15:00:07 <mlavalle> #startmeeting neutron_l3
15:00:08 <openstack> Meeting started Thu Nov 10 15:00:07 2016 UTC and is due to finish in 60 minutes.  The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:09 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:12 <openstack> The meeting name has been set to 'neutron_l3'
15:00:21 <mlavalle> Hi
15:00:31 <mlavalle> who's here for the L3 meeting?
15:00:38 <janzian> o/
15:00:44 <carl_baldwin> o/
15:00:45 <john-davidge> o/
15:00:48 <baoli_> hi
15:01:15 <mlavalle> #topic Announcements
15:01:55 <mlavalle> First of, Ocata-1 is almost here:
15:02:09 <mlavalle> #link https://releases.openstack.org/ocata/schedule.html
15:02:36 <mlavalle> Any other annoucements from the team?
15:03:19 <mlavalle> ok, let's move on....
15:03:24 <mlavalle> #topic bugs
15:03:50 <mlavalle> First one is https://bugs.launchpad.net/neutron/+bug/1509004
15:03:50 <openstack> Launchpad bug 1509004 in neutron ""test_dualnet_dhcp6_stateless_from_os" failures seen in the gate" [High,Confirmed]
15:04:04 <mlavalle> I tried to work on this one yesterday
15:04:41 <mlavalle> Problem I have is that http://logstash.openstack.org is not working for me
15:04:59 <mlavalle> I have tried from several machines
15:05:14 <mlavalle> The page starts but it doesn't show any results
15:05:26 <mlavalle> Are others experiencing a similar problem?
15:06:01 <carl_baldwin> Hmmm. I haven't tried in a little while.
15:06:37 <carl_baldwin> Yeah, nothing comes up so far. It is just churning.
15:06:45 <mlavalle> Exactly!
15:06:48 <john-davidge> Yeah, I'm seeing the same thing
15:06:58 <baoli_> same thing here
15:07:32 <mlavalle> In that case I will reach out to the infra team and have this fixed. Thanks
15:07:58 <mlavalle> As for the bug, I am assigning to me and will debug with the cases already logged there
15:08:17 <mlavalle> any more comments on this one?
15:09:24 <mlavalle> ok, let's move on
15:09:40 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1540983
15:09:40 <openstack> Launchpad bug 1540983 in OpenStack-Gate "Gate failures for neutron in test_dualnet_multi_prefix_slaac" [Undecided,Expired]
15:10:30 <mlavalle> Oleg was trying to reproduce this one. I will reach out to him to see whether he's had any success
15:11:11 <mlavalle> I don't see him today in the Neutron channel
15:11:22 <mlavalle> Any other comments?
15:11:57 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1570122
15:11:57 <openstack> Launchpad bug 1570122 in neutron "ipv6 prefix delegated subnets are not accessable external of the router they are attached." [High,Confirmed] - Assigned to John Davidge (john-davidge)
15:12:45 * haleyb_ strolls in late
15:12:57 <mlavalle> haleyb_: welcome!
15:13:13 <mlavalle> any comments john-davidge?
15:13:40 <john-davidge> mlavalle: Intended to comment with my thoughts from last meeting on launchpad but got pulled onto ther things this week. Will attempt to visit it soon. I don't anticipate this being fixed before o-1
15:14:39 <mlavalle> john-davidge: Thanks. Since we are with you, could we comment also on https://bugs.launchpad.net/neutron/+bug/1636963
15:14:39 <openstack> Launchpad bug 1636963 in neutron "Service Subnets - DHCP port creation fails" [High,In progress] - Assigned to John Davidge (john-davidge)
15:15:13 <john-davidge> mlavalle: Yep, this one is still in need of reviews: https://review.openstack.org/#/c/392183/
15:15:39 <mlavalle> ok, thanks
15:16:11 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1610483
15:16:11 <openstack> Launchpad bug 1610483 in neutron "Pluggable IPAM rollback mechanism is not robust" [High,In progress] - Assigned to Aliaksandr Dziarkach (aliaksandr-dziarkach)
15:16:33 <mlavalle> For this one we have a proposed fix:
15:16:39 <mlavalle> #link https://review.openstack.org/#/c/390594/
15:17:01 <mlavalle> I intend to take a look later today and encourage the rest of the team to do the same
15:17:29 <mlavalle> any other comments?
15:19:22 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1627424
15:19:22 <openstack> Launchpad bug 1627424 in neutron "FlushError on IPAllocation" [High,Confirmed]
15:19:44 <mlavalle> We don't have an owner for this one
15:20:33 <mlavalle> If no one volunteers I will assign it to me
15:21:10 <mlavalle> Moving on
15:21:56 <mlavalle> I see someone adding 3 bugs about prefix delegation:
15:22:00 <mlavalle> https://bugs.launchpad.net/neutron/+bug/16390332.
15:22:00 <openstack> mlavalle: Error: malone bug 16390332 not found
15:22:06 <mlavalle> https://bugs.launchpad.net/neutron/+bug/16390332
15:22:11 <baoli_> mlavalle: I added them
15:22:25 <mlavalle> baoli_: thanks. do you care to comment?
15:22:37 <baoli_> basically: PD is broken
15:22:45 <john-davidge> :(
15:23:04 <mlavalle> :(
15:23:06 <carl_baldwin> Need the correct bug number.
15:23:13 <john-davidge> remove the 2
15:23:30 <john-davidge> https://bugs.launchpad.net/neutron/+bug/1639033
15:23:30 <openstack> Launchpad bug 1639033 in neutron "Traces seen in l3 agent with prefix delegation enabled subnet" [Undecided,New] - Assigned to Baodong (Robert) Li (baoli)
15:24:02 <carl_baldwin> I'm not really surprised since PD has never really had any testing.
15:24:02 <baoli_> sorry about the typo
15:24:27 <baoli_> I started looking at PD since last week
15:25:29 <baoli_> I actually have a fix for this one
15:25:50 <mlavalle> baoli_: have you pushed it to gerrit?
15:26:01 <baoli_> mlavalle, not yet
15:26:32 <baoli_> mlavalle, will do it soon once I have the unit test
15:26:48 <mlavalle> Should we expect fixes from you for the other 2 bugs?
15:27:06 <baoli_> malvalle, I'd like to discuss those two a little bit
15:27:13 <mlavalle> ok
15:27:40 <mlavalle> Next one is https://bugs.launchpad.net/neutron/+bug/1640271
15:27:40 <openstack> Launchpad bug 1640271 in neutron "Prefix delegation: bind LLA is removed from the gateway port with a regular neutron router" [Undecided,New]
15:28:24 <baoli_> basically init_router_port would remove any IPs that's not passed in or presevered. And the method has been seen to be called a couple of times while adding PD-enabled subnet
15:29:23 <baoli_> PD works by adding a LLA to the gateway port that a dibbler is bound to and receives prefix from the server
15:29:29 * john-davidge is being pulled into another meeting, will keep my eye here where i can
15:29:46 <baoli_> But the later call to init_router_port has the LLA removed
15:30:42 <haleyb_> Did we break this in newton or right after?
15:30:45 <baoli_> So the fix is to pass the LLA as part of the preserve_ips
15:31:17 <baoli_> haleyb_: I am not sure.
15:32:10 <haleyb_> ok, it will probably be obvious once you fix it, just wanted to keep it on the backport radar
15:33:27 <mlavalle> any other comments on this one baoli_ ?
15:33:31 <baoli_> But the tricky part is that in order to pass the LLA, agent handle has to be passed in to methods (such as external_gateway_added()), etc, and it doesn't seem to be desriable given that classes like HARouter, DVRRouter, etc has to be modified accordingly
15:34:38 <baoli_> so I'm thinking to refactor the PrefixDelegation Class so that the change is restrictted to the agent, router info and PD classes.
15:35:00 <baoli_> Just want to get some option on whether I'm going on the right direction.
15:35:17 <mlavalle> opinion^^^^
15:35:50 <haleyb_> yes, doing that would make it easy to backport.  once we can do that we could potentially refactor going forward
15:35:56 <baoli_> thanks mlavalle.
15:37:02 <mlavalle> baoli_: any other comments on this one
15:37:29 <baoli_> mlavalle, that's all from my side.
15:37:51 <mlavalle> baoli_: so we will keep eyes open to fix
15:38:02 <baoli_> mlavalle, thanks
15:38:17 <mlavalle> Last one is https://bugs.launchpad.net/neutron/+bug/1639042
15:38:17 <openstack> Launchpad bug 1639042 in neutron "Prefix delegation doesn't work in the described scenario" [Undecided,New] - Assigned to Baodong (Robert) Li (baoli)
15:38:40 <baoli_> mlavalle, basically this one has the same root cause as the previous one.
15:40:32 <mlavalle> baoli_: thank you so much for your work on this subject
15:41:17 <mlavalle> baoli_: do you need any other form of help?
15:41:22 <baoli_> mlavalle, my pleasure
15:42:01 <baoli_> mlavalle, I' fine for the time being. will let you know if I need further help
15:42:10 <mlavalle> Thanks
15:42:38 <mlavalle> Any other bugs from the team?
15:44:31 <mlavalle> Ok, moving on
15:44:40 <mlavalle> #topic Routed Networks
15:45:27 <mlavalle> On this topic, I have continued making steady progress with the updating of IPv4 inventories for the Nova scheduler
15:45:38 <mlavalle> The patchset is here:
15:45:44 <mlavalle> #link https://review.openstack.org/#/c/358658/
15:46:21 <carl_baldwin> mlavalle: Sounds great, mlavalle . I enjoyed seeing your demo.
15:46:42 * mlavalle enjoyed presenting with carl_baldwin also
15:46:57 <mlavalle> I am adding unit tests
15:47:30 <mlavalle> and only waiting for this patchset to merge on the Nova side: https://review.openstack.org/#/c/358658/
15:48:05 <mlavalle> That adds the association of hosts aggregates to the ipv4 inventories
15:48:35 <mlavalle> Discussed it with the Nova team in Barcelona and they are moving as fast as they can with it
15:49:15 <mlavalle> I also discussed with johnthetubaguy the work on allocate_for_instance in Nova's APU with Neutron
15:49:31 <mlavalle> he was very nice and got this work to be prioritized in Ocata
15:49:52 <mlavalle> so I think we are in good shape in that front
15:50:12 * john-davidge has just heard him mention its priority as well :)
15:50:45 <mlavalle> john-davidge: thanks!
15:51:10 <mlavalle> janzian has continued his work on the tempest api tests
15:51:20 <mlavalle> any comments to the team janzian?
15:52:22 <janzian> Nope :)
15:52:40 <janzian> Not much to update
15:53:04 <mlavalle> ok, thanks
15:54:02 <mlavalle> I don't see Ryan Tidwell on-line. I'll reach out to him today to check on how we plan to continue working on BGP Dynamic routing
15:54:12 <mlavalle> #topic Open Agenda
15:54:22 <mlavalle> Any other topics from the team?
15:54:23 <jschwarz> hey guys
15:54:27 * john-davidge has to run, thanks all o/
15:54:36 <jschwarz> I have a bunch of patches that deal with l3 scheduler refactor, starting with https://review.openstack.org/#/c/365326/
15:54:43 <jschwarz> it's in desperate need of reviews :)
15:55:24 <mlavalle> jschwarz: do you have links available? I will take a look over the next 2 or 3 days
15:55:47 <jschwarz> mlavalle, https://review.openstack.org/#/c/365326/13, https://review.openstack.org/#/c/317949/24 and https://review.openstack.org/#/c/386077/2
15:55:51 <jschwarz> in that order :) thanks
15:56:01 <mlavalle> jschwarz: thanks!
15:56:11 <mlavalle> any other topics?
15:56:21 <haleyb_> jschwarz: i will look again as well, i know things have been getting re-spun
15:56:44 <jschwarz> haleyb, mlavalle, much appreciated
15:57:15 <mlavalle> Ok team, thanks for attending
15:57:23 <mlavalle> have a nice rest of your day
15:57:28 <mlavalle> #endmeeting