16:03:08 <Sukhdev> #startmeeting networking_ml2
16:03:09 <openstack> Meeting started Wed Jan 14 16:03:08 2015 UTC and is due to finish in 60 minutes.  The chair is Sukhdev. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:03:10 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:03:12 <openstack> The meeting name has been set to 'networking_ml2'
16:03:41 <Sukhdev> #topic: Agenda
16:03:53 <Sukhdev> https://wiki.openstack.org/wiki/Meetings/ML2#Agenda
16:04:25 <Sukhdev> Folks agenda is light today and many members are not available - so, we will have a short meeting today
16:04:37 <Sukhdev> #topic: Announcements
16:05:03 <Sukhdev> Kilo is only 3 weeks away
16:05:23 <Sukhdev> if you have anything need to go into kilo make sure it gets tagged correctly
16:05:30 <shivharis> you mean k2
16:05:31 <Sukhdev> #link: https://launchpad.net/neutron/+milestone/kilo-2
16:05:41 <Sukhdev> shivharis: yes
16:06:46 <Sukhdev> Have noticed that most of the patches are failing jenkins
16:07:31 <Sukhdev> grenade tests are failing in most of the neutron patches - I have one patch stuck waiting to pass
16:07:48 <Sukhdev> Anybody has any announcement?
16:08:29 <Sukhdev> #topic: ML2 drivers decomposition
16:08:50 <shivharis> I am actually looking for some documentation on this, any pointers?
16:09:04 <Sukhdev> Last week we decided to keep this on the agenda
16:09:38 <Sukhdev> shivharis: there are no pointers or documents other than the spec
16:09:43 <shivharis> Sukhdev: Any progress on your end? Actually I will be happy with any sparse pointers as well
16:10:13 <shivharis> Sukhdev: Ok, will see what I can do with that
16:10:53 <Sukhdev> shivharis: I initiated the work and got pulled into something. Plan on getting back this week - you can see what I have on my WIP patch
16:11:01 <Sukhdev> my WIP patch is out there
16:11:12 <shivharis> link?
16:11:14 <banix> (sorry for being late.)
16:11:32 <Sukhdev> I have also created a stackforge project - have not populated it yet
16:11:43 * Sukhdev looking
16:12:10 <Sukhdev> https://review.openstack.org/#/c/140484/
16:12:27 <shivharis> Ah.. thanks will look at that
16:12:41 <shivharis> Also any pointers on how to create stackforge project?
16:13:10 <Sukhdev> mestery is bit more ahead of me on this
16:13:16 <banix> Sukhdev: sorry if this was mentioned earlier, where do you plan to keep the library on stackforge?
16:13:22 <mestery> Sukhdev: lol
16:13:58 <Sukhdev> hopefully, we will make progress this week
16:14:38 <Sukhdev> banix: not sure if I understood your question correctly
16:14:48 <Sukhdev> which library are you thinking about?
16:15:05 <banix> Sukhdev: arista library? or code that the plugin will talk to
16:15:39 <Sukhdev> banix: Oh that will go on the stackforge project
16:15:49 * Sukhdev looking for link
16:15:50 <banix> Sukhdev: ok thanks
16:16:15 <shivharis> banix: you meant mech driver or the library, please clarify
16:16:26 <Sukhdev> https://github.com/stackforge/networking-arista
16:16:46 <banix> shivharis: what Sukhdev mentioned above
16:16:59 <Sukhdev> So, all of the backend code will go here in this repo
16:17:33 <Sukhdev> I have set up the core team for the project which will review all the patches on this project - not the neutron core team
16:18:38 <Sukhdev> most of the code will move here and the core-facing APIs will remain in the tree - you will see it in my WIP patch
16:19:05 <Sukhdev> I am almost there - only the issue is unit tests
16:19:43 <Sukhdev> I am splitting them - some of them will remain in the tree, rest of them will go in stackforge project
16:20:16 <Sukhdev> this splitting part and making Gerritt happy is where I am -
16:21:00 <Sukhdev> mestery is ahead of me - he mentioned that he has made progress on the tests
16:21:30 <Sukhdev> As soon as either one of is ready, we will post the whole thing for the full review and you can see it
16:22:16 <Sukhdev> rkukura mentioned yesterday that he is looking at decomosing APIC driver
16:22:26 <Sukhdev> so, soon there will be lots of examples
16:22:43 <Sukhdev> Hope this helps
16:23:14 <Sukhdev> Anything else on this topic?
16:23:19 <moshele> sukhdev: regarding agent we need to keep the main in the tree right?
16:23:45 <Sukhdev> moshele: yes for now it is not impacted
16:24:07 <moshele> and also config are ok right?
16:24:10 <Sukhdev> moshele: unless it is vendor specific
16:24:26 <Sukhdev> moshele: Correct
16:24:35 <Sukhdev> I forgot to mention couple of things
16:24:45 <moshele> I remove all the vendor  specific  is just agent.run()
16:25:08 <Sukhdev> the config and DB remains in the tree - to deal with the migrations
16:25:29 <Sukhdev> moshele: that is fine
16:25:32 <moshele> ok thanks
16:26:14 <Sukhdev> So, it is left up to vendors as to how to deal with the action functions to access the DB.
16:27:03 <Sukhdev> Ideally, what remains in the tree is very small footprint
16:27:33 <Sukhdev> Anything else?
16:28:12 <Sukhdev> #topic: ML2 Sync and error handling
16:28:38 <Sukhdev> I put this topic back on Agenda - as we did not discuss it after Paris summit
16:28:57 <Sukhdev> manish was looking at the taskflow
16:29:14 <Sukhdev> He is not here today - was hoping to pick it up
16:29:58 <Sukhdev> perhaps we can postpone it to next week.
16:30:10 <shivharis> sounds fine
16:30:26 <Sukhdev> #topic: Blueprint consolidation
16:30:41 <Sukhdev> #link https://wiki.openstack.org/wiki/Tracking_ML2_Subgroup_Reviews#Under_Review
16:30:54 <Sukhdev> shivharis: you were working with manish on this one
16:31:02 <shivharis> we are done with the consolidation work, however we still have one item to take care of
16:31:18 <shivharis> that is the split up the vendor and infra stuff.
16:31:27 <shivharis> we will take care of that this week
16:31:54 <shivharis> other than that, i urge folks to verify the correctness of it from their perspective
16:32:09 <Sukhdev> everybody - please have a look and try to keep it up to date - it helps everybody
16:32:10 <shivharis> i is possibe we may have missed something.
16:32:30 <Sukhdev> shivharis: thanks
16:32:41 <Sukhdev> anything else on this?
16:32:42 <shivharis> so, we will make additional progess this week. thats all from me and manishg
16:33:13 <Sukhdev> OK then lets move to next topic
16:33:18 <Sukhdev> #topic: Bugs
16:33:37 <Sukhdev> shivharis: you want to go over them?
16:33:51 <shivharis> the bugs state is moving at a snails pace.
16:34:14 <shivharis> most bugs are assigned to individuals.
16:34:25 <Sukhdev> we have 3 weeks to K2
16:34:36 <shivharis> I am hoping to see some progress in the bugs and k2 is 3 weeks away
16:34:46 <shivharis> as Sukhdev mentioned.
16:35:21 <shivharis> I have left messages for folks in the bugs themselves. Please update these.
16:35:55 <shivharis> the 4 high priority bugs need to be addressed soon. I have also pings folks on those.
16:36:04 <Sukhdev> rkukura is not here today - he gave me update on his two bugs yesterda
16:36:28 <Sukhdev> he is planning on working on them this week
16:36:44 <banix> This one #link https://review.openstack.org/#/c/113999/  got very close to go through in Juno but didn’t get done before the cut; Now it is targeted for kilo-2. amotoki and others had reviewed it closely during the last cycle. it has a +2 now. anychance amotoki can have a second look if others also find it reasonable
16:37:19 <Sukhdev> amotoki: are you here?
16:37:46 <Sukhdev> banix: I think we should get this in
16:38:13 <banix> Sukhdev: yeah it has been reviewed for quite some time
16:38:31 <shivharis> how was the test setup for this done?
16:39:01 <banix> shivharis: unit tests?
16:39:09 <shivharis> i thought in Juno the testing of this got in the way
16:39:20 <shivharis> unit/functional test
16:39:44 <banix> unit tests were updated to cover the added code
16:40:01 <shivharis> was there any tempeststuff added to this?
16:40:41 <shivharis> s/to/for/
16:40:56 <banix> shivharis: we concluded (during these meetings) that no need for extra coverage in tempest
16:41:04 <Sukhdev> shivharis: I thought we had decided to create a separate patch to write addiitional tempest test if needed
16:41:27 <shivharis> banix, Sukhdev: ok, thanks.
16:41:29 <banix> shivharis: that was discussed during the meetings towards the end of juno
16:41:42 <shivharis> i might have missed it.
16:41:46 <shivharis> np
16:41:57 <banix> shivharis: it has been a while :)
16:42:12 <Sukhdev> banix: correct
16:42:47 <Sukhdev> banix: so, can you ping folks to get this moving again and get it into k2
16:42:53 <shivharis> ok, then it is a good idea to move this forward.
16:43:06 <banix> Sukhdev: shivharis: will do. thanks.
16:43:26 <Sukhdev> Anything else on the bugs?
16:43:34 * Sukhdev waiting
16:43:48 <shivharis> no thats all from my end
16:43:57 <Sukhdev> #topic: Open Discussion
16:44:28 <Sukhdev> floor is open for any discussion
16:44:44 <Sukhdev> How is weather outside :-)
16:44:57 <banix> Sukhdev: freezing
16:45:13 <Sukhdev> ha ha - it is sunny and bright here :-)
16:45:26 <banix> :)
16:46:03 <Sukhdev> OK folks - if there is nothing else, lets enjoy extra minutes
16:46:10 <Sukhdev> Thanks
16:46:14 <Sukhdev> we are done
16:46:21 <banix> thanks
16:46:21 <shivharis> thanks Sukhdev, bye all
16:46:24 <Sukhdev> #endmeeting