15:00:16 <mlavalle> #startmeeting neutron_l3
15:00:17 <openstack> Meeting started Thu Feb  2 15:00:16 2017 UTC and is due to finish in 60 minutes.  The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:18 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:20 <openstack> The meeting name has been set to 'neutron_l3'
15:00:47 <john-davidge> o/
15:01:10 <mlavalle> hey john-davidge
15:01:22 <mlavalle> I saw janzian joining the meeting also
15:01:32 <mlavalle> haleyb: you around?
15:01:34 <janzian> Morning everyone
15:01:56 <haleyb> yes, i'm here
15:02:05 <mlavalle> Agenda for today is here:
15:02:16 <mlavalle> #link https://etherpad.openstack.org/p/neutron-l3-subteam
15:02:36 <mlavalle> #topic Announcements
15:02:59 <mlavalle> First one is we are cutting RC-1 this week
15:03:12 <mlavalle> so we need to be careful merging patchsets
15:04:12 <mlavalle> Second announcements, this week the community is holding PTL elections:
15:04:34 <mlavalle> #link https://governance.openstack.org/election/
15:05:13 <mlavalle> In the case of Neutron, we have two candidates. Please don't forget to cast your votes
15:06:03 <mlavalle> Third, the PTG is 3 weeks. You can find the Neutron etherpad here:
15:06:21 <mlavalle> #link https://etherpad.openstack.org/p/neutron-ptg-pike
15:06:48 <mlavalle> If you have subjects you want discussed during the PTG, add them to the etherpad
15:07:14 <mlavalle> any other annoucements from the team?
15:08:02 <mlavalle> #topic Bugs
15:09:01 <mlavalle> First up today is https://bugs.launchpad.net/neutron/+bug/1610483
15:09:01 <openstack> Launchpad bug 1610483 in neutron "Pluggable IPAM rollback mechanism is not robust" [High,In progress] - Assigned to Aliaksandr Dziarkach (aliaksandr-dziarkach)
15:09:21 <mlavalle> I see that Launchpad wants to assign it to haleyb
15:09:33 <mlavalle> and haleyb is pushing it back ;-)
15:09:55 <mlavalle> haleyb: thanks for the review
15:10:28 <mlavalle> I'll probably ping the author or johnbelamaric to see what is the next step
15:10:33 <mlavalle> any other comments?
15:11:19 <haleyb> yes, i updated, but then ?? fixed it
15:12:26 <mlavalle> haleyb: i'll ping John, if they don't follow up, we can fix it. ok haleyb?
15:12:46 <haleyb> ok
15:13:06 <mlavalle> next up https://bugs.launchpad.net/neutron/+bug/1570122
15:13:06 <openstack> Launchpad bug 1570122 in neutron "ipv6 prefix delegated subnets are not accessable external of the router they are attached." [High,In progress] - Assigned to John Davidge (john-davidge)
15:13:42 <john-davidge> mlavalle: The patch has a +2 from haleyb, but is getting caught in the gate
15:14:08 <haleyb> it needs another +2, and i got it on the RC1 list
15:14:09 <mlavalle> john-davidge: but is ready to go, right?
15:14:17 <john-davidge> mlavalle: Wait, no, that's the other one getting caught in the gate :)
15:14:29 <haleyb> that nova bug fix has almost merged...
15:14:31 <john-davidge> mlavalle: Yes, ready to go, needs another +2
15:14:45 <john-davidge> haleyb: Thanks@
15:14:46 <mlavalle> ah ok, in that case I'll review it later today and merge it
15:14:47 <john-davidge> !
15:14:56 <mlavalle> john-davidge: ^^^^
15:15:04 <john-davidge> mlavalle: perfect, thank you
15:15:12 <haleyb> until https://review.openstack.org/#/c/427775/ merges (nova) it's probably going to fail recheck
15:15:26 <haleyb> i hope
15:16:10 <mlavalle> Next up is https://bugs.launchpad.net/neutron/+bug/1660607
15:16:10 <openstack> Launchpad bug 1660607 in neutron "Newton: Changing fixed ip doesn't work when a subnet is using service type" [High,In progress] - Assigned to John Davidge (john-davidge)
15:17:00 <haleyb> that's stuck in the gate queue as well
15:17:03 <mlavalle> john-davidge: this is the one caught in the gate, right?
15:17:10 <mlavalle> #link https://review.openstack.org/#/c/427223/
15:17:10 <john-davidge> mlavalle: That's the one getting caught in the gate, alreayd has 2 +2s and workflow
15:17:16 <haleyb> mlavalle: i just added two other IPv6 PD bugs to the wiki as well
15:17:32 <mlavalle> haleyb: yeah I saw that
15:17:42 <haleyb> or maybe just the one, broken w /dvr
15:18:08 <haleyb> https://review.openstack.org/#/c/277657/
15:18:29 <mlavalle> this is the bug: https://bugs.launchpad.net/neutron/+bug/1541406
15:18:29 <openstack> Launchpad bug 1541406 in neutron "IPv6 prefix delegation does not work with DVR" [Medium,In progress] - Assigned to Brian Haley (brian-haley)
15:18:35 <haleyb> john-davidge: can you take a look at ^^, might as well fix all the PD issues we can
15:18:55 <haleyb> HA is the only remaining after that
15:18:55 <john-davidge> haleyb: Sure
15:19:30 <mlavalle> haleyb: I'll take a look as well. Ideally, we want thos for Rc-1, right?
15:19:46 <haleyb> mlavalle: yes, they have been approved for RC1
15:19:52 <mlavalle> cool
15:20:08 <mlavalle> already in my pile of reviews
15:20:10 <haleyb> https://launchpad.net/neutron/+milestone/ocata-rc1
15:20:35 <john-davidge> haleyb: This is a pretty big chnage that hasnt been looked at since Feb, do you think its realistic to merge for rc1?
15:20:42 <haleyb> people should make sure their bug is on that list
15:21:23 <haleyb> john-davidge: did i paste the wrong link?  i didn't think the change was large
15:21:37 <haleyb> https://review.openstack.org/#/c/277657/
15:21:53 <john-davidge> haleyb: Ah, yeah, my bad, looking at the wrong patch
15:22:05 <haleyb> phew!
15:22:14 <mlavalle> LOL
15:22:42 <mlavalle> ok, next up is https://bugs.launchpad.net/neutron/+bug/1627424
15:22:42 <openstack> Launchpad bug 1627424 in neutron "FlushError on IPAllocation" [High,Confirmed] - Assigned to Miguel Lavalle (minsel)
15:23:05 <haleyb> you didn't fix that yet? :)
15:23:12 <mlavalle> I made great progress with this one. I figured out what the problem is
15:23:27 <mlavalle> no fix yet, but now I know what is going on
15:23:47 <haleyb> nice, was being sarcastic
15:24:02 <mlavalle> the FlushError is really just a symptom of a deeper problem with our retry logic
15:25:04 <mlavalle> we have a port P with ip addresses ip1 (subnet S1), ip2 (subnet S2) and more ip's on other subnets
15:25:27 <mlavalle> we get a request to delete S1 and another request to delete S2
15:26:02 <mlavalle> each one of those requests goes and tries to update P, removing the respective ip address
15:26:23 <mlavalle> request to delete S2 succeeds updating P
15:27:04 <mlavalle> when request to delete S1 tries to update P, it gets stale data from common attributes of the P
15:27:57 <mlavalle> the update_port for P is retried, but that update_port still has ip2 as a fixed ip of P
15:28:17 <mlavalle> so it ends up adding back ip2 and trying to delete ip1
15:28:29 <mlavalle> and that is where we get the FlushError
15:28:54 <mlavalle> howpefully I didn't confuse you
15:29:12 <mlavalle> I'll update the bug with my findings and propose a fix soon
15:29:23 <mlavalle> any questions or advice?
15:29:52 <john-davidge> mlavalle: Sounds like you have a good handle on it :)
15:29:58 <haleyb> makes sense, looking forward to seeing fix.  i'd advise pinging kevin benton when you get a patch out is all
15:30:42 <mlavalle> haleyb: yeah, that is my plan. he is the one who really knows the retry logic plumbing
15:32:11 * mlavalle as a English speaker as a second language is always confused between advice and advise :-)
15:32:29 <mlavalle> Next up is https://bugs.launchpad.net/neutron/+bug/1627480
15:32:29 <openstack> Launchpad bug 1627480 in neutron "create_port can succeed without returning fixed_ips on all requested subnets" [High,Confirmed] - Assigned to James Anziano (janzian)
15:32:50 <mlavalle> janzian: any updates?
15:33:38 <janzian> I've been looking through the test case for this bug, there's one written specifically for it
15:34:02 <janzian> It's passing in my environment, where I think the test case was written to fail as long as the bug is present
15:34:20 <janzian> So I'm not quite sure what's going on yet, but still figuring things out
15:34:39 <mlavalle> janzian: most of this failures show up with high level of concurrency, like in the gate
15:35:06 <mlavalle> janzian: very seldom they show up just running a test case individually
15:35:18 <mlavalle> like the one I explained above
15:35:34 <mlavalle> janzian: let's spend some time together looking at the logs
15:35:54 <janzian> mlavalle: Sounds like a plan
15:36:09 <mlavalle> janzian: thanks for working on this, though
15:37:05 <mlavalle> Last one today is https://bugs.launchpad.net/neutron/+bug/1509004
15:37:05 <openstack> Launchpad bug 1509004 in neutron ""test_dualnet_dhcp6_stateless_from_os" failures seen in the gate" [High,Confirmed]
15:37:46 <mlavalle> I haven't had time to track in down in Kibana. Has anyone taken a look?
15:38:26 <haleyb> it's been a while since i looked
15:38:54 <mlavalle> no problem. Now that I fiugured out the FkushEroor one, I'll start tracking this one
15:39:08 <mlavalle> Any other bugs from the team?
15:39:09 <haleyb> couple of times a day it looks like
15:39:59 <mlavalle> haleyb: thanks for updating the etherpad with those bugs :-)
15:40:35 <mlavalle> #topic Prefix Delegation
15:40:49 <mlavalle> baoli: hi! Any updates?
15:41:55 <mlavalle> I think that means no.... :-)
15:42:07 <john-davidge> :)
15:42:16 <mlavalle> #topic BGP Dynamic Routing
15:42:37 <mlavalle> I was able to exchange some messages with tidwellr
15:42:58 <mlavalle> He is with Target now, leading a team that is deploying OpenStack for Target
15:43:11 <mlavalle> he plans to continue playing with us
15:43:24 <mlavalle> but he's been too busy lately
15:43:38 <mlavalle> any other comments?
15:44:11 <mlavalle> #topic Routed Networks
15:44:29 <mlavalle> All the code for this was merged last week, as well as the doc update
15:44:59 <john-davidge> mlavalle: Great work! Thanks for your efforts
15:45:02 <mlavalle> john-davidge: I'll keep this subject in the agenda, since it seems we will come back to it during Pike for floating IPs
15:45:34 <john-davidge> mlavalle: Yeah, I'll post an RFE before the PTG, but I'm sure it will warrant plenty of discussion
15:45:50 <haleyb> mlavalle: yes, great job.  Did you see the recent tripleo email rgarding routed networks?
15:46:54 <haleyb> http://lists.openstack.org/pipermail/openstack-dev/2017-February/111385.html
15:47:08 <haleyb> https://review.openstack.org/#/c/425464/ in particular
15:47:26 <haleyb> i don't know if they need some help undertanding where things currently are
15:47:48 <john-davidge> haleyb: I hadn't seen that either, will take a look
15:48:00 <mlavalle> ok?
15:48:00 <mlavalle> #topic Open Agenda
15:48:00 <mlavalle> Any other topics from the team?
15:48:31 <mlavalle> haleyb: I'll take a look
15:49:07 <mlavalle> any other topics?
15:49:40 <john-davidge> not from me
15:49:58 <mlavalle> ok, team, thanks for your attendance and your follow up. Enjoy the rest of your days!
15:50:04 <mlavalle> #endmeeting