21:01:15 <HenryG> #startmeeting networking
21:01:16 <openstack> Meeting started Mon Jan 25 21:01:15 2016 UTC and is due to finish in 60 minutes.  The chair is HenryG. Information about MeetBot at http://wiki.debian.org/MeetBot.
21:01:17 <mestery> hi!
21:01:17 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
21:01:18 <manjeets1> hi
21:01:20 <openstack> The meeting name has been set to 'networking'
21:01:26 <njohnston> \o/
21:01:28 <ZZelle__> Hello everyone
21:01:29 <HenryG> #chair mestery
21:01:30 <openstack> Current chairs: HenryG mestery
21:01:46 * mestery straps into the shotgun seat of HenryG's meeting
21:01:55 <HenryG> armax is traveling, so I will run the meeting and mestery will help
21:01:59 <Sam-I-Am> did you bring a shotgun?
21:02:05 <carl_baldwin> o/
21:02:11 <ihrachys> o/
21:02:17 <mhickey> o/
21:02:17 <HenryG> Some objections were voiced about the change to Mondays only for this meeting.
21:02:24 <HenryG> But there has been no response to this invitation:
21:02:31 <HenryG> #link http://lists.openstack.org/pipermail/openstack-dev/2016-January/084556.html
21:02:39 <HenryG> Please read that email (and thread) and respond on the ML or in gerrit.
21:03:01 <HenryG> #topic Announcements
21:03:08 <ihrachys> HenryG: I hoped someone else will raise his hand. I can step forward if no one else is ready for that...
21:03:29 <rossella_s> go for it ihrachys
21:03:39 <ihrachys> ok, will send a patch tomorrow.
21:04:07 <HenryG> sounds good ihrachys
21:04:15 <HenryG> First announcement ...
21:04:17 <kevinbenton> #midcycle #midcycle #midcycle #midcycle!
21:04:26 <HenryG> Miktaka Milestone 2 has been released
21:04:32 <HenryG> #link https://launchpad.net/neutron/+milestone/mitaka-2
21:04:42 <HenryG> Second announcement ...
21:04:49 <HenryG> The Nova mid-cycle sprint is in progress
21:04:56 <HenryG> #link https://etherpad.openstack.org/p/mitaka-nova-midcycle
21:05:06 <HenryG> Third announcement ...
21:05:16 <HenryG> The Neutron mitaka mid-cycle sprint will be on Feb 23-26 in Rochester, MN, USA
21:05:19 * salv-orlando loves this new meeting chair ;)
21:05:24 <HenryG> #link https://etherpad.openstack.org/p/neutron-mitaka-midcycle
21:05:32 <HenryG> (brrrr)
21:05:53 <mhickey> HenryG: ice fishing all the way! :)
21:06:04 <HenryG> salv-orlando: armax wanted me to get to the backlog
21:06:29 <HenryG> Any other announcements?
21:06:45 <ihrachys> one preliminary here
21:07:14 <ihrachys> we will probably have a code sprint for "objectification" some time in mid-March somewhere in Europe
21:07:26 <ihrachys> stay tuned for more updates on the ML
21:07:45 <HenryG> thanks ihrachys
21:07:55 <salv-orlando> ihrachys: what would objectification be? The stuff you did for QoS applied to everything?
21:08:16 <ihrachys> salv-orlando: that's the end goal, but we'll be tactic and do only half of everything ;)
21:08:51 <ihrachys> we are currently looking into some core resources in context of upgrades subteam (port, network)
21:08:53 <salv-orlando> ihrachys: that would still be a rich goal
21:08:57 <ihrachys> we'll see what we will get to
21:09:28 <HenryG> Let's move on to bugs
21:09:37 <HenryG> #topic Bugs
21:09:52 <HenryG> ZZelle__: Anything to report from bugs last week?
21:10:14 <ZZelle__> HenryG, 1 critical bug
21:10:34 <ZZelle__> https://bugs.launchpad.net/neutron/+bug/1537746 about to be solved
21:10:35 <openstack> Launchpad bug 1537746 in neutron "l3_db should not assume dns-integration" [Critical,In progress] - Assigned to Russell Bryant (russellb)
21:11:58 <ZZelle__> a potentially critical bug: https://bugs.launchpad.net/neutron/+bug/1536540 (i wait for ajo feedback before setting this importance to critical if it's confirmed)
21:11:59 <openstack> Launchpad bug 1536540 in neutron "AttributeError: 'NoneType' object has no attribute 'port_name' when deleting an instance with QoS policy attached" [Undecided,New] - Assigned to Miguel Angel Ajo (mangelajo)
21:12:18 <amuller> ZZelle__: even if QoS is broken entirely it is not a 'critical' bug
21:12:21 <amuller> t's not a core feature
21:12:30 <ZZelle__> amuller, oups, ok
21:12:33 <ihrachys> + to amuller
21:13:05 <amuller> (even though it's the best feature anyone could possibly ever imagine and looking at its code is like being hugged by a thousand unicorns)
21:13:18 <ZZelle__> huhum
21:13:19 * njohnston spits out his soda
21:13:55 <HenryG> thanks ZZelle__
21:13:57 <ihrachys> also does not look like actual behavioral issue. more like a nasty traceback in lots.
21:14:11 <ihrachys> *in logs
21:14:35 <HenryG> I noticed a rise in the jobs failure rates over the weekend, but I think that was a blip due to the pip 8.0.0 release. The failure rates seem to be dropping back to "normal".
21:15:01 <HenryG> This week mlavalle is the bug deputy.
21:15:08 <mlavalle> HenryG: Yes I am
21:15:18 <HenryG> Any volunteers for next week?
21:16:34 <HenryG> OK, after the meeting I'll have to guilt someone into volunteering :)
21:16:38 <njohnston> I will volunteer, though I am a little green for it.
21:17:09 <HenryG> njohnston: great! and no problem, you have a ton of people to pester for help
21:17:26 <ihrachys> yay to volunteer!
21:17:40 <HenryG> njohnston: see the meeting wiki for the table of past deputies:
21:17:44 <HenryG> #link https://wiki.openstack.org/wiki/Network/Meetings
21:17:45 * russellb wanders in late and pretends he was here the whole time
21:17:48 <kevinbenton> just mark them all as opinion and then change the project to nova. that should throw a week of confusion onto it so you wont have to deal with it :)
21:18:04 <njohnston> HenryG: Thanks!
21:18:06 <Sam-I-Am> kevinbenton: hahah
21:18:22 <njohnston> kevinbenton: That's a hanging offense in Texas
21:18:46 <HenryG> Let
21:18:50 <HenryG> oops
21:18:54 <salv-orlando> kevinbenton: if armax ever needs what you just wrote, you're dead.
21:19:01 <salv-orlando> You know that, don't you
21:19:17 <kevinbenton> he'll have to find me first. i'm hiding in the desert
21:19:26 <HenryG> Let's move on to the backlog for M
21:19:37 <HenryG> #topic Blueprints
21:20:04 <HenryG> #link https://launchpad.net/neutron/+milestone/mitaka-3
21:20:45 <HenryG> First, https://blueprints.launchpad.net/neutron/+spec/tenant-delete
21:21:43 <kevinbenton> related question, is it okay if an HTTP DELETE request returns a body?
21:21:56 <kevinbenton> i was thinking it would be nice if this thing returned a list of all of the things it deleted
21:21:57 <ZZelle__> kevinbenton, nope
21:22:00 <kevinbenton> ok
21:22:51 <HenryG> Looks like that BP is out of M?
21:23:07 <HenryG> Most likely, I mean.
21:23:42 <hichihara> HenryG: +1
21:24:07 <ihrachys> looks like it. well, it's common for High prio BPs to fall thru cracks, ain't it?
21:24:55 <HenryG> Still more discussions needed to agree on how and where to orchestrate, I guess.
21:25:09 <salv-orlando> kevinbenton: you should check the api-wg guidelines and if no answer there RFC 7231. But if I'm not mistaken it's a no
21:25:45 <ZZelle__> kevinbenton, after checking HTTP RFC allows response bodies in DELETE
21:25:47 <HenryG> next
21:25:51 <HenryG> https://blueprints.launchpad.net/neutron/+spec/multi-l3-backends
21:25:57 <HenryG> kevinbenton ?
21:26:37 <kevinbenton> in progress. should have some PoC code and a spec pushed up at the end of this week
21:26:50 <kevinbenton> still figuring out the least invasive way to do it
21:26:53 <HenryG> sounds good
21:26:55 <HenryG> next
21:27:06 <HenryG> https://blueprints.launchpad.net/neutron/+spec/push-notifications
21:27:11 <HenryG> kevinbenton: also you
21:27:27 <kevinbenton> still in spec review
21:27:36 <kevinbenton> will need a volunteer core reviewer to be my approver also
21:28:19 <ihrachys> kevinbenton: would be glad, though there is no agreement for what I see
21:28:35 <kevinbenton> ihrachys: yes :(
21:28:46 <ihrachys> I assume it's not Mitaka deliverable
21:28:50 <HenryG> chances of making M = slim?
21:28:59 <kevinbenton> yeah, partial at best
21:29:05 <kevinbenton> i assume it will need a proof of concept
21:29:11 <kevinbenton> so we can at least fix the race condition in QoS with it
21:29:23 <kevinbenton> and then worry about more objects next cycle
21:29:43 <ihrachys> and it gives us some time to proceed with objectification first ;)
21:29:55 <HenryG> thanks kevinbenton and ihrachys
21:29:59 <HenryG> next
21:30:13 <HenryG> https://blueprints.launchpad.net/neutron/+spec/get-me-a-network
21:30:33 <HenryG> That's me, and armax and jaypipes
21:30:49 <ihrachys> chances?..
21:30:59 <HenryG> Some good feedback has been given on the patch, which I updated
21:31:19 <HenryG> It is being discussed and sprinted at the nova mid-cycle this week
21:31:46 <HenryG> We hope nova will pick up the orchestration/policy bits on their side
21:32:13 <HenryG> I have no idea of the feasibility of nova getting that done for M
21:32:24 <HenryG> We'll hear a report next week
21:32:53 <ihrachys> sounds promising nevertheless
21:33:12 <HenryG> The neutron work will proceed full steam for now
21:33:38 <HenryG> next
21:33:48 <HenryG> https://blueprints.launchpad.net/neutron/+spec/external-dns-resolution
21:33:52 <mlavalle> hi
21:34:07 <HenryG> mlavalle: hi
21:34:29 <mlavalle> code merged last week: https://review.openstack.org/#/c/212213/
21:34:59 <HenryG> mlavalle: is anything left? (sorry, have not followed the BP)
21:35:22 <mlavalle> armax indicated in the blueprint that operator docs are needed. So I have already coordinated with Sam-I-Am as to what needs to be done
21:35:48 <mlavalle> I'll update the networking guide over the next few days
21:35:53 <mlavalle> that's it
21:35:55 <HenryG> mlavalle: great! and thanks as always to superdoc Sam-I-Am
21:36:01 <Sam-I-Am> woot
21:36:14 <Sam-I-Am> if you document it, they will come
21:36:28 <Sam-I-Am> this is a thing a lot of people have been asking about
21:36:29 <HenryG> too true
21:36:46 <HenryG> next
21:36:59 <HenryG> https://blueprints.launchpad.net/neutron/+spec/l2-api-extensions
21:37:18 <ihrachys> that one is mine. the spec in review: https://review.openstack.org/263819 (armax requested some clarifications, respinned, beleive it's ready)
21:37:33 <ihrachys> and David pushed a patch for review: https://review.openstack.org/267591
21:37:47 <ihrachys> that one may require some respins, I need to get back to it.
21:37:56 <ihrachys> but overall, should easily land in M
21:38:11 <HenryG> nice, thanks ihrachys
21:38:33 <HenryG> next
21:38:35 <HenryG> https://blueprints.launchpad.net/neutron/+spec/modular-l2-agent
21:38:45 <ihrachys> ok, I am approver there.
21:38:54 <ihrachys> Andreas landed some preparation patches.
21:39:09 <ihrachys> there is also the main patch up for review: https://review.openstack.org/246318
21:39:18 <ihrachys> I think it should be almost ready for landing
21:39:33 <ihrachys> then there will be another patch that just moves code in another file
21:39:33 <ihrachys> and that should be it
21:39:37 <ihrachys> so will also land in M
21:40:12 <HenryG> good stuff, thanks ihrachys and Andreas
21:40:30 <HenryG> next
21:40:34 <HenryG> https://blueprints.launchpad.net/neutron/+spec/neutron-lib
21:40:42 <HenryG> dougwig
21:40:51 <kevinbenton> i think he's afk
21:40:51 <HenryG> I can report
21:41:02 <HenryG> Patches are landing
21:41:34 <HenryG> But it is a long-term ongoing effort that can be adopted gradually
21:41:53 <ihrachys> will we have first release in M?
21:42:01 <ihrachys> I mean, consumed by neutron
21:42:19 <HenryG> You can start now, it is on PyPI
21:42:27 <ihrachys> but is it in g-r?
21:42:59 <HenryG> I believe so
21:43:11 * HenryG scrambles to find the review
21:43:22 <ihrachys> https://github.com/openstack/requirements/blob/master/global-requirements.txt#L88
21:43:35 <HenryG> thanks :)
21:44:11 <HenryG> So knock yourselves out :/
21:44:35 <HenryG> It has some constants you can use :)
21:45:04 <HenryG> next
21:45:28 <HenryG> https://blueprints.launchpad.net/neutron/+spec/dhcp-options-per-subnet
21:46:14 <HenryG> Looks like this is being re-thought
21:46:33 <HenryG> I'll talk to the owners later, since they are not online
21:46:33 <pc_m> #link https://review.openstack.org/#/c/268232/ for neutron-lib
21:46:34 <amotoki> yeah I think so.
21:46:44 <amotoki> i will follow up.
21:46:57 <HenryG> sorry amotoki, I expected you to be asleep :)
21:47:13 <HenryG> thanks amotoki
21:47:53 <HenryG> next
21:47:54 <HenryG> https://blueprints.launchpad.net/neutron/+spec/keystone-v3
21:48:01 <HenryG> I am the approver
21:48:16 <HenryG> The spec is being reviewed
21:48:33 <HenryG> The spec has the projected deliverables with timelines
21:49:14 <HenryG> The spec: https://review.openstack.org/257362
21:49:28 <HenryG> next
21:49:30 <amotoki> doesn't migration to keystoneauth help us? I think most of the work has been done actaully.
21:50:51 <amotoki> it seems it is trying more than that. let's move to the next one.
21:51:43 <HenryG> amotoki: Let's follow up with Dariusz on that
21:51:51 <HenryG> next
21:51:55 <amotoki> HenryG: ok
21:51:58 <HenryG> https://blueprints.launchpad.net/neutron/+spec/performance-counters
21:52:12 * regXboi wakes up
21:52:16 <HenryG> Hi regXboi
21:52:50 <regXboi> this is not likely to make M - what PoC code is there is the low level agent side and transport - still need to work out the tables and storage
21:53:06 <regXboi> and then passing up to the northbound consumer
21:54:07 <regXboi> that's all I've got
21:54:22 <HenryG> regXboi: thanks, that's fine as long as no one is expecting otherwise
21:54:46 <regXboi> HenryG: if folks are expecting otherwise, they haven't dropped me the hint
21:55:27 <HenryG> we'll make sure it's noted in the BP
21:55:32 <HenryG> next
21:56:20 <HenryG> actually I'm gonna stop there for today
21:56:29 <HenryG> #topic Open Discussion
21:56:35 <HenryG> 5 minutes
21:56:46 <Sam-I-Am> i had something...
21:56:54 <Sam-I-Am> or is it more important to hit the docs stuff?
21:57:24 <Sam-I-Am> HenryG: ?
21:57:26 <HenryG> Sam-I-Am: we skipped the docs today to churn through the BP backlog
21:57:32 <Sam-I-Am> ok
21:57:40 <HenryG> but you can bring it up now if you want
21:57:50 <Sam-I-Am> quick stuff - the networking guide is now versioned
21:58:01 <HenryG> nice!
21:58:08 <Sam-I-Am> so we don't need to worry about 'in version X, you can do... in version Y, you can do..."
21:58:09 <amotoki> "versioned" means it has stable branches.
21:58:12 <Sam-I-Am> yes
21:58:28 <Sam-I-Am> we tried making it CD, but it didnt work out well
21:58:44 <Sam-I-Am> i'm working on updates for liberty (yeah, i'm behind)
21:58:53 <Sam-I-Am> most of them apply to mitaka too, which is aoog
21:58:54 <Sam-I-Am> good
21:59:20 <Sam-I-Am> always looking for more contributions. i can help you figure out where stuff goes.
21:59:38 <Sam-I-Am> the other open discussion topic - please read the 'mtu' discussion on the -dev mailing list if you haven't
21:59:43 <Sam-I-Am> its pretty enlightening
21:59:48 <Sam-I-Am> i'll be posting my results with OVS soon
22:00:15 <HenryG> Sam-I-Am: thanks
22:00:19 <Sam-I-Am> #link http://lists.openstack.org/pipermail/openstack-dev/2016-January/084241.html
22:00:37 <Sam-I-Am> lets solve this once and for all
22:00:38 <HenryG> If there is nothing else, then ...
22:00:40 <ihrachys> Sam-I-Am: hopefully the thread will get to some actionable items
22:00:48 <Sam-I-Am> hope so
22:01:02 <HenryG> #endmeeting