15:00:42 <tidwellr> #startmeeting neutron_l3
15:00:43 <openstack> Meeting started Thu Jul  7 15:00:42 2016 UTC and is due to finish in 60 minutes.  The chair is tidwellr. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:45 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:46 <carl_baldwin> Hi
15:00:47 <openstack> The meeting name has been set to 'neutron_l3'
15:00:47 <njohnston> \o/
15:01:08 <pavel_bondar> hi
15:01:37 <tidwellr> #topic Announcements
15:02:13 <tidwellr> Newton-2 is next week
15:03:01 <tidwellr> not sure about everyone else but it doesn't seem to be flying by as fast this cycle, maybe I'm just out of touch :)
15:03:20 <carl_baldwin> Just you tidwellr
15:03:24 <mlavalle> lol
15:03:36 <tidwellr> carl_baldwin: I figured :)
15:04:03 <tidwellr> mid-cycle is coming up 17-19 August 2016
15:04:17 <tidwellr> #link https://etherpad.openstack.org/p/newton-neutron-midcycle
15:04:29 <tidwellr> etherpad ^^^
15:04:42 <mlavalle> I got my travel approval yesterday and I am all booked now!
15:05:11 <john-davidge> woop!
15:05:49 <tidwellr> oh yeah, the agenda for this meeting is here https://etherpad.openstack.org/p/neutron-l3-subteam
15:06:12 <tidwellr> any other announcements I missed?
15:06:28 <mlavalle> Talks submission deadline for Barcelona is July 13th
15:06:58 <tidwellr> mlavalle: thanks, that one slipped my mind
15:07:23 <mlavalle> I got your back, partner :-)
15:07:36 <tidwellr> I need all the backup I can get!
15:07:55 <tidwellr> anything else before we move on?
15:08:24 <tidwellr> #topic Bugs
15:08:37 <tidwellr> https://bugs.launchpad.net/neutron/+bug/1596075
15:08:37 <openstack> Launchpad bug 1596075 in neutron "Neutron confused about overlapping subnet creation" [High,Incomplete]
15:08:56 <mlavalle> I tried to replicate that one a few days ago
15:09:03 <mlavalle> without luck
15:09:23 <mlavalle> the reporter then wrote a python script that repliactes the issue
15:09:35 <mlavalle> that remove heat out of the equation
15:09:42 <mlavalle> removes^^^
15:10:27 <tidwellr> I see it, is that the code included in the second to last comment?
15:10:49 <mlavalle> he ha 3 api servers, which I don't. Can someone here can test this easily with more than 1 api server?
15:11:13 <mlavalle> the last comment actually has a gist
15:11:25 <mlavalle> with the whole script. So it can be easily downloaded
15:11:36 <tidwellr> oh, ok
15:12:26 <carl_baldwin> I'd like to look in to how nginx is doing round-robin.  I wonder if it is doing something funny.
15:13:17 <mlavalle> We can also ask him to repliacte the failure and post the logs from the 3 servers while the test was running
15:14:07 <carl_baldwin> mlavalle: That could help.
15:14:43 <mlavalle> ok, I'll post a comment later today. I will also ask him for his nginx setup
15:15:23 <johndperkins> I'm working on https://bugs.launchpad.net/neutron/+bug/1493945, l3 agent router scheduling fails under load, I haven't quite replicated the failure but I'm close if anyone has time to take a peek.
15:15:23 <openstack> Launchpad bug 1493945 in neutron "Router scheduling at network node fails under scale" [Medium,In progress] - Assigned to John Perkins (john-d-perkins)
15:16:00 * mlavalle waves at Perks
15:16:07 <johndperkins> hey hey miguel!
15:16:23 <tidwellr> johndperkins: thanks for pointing that out
15:16:30 <johndperkins> hopefully this was an appropriate time to mention that
15:16:52 <tidwellr> we are going through bugs right now, so no worries
15:17:37 <tidwellr> johndperkins: anything else you wanted to discuss before chat about the next bug?
15:17:48 <johndperkins> I'm good
15:17:57 <tidwellr> https://bugs.launchpad.net/neutron/+bug/1599329
15:17:57 <openstack> Launchpad bug 1599329 in neutron "Potential regression on handing over DHCP addresses to VMs" [High,Confirmed] - Assigned to Kevin Benton (kevinbenton)
15:19:16 <tidwellr> this seems to be showing up in the gate from time to time, not sure what else there is to say about it at the moment
15:20:10 <tidwellr> anybody have anything to add?
15:20:37 <tidwellr> if not we can move on
15:21:09 <tidwellr> #topic Routed Networks
15:21:33 <carl_baldwin> Hi
15:21:49 <carl_baldwin> I think we've made some progress on deferred port binding.
15:21:58 <carl_baldwin> I'll have a new patch up today.
15:22:42 <carl_baldwin> Still need to get DHCP in shape.
15:22:47 <carl_baldwin> Is blogan around?
15:23:36 <carl_baldwin> mlavalle: You have anything?
15:23:56 <mlavalle> I just had a conversation with Chris Dent of the Nova team
15:24:25 <mlavalle> I requested his guidance of how to deploy g-r-p patches in my test environment
15:24:33 <mlavalle> So I will start playing with it
15:24:55 <carl_baldwin> Great.
15:25:03 <mlavalle> He indicated that the api itself and the inventories functionality should be ok
15:25:14 <mlavalle> no integration with aggregates yet
15:25:27 <mlavalle> I'll keep pushing forward on that front
15:25:33 <mlavalle> that's all
15:25:34 <carl_baldwin> Make senses.  WIP
15:26:05 <mlavalle> Yeah, I want to push a patchset on the Neutron side with atht interaction as early as possible
15:26:33 <mlavalle> The arliest, the more time we will have to mature it
15:26:46 <carl_baldwin> agreed
15:27:46 <carl_baldwin> john-davidge: haleyb:  I read through the service subnets patch yesterday.  Any questions about that?
15:28:30 <john-davidge> carl_baldwin: Yeah, thanks for your comments! I've been working on a new patch since haleyb posted it, looking to put it up for review by the end of the week
15:28:53 <john-davidge> carl_baldwin: So far progress is good, already quite different to what's there now
15:28:54 <haleyb> carl_baldwin: i need to respond to your comments, but i really just threw something over the wall (it doesn't work as-is)
15:29:18 <haleyb> john-davidge: good to see you're making progress
15:29:19 <carl_baldwin> john-davidge: great, let me know when something else needs review.
15:29:55 <john-davidge> carl_baldwin: Will do, and thanks to haleyb for getting the ball rolling
15:30:29 <haleyb> john-davidge: i will work on the OSC changes based on my neutronclient patch
15:31:00 <john-davidge> haleyb: Cool, let me know when you need reviews
15:31:15 <carl_baldwin> +1
15:31:57 <carl_baldwin> I think that's all we have going on with routed networks.
15:32:33 <tidwellr> carl_baldwin: thanks for the update
15:32:46 <tidwellr> lots going on here
15:32:55 <tidwellr> #topic Conversion to Pluggable IPAM
15:33:09 <pavel_bondar> hi
15:33:10 <tidwellr> it seems like we haven't touched on this in a while
15:33:32 <tidwellr> pavel_bondar: I wanted to give you some air time
15:33:41 <carl_baldwin> We've got to get this done.  I think we might have some test failures to look in to in the API tests.
15:34:12 <pavel_bondar> The review to switch to pluggable ipam is here #link https://review.openstack.org/#/c/181023/
15:34:29 <pavel_bondar> I have not looked deeply into this API test failures
15:34:44 <tidwellr> pavel_bondar: could you use some help?
15:35:26 <pavel_bondar> yes, if someone could look into it, it would be great
15:36:22 <pavel_bondar> I am working on openstack just part time right now, so can spend a lot of time on it
15:36:47 <carl_baldwin> I can probably carve out some time tomorrow to look in to it.
15:36:51 <tidwellr> pavel_bondar: is it neutron.tests.tempest.api.test_auto_allocated_topology.TestAutoAllocatedTopology that seems to be failing consistently?
15:37:11 <tidwellr> or do we have different tests bombing out on each run?
15:37:22 <pavel_bondar> yes, looks like that one
15:37:53 <pavel_bondar> I see there is a fix merged for it: https://review.openstack.org/#/c/337863/
15:38:15 <tidwellr> are we just hung up on the API test failures then?
15:38:52 <pavel_bondar> yes
15:39:13 <pavel_bondar> by the way there were several succesful passes before this failure
15:39:36 <carl_baldwin> pavel_bondar: Thanks.
15:40:04 <pavel_bondar> so that is probably it from my side
15:40:07 <tidwellr> pavel_bondar: thanks for the update, I'll see if I can carve out some time as well
15:40:32 <tidwellr> #topic FWaaS
15:40:47 <tidwellr> I'm jumping around the agenda today, sorry
15:40:52 <njohnston> No problem
15:41:22 <njohnston> So, the l3 agent extension spec is merged, and work is coming along on the change to generalize the agent extension framework: https://review.openstack.org/329701
15:41:27 <njohnston> I think that is not far from landing.
15:42:14 <njohnston> In order to make use of the l3 agent extension, we are getting going working on the (WIP) FWaaS L3 agent extension implementation: https://review.openstack.org/337699
15:42:55 <njohnston> After the generic agent extension interface stabilizes, will work with boden to migrate applicable agent code to neutron-lib, per discussion in the neutron-lib meeting yesterday.
15:43:48 <njohnston> So progress is being made, and soon I think we'll be able to have working code for an l3 agent extension manager and FWaaS being loaded as an l3 agent extension.
15:44:05 <carl_baldwin> njohnston: sounds like good progress
15:44:07 <tidwellr> njohnston: the generalized agent extension framework is good stuff, glad to see it coming along
15:44:38 <njohnston> I'm feeling optimistic about it.
15:45:21 <tidwellr> njohnston: thanks, anything else to share?
15:45:31 <njohnston> Nope, that'
15:45:34 <njohnston> that's it for me
15:45:48 <tidwellr> #topic BGP Dynamic Routing
15:46:46 <tidwellr> I'm slowly making progress on getting the tempest plugin and gate hook in place for neutron-dynamic-routing
15:47:18 <tidwellr> https://review.openstack.org/#/c/335310/
15:47:47 <tidwellr> we've got got a patch for getting BGP peering stats in progress as well
15:48:16 <tidwellr> https://review.openstack.org/#/c/334397/
15:49:06 <tidwellr> we're just working through what data should exposed and the mechanics of the RPC interaction with the BGP agent
15:49:28 <carl_baldwin> tidwellr: Are you getting enough reviews on this repo?
15:50:02 <tidwellr> I think so, the activity with bugs and patches being proposed is trending up
15:50:18 <carl_baldwin> good
15:50:42 <tidwellr> still, my first priority is the CI
15:51:01 <tidwellr> we need it in place ASAP
15:51:32 <tidwellr> I also have been meaning to ping armax about the status neutron-api, that's another item we'll need to meet the stadium criteria
15:52:09 <carl_baldwin> ok
15:52:12 <tidwellr> steve_ruan: anything to add?
15:52:13 * njohnston is interested in what he says about neutron-api as well
15:52:47 <steve_ruan> no update
15:53:13 <tidwellr> alright
15:53:23 <tidwellr> #topic Open Discussion
15:53:45 <tidwellr> if we missed anything, now's the time to chat about it
15:54:15 <tidwellr> going once...
15:54:17 <mlavalle> carl_baldwin: could you take a look at https://review.openstack.org/#/c/313291/. It is essentially what you had already approved after some rebases
15:54:32 <carl_baldwin> mlavalle: user
15:54:35 <carl_baldwin> mlavalle: sure
15:54:40 <mlavalle> :-)
15:54:58 <carl_baldwin> dyslexic fingers
15:55:13 <mlavalle> I do that all the time, so no worries
15:55:21 <pavel_bondar> I have addressed comments for #link https://review.openstack.org/#/c/323345/  today, so feel free to review
15:56:07 <carl_baldwin> mlavalle: Sorry that dropped off the radar.
15:56:13 <carl_baldwin> pavel_bondar: ok
15:56:40 <mlavalle> carl_baldwin: np whatsoever
15:58:14 <tidwellr> anything else?
15:58:37 <tidwellr> thanks everyone!
15:58:45 <tidwellr> #endmeeting