16:01:49 <Sukhdev> #startmeeting networking_ml2
16:01:49 <openstack> Meeting started Wed Feb 25 16:01:49 2015 UTC and is due to finish in 60 minutes.  The chair is Sukhdev. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:50 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:01:53 <openstack> The meeting name has been set to 'networking_ml2'
16:02:16 <Sukhdev> #topic: Agenda
16:02:38 <Sukhdev> #link: https://wiki.openstack.org/wiki/Meetings/ML2
16:02:50 <Sukhdev> #topic: Announcements:
16:03:11 <Sukhdev> Feature freeze for Kilo is next week
16:03:41 <Sukhdev> If you are working on a new feature, keep it in mind
16:04:11 <Sukhdev> shivharis is absent - not feeling well
16:04:31 <Sukhdev> That is all I have
16:04:47 <Sukhdev> Anybody has any thing which team should be aware of?
16:05:37 <Sukhdev> #topic: ML2 Driver Decomposition Discussion
16:06:03 <Sukhdev> We have been having good discussion on this topic for past few weeks
16:06:12 <banix> Sukhdev: were you planning to post the info on a wiki?
16:06:16 <Sukhdev> How are your drivers coming along?
16:06:52 <Sukhdev> banix: I have been bad (actually swamped) - will try to do it this week
16:07:15 <Sukhdev> banix: but, I am available to help out -
16:07:24 <banix> Sukhdev: np. it would be most helpful now that you have gone through the hard work of figuring things out
16:07:27 <Sukhdev> feel free to ping me on irc
16:07:38 <banix> great thanks
16:08:27 <Sukhdev> There was a great discussion on this topic in Neutron Core meeting earlier this week - related to distros
16:09:49 <Sukhdev> The process is being documented - but, in general one has to have CI functional and the library uploaded at pypi to be included
16:10:26 <Sukhdev> armax is putting together a list of "ready" drivers
16:10:50 <Sukhdev> these drivers will become candidates for inclusion
16:11:25 <Sukhdev> Any body has any question or require any clarification on this?
16:11:42 <moshele> is it the open for review? (the process I mean)
16:12:15 <Sukhdev> moshele: I do not believe it is yet
16:12:25 <moshele> ok
16:12:29 <Sukhdev> It really is not a big process per se
16:13:08 <Sukhdev> As soon as your driver is ready and meets those two criteria you make sure to add the etherpad (or spreadsheet) that armax will create
16:13:43 <Sukhdev> that spreadsheet or etherpad will be used to track the ready drivers/plugins
16:14:29 <Sukhdev> One last thing I want to say on this topic is:
16:14:54 <Sukhdev> Please share your expeiences with the team - so that others can benefit
16:15:08 <Sukhdev> this process will continue through the Liberty cycle
16:15:40 <Sukhdev> #topic: ML2 Sync and Error Handling
16:15:58 <Sukhdev> manish does not seem to be here
16:16:18 <Sukhdev> I reviewed his patch - https://review.openstack.org/#/c/154333/
16:16:33 <Sukhdev> This is looking good
16:16:48 <Sukhdev> He is beefing it up - but, not there yet
16:17:15 <Sukhdev> Please do spend some time to become familiar with this - and provide your feedback/suggestions to him
16:17:41 <Sukhdev> Since he is not here to provide update, we can move on...
16:18:03 <Sukhdev> #topic: Potential ML2 impacting BPs
16:18:33 <Sukhdev> rkukura made a note that there are couple of BPs that are being worked on which could potentially impact ML2
16:19:01 <Sukhdev> So, we thought we bring it to your attention so that you are aware of it and you should review them
16:19:18 <Sukhdev> #link: https://blueprints.launchpad.net/neutron/+spec/nfv-vlan-trunks
16:19:36 <Sukhdev> #link: https://blueprints.launchpad.net/neutron/+spec/restructure-l2-agent
16:20:05 <Sukhdev> banix: second one is about L2 agent refactoring - you may want to look at it
16:20:40 <yamahata> Is that different from modular l2 agent?
16:20:56 <Sukhdev> yamahata: I believe so
16:21:03 <banix> Sukhdev: yamahata: yes different
16:21:10 <yamahata> okay thanks
16:21:43 <Sukhdev> rkukura want to say anything about the first one?
16:22:28 <rkukura> The two patches so far don’t impact ML2, but I’m assuming ML2, or at least some drivers, will need some changes to support trunking
16:23:42 <rkukura> I admit I haven’t taken time to re-read the spec, so details on what’s coming could be in there
16:24:43 <Sukhdev> I think it is worth for all of us to give it a read
16:24:53 <Sukhdev> hence, we brought it up here
16:24:53 <rkukura> Sukhdev: right - I plan to
16:25:13 <Sukhdev> anything else on this?
16:25:51 <Sukhdev> #topic: ML2 Extension Driver API enhancements
16:26:03 <Sukhdev> This one is in good shape -
16:26:27 <Sukhdev> I reviewed https://review.openstack.org/#/c/152759/ - this is ready to merge
16:26:39 <yamahata> +1
16:26:55 <Sukhdev> https://review.openstack.org/#/c/129178/ - is also moving along well
16:26:58 <rkukura> I just posted a comment on the 2nd patch suggesting to fix the consistency issue the opposite way.
16:27:25 <yamahata> rkukura: I'm fine with that.
16:27:42 <yamahata> any objections?
16:28:14 <yamahata> seems no. I'll update it today.
16:28:26 <rkukura> yamahata: thanks
16:28:38 <Sukhdev> yamahata: please do
16:28:59 <yamahata> and https://review.openstack.org/#/c/126552/
16:29:05 <yalie> for the another patchset https://review.openstack.org/#/c/126552/
16:29:08 <yalie> yes
16:29:30 <yalie> I am not sure if we need a functional test on it
16:30:16 <yalie> it's a patch to add port-security feature
16:30:55 <yalie> it will modify the iptables rules, Muller leave comments to add functional test
16:31:25 <yalie> to test the connectivy before and after set the port-sec attr
16:31:33 <Sukhdev> yalie: past experience suggests UTs should do - and you can propose a separate patch for tempest tests in addiition
16:31:34 <yalie> but I think that looks like a tempest case.
16:31:58 <rkukura> yalie: Since the extension itself is being integrated as an extension driver, I’m wondering if the changes you are making to plugin.py could be integrated as a mechanism driver?
16:32:06 <yalie> Sukhdev: thanks
16:32:55 <yalie> I think not
16:32:59 <yamahata> rkukura: the issue is the dependency on security group.
16:33:24 <yalie> yes ,and address-pairs
16:33:35 <yamahata> rkukura: to address it, security group needs to be converted to extension driver and the sg ED needs to be processed first.
16:33:59 <yamahata> So I suppose the conversion takes a while, so for L cycle.
16:34:09 <rkukura> yamahata: I see what you mean
16:34:27 <banix> modular agents!
16:34:32 <yamahata> We are willing to work on the conversion
16:34:36 <yamahata> for L cycle.
16:34:53 <rkukura> I’d suggest adding TODO comments for followup work where appropriate
16:35:06 <yalie> I see
16:35:13 <yamahata> Makes sense.
16:35:37 <yalie> rkukura: thanks
16:35:38 <rkukura> OK, thanks
16:35:52 <Sukhdev> on to the next topic -
16:36:17 <Sukhdev> BTW, L cycle now has a name - and it is Liberty
16:36:41 <Sukhdev> #topic: Bugs
16:36:53 <Sukhdev> Our bug czar is under weather -
16:37:06 <Sukhdev> any specific bug we want to discuss here?
16:37:19 <Sukhdev> #link: https://bugs.launchpad.net/neutron?field.searchtext=ml2
16:37:38 * Sukhdev waiting
16:38:00 <Sukhdev> #Open Discussion
16:38:13 <Sukhdev> Any thing to discuss?
16:38:42 <Sukhdev> If not, we are done -
16:39:03 <Sukhdev> it is good to keep it short and simple :-)
16:39:09 * Sukhdev waiting
16:39:13 <rkukura> +1
16:39:21 <rkukura> Thanks Sukhdev!
16:39:26 <banix> sounds good!
16:39:31 <Sukhdev> Thanks folks
16:39:36 <Sukhdev> see you next week
16:39:37 <Sukhdev> bye
16:39:40 <banix> bye
16:39:41 <Sukhdev> #endmeeting