14:00:33 <mlavalle> #startmeeting neutron_drivers 14:00:34 <openstack> Meeting started Fri Sep 7 14:00:33 2018 UTC and is due to finish in 60 minutes. The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:35 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:37 <openstack> The meeting name has been set to 'neutron_drivers' 14:02:00 <amotoki> hi 14:02:09 <mlavalle> hi amotoki 14:02:19 <mlavalle> when are you leaving for Denver? 14:02:25 <slaweq> hi 14:02:30 <slaweq> sorry for being late 14:02:42 <amotoki> I leave Japan on Sunday. there is no spare day this time :( 14:02:47 <njohnston> o/ 14:03:00 <mlavalle> slaweq: no problem. I was pinging you in the channel just to make sure my connection was properly set up 14:03:12 <slaweq> yes, it is fine :) 14:03:47 <mlavalle> let's wait a minute and then we get going 14:04:02 <haleyb> hi 14:04:24 <amotoki> I had a trouble in connecting freenode last week.... my IRC bouncer was kicked out automatically :( 14:04:38 <mlavalle> ok, let's get going 14:05:01 <mlavalle> we have a short agenda for today, which seems appropriate give that we have the PTG next week 14:05:29 <mlavalle> first one is https://bugs.launchpad.net/neutron/+bug/1774463 14:05:29 <openstack> Launchpad bug 1774463 in neutron "RFE: Add support for IPv6 on DVR Routers for the Fast-path exit " [Wishlist,Triaged] 14:06:01 <mlavalle> yesterday I pinged Swami in the Neutron channel and asked him to answer the questions we left last week 14:06:07 <mlavalle> so let's take a look 14:10:21 <slaweq> I basically agree with amotoki's last comment there 14:10:29 <mlavalle> yeah, me too 14:10:59 <yamamoto> hi, sorry late 14:11:13 <slaweq> hi yamamoto 14:11:21 <mlavalle> yamamoto: no problem, we are considering https://bugs.launchpad.net/neutron/+bug/1774463 14:11:21 <openstack> Launchpad bug 1774463 in neutron "RFE: Add support for IPv6 on DVR Routers for the Fast-path exit " [Wishlist,Triaged] 14:11:22 <amotoki> In this RFE, I think we assume tenant IPv6 subnet is routable. 14:11:43 <mlavalle> I would say so 14:12:14 <haleyb> yes, it has to be routable 14:12:25 <amotoki> so it is reasonable to assume two subnets connected by a dvr router belong to a same address scope 14:13:09 <amotoki> we need to check conditions that they are routable but the RFE itself totally makes sense and the approach looks good to me. 14:13:54 <haleyb> yes, and swami seemed to confirm same scope in his response 14:14:39 <amotoki> yes, it is the usual IPv6 use case :) 14:14:55 <haleyb> right, no NAT, routable :) 14:15:15 <mlavalle> what IPv6 is for 14:15:27 <slaweq> I agree :) 14:16:17 <amotoki> do we need to care prefix delegation in this RFE? 14:16:49 <amotoki> IMHO it can be tackled later 14:17:28 <mlavalle> we could leave it as optional 14:17:40 <mlavalle> for the first implementation 14:18:31 <amotoki> totally agree. I don't know how PD can handle DVR at the moment. 14:18:36 <haleyb> yes, as it would require additional work i would think, so he should say it's nod supported 14:19:12 <haleyb> amotoki: well, PD is centralized along with all IPv6 routing, so no DVR support for North/South 14:19:38 <amotoki> haleyb: agree 14:20:27 <mlavalle> so any oppostion left on this RFE? or can we move ahead and approve it. I am ++ on it 14:20:37 <slaweq> ++ 14:20:49 <amotoki> ++ for moving this forward 14:20:52 <haleyb> +1 14:20:55 <yamamoto> i see no problem 14:21:02 <mlavalle> ok, hang on 14:22:19 <mlavalle> Done 14:22:27 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1785213 14:22:27 <openstack> Launchpad bug 1785213 in neutron "[RFE] Automatically allow incoming DHCP traffic for networks which uses external dhcp server" [Wishlist,Triaged] - Assigned to Slawek Kaplonski (slaweq) 14:24:07 <amotoki> basically it looks good to me. I post two questions. 14:24:45 <slaweq> amotoki: about first question, I wanted to add it to network, as ports are connected to network, not to subnet 14:25:02 <slaweq> and such network can be "provider" network with some external dhcp server connected to it 14:25:36 <amotoki> slaweq: it is understandable. we are always in a dillemma 14:25:57 <slaweq> about second question, I need to read Your reference document first and think about it 14:26:13 <amotoki> slaweq: it is a headache to mix L2 and L3 stuffs in "network" resource 14:26:41 <slaweq> amotoki: I agree, so maybe indeed subnet may be better place 14:27:33 <amotoki> honestly I am not sure if it works to enable_dhcp for partial subnets on a network 14:28:08 <slaweq> network is scheduled to some dhcp agent normally 14:28:16 <amotoki> if we assume enable_dhcp consistently for all subnets, I think it is better to have 'external_dhcp' in 'subnet'. 14:28:26 <amotoki> that's the only reason 14:29:01 <slaweq> ok, so I will check if there is possibility to have in one network, subnet with enabled and subnet with disabled dhcp together 14:29:05 <slaweq> and how it will then work 14:29:44 <mlavalle> so should we expect a next iteration in the RFE? 14:30:02 <slaweq> mlavalle: yes, I will check that and update rfe in comment 14:30:03 <amotoki> sounds reasonable to me 14:30:18 <mlavalle> sounds good to me 14:30:35 <mlavalle> thanks for the proposal slaweq :-) 14:30:51 <mlavalle> and that allows us to adjourn our meeting for today 14:31:07 <mlavalle> since most of you will be in Denver, have safe trips ! 14:31:20 <slaweq> see You in Denver :) 14:31:21 <mlavalle> Looking forward to see you there 14:31:21 <njohnston> see you there! 14:31:29 <slaweq> I will be there on Sunday afternoon 14:31:33 <mlavalle> we will miss you yamamoto! 14:31:45 <amotoki> me too 14:31:47 <mlavalle> I get to the hotel arounf 11:30 am 14:31:49 <njohnston> I will arrive very late on Sunday 14:32:03 <yamamoto> i'm not going this time. enjoy the train horn. 14:32:06 <haleyb> i'm late Sunday as well, probably not at hotel until after 9pm 14:32:26 <amotoki> perhaps I will arrive at the hotel around 6pm 14:32:27 <mlavalle> yamamoto: yeah, I bet you won't miss that.... LOL 14:32:37 <mlavalle> o/ 14:32:42 <mlavalle> #endmeeting