17:00:21 <tjones> #startmeeting vmwareapi
17:00:22 <openstack> Meeting started Wed Oct 29 17:00:21 2014 UTC and is due to finish in 60 minutes.  The chair is tjones. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:24 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:27 <openstack> The meeting name has been set to 'vmwareapi'
17:00:34 <browne> hi
17:00:36 <tjones> hi folks
17:00:41 <rgerganov> Bonjour
17:00:50 <tjones> Bonjour!!
17:01:08 <rgerganov> this is one of the five words that I know in french
17:01:12 <tjones> we could hold this meeting in french but i would only be able to say a few words
17:01:33 <tjones> parles vou angleas (pardon my spelling)
17:01:45 <tjones> my 1 sentence
17:01:56 <tjones> good thing arnaud is not here yet
17:02:00 <rgerganov> lol
17:02:06 <tjones> so only 3 of us today?
17:02:09 <thangp> o/
17:02:15 <rgerganov> gary is lurking
17:02:19 <rgerganov> I will ping him
17:02:31 <tflower> o/
17:02:52 <tjones> hi thangp and tflower.  will we see you in person next week?
17:02:52 <garyk> hi
17:02:54 <mateescu> o/
17:03:04 <thangp> yes
17:03:15 <tjones> hi mateescu too
17:03:16 <tflower> Hi, unfortunately no :(
17:03:25 <tjones> ok lets get started (bummer tflower)
17:03:28 <thangp> looking forward to some french toast :)
17:03:40 <tjones> and pomme frittes!
17:03:43 * mdbooth_ waves
17:03:46 <tjones> #topic BP
17:03:47 <tjones> https://review.openstack.org/#/q/status:open+project:openstack/nova-specs+message:vmware,n,z
17:03:56 <tflower> Dims will be there and let me know about all the croissants I'm missing
17:04:05 <tjones> so specs under review is in that link
17:04:28 <tjones> rado im really excited about the console/vnc changes you are proposing.
17:04:54 <rgerganov> tjones: thanks, I will try to reach agreement with the core team about the API
17:04:56 <tjones> will you take some time to speak with jogo and johnthetubaguy about their ideas?
17:04:58 <mdbooth_> That's in danger of scope creep
17:05:20 <johnthetubaguy> can I help?
17:05:22 <tjones> this is a really important improvement cause our vnc is completely bad
17:05:39 <rgerganov> johnthetubaguy: this is about https://review.openstack.org/#/c/127283/
17:05:47 <tjones> johnthetubaguy: i was hoping than rgerganov can connect with you about ^^
17:05:51 <tjones> next week
17:06:08 <rgerganov> yeah, we can discuss in person at the summit
17:06:24 <johnthetubaguy> cool
17:06:30 <johnthetubaguy> do you want an unconference slot?
17:06:43 <rgerganov> johnthetubaguy: not really sure what this is :)
17:07:05 <rgerganov> tjones: are we going to have a VMW slot at the design summit?
17:07:11 <johnthetubaguy> rgerganov: basically 10 mins to pitch your idea and discuss with everone
17:07:12 <rgerganov> like the last summit
17:07:22 <mdbooth_> Hopefully only needs 30 minutes max with the right people
17:07:45 <jogo> so I disagree about scope creep, I think that is the wrong concern
17:07:46 <johnthetubaguy> rgerganov: we ditched those, so unconference the bits you want people to talk about
17:07:48 <tjones> rgerganov: im hoping we do on friday. it's not part of the official schedule this time -
17:07:52 <jogo> I think these should be two blueprints though
17:08:12 <johnthetubaguy> jogo: yeah, its more "fix the api", extent the "api"
17:08:21 <jogo> johnthetubaguy: yup
17:08:24 <johnthetubaguy> bad api stuff is very expensive to fix
17:08:41 <jogo> and it just gets more expensive later
17:08:50 <johnthetubaguy> sorry if you get stuck with extra work, its just a case of it being the third thing, and we spot a pattern that needs fixing
17:09:08 <johnthetubaguy> its our bad for not spotting the previous addition
17:09:22 <tjones> so sounds like we need an unconference slot for this.  How do we go abuot getting one?
17:09:22 <jogo> yup
17:09:23 <johnthetubaguy> addition(s)
17:09:33 <johnthetubaguy> I just added you: https://etherpad.openstack.org/p/kilo-nova-summit-unconference
17:09:37 <tjones> thanks!
17:09:43 <rgerganov> ok, great
17:10:07 <tjones> ok lets move to the next one #link https://review.openstack.org/#/c/126866/
17:10:12 <johnthetubaguy> we might have to vote about what to cover
17:10:38 <tjones> kanagaraj is not here
17:11:08 <tjones> #link https://review.openstack.org/#/c/128691/
17:11:27 <tjones> tflower: did you rebase as jogo recommended?
17:11:32 <tflower> tjones, yes I rebased
17:11:51 <tflower> I don't actually see the ova support blueprint in the list
17:11:53 <tjones> thanks - guys we need more reviews on this - only jogo has reviewed it.
17:12:04 <tflower> and mine depends on it
17:12:22 <tjones> that has merged
17:12:26 <tjones> it is accepted
17:12:44 <tjones> these are the unaccepted specs
17:12:57 <tjones> can people here take some time this week to review this BP?
17:12:58 <tflower> ah, so the other one is already accepted
17:12:59 <tjones> spec
17:13:43 <tjones> the last one #link https://review.openstack.org/#/c/128697/ is blocked on a cinder spec
17:13:58 <mateescu> tjones: that's correct
17:14:15 <tjones> but also needs more eyes from this team
17:14:23 <mateescu> I got a few comments from Duncan about the minimum interface a cinder driver should implement
17:14:45 <mateescu> looks like VMware RDM does not support extending disks though, according to Vipin
17:15:16 <tjones> is there another way to do it?
17:15:56 <mateescu> tjones:by getting down to the underlying storage protocol (FC or iSCSi)
17:16:18 <mateescu> but my proposal was trying to avoid exactly that and use the VMware abstractions for RDM, regardless of th eprotocol
17:16:53 <tjones> ok so it looks like you have some work to do there before the nova spec is likely to land.  Just trying to prioritize people's time
17:17:06 <mateescu> I need to look up more and see if there's any other way to achieve this while still keeping it protocol agnostic
17:17:13 <tjones> anything else about *unapproved* specs?
17:17:46 <rgerganov> I have a blueprint for implementing console logs
17:17:54 <rgerganov> but I think that it doesn't need a spec
17:18:06 <tjones> oh - i am confused about what does and does not need a spec
17:18:19 <rgerganov> there was a wiki page about this
17:18:20 <rgerganov> sec.
17:18:56 <tjones> i'll educate myself after this meeting
17:19:02 <rgerganov> http://docs.openstack.org/developer/nova/devref/kilo.blueprints.html#when-is-a-blueprint-needed
17:19:07 <tjones> lets move on to blueprints
17:19:13 <tjones> #topic blueprints
17:19:21 <tjones> https://blueprints.launchpad.net/openstack?searchtext=vmware
17:19:39 <mdbooth_> tjones: The advice I heard recently is that nothing needs a spec. However, things without a spec may be rejected on the basis of their design. A spec allows you to get approval on the design before writing code.
17:20:05 <tjones> rgerganov: why don't you go 1st.  you have some legacy work and some new work.  can you update us and do you need any help?
17:20:13 <garyk> the line between a spec and a bp is very fin. i really do not understand it
17:20:57 <rgerganov> tjones: sure. I am currently resuming the vSAN work that Vui has started
17:21:22 <rgerganov> there are some issues wrt image transfer which were raised by jaypipes
17:21:38 <rgerganov> I am currently working to address them
17:22:07 <tjones> and also #link https://blueprints.launchpad.net/nova/+spec/vmware-console-log
17:22:15 <rgerganov> right
17:22:25 <rgerganov> so in Juno I have tried to implement this with serial ports
17:22:28 <rgerganov> with file backing
17:22:46 <rgerganov> and it was not accepted because there is no way to limit the file size which could lead to security issues
17:23:10 <tjones> funny - there is a VERY old bug about other drivers that do this
17:23:11 <rgerganov> now I am proposing to implement this using serial ports with network backing
17:23:45 <rgerganov> tjones: correct
17:23:58 <tjones> looks like you john g is asking for more into in the BP
17:24:20 <rgerganov> yes, I will update it shortly
17:24:59 <rgerganov> the down side is that this will require one more process for handling the network traffic
17:25:06 <tjones> this will not make k-1 in this state - it's status is not started, etc.  will you be able to do that before the summit?  would be good to get it closer and we can perhaps talk about it with core folks
17:25:26 <rgerganov> ok, will do
17:25:33 <tjones> anything else rgerganov?
17:25:36 <rgerganov> nope
17:25:47 <tjones> ok garyk how about your stuff?
17:26:01 <garyk> it is all in review at the moment
17:26:18 <tjones> great!
17:26:31 <garyk> i still need to do some work on resize of the ephemeral disks - need to see how it mixes with block devies
17:26:41 <garyk> hopefully that will be done tomorrow or by end of the week
17:26:44 <tjones> anyone else have an BP for discussion?
17:26:49 <garyk> which is one and the same here :)
17:27:00 <tjones> :-D
17:27:22 <garyk> we need to get jogo to see if you saw comments that rgerganov added to the cpu allocations patch
17:27:31 <garyk> i will try and catch him tomorrow
17:27:39 <garyk> that is blocking the spbm bp
17:27:48 <tjones> do you have a link handy?
17:28:19 <garyk> sec
17:29:20 <garyk> https://review.openstack.org/#/c/128931/
17:30:10 <tjones> looks close
17:30:22 <garyk> yeah, just need jogo
17:30:43 <tjones> ok looks like we are done with BP
17:30:59 <tjones> #topic open discussion
17:31:03 <tjones> anything else?
17:31:04 <mdbooth_> garyk: Well danpb already gave you a +2 on that. I came down in the middle. I didn't really feel like it was a substantive issue.
17:32:05 <tjones> end early?
17:32:19 <mdbooth_> Do we have plans for our own informal meetup at the summit?
17:32:34 <tjones> we have not discussed it - lets do it now :-)
17:32:38 <garyk> yeah, each person brings a $100 and we drink till it is finished
17:32:41 <tjones> where and when?
17:32:42 <mdbooth_> lol
17:32:44 <rgerganov> somewhere with beer
17:32:46 <tjones> LOL
17:33:04 <tjones> ok wed evening somewhere with beer?
17:33:13 <tjones> just tossing that out there
17:33:17 * mdbooth_ hasn't seen the evening schedule, yet
17:33:25 <mdbooth_> But potentially any evening other than Monday works
17:34:14 <mdbooth_> tjones: Do you mind being the central point of communication?
17:34:23 <tjones> mdbooth_: no that is fine
17:34:47 <tjones> why don't we put our email on https://etherpad.openstack.org/p/VMwareapi-kilo so i have an easy way to communicate?
17:34:50 * mdbooth_ is fine with a evening of beer :)
17:35:36 <mdbooth_> tjones: Do you want to add a section for it?
17:35:41 <tjones> just did
17:35:43 <tjones> at the end
17:36:00 <mdbooth_> Got it
17:36:28 <tjones> ok i think we are done
17:36:34 <mdbooth_> Ok
17:36:34 <tjones> *listens*
17:36:38 <garyk> cool. have a good one and travel safe
17:36:45 <tjones> you too
17:36:47 <tjones> c ja
17:36:50 <tjones> #endmeeting