15:59:28 <primeministerp> #startmeeting hyper-v
15:59:29 <openstack> Meeting started Tue Jan 15 15:59:28 2013 UTC.  The chair is primeministerp. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:59:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:59:32 <openstack> The meeting name has been set to 'hyper_v'
15:59:44 <primeministerp> flaky webchat today
15:59:51 <primeministerp> alexpilotti: morning
16:00:01 <primeministerp> pnavarro: er afternoon
16:00:11 <primeministerp> lots to discuss today
16:00:14 <alexpilotti> morning!
16:00:22 <pnavarro> good morning primeministerp !
16:00:29 <primeministerp> how's everyone today
16:00:42 <alexpilotti> hangovered!
16:00:47 <primeministerp> alexpilotti: good
16:00:49 <alexpilotti> kidding ;-)
16:00:50 <primeministerp> alexpilotti: you deserve it
16:00:54 <primeministerp> ;)(
16:00:58 <primeministerp> ok
16:01:00 <alexpilotti> I still have to properly party for the release
16:01:03 <primeministerp> on that note
16:01:12 <primeministerp> #topic Hyper-V/Quantum
16:01:51 <alexpilotti> I guess that I start this one :-)
16:01:57 <ociuhandu> morning all
16:01:57 <primeministerp> alessandro's hard work paid off and hyper-v quantum bits were accepted in both the quantum project as well as nova
16:02:05 <primeministerp> ociuhandu: tavi!
16:02:40 <primeministerp> alexpilotti: feel free to continue
16:02:50 <alexpilotti> primeministerp: ok, was just waiting ;-)
16:03:10 <alexpilotti> so, fater one week of reviews
16:03:26 <alexpilotti> 99% of them just spaces, comments, syntax, etc
16:03:48 <alexpilotti> actually nothing really drastic changed from the proposed stuff
16:04:01 <alexpilotti> s/fater/after/ lol
16:04:16 <pnavarro> hehehe
16:04:16 <primeministerp> hehe
16:04:19 <alexpilotti> it was a very hard work, but it really paid off
16:04:26 <primeministerp> yes excellent
16:04:34 <pnavarro> yeah, congrats alexpilotti !
16:04:36 <primeministerp> major milestone achieved
16:04:54 <alexpilotti> gongysh, a core reviewer from IBM china helped a lot
16:05:00 <primeministerp> is he on
16:05:16 <alexpilotti> he's actually the first core dev to really run our stuff on Hyper-V!
16:05:17 <primeministerp> was hoping to see if he got his msdn acct
16:05:27 <primeministerp> that's great news
16:05:27 <alexpilotti> no, he's offline, I looked for him
16:05:43 <alexpilotti> it's 00.00 am in Beijing
16:05:47 <alexpilotti> so
16:06:04 <primeministerp> well offically if anyone else wants to run on hyper-v and needs msdn access, please contact me  directly
16:06:25 <primeministerp> and I'll do my best to get a subscription
16:06:48 <primeministerp> sorry
16:06:51 <primeministerp> alexpilotti: continue pls
16:06:58 <alexpilotti> he worked with me and I helped him on getting his way around on HyperV
16:07:21 <alexpilotti> it's very good that he did that, so we have somebody else that can test our code
16:07:41 <alexpilotti> another great news is that he tested it also connected to OVS
16:07:43 <primeministerp> second eyes are always key
16:07:52 <alexpilotti> including L3 tests
16:08:12 <alexpilotti> and everything works very well I have to say
16:08:29 <alexpilotti> thanks pnavarro for your help in the initial stage of the project! :-)
16:08:41 <alexpilotti> It was very helpful to finish this in time!
16:08:50 <primeministerp> haha no hard push like last time
16:09:05 <primeministerp> alexpilotti: ok to move on?
16:09:14 <primeministerp> alexpilotti: or do you want to get into any more detail
16:09:27 <alexpilotti> last line as usual: please tests and feedback! :-)
16:09:31 <pnavarro> it's a pleasure to work with you !
16:09:44 <primeministerp> ok
16:10:03 <primeministerp> #topic RDP status
16:10:09 <primeministerp> so moving on
16:10:18 <alexpilotti> yep
16:10:19 <primeministerp> is there a general status of the rdp-console-proxy
16:10:29 <primeministerp> how far off until testable bits
16:10:34 <alexpilotti> the FreeRDP-WebConnect gateway is taking shape
16:10:45 <alexpilotti> 3 weeks
16:10:49 <primeministerp> great
16:11:00 <alexpilotti> from a testable alfa
16:11:24 <alexpilotti> so the bad part is that the WebConnect gateway is not mantained anymore
16:11:35 <alexpilotti> so we updated it to support the latest FreeRDP code
16:11:58 <primeministerp> alexpilotti: were you able to push it upstream?
16:12:04 <alexpilotti> The FreeRDP itslef was broken, so I worked with Marc, the FreeRDP lead to fix it
16:12:28 <alexpilotti> We agreed with Marc to push it in the official code base
16:12:42 <alexpilotti> which is really cool, as we don't have to mantain a separate fork
16:12:47 <primeministerp> gotcha
16:12:53 <alexpilotti> which is what killed nova-novnc
16:12:58 <primeministerp> that works out well then
16:13:03 <alexpilotti> yes
16:13:22 <alexpilotti> so, now we are adding the support for Hyper-V console connections to the gateway
16:13:41 <primeministerp> so moving on
16:13:53 <alexpilotti> once that is done, we'll add the Keystone bits for auth
16:14:14 <alexpilotti> we have a dedicated developer on it, Marius, next time I'll ask him to join
16:14:15 <primeministerp> alexpilotti: i need to resurrect the customssp provider discussion
16:14:35 <primeministerp> alexpilotti: that would great to get him involved
16:14:37 <alexpilotti> yeah, for that one we definitely need support from MS
16:14:41 <primeministerp> yes
16:14:56 <primeministerp> alexpilotti: i'm fire in additional requests today
16:15:00 <primeministerp> er i'll
16:15:08 <alexpilotti> ah, BTW Marc (FreeRDP) is implementing a porting of CredSSP on Linux
16:15:18 <alexpilotti> that will also be used by ReactOS
16:15:53 <alexpilotti> we agreed to contribute on that project (CredSSP), with it as a base we will be able to have a starting point for the SSH SSP
16:16:03 <primeministerp> that's good
16:16:12 <alexpilotti> this is a mid term project anyway
16:16:28 <alexpilotti> Ah, one small extra:
16:16:47 <alexpilotti> 2 of our guys are woring on a .Net interface on the OPenstack RESTful APIs
16:16:56 <alexpilotti> as we need it for on of our projects
16:16:57 <primeministerp> alexpilotti: interesting
16:17:10 <alexpilotti> we'll release that as Apache code of course
16:17:27 <primeministerp> perfecto
16:17:41 <alexpilotti> we should speak with ttx to see if they fit somewhere in inclubation
16:17:52 <alexpilotti> lol "incubation"
16:17:56 <primeministerp> hehe
16:18:08 <alexpilotti> my fingers needs some debug :-)
16:18:11 <primeministerp> alexpilotti: maybe address that in an email thread?
16:18:21 <primeministerp> alexpilotti: start the discussion
16:18:32 <alexpilotti> good point
16:18:47 <alexpilotti> it's fresh news anyway, we started this week
16:18:58 <primeministerp> that's good to hear
16:19:12 <primeministerp> anything else on rdp?
16:19:15 <alexpilotti> I'll ask Adi and Gerry to join this meeting next time as well
16:19:22 <primeministerp> perfect
16:19:29 <alexpilotti> that's it for now
16:19:32 <primeministerp> ok
16:19:41 <primeministerp> #topic general updates
16:19:58 <primeministerp> Cirrius
16:20:01 <primeministerp> er
16:20:24 <alexpilotti> about typos and autocorrection:
16:20:37 <primeministerp> at least i have an excuse
16:20:47 <alexpilotti> google suggested for "CirrOS" the term "Cirrosis"
16:21:07 <pnavarro> yeah, cirrosis is not cool
16:21:10 <primeministerp> alexpilotti: thx
16:21:16 <primeministerp> lost my window
16:21:31 <primeministerp> so Linux integration services were added to CirrOS
16:21:38 <alexpilotti> let me paste the link
16:21:42 <primeministerp> perfect
16:21:52 <alexpilotti> to cut it short, I worked with smoser to make Cirros working
16:21:57 <alexpilotti> on Hyper-V
16:22:02 <primeministerp> great news
16:22:09 <alexpilotti> we added the LIS and it works perfectly
16:22:22 <primeministerp> alexpilotti: what's the base os on cirros
16:22:28 <alexpilotti> now we have to package it in a way that the LIS don't get loaded on other hypervisors
16:22:42 <alexpilotti> the kernel is the one from 12.04
16:22:48 <primeministerp> gotcha
16:22:54 <alexpilotti> #link https://bugs.launchpad.net/cirros/+bug/1099611
16:22:55 <smoser> (cirros is the base os for cirros. built from buildroot, with ubuntu kernel).
16:22:56 <uvirtbot> Launchpad bug 1099611 in cirros "Missing Hyper-V modules" [Undecided,New]
16:23:15 <alexpilotti> hi smoser :-)
16:23:18 <primeministerp> smoser: thx
16:23:27 <primeministerp> smoser: thanks for the help btw
16:23:33 <primeministerp> smoser: both w/ this and cloudinit bits
16:23:37 <primeministerp> smoser: greatly appreciated
16:23:39 <alexpilotti> primeministerp: smoser was super fast in packaging it for testing
16:24:14 <alexpilotti> smoser: I was updating the masses here with what we did :-)
16:24:49 <primeministerp> alexpilotti: shall we discuss resize?
16:24:56 <alexpilotti> sure
16:25:07 <primeministerp> alexpilotti: will be finished prior to G3
16:25:45 <primeministerp> alexpilotti: or is it going to pushed to h
16:25:49 <alexpilotti> "resize" will probably use CirrOS as a working VM for managing the actual resizes of Linux partitions (extx)
16:26:08 <alexpilotti> resize depends on the refactoring
16:26:14 <primeministerp> ok
16:26:21 <alexpilotti> I'll do anything possible to finish refactoring in time
16:26:39 <alexpilotti> is not IMO a mandatory feature for now, as not so many customers really asked for it
16:26:49 <primeministerp> alexpilotti: and after the refactor how much additional is needed to rework the original cloud.com code to be in the v2 wmi namespace?
16:26:50 <alexpilotti> but it's still an important feature in Nova
16:27:03 <primeministerp> danwent: btw thanks for your help as well
16:27:13 <primeministerp> danwent: greatly appreciated
16:27:14 <alexpilotti> primeministerp: not much, between 4 and 7 days
16:27:19 <primeministerp> alexpilotti: ok
16:27:41 <primeministerp> alexpilotti: still attainable for G?
16:27:49 <alexpilotti> V2?
16:27:53 <alexpilotti> very hard
16:27:58 <primeministerp> ok
16:28:00 <hanrahat> alexpilotti: yes thanks on the v2 wmi  ... good news
16:28:14 <primeministerp> hanrahat: huh?
16:28:21 <alexpilotti> hanrahat: hi Tom!
16:28:32 <primeministerp> hanrahat: we're discussing moving the original cloud.com code from v1 to v2
16:28:41 <primeministerp> hanrahat: it's not complete yet
16:28:50 <primeministerp> hanrahat: that's work to be done
16:29:03 <hanrahat> primeministerp: understood.  i was referring to the time estimeate
16:29:06 <hanrahat> estimate
16:29:11 <primeministerp> hanrahat: perfect
16:29:15 <alexpilotti> th eoriginal cloud.com code is not exactly a show of best coding principles, to be eufemistic
16:29:43 <primeministerp> alexpilotti: we should probabably talk about the renaming
16:29:44 <alexpilotti> so the sooner we clean it up and refactor it the better it will be
16:29:45 <primeministerp> as well
16:29:57 <alexpilotti> renaming?
16:30:09 <primeministerp> alexpilotti: the suggestion
16:30:13 <primeministerp> on renaming the hyper-v
16:30:22 <alexpilotti> ah yeah.
16:30:26 <primeministerp> in the refactor
16:30:34 <alexpilotti> should we change topic?
16:30:38 <primeministerp> yes
16:30:43 <alexpilotti> ok
16:30:50 <primeministerp> to what specifically
16:30:52 <primeministerp> ;)
16:31:01 <alexpilotti> russellb asked to move all the Hyper-V specific options
16:31:07 <primeministerp> #topic Hyper-V Specific Options
16:31:09 <alexpilotti> to a [HYPERV] group
16:31:21 <primeministerp> so
16:31:24 <alexpilotti> it will break compatibility with older versions of course
16:31:24 <primeministerp> I'm all for it
16:31:28 <primeministerp> well
16:31:35 <primeministerp> we're going to break compat once we move to v2
16:31:43 <alexpilotti> but the more we wait the worse it will be
16:31:50 <primeministerp> that's correct
16:31:55 <primeministerp> I'm in agreement
16:32:01 <alexpilotti> also the installer will take care of it for most customers
16:32:07 <primeministerp> well
16:32:17 <alexpilotti> and for manual deployments we will make sure to document it well
16:32:28 <primeministerp> considering we already got a backport request for the yet to be completed RDP code
16:32:40 <primeministerp> I would imagine we would also get a back port request
16:32:47 <primeministerp> for this
16:33:07 <primeministerp> cern is moving to folsom, not grizzly
16:33:08 <alexpilotti> hmm
16:33:12 <primeministerp> i believe
16:33:25 <primeministerp> could cause some issue for thier deployment
16:33:33 <primeministerp> no
16:33:35 <primeministerp> er now
16:33:36 <alexpilotti> our Folsom code contains 10% of the features we have
16:33:48 <russellb> thanks guys :)
16:33:59 <alexpilotti> If they are serious about it, we should backport the entire HyperV driver
16:33:59 <primeministerp> russellb: np
16:34:07 <primeministerp> russellb: thanks for your help
16:34:19 <alexpilotti> russellb: we didn't say yes yet! :-D
16:34:29 <primeministerp> so
16:34:36 <primeministerp> I'm in agreement that it should be renamed
16:34:37 <alexpilotti> russellb: kidding
16:34:46 <alexpilotti> +1
16:34:54 <ociuhandu> +1
16:34:55 <russellb> happy to help with ideas on backwards compatibility if needed
16:34:57 <russellb> that's up to you guys
16:34:58 <pnavarro> +1
16:34:59 <alexpilotti> pnavarro: ?
16:35:02 <russellb> easier code change if you just don't bother
16:35:04 <primeministerp> russellb: that would be great
16:35:17 <primeministerp> russellb: well that is true
16:35:24 <russellb> thing is ... we already have code to easily support the old option name ... but I don't think it supports the case of moving options to a new group
16:35:40 <russellb> wouldn't be too hard to add support for it i'm guessing ... but it's just another step that needs to be done
16:35:51 <russellb> would be a patch to cfg in oslo
16:36:01 <alexpilotti> russellb: I don't think it's worth for us now
16:36:12 <primeministerp> alexpilotti: let's push it to H
16:36:20 <alexpilotti> the rename?
16:36:25 <primeministerp> well
16:36:27 <alexpilotti> primeministerp: ^
16:36:35 * primeministerp thinks
16:36:41 <primeministerp> is it worth doing it now?
16:36:49 <russellb> why not now?
16:36:52 <primeministerp> maybe i didn't understand where you were going
16:36:56 <primeministerp> w/ your comment
16:36:57 <alexpilotti> pushing it for H means having a lot of customers having backwards compatibility issues
16:37:05 <primeministerp> alexpilotti: yes I agree
16:37:06 <primeministerp> ok
16:37:10 <primeministerp> we're on the same page
16:37:15 <primeministerp> tha'ts the only issue i have
16:37:33 <alexpilotti> ok, so option rename goes for G3, ok?
16:37:40 <russellb> it would be all the same option names, just in [HYPERV] instead of [DEFAULT]
16:37:55 <russellb> to tidy things up a bit :)
16:38:00 <primeministerp> ok
16:38:10 <russellb> in general, we want to move more and more things into groups as it makes sense
16:38:48 <alexpilotti> russellb: what do we do with the "cross hypervisor" ones?
16:38:58 <alexpilotti> like "force_config_drive"?
16:39:10 <russellb> if it's cross hypervisor ... leave in [DEFAULT] for now
16:39:10 <alexpilotti> which is also available for KVM?
16:39:25 <alexpilotti> russellb: ok
16:39:47 <russellb> just the stuff that you know is hyperv specific
16:39:59 <primeministerp> that makes more sense
16:40:20 <alexpilotti> russellb: ok tx, I'll ping you if I have doubts about specific options!
16:40:28 <russellb> sounds good
16:40:34 <primeministerp> russellb: thanks again
16:40:40 <russellb> sure, np
16:40:49 <primeministerp> #topic CI progress
16:41:03 <primeministerp> so tavi and i have been working on our puppet bits
16:41:29 <primeministerp> hopefully we'll have our first control ring up in a couple weeks
16:41:39 <primeministerp> maybe sooner
16:41:58 <primeministerp> mordred: ping
16:42:22 <primeministerp> so
16:42:41 <primeministerp> anyone have anything additional to add?
16:43:09 <pnavarro> yes
16:43:11 <primeministerp> ociuhandu: I think i know what the problem is w/ the puppet registration
16:43:14 <primeministerp> pnavarro: shoot
16:43:27 <ociuhandu> primeministerp: cool
16:43:33 <pnavarro> what are  next steps for quantum?
16:43:35 <ociuhandu> primeministerp: keystone is 99% done
16:43:48 <ociuhandu> and all "clients" are working fine
16:43:51 <primeministerp> ociuhandu: good
16:44:00 <ociuhandu> i expect the other ones to go faster now
16:44:19 <primeministerp> ociuhandu: i need to get moving on the other windows pieces as well
16:44:26 <ociuhandu> primeministerp: also a first "draft" for the compute node is working
16:44:37 <primeministerp> ociuhandu: i've been testing that this week
16:44:49 <primeministerp> ociuhandu: i'm sure we're missing something
16:45:04 <primeministerp> we'll find out shortly
16:45:27 <primeministerp> pnavarro: that's a good question
16:45:46 <primeministerp> pnavarro: there are bunch of different discussions going on
16:46:15 <alexpilotti> pnavarro: I think we should discuss it in the general discussion after the CI!
16:46:19 <pnavarro> CERN guys were interested in ceilometer project, do you know if they are coding something?
16:46:22 <primeministerp> alexpilotti: yes
16:46:42 <primeministerp> pnavarro: they mentioned interest to me a while back
16:46:50 <primeministerp> pnavarro: however I don't know if they are
16:47:02 <pnavarro> ok, sorry, I thought primeministerp was going to close the meeting
16:47:19 <primeministerp> pnavarro: I've always looked at cielometer as a H project
16:47:31 <pnavarro> got it !
16:47:36 <primeministerp> pnavarro: we can move the discussion
16:47:40 <primeministerp> ok
16:47:45 <primeministerp> I'll close
16:47:49 <primeministerp> the meeting
16:47:52 <primeministerp> thanks for the time everyone
16:47:56 <primeministerp> great work
16:48:02 <primeministerp> as usual alexpilotti
16:48:12 <primeministerp> ociuhandu: i'll be in touch
16:48:18 <primeministerp> #closemeeting
16:48:27 <alexpilotti> primeministerp: tx! and tx pnavarro as well!
16:48:31 <primeministerp> #endmeeting