16:02:44 <Sukhdev> #startmeeting networking_ml2
16:02:44 <openstack> Meeting started Wed Jul  2 16:02:44 2014 UTC and is due to finish in 60 minutes.  The chair is Sukhdev. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:02:46 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:02:47 <rcurran> vacation week for folks
16:02:48 <openstack> The meeting name has been set to 'networking_ml2'
16:02:51 <emagana> Sukhdev: Summer and Worldcup!
16:03:23 <Sukhdev> emagana: yes - I figured :-)
16:03:38 <Sukhdev> Lets get started - we'll keep it brief today
16:03:48 <Sukhdev> #topic: Agenda
16:04:10 <Sukhdev> #link https://wiki.openstack.org/wiki/Meetings/ML2#Meeting_July_2.2C_2014
16:04:19 <Sukhdev> #topic Announcements
16:04:41 <Sukhdev> #topic Nova/Neutron Parity Sprint next week in Minneapolis, MN
16:05:13 <Sukhdev> We have a sprint next week - I am sure I will see some of you there
16:06:01 <Sukhdev> I do not if you heard about the progress that DVR team has made
16:06:32 <Sukhdev> you can check out pretty cool demo at
16:06:40 <Sukhdev> #link https://www.youtube.com/watch?v=p4BwAjLHd0M
16:07:08 <Sukhdev> #topic: Update on ML2 sync meeting
16:07:26 <banix> Hi. Sorry for being late.
16:07:48 <Sukhdev> Bunch of ML2 team member met last Friday
16:07:59 <Sukhdev> banix: Hi - perfect timing
16:08:44 <Sukhdev> We covered Bulk operations and ML2 Sync in that meeting
16:09:21 <Sukhdev> banix, chuckC, Shiv and I participated in the discussion
16:09:41 <Sukhdev> https://bugs.launchpad.net/neutron/+bug/1193861
16:09:43 <uvirtbot> Launchpad bug 1193861 in neutron "ML2 plugin needs to override bulk operations" [Medium,In progress]
16:10:30 <Sukhdev> We have a long way to go to plan the ML2 sync activities - but, we all felt that it is the right way to go from the long term solution point of view
16:10:40 <Sukhdev> banix: want to add something?
16:11:40 <banix> Sukhdev: nothing but to say that we thought we would fix the bulk bug mentioned above independent of adding sync
16:12:20 <Sukhdev> correct - this will give us the short term solution towards the long term goal
16:12:28 <banix> The sync requires a more significant amount of time (discussion, and coding) and is needed for non bulk ops as well
16:12:33 <Sukhdev> Any body have any questions?
16:12:35 <banix> yes that’s all.
16:12:51 <Sukhdev> banix: correct
16:13:12 <Sukhdev> We will have follow on discussion on the topic
16:13:22 <Sukhdev> Moving right along
16:13:34 <Sukhdev> #topic: Action Items from last weel
16:13:42 <Sukhdev> s/weel/week
16:14:09 <Sukhdev> There were no action items, other than the meeting that took place last Friday - that I already mentioned
16:14:28 <Sukhdev> #topic: Bugs
16:15:07 <Sukhdev> Looks like our bug czar is missing - so, we'll skip it over - unless somebody wants to discuss any specific one
16:15:53 <Sukhdev> Ok Moving right along ---
16:16:11 <Sukhdev> #topic: Spec Reviewes
16:16:40 <Sukhdev> Lets get into some of the specs that are ready to merge and require some minor fixes
16:17:08 <Sukhdev> I presented three specs to the core team for core approvals -
16:17:17 <lukego> I added my existing spec for the “Snabb NFV” mech driver to the wiki page now btw: https://review.openstack.org/#/c/95707/
16:17:40 <Sukhdev> asomya banix: did you see the comments on your patches?
16:17:55 <banix> emagana: reviewed the specs (at least this one: #link https://review.openstack.org/#/c/88101/ ) as promissed. I made the requested changes
16:18:05 <asomya> yes i saw them this morning, i'll address the comments asap
16:18:39 <Sukhdev> asomya: I am trying to get this merged - need you to reply to the comments - so that we can wrap it up
16:18:57 <Sukhdev> emagana: Thanks a ton for jumping in to help
16:19:35 <Sukhdev> lukego: good - I will review it
16:19:46 <lukego> (Thanks!)
16:19:54 <asomya> Sukhdev: Sorry i didn't know it had comments on it
16:19:56 <Sukhdev> banix: you have some comments to address as well
16:19:58 <banix> lukego: will review
16:20:08 <banix> Sukhdev: i believe i did address them
16:20:11 <Sukhdev> asomya: it happens :-)
16:20:16 <banix> Sukhdev: uploaded a new version
16:20:33 <Sukhdev> banix: cool - I will make sure to follow up
16:20:41 <banix> yeah comments with no votes some times don’t get noticed
16:20:47 <banix> Sukhdev: thanks
16:21:12 <Sukhdev> Please be sure to update the wiki
16:22:12 <Sukhdev> Are there any other specs which the team (or the owner) believe it is ready?
16:22:38 <rcurran> Sukhdev, the cisco_nexus vxlan spec
16:23:00 <yamamoto> anyone investigated a way to update the wiki mechanically?
16:23:06 <rcurran> Sukhdev, and l3 service plugin - but i'll wait to discuss this at end of meeting since it's not technically ml2
16:23:09 <Sukhdev> rcurran: I looked at it Monday - it was looking good -
16:23:28 <yamamoto> "gerrit query" can show necessary info.  i'm not sure about wiki side.
16:23:49 <Sukhdev> rcurran: That is a bigger issue - I had a big debate with Mark on this one - lets wait until the end of the meeting
16:24:03 <rcurran> Sukhdev, also i have a cisco_nexus provider segment spec out for review - we need provider + l3 to get complete parity w/ the core cisco plugin
16:24:15 <banix> yamamoto: havn’t looked at it; it would be really good if it could be automated at some level at least
16:24:21 <rcurran> Sukhdev, yeah i saw the openstack-neutron log :-)
16:25:25 <Sukhdev> banix yamamoto: if you can figure out automated way - that will be fantastic - I am not savvy with this kind of stuff :-(
16:26:17 <banix> yamamoto: not sure if we can add extensions to the wiki. that could be one way.
16:26:53 <Sukhdev> rcurran: cisco vxlan gateway spec is ready to go? does it have approvals from this team?
16:27:30 <yamamoto> probably automate gerrit side only and use manual copy-and-paste for wiki side?
16:27:34 <Sukhdev> rcurran: I will look later today - if, yes, will put it on the list for approvals for next week
16:27:35 <rcurran> Sukhdev, it had one, needs another (asomya can review) - latest version needs to be reviewed again
16:27:57 <Sukhdev> rcurran: I will review it as well - others, please jump in
16:28:07 <asomya> rcurran: will review
16:28:31 <Sukhdev> Anything else on the specs? Any other one? that needs to brought up on the list?
16:28:45 <rcurran> yes, anyone interested in vxlan support on external devices would want to look at this and asomya's bp's
16:28:50 <banix> yamamoto: that would be reasonable step forward
16:29:07 <padkrish_> sukhdev# just would like to add VDP spec is ready to be merged https://review.openstack.org/#/c/89728/
16:29:25 <padkrish_> #sukhdev# just need to submit a new patch with minor comments addressed
16:29:38 <Sukhdev> padkrish_: good you are here - yours is on the list to ready to go
16:29:49 <yamamoto> i'll investigate gerrit side if/when i find spare time
16:30:05 <Sukhdev> padkrish_: Akihoro was asking for some clarifications - did you take care of it?
16:30:06 <padkrish_> sukhdev# thanks
16:30:07 <banix> yamamoto: great
16:30:35 <padkrish_> #sukhdev# yes i did.... need to expand on MSDC and privide some references in the new updated spec
16:31:06 <Sukhdev> padkrish_: good
16:31:36 <Sukhdev> emagana: can you look at this one as well? This is pretty much ready to go?
16:32:08 <Sukhdev> OK - anything else on the specs?
16:32:35 <Sukhdev> #topic: Code Reviews
16:33:05 <Sukhdev> #link: https://review.openstack.org/#/q/status:open+project:openstack/neutron,n,z
16:33:40 <Sukhdev> I browsed through some of the code - but not as much as I wanted to…
16:34:01 <Sukhdev> Please review these as well so that we can get them merged as well
16:34:42 <Sukhdev> #topic: Third Party Test System requirements
16:35:11 <Sukhdev> Folks, there is lot of discussion  on this topic on ML
16:35:31 <Sukhdev> The deadline is July 24 - we all are impacted by this
16:35:40 <Sukhdev> make sure to remain on top of this
16:36:10 <Sukhdev> did you guys see this http://stackalytics.com/report/ci/neutron/7
16:36:16 <Sukhdev> pretty cool..
16:36:43 <mxu> Hi Sukhdev: can you pls elaborate the 7/24 deadline?
16:37:25 <Sukhdev> mxu: It is tied to Juno-2, which is scheduled for 7/24
16:37:51 <mxu> Sukhdev:  Thanks!
16:38:09 <Sukhdev> Down to last topic -
16:38:20 <Sukhdev> #topic: Open Discussion
16:38:48 <Sukhdev> rcurran: so, you saw the discussion between Mark and me
16:38:59 <rcurran> Sukhdev, yes
16:39:38 <Sukhdev> It turns out there is an issue with partial implementation - your and mine BPs are impacted (along with others)
16:40:26 <Sukhdev> So, we have agreed to have a wider debate with the core team about handling of 501 codes
16:40:44 <rcurran> right, not sure what vendor plugins can do for events we can't handle
16:41:04 <Sukhdev> I wil be working with Kyle to get this issue on the agenda towards the later part of July
16:41:22 <Sukhdev> rcurran: correct -
16:41:43 <Sukhdev> rcurran: be sure to attend the debate - we need more vendor presence in that debate
16:42:12 <rcurran> yeah, i think we'll need to see how kyle feels - as i mentioned earlier - to get complete parity w/ the core cisco plugin (kyle wanted to start deprecated this plugin) we need this plugin
16:42:30 <Sukhdev> I am sure more and more vendors will be having these issues, as the scope of openstack deployments widen
16:43:15 <yamamoto> what issues are you talking about?
16:43:24 <yamamoto> pointer?
16:43:25 <rcurran> Sukhdev, so was the last interaction w/ you and mmclain/salvatore from that openstack-neutron log
16:43:44 <Sukhdev> rcurran:correct
16:44:02 <Sukhdev> yamamoto: dealing with partial implementation of L3 plugin
16:44:20 <Sukhdev> rcurran: do you have link handy?
16:44:32 <rcurran> http://eavesdrop.openstack.org/irclogs/%23openstack-neutron/%23openstack-neutron.2014-06-27.log
16:44:57 <rcurran> start here "2014-06-27T19:47:40  <markmcclain> Sukhdev: ping"
16:45:18 <Sukhdev> yamamoto: here is the link to the bp https://review.openstack.org/#/c/95910/
16:45:49 <yamamoto> thank you
16:46:15 <Sukhdev> rcurran: I will try to get on the agenda on 7/21 or 7/27 - keep an eye for it
16:46:20 <rcurran> basically we need some l3 support for the cisco devices; for the cisco core plugin we processed these events (add_router_interface as ex) right in the core plugin
16:46:28 <rcurran> we're not suppose to do this under ml2
16:46:34 <Sukhdev> s/27/28
16:46:58 <rcurran> so Sukhdev, asomya and i have created new services/l3_router plugins to handle these events
16:47:25 <Sukhdev> rcurran: any pointer?
16:47:36 <rcurran> on? my spec?
16:48:13 <Sukhdev> rcurran: oh I misunderstood - I thought you meant you have created a new spec
16:48:30 <rcurran> Sukhdev, well i do have a spec for my l3 service plugin - https://review.openstack.org/#/c/95910/
16:48:38 <Sukhdev> rcurran: yes both yours and asomya specs -
16:48:43 <banix> rcurran: that is fine; the issues is essentially partial implementation of a given extension
16:48:49 <banix> rcurran: right?
16:48:59 <rcurran> banix, correct
16:49:36 <Sukhdev> banix is the trouble maker - he started all this and it bubbled up :-):-)
16:49:38 <rcurran> Sukhdev, apic spec = https://review.openstack.org/#/c/96187/5
16:49:43 <Sukhdev> just kidding banix
16:50:21 <banix> Sukhdev: I felt bad after I saw how it got escelated
16:50:46 <Sukhdev> rcurran: I know he lucked out - went under the radar - but, same issue applies to him as well
16:51:09 <rcurran> Sukhdev, yes, seems that way
16:51:26 <Sukhdev> rcurran: hopefully, we will get a good clean resolution - Mark agrees with my view point and he stated to proceed with the implementation
16:51:52 <rcurran> Sukhdev, yeah i liked those comments; keep implementing ... but the code might not get in
16:51:57 <banix> Sukhdev: but this would have been noticed ny Mark at some point; so better if it is early on
16:51:58 <Sukhdev> banix: no worries - I was just  joking :-)
16:52:02 <rcurran> Sukhdev, note that i'm done w/ my code :-)
16:52:53 <Sukhdev> rcurran: well, I am no where close to finish line yet :-):-)
16:53:07 <Sukhdev> rcurran: we have time until Juno-3
16:53:11 <banix> it is the journey guys :)
16:53:22 <rcurran> Sukhdev, mine was easier to implement since it was already done under the core cisco plugin
16:53:24 <Sukhdev> banix: correct
16:53:57 <Sukhdev> Folks, anything else?
16:54:18 <Sukhdev> We can end early
16:54:27 <yamamoto> nothing from me
16:54:41 <Sukhdev> going once ----
16:54:46 <Sukhdev> going twice
16:54:49 <Sukhdev> Gone
16:54:50 <banix> by everybody
16:54:53 <banix> bye
16:54:54 <rcurran> bye
16:54:57 <Sukhdev> Have a great 4th of july
16:54:57 <yamamoto> good night
16:55:05 <chuckC> bye
16:55:05 <Sukhdev> #endmeeting