21:00:14 <danwent> #startmeeting quantum
21:00:14 <openstack> Meeting started Mon Dec 31 21:00:14 2012 UTC.  The chair is danwent. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:00:15 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:00:16 <arosen> hi
21:00:18 <openstack> The meeting name has been set to 'quantum'
21:00:21 <danwent> arosen: hello!
21:00:32 <danwent> greatings from 10,000 ft folks.
21:00:37 <garyk> hi
21:00:41 <alexpilotti> Hi guys
21:00:47 <danwent> i'm flying above western nebraska :)
21:01:02 <danwent> garyk, salv-orlando, both of you should DEFINIITELY not be here...
21:01:13 <garyk> danwent: :)
21:01:28 <danwent> ok, let's do a roll call. of who is here.  if we don't have many people, we can just addrss urgent business and call it a meeting
21:01:35 <zyluo> hi
21:01:42 <salv-orlando> great. So I'm not here :)
21:01:48 <Nachi> ha ha ha
21:01:53 <danwent> I did my best to put a lot of info in the agenda, so people could more or less read it and understand what's going on
21:02:18 <danwent> #info detailed info about meeting topics is at: http://wiki.openstack.org/Network/Meetings
21:02:30 <mlavalle> Hi
21:02:55 <danwent> ok, let's try to move quickly.
21:03:21 <danwent> #info thanks to nati_ueno and team, we now have a gate running on devstack as well as devstack + tempest
21:03:35 <nati_ueno> Thanks
21:03:55 <danwent> #info right now these gates are in trial mode, so failures to do prevent comments, so keep an eye on them.  If we don't see any false-positives, we will soon switch to actually gating on them.
21:04:21 <garyk> danwent: is this the reason why jenkins takes quite a long time to give his +1
21:04:23 <danwent> #info grizzly-2 branch point is one week from today.  So all features should be in code review with two core dev reviewers actively reviewing.
21:04:39 <danwent> garyk: yes, I suspect it takes longer with gate than when it just had to run unit tests.
21:04:51 <garyk> danwent: ok
21:05:19 <danwent> #info next stable release is planned for end of January.  Will be 2012.2.3
21:05:26 <danwent> garyk: anything to add on stable announcement?
21:05:40 <danwent> #info no openstack project meeting tomorrow due to holiday.
21:05:45 <danwent> any other announcements?
21:05:48 <garyk> danwent: no, just business as usual. decent progress
21:06:04 <danwent> Ok, i'm going to skip the docs section, as there's nothing urgent.
21:06:10 <danwent> please see agenda for details.
21:06:16 <danwent> #topic grizzly-2 release
21:06:24 <danwent> #link current status: https://launchpad.net/quantum/+milestone/grizzly-2
21:06:59 <danwent> we have a lot of stuff in review, so I'd like to focus on what we think we have a good chance of merging by next tuesday, considering the holidays this week.
21:07:21 <danwent> on gating, there's one more devstack review to get tempest tests passing: https://review.openstack.org/#/c/18665/
21:07:38 <danwent> we're waiting for input from tempest/devstack people.
21:07:51 <danwent> but i'm super pumped about the fact that gating is now up and running.
21:07:54 <danwent> nati_ueno: anything to add?
21:08:21 <danwent> #info db migration patch seems close to merge.  has several core devs on it.  https://blueprints.launchpad.net/quantum/+spec/quantum-db-upgrades
21:08:23 <nati_ueno> danwent: That's all for now. In January, I will fix tempest & tenant problems
21:09:09 <danwent> i'm not aware of any fundemental concerns with db-migration stuff.  if anyone is, please raise it now.
21:09:24 <danwent> otherwise, i expect the review to complete by next tuesday as scheduled
21:10:09 <danwent> nati_ueno: comments on security group branch? https://review.openstack.org/#/c/16210/
21:10:17 <danwent> seems like its back out of WIP
21:10:35 <danwent> other than gary's bug, are there major outstanding issues (i assume the additional tests were added?)
21:10:42 <nati_ueno> danwent: Yes I discussed with Akihiro, and I added unit test he mentioned
21:11:01 <garyk> danwent: nati_ueno : its looking a lot better. i had a few questions with last patch
21:11:21 <nati_ueno> danwent: garyk: Thanks I'll fix it in Today
21:12:01 <danwent> nati_ueno: ok, can you send a note to garyk, amotoki, and arosen, just to make sure we have two core devs with availability this week to review.
21:12:20 <nati_ueno> danwent: I have already send it
21:12:48 <danwent> #info we expect security groups review to merge this week, with garyk and either amotoki/arosen as core devs https://review.openstack.org/#/c/16210/
21:12:48 <nati_ueno> danwent: Ah but i should confirm availability of this week. THanks
21:13:25 <danwent> salv-orlando: service type stuff is back out of WIP https://review.openstack.org/#/c/17686/
21:13:35 <danwent> salv-orlando: i still owe you a review on this.
21:14:17 <danwent> salv-orlando: hopefully salv-orlando's vision is too blurry to see the screen right now :P
21:15:01 <danwent> #info service type should merge by next tuesday, but will need two core devs (garyk + danwent)? https://review.openstack.org/#/c/17686/
21:15:23 <danwent> #info lbaas API code merged since last meeting
21:15:30 <garyk> danwent: salv-orlando: yup i am in on reviewing this one
21:15:44 <danwent> focus is now on getting lbaas crud code merged. https://review.openstack.org/#/c/16919/
21:16:23 <danwent> anyone her who can comment on status of review?
21:16:47 <danwent> looks like the mirantis team has been working closely with the review, which is good.
21:17:14 <danwent> garyk (as always) has been helping too.  do we have another core dev active on this?
21:17:25 <danwent> ah, i see salv-orlando as well.
21:18:22 <danwent> #info likelihood of lbaas crud review for G-2 is unclear. https://review.openstack.org/#/c/16919/
21:18:31 <danwent> #todo #danwent follow-up with leon on status.
21:18:42 <danwent> i'm a bit worried that with leon being in china, they may be off the next couple of days.
21:19:20 <danwent> garyk: any comments you wanted to make on vif-plugging stuff?  only current patch is in nova, right?
21:19:39 <garyk> danwent: correct. still have to resolve the design issues :(
21:20:06 <garyk> danwent: hopefully in the coming week. saw you moved this to g3 which is cool with me
21:20:18 <danwent> yup, was just typing that.
21:20:36 <danwent> we shoudl also figure out if we need something filed in nova for this, as likely will be a non-trivial nova change.
21:20:46 <danwent> and I want to make sure its on vish's radar
21:20:54 <garyk> thanks
21:21:26 <danwent> #info vif-plugging improvements are still in design discussions.  nova team has some ideas on it as wel :) https://review.openstack.org/#/c/17747/
21:21:47 <danwent> garyk: would be good to bring tr3buchet into the discussion as well.  he may have a good viewpoint, as they are using xenserver
21:22:07 <garyk> danwent: good idea. i'll be in touch with him
21:22:18 <danwent> yong isn't here, but I just wanted to highlight that we're in design discussions around the multi-agent work he's been doing
21:22:48 <danwent> #info design spec for multi-dhcp/l3-agent work by yong: https://docs.google.com/document/d/1TJlW0_tMpeENA_ia38fvRu7ioKRt9fsWXBjivwd1mMw/edit
21:23:09 <danwent> #info multi-agent stuff will be g-3, but would be good to continue design discussion in G-2
21:23:23 <danwent> final item: pagination.  https://blueprints.launchpad.net/quantum/+spec/support-pagination-in-api-v2
21:23:42 <danwent> this is a monster patch, doesn't seem to have active core devs reviewing.  this is ok, as it is medium priority.
21:23:54 <danwent> just wanted to confirm this before i bump to G-3.  Any concerns?
21:24:47 <danwent> ok, any other G-2 features we need to discuss?  by now, hopefully there aren't any surprises :)
21:25:21 <danwent> remember, all code must be posted by tomorrow for any G-2 feature AND there should be two core devs committed to reviewing it this week.
21:25:33 <alexpilotti> There's the Hyper-V plugin :-)
21:25:50 <danwent> alexpilotti: i have a slot to discuss that at the end of the agenda
21:26:04 <alexpilotti> danwent: Cool!
21:26:24 <danwent> #topic folsom stable
21:26:42 <danwent> garyk, anything to comment on specifically, or just to have people read the email you sent out?
21:26:56 <danwent> i copy/pasted from the email into the agenda, but we probably don't need to go through each in detail.
21:27:10 <garyk> danwent: no comments. i think i just sent the mail to quantum core. tomorrow i'll send a mail to the list
21:27:36 <danwent> the only general thing I want to say is that garyk has been carrying quite a load leading the folsom/stable stuff, and we all need to do a better job of pitching in and helping him out.  So please look for bugs + backports to pick off in emails that garyk sends out.
21:27:39 <garyk> danwent: just importnat for people to give karma to the reviews in stable
21:28:07 <danwent> #topic tempest and quantum
21:28:28 <danwent> mnewby is not here, but others are
21:28:53 <danwent> can someone point out the active blueprints for tempest in quantum now that maru's base patch is in?  I want to make sure people are able to coordinate efforts.
21:29:17 <nati_ueno> I'm working on tempest also.
21:29:20 <mlavalle> danwent: I have submitted to BP's for approval
21:29:31 <danwent> https://blueprints.launchpad.net/quantum/+spec/tempest-tests-api
21:29:32 <danwent> https://blueprints.launchpad.net/quantum/+spec/extended-tempest-tests-api
21:29:36 <danwent> mlavalle: yes.
21:29:47 <zyluo> https://blueprints.launchpad.net/tempest/+spec/add-basic-quantum-system-tests
21:29:49 <nati_ueno> danwent:  https://blueprints.launchpad.net/tempest/+spec/quantum-tempest this one
21:29:50 <mlavalle> danwent: Yes, those two
21:29:57 <danwent> I have set maru to be approver. as i mentioned, these BPs should probably be filed under tempest
21:30:04 <danwent> does anyone here attend the tempest team meetings?
21:30:17 <mlavalle> danwent: I can attend from now on
21:30:26 <danwent> I'm fine having a quantum meeting slot for this as well, but we should at least be linked to the main project
21:30:30 <danwent> mlavalle: great, thanks.
21:30:39 <mlavalle> and report back to this meeting on Mondays
21:30:50 <nati_ueno> mlavalle: Thanks!
21:31:05 <danwent> nati_ueno, zyluo, mlavalle are you already in sync on who is doing what, or would some openstack-dev discussion time be useful?
21:31:30 <mlavalle> danwent: I think some discussion will be helpful
21:31:30 <nati_ueno> danwent: I agree.
21:31:34 <zyluo> Yeah we need to sync up I think
21:31:54 <danwent> ok, email or openstack-dev, either if fine with me.  just want to make sure we're coordinated.
21:32:23 <mlavalle> danwent; I'll start an email thread on this
21:32:33 <danwent> mlavalle: thx.
21:32:58 <danwent> #todo mlavalle to start email thread to coordinate quantum tempest next steps now that main patch has merged
21:33:17 <mlavalle> and copy nati_ueno and zyluo
21:33:24 <mlavalle> and maru
21:33:30 <nati_ueno> mlavalle +1
21:33:33 <danwent> actually, i'd just copy the ML
21:33:42 <danwent> as i believe there are others interested as well.
21:33:48 <mlavalle> danwent: ok, will do
21:34:03 <mlavalle> all the ML
21:34:04 <danwent> please put [Quantum] in subject, otherwise many of us will miss it :P
21:34:17 <danwent> #topic quantum + horizon
21:34:17 <mlavalle> danwent: will do
21:34:22 <danwent> amotoki is not here, but nati_ueno is
21:34:37 <danwent> nati_ueno: i noticed that several of the G-2 horizon blueprints are only in 'started' status
21:34:49 <danwent> is status out of date, or are these likely to slip from G-2?
21:35:01 <nati_ueno> I'll push the review of l3 stuff by 1/4.
21:35:11 <danwent> if slip, we probably need to try and recruit more resources, as i know there was a fair amount of work already slotted for G-3
21:35:26 <nati_ueno> but if the situation is not good, I'll slip the review
21:35:31 <danwent> nati_ueno: ok, please ping amotoki and gabriel on this, so the reviews do not come as a surprise
21:35:40 <nati_ueno> danwent: sure
21:35:50 <danwent> nati_ueno: are the other two issues from amotoki?
21:35:57 <nati_ueno> yes
21:36:04 <danwent> #todo danwent check with amotoki about his G-2 horizon + quantum issues
21:36:14 <danwent> #topic open discussion
21:36:18 <danwent> alexpilotti: ping
21:36:31 <alexpilotti> dansmith: Tx
21:36:42 <alexpilotti> danwent: Ops :-)
21:36:59 <alexpilotti> I'm writing on the iPhone and it's 23.30 here
21:37:21 <alexpilotti> Not my best sober status :-)
21:37:22 <garyk> alexpilotti: it is easier to drink beer at this hour
21:37:26 <danwent> #info alexpilotti has been working on a hyper-v plugin.  they are looking for a core dev to work with them as the official maintainer
21:37:47 <alexpilotti> Yep, we have the VLAN implementation ready
21:38:01 <alexpilotti> And NVGRE in the works
21:38:11 <danwent> they are going to keep the initial version pretty simple to make it easier to maintain, as they get up to speed on contributing to the committy
21:38:12 <danwent> with the goal of eventually having their own core dev to maintain it.
21:38:25 <danwent> alexpilotti: feel free to add anything else.
21:38:32 <alexpilotti> The plugin side is largely based on the OVS one
21:38:47 <danwent> alexpilotti will be sending out link to code in a public repo so people can take a look.
21:38:51 <alexpilotti> While the agent is of course very platform dependent
21:39:02 <danwent> alexpilotti: it will also be important that people have a way to test.
21:39:09 <alexpilotti> Sure
21:39:10 <garyk> danwent: alexpilotti: i seem to be missing something. why not just have hyperv support for all plugins (kind of like a hyp[re v directory)
21:39:15 <danwent> alexpilotti: is there devstack support for hyper-v?
21:39:31 <alexpilotti> No
21:39:50 <danwent> alexpilotti: ok, that will make things trickier
21:39:54 <alexpilotti> We use devstack for the controller
21:40:10 <alexpilotti> And quantum-server runs there
21:40:17 <danwent> garyk: this is where something like a generic vlan plugin coudl come into play
21:40:18 <alexpilotti> With the HyperV plugin
21:40:28 <alexpilotti> Only the agent runs on HyperV
21:40:51 <garyk> ok, understood.
21:40:53 <danwent> garyk: and in the future, perhaps a plugin that supports nvgre.
21:40:58 <alexpilotti> Having quantum-server running on windows is also on the roadmap
21:41:10 <danwent> garyk: i think alexpilotti may also be working on getting OVS integrated into hyper-v, which would be very cool.
21:41:20 <alexpilotti> Correct
21:41:36 <danwent> this woudl actually let a lot of plugins work with hyper-v
21:41:42 <danwent> without significant changes
21:41:50 <danwent> ok, any other open discussion before we wrap it up?
21:41:51 <alexpilotti> Yes
21:42:07 <danwent> alexpilotti: was that yes to additional discussion?
21:42:08 <mlavalle> alexpilotti: Cin cin
21:42:15 <garyk> as we say in africa - happy happy happy!
21:42:22 <danwent> garyk: indeed
21:42:28 <danwent> have a very happy new year folks
21:42:30 <alexpilotti> Yes to OVS :-)
21:42:34 <danwent> here's to 2013, the year of Quantum :)
21:42:41 <danwent> #endmeeting