21:22:05 <mtaylor> jaypipes: just commented on the glance config devstack bug ... does applying the nova fix seem workable to you? 21:22:21 <jaypipes> ttx: :) 21:22:34 <jaypipes> mtaylor: will take offline 21:22:34 <ttx> Other questions on Glance ? 21:23:28 <ttx> #topic Nova status 21:23:32 <ttx> vishy: o/ 21:23:41 <vishy> hi 21:23:42 <ttx> #link https://launchpad.net/nova/+milestone/essex-3 21:23:51 <ttx> Looks like we'll have a few high-prio misses: 21:23:59 <ttx> * https://blueprints.launchpad.net/nova/+spec/separate-nova-adminapi (bcwaldon) 21:24:07 <ttx> (unless it's unblocked ?) 21:24:17 <vishy> it is unblocked 21:24:20 <bcwaldon> ttx: unblocked this morning, we've got a clear path forward 21:24:23 <bcwaldon> ttx: it will land this week 21:24:31 <vishy> we're going to do it by adding policy checks to the extensions 21:24:49 <ttx> marking good progress 21:25:02 <ttx> * https://blueprints.launchpad.net/nova/+spec/volume-type-scheduler (vladimir3p) 21:25:30 <ttx> vishy: we'll miss this one, right ? 21:25:47 <ttx> What is plan B for this one ? E4 or F ? 21:25:55 <vishy> yes that one will be missed 21:26:04 <vishy> i guess vladimir is too busy 21:26:10 <vishy> I think we will have to do without it 21:26:16 <ttx> so F it is ? 21:26:22 <vishy> looks like F to me 21:26:33 <vishy> I'm fiddling with scheduler stuff this week for a couple other blueprints 21:26:40 <ttx> ok, we'll defer next week if nothing landed 21:26:44 <vishy> so I will try and see if it can be done externally 21:26:54 <ttx> Also going slowly and in jeopardy (medium-prio): 21:26:55 <vishy> I think there is a way to do it with an external scheduler 21:27:01 <ttx> * https://blueprints.launchpad.net/nova/+spec/openstack-api-ssl (HP) 21:27:06 <ttx> * https://blueprints.launchpad.net/nova/+spec/scaling-zones (comstud) 21:27:15 <ttx> unless someone has good news. 21:27:23 <vishy> a) is in jeopardy, I'm not sure if they will get to it but I think that is e4 no problem 21:27:33 <vishy> b) comstud assures me it will be complete 21:27:41 <vishy> the majority of the work is in a patch under review 21:27:47 <vishy> which is not linked properly 21:28:22 <ttx> ok. All in all it sounds like a fair feature freeze waiting for us next week. 21:28:56 <ttx> vishy: we'll try to find a way to get popular bugs some attention, starting next week and throughout E4 21:29:01 <vishy> yes 21:29:12 <vishy> I think the bugsquash day will be a good time to kick that off 21:29:17 <ttx> I want people to know where they can help and make a difference 21:29:27 <jog0> any plans to fix up the numerous HACKING violations? 21:29:33 <ttx> the bugsquash day will definitely help in getting legacy stuff out of the way 21:29:47 <Vek> jog0: wasn't someone working on a tool to help us find them? 21:30:08 <ttx> vishy: would be also good to push nova-core to get a bit more anal retentive as we make progress towards Essex 21:30:09 <Vek> that would probably be the first step toward fixing those up... 21:30:12 <jog0> Vek: Yes I am, but I can use some help fixing the problems 21:30:19 <ewindisch> The ZeroMQ blueprint is actually making considerable progress and should be ready for E3. I've been emailing Duncan on it, but he hasn't been responding for nearly a month - Duncan owns the blueprint, officially. 21:30:21 <vishy> yessir 21:30:39 <vishy> ewindisch: if you can get a proposal in we will look at it 21:30:41 <ttx> ewindisch: the deadline is Tuesday next week 21:30:52 <vishy> but I'm not sure it is a huge deal, you could easily ship the driver separately if need be 21:31:19 <ttx> vishy: Anything else ? 21:31:24 <ewindisch> vishy, true, but I'd like to get it included if possible. 21:31:36 <russellb> same with the qpid equivalent :-) 21:31:38 <vishy> ewindisch: sure 21:31:44 <russellb> but i think that one is close 21:32:27 <ttx> Nova subteam leads: anything on your side ? 21:33:03 <ttx> Questions on Nova ? 21:33:28 <ttx> #topic Horizon status 21:33:35 <ttx> devcamcar: around ? 21:33:42 <ohnoimdead> <- filling in for devcamcar (currently stuck in his car on snowy seattle roads) 21:34:00 <ttx> ohnoimdead: welcome :) 21:34:04 <ohnoimdead> :D 21:34:05 <ttx> #link https://launchpad.net/horizon/+milestone/essex-3 21:34:11 <ttx> So... ext-roles was deferred. Looks like you're preparing yourself for a busy E4 :) 21:34:30 <ttx> Prioritization will be key. 21:34:40 <ohnoimdead> yeah. we punted a few things to e4. still a nasty bunch of bugs for e3 too. 21:34:58 <ttx> ohnoimdead: from what I'm seeing you have a good handle on them 21:35:14 <ttx> ohnoimdead: so I have no question. Anything you wanetd to mention ? 21:35:45 <ohnoimdead> nope 21:35:53 <ttx> Questions for Horizon ? 21:35:55 <ohnoimdead> (not that i can think of anyway) 21:36:28 <ttx> #topic Incubated projects and other Team reports 21:36:37 <ttx> danwent: o/ 21:36:41 <danwent> hello 21:36:44 <danwent> https://launchpad.net/quantum/+milestone/essex-3 21:36:44 <ttx> troytoman is stuck in some training 21:36:51 <danwent> we're a bit behind 21:36:53 <wwkeyboard> 7 21:37:01 <danwent> though a lot of the code for review just landed. 21:37:21 <danwent> in terms of nova code, I think the floating IP stuff should make it (enable floating IPs for quantum manager) 21:37:32 <danwent> the VPN stuff may not (debo has been fighting with devstack today) 21:37:39 <ttx> danwent: looks like a busy week ahead 21:37:43 <danwent> sigh.... 21:37:48 <danwent> indeed 21:37:59 <danwent> we'll bump one or two things out 21:38:06 <ttx> danwent: we shall branch milestone-proposed at the end of Tuesday (i.e. Wednesday morning European time) 21:38:14 <ttx> danwent: would that work for you ? 21:38:31 <danwent> yup. thanks again for getting quantum into the standard release processing 21:38:40 <ttx> Quantum and Melange now have working -tarball jobs 21:38:43 <danwent> i think all issues monty + james have had are working now 21:38:46 <danwent> yup 21:38:48 <ttx> so everything should be ready for getting E3 released under release team management 21:39:15 <danwent> we're also going to push to get something whipped up with tempest… that's all for quantum. 21:39:18 <ttx> Any other team lead with a status report ? annegentle (docs), mtaylor (CI), jaypipes (QA) ? 21:39:25 <annegentle> 0/ 21:39:32 <ttx> annegentle: floor is yours 21:39:34 <annegentle> or is it o/ ? anyway 21:39:43 <mtaylor> ttx: o/ after anne 21:39:46 <annegentle> I sent the note to the mailing list about a "diablo stable" install doc 21:39:48 <vishy> depends if you are a cyclops... 21:40:01 <jaypipes> ttx: I'll be sending out a status report on Tempest tomorrow. 21:40:18 <annegentle> sounds like the community is on board with that approach for now 21:40:37 <annegentle> but I'm happy to entertain other ideas 21:41:02 <ttx> annegentle: that doc points to cloudbuilders packages ? 21:41:07 <annegentle> we're also making progress on building an API reference site (though we're tracking down api.openstack.org, if you registered it please let me know) 21:41:25 <annegentle> ttx: it points to Cloud Builders packages and ones from Managed IT (Kiall) 21:41:40 <ttx> annegentle: any reason why we don't use Oneiric official Ubuntu ones ? 21:41:56 <russellb> stable/diablo is also packaged in Fedora 21:42:17 <annegentle> ttx: the OpenStack Starter Guide does (point to Ubuntu official). This is an additional installation guide that gets Keystone and Dashboard on Ubuntu. 21:42:34 <ttx> annegentle: ok 21:42:40 <annegentle> russellb: this guide is built so that additional distros instructions can be added easily. 21:42:51 <ttx> annegentle: anything else ?* 21:43:00 <annegentle> It's really a starting point to build from - a first step in separating install from admin also, which will take more work. 21:43:08 <russellb> annegentle: cool. we don't have dashboard packaged yet. we'll look at helping with the docs though. 21:43:16 <annegentle> That's all from the land of docs. 21:43:20 <ttx> mtaylor: go ahead 21:43:22 <annegentle> russellb: much appreciated! 21:43:23 <mtaylor> I'm planning on having multi-python testing rolled out next week- we've had several issues with this recently. 21:43:46 <ttx> mtaylor: I'd like to have tarball/repo diff jobs as well, we keep getting hit by surpises in that area 21:43:48 <mtaylor> but that'll be a dev-visible change 21:43:55 <mtaylor> ttx: funny story... 21:44:08 <mtaylor> ttx: the new hotness for doing the multi-version testing gets you that as well 21:44:25 <mtaylor> as it makes an sdist and then installs _that_ into the virtualenv, and then does the testing 21:44:25 <ttx> mtaylor: I need to see that with my own eyes 21:44:28 <Vek> heh. 21:44:30 <dragondm> re mult-python testing: ...and there was much rejoicing. 21:44:38 <mtaylor> so we actually start testing the tarball, rather than the repo 21:44:40 <Vek> +1 21:44:47 <soren> mtaylor: Sadly, many things missing from tarball won't cause test failures. 21:44:50 <ttx> mtaylor: you'll still miss the "missing binaries" issues, I guess 21:44:58 <soren> mtaylor: like that ^ 21:45:21 <ttx> I'm pretty sure tests pass if bin/nova-api-ec2 is missing for example. 21:45:22 <mtaylor> soren, ttx: those should be filed as missing test cases or something 21:45:24 <soren> mtaylor: But yes, many things will be caught by that. 21:45:35 <mtaylor> it's a start at least :) 21:45:38 <ttx> ok 21:45:43 <ttx> mtaylor: anything else ? 21:45:48 <mtaylor> but yes - I agree with both of you 21:46:03 <soren> mtaylor: Not everything will have test cases like that. 21:46:11 <mtaylor> nope, that's it - I just wanted to bring that up because it'll be a slight change that people will notice 21:46:14 <ttx> From release management land, markmc has worked on 2011.3.1 tarballs for Nova and Glance 21:46:17 <soren> mtaylor: We have no unit tests for the existence of docs, for instance. 21:46:43 <ttx> #help Would be good to get those 2011.3.1 candidates out for a date 21:47:03 <Vek> the one downside of the multipython testing will probably be that the unit tests will take twice as long to run once a mp is approved... 21:47:03 <mtaylor> soren: ++ 21:47:19 <mtaylor> Vek: not necessarily - we can parallelize that 21:47:29 <ttx> and without further ado... 21:47:33 <ttx> #topic Open discussion 21:47:36 <Vek> good :) 21:47:40 <ttx> #info So the winner is... Folsom 21:48:10 <danwent> yes! 21:48:21 <ttx> by an overwhelming 50 votes, Freedom was second with 35 votes, Fortuna had 33 21:48:41 <ttx> In related news, the F design summit location and date was announced: 21:48:49 <ttx> #info F design summit: April 16-18, San Francisco, CA 21:48:56 <ttx> #link http://www.openstack.org/conference/san-francisco-2012/ 21:49:08 <ttx> It will be followed by an OpenStack Conference, same place, April 19-20. 21:49:22 <ttx> Anything else, anyone ? 21:49:42 <ttx> Bug squashing day is still on Groundhog day, Feb 2 21:49:46 <annegentle> Bug Squash day links: 21:49:58 <annegentle> #link San Fran http://www.meetup.com/openstack/events/48362422/ 21:50:09 <annegentle> #link Austin http://www.meetup.com/OpenStack-Austin/events/48406252/ 21:50:18 <ttx> Is anyone interested in tracking stats for the bug squashing day ? 21:50:27 <ttx> I think reed will be on a plane that day 21:51:21 <annegentle> ttx: with an already exiting tool or one to be made soon? 21:52:10 <ttx> annegentle: probably with custom made tools. 21:52:18 <ttx> annegentle: Ubuntu used things like http://people.canonical.com/~brian/complete-graphs/nautilus/plots/nautilus-1day-triaging.png 21:52:23 <ttx> to track progress during the day 21:52:52 <ttx> I guess I could ask brian for his magic scripts 21:53:59 <ttx> #help Anyone interested in coordinating the Bug Squashing day should contact annegentle or ttx 21:54:14 <ttx> and if nobody has anything else... 21:54:33 <ttx> #endmeeting