15:00:23 <mlavalle> #startmeeting neutron_l3
15:00:29 <openstack> Meeting started Thu Oct 26 15:00:23 2017 UTC and is due to finish in 60 minutes.  The chair is mlavalle. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:00:33 <openstack> The meeting name has been set to 'neutron_l3'
15:00:39 <davidsha> o/
15:00:40 <Swami> hi
15:00:43 <haleyb> hi
15:01:34 <mlavalle> #topic Announcements
15:02:27 <mlavalle> Queens-1 was cut yesterday
15:03:55 <mlavalle> Next milestone is Dec 4 - 8
15:04:04 <mlavalle> about 6 weeks from now
15:04:24 <mlavalle> #link https://releases.openstack.org/queens/schedule.html
15:04:49 <mlavalle> Summit is a litlle more that a week from now
15:05:03 <mlavalle> will miss haleyb and davidsha :-(
15:05:15 <haleyb> have a beer or three for me :)
15:05:27 <mlavalle> will do
15:05:27 <davidsha> :(
15:05:28 <haleyb> and not that Fosters junk
15:05:31 <Swami> mlavalle: getting ready
15:05:40 <davidsha> Lol
15:05:40 <mlavalle> cool
15:05:47 <mlavalle> any other announcements?
15:05:51 <Swami> mlavalle: I got my visa at last after 25 days
15:05:58 <mlavalle> wow!
15:06:13 <mlavalle> is it a hard visa?
15:06:23 <mlavalle> in my case it is just an email
15:06:35 <Swami> mlavalle: the odd part is there is no communication and everything is silent and you don't know whom to contact.
15:06:59 <Swami> It seems they basically take at least 17 working days to process visa for other countries.
15:07:02 <mlavalle> ahhh, in the end is the same as mine, the airline has access to a system where they verify
15:07:20 <Swami> mlavalle: yes it is online approval.
15:07:25 <mlavalle> ok moving on
15:07:31 <mlavalle> #topic Bugs
15:07:36 <mlavalle> Swami: take it away
15:07:42 <Swami> mlavalle: thanks
15:07:51 <Swami> I have not seen any new bugs this week.
15:08:10 <Swami> I saw one bug, but that was an invalid bug with respect to MTU setting and FIPs.
15:08:20 <Swami> Most of our patches have merged.
15:08:30 <Swami> But we still have the Tempest test failure.
15:08:50 <Swami> I did not have time this week to work on it since I was busy with my downstream work.
15:08:54 <haleyb> and scenario test failure
15:09:12 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1717302
15:09:14 <openstack> Launchpad bug 1717302 in neutron "Tempest floatingip scenario tests failing on DVR Multinode setup with HA" [High,Confirmed]
15:09:23 <Swami> This is the bug that we need to address.
15:09:26 <haleyb> ah, the same, tempest and scenario
15:09:32 <Swami> haleyb: yes.
15:09:53 <haleyb> east-west test failures are all that's left
15:09:57 <Swami> Probably this week I can focus on this one.
15:10:15 * mlavalle and haleyb had the impression earlier Swami was busy doing something else :-)
15:10:21 <haleyb> Swami: there was also the ml2 dvr port bug, right?  needs an update?
15:10:39 <Swami> mlavalle: I was still working on openstack but on LBaas octavia.
15:10:52 <Swami> haleyb: yes, I am coming to that next.
15:11:07 * haleyb sits down and gets his popcorn
15:11:19 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1718345
15:11:20 <openstack> Launchpad bug 1718345 in neutron "ml2_distributed_port_bindings not cleared after migration from DVR" [Medium,In progress] - Assigned to Swaminathan Vasudevan (swaminathan-vasudevan)
15:11:26 * mlavalle is going to have a chat with Mike Johnson. Swami is only our
15:11:33 <Swami> I had a patch for it.
15:11:34 <mlavalle> ours^^^
15:12:20 <Swami> #link https://review.openstack.org/#/c/508808/
15:12:42 <Swami> While I was testing this patch, I am seeing that there is a mysql timeout error.
15:13:05 <Swami> So I need to revisit this patch. I have captured the error and provided a comment in the patch.
15:13:09 <mlavalle> so, WIP still
15:13:22 <Swami> haleyb: if you have any ideas please provide your comments on the patch.
15:13:42 <haleyb> i will take another look, hadn't gone through logs
15:13:59 <Swami> mlavalle: yes WIP still, since I wanted to test it and add a functional test to it, but it seems it fails functional test.
15:14:33 <Swami> #link https://bugs.launchpad.net/neutron/+bug/1716401
15:14:34 <openstack> Launchpad bug 1716401 in neutron "FWaaS: Ip tables rules do not get updated in case of distributed virtual routers (DVR)" [Undecided,New]
15:14:57 <Swami> mlavalle: reedip mentioned that he is busy with something else and he might not have time to fix this issue.
15:15:22 <mlavalle> yeah, I think he is oversubscribed
15:15:22 <Swami> mlavalle: So I will take a look at the FWaaS and DVR and see if I can fix this problem,
15:15:45 <Swami> Ok, that's all for today.
15:15:57 <Swami> My priority for this week is to fix the Tempest test and find out the issue.
15:16:03 <Swami> mlavalle: back to you
15:16:10 <mlavalle> Thanks for the update
15:16:26 <mlavalle> I don't have any other high priority bugs
15:16:29 <Swami> mlavalle: BTW, I will be out on a flight next week. So might not be able to attend the meeting next week.
15:16:46 <mlavalle> duly noted, thanks for the heads up
15:16:55 <mlavalle> any other bugs from the team
15:16:57 <mlavalle> ?
15:17:10 <davidsha> Not from me.
15:17:43 <mlavalle> ok, moving on
15:18:02 <mlavalle> #topic dvr_bridged
15:18:17 <mlavalle> did I use the correct term?
15:18:24 <mlavalle> davidsha: ^^^^
15:18:40 <davidsha> Close enough, think kickass is spelt wrong though :P
15:18:50 <mlavalle> LOL
15:18:57 <mlavalle> any updates on this?
15:19:22 <davidsha> I have floating IP working now and fixed the other issue I was having.
15:19:38 <davidsha> I'm just wrapping it up now and making it pretty for review
15:20:12 <davidsha> I'll move onto writing tests for it after the next PS
15:21:01 <davidsha> Should have the next patch set up tomorrow, pretty or not.
15:21:49 <davidsha> Thats, all from me. does anyone have any questions?
15:22:18 <Swami> davidsha: we are fine.
15:22:23 <mlavalle> not from me. will try to look at your patchset as soon as you push it
15:22:31 <davidsha> Thanks!
15:22:46 <mlavalle> #topic Multiple port bindings
15:22:59 <mlavalle> is sean-k-mooney around?
15:23:03 <davidsha> I have seans updates
15:23:09 <mlavalle> great
15:24:22 <davidsha> He's working on bindings atm, He's put up the nova vif objects and he'll be working on how to check the version of source and destination compute nodes
15:25:56 <mlavalle> Great! anything else from him
15:26:35 <davidsha> Not that I'm aware of.
15:27:33 <mlavalle> ok, on my side I merged in neutron-lib the resource defintion I need to notify event from the multiple port bindings service plugin
15:28:19 <mlavalle> this is being used in the code I am writing to notify ML2 that that bindings have been created / updated / deleted / activated
15:28:32 <mlavalle> making good progress
15:28:46 <mlavalle> that's all I have for this week
15:28:55 <mlavalle> #topic Open Agenda
15:29:38 <mlavalle> I couple of days ago I pinged haleyb and Swami about this spec: https://review.openstack.org/#/c/470596/
15:29:53 <mlavalle> seeking input to make sure we can implement this with DVR
15:30:17 <Swami> mlavalle: ok will take a look at it.
15:30:21 <mlavalle> The summary of it is that it proposes to add port fprwarding to floating ips
15:30:21 <haleyb> mlavalle: sorry, didn't get to it yet, will read today
15:30:26 <mlavalle> np
15:30:39 <mlavalle> let's spend a few minutes, if that's ok
15:31:16 <Swami> mlavalle: sure
15:31:31 <mlavalle> so we would create a fip and then add port forwarding rules to it
15:32:13 <mlavalle> to forward the traffic to a fixed ip / port combination
15:32:41 <mlavalle> could we use the new centralized fips for this?
15:32:46 <haleyb> it looks a little like load-balancing?
15:33:19 <mlavalle> it is more an effort to reduce the number of public IPs necessary
15:34:15 <Swami> mlavalle: If concerned about the public IP consumption then using the centralized fip will be the optimal one.
15:35:24 <Swami> mlavalle: haleyb: Yes looks to me like the lbaas where we reuse the same vrrp ip to contact different instances.
15:35:50 <mlavalle> ok, I just wanted to share my thoughts. Please give feedback in the spec
15:36:10 <mlavalle> I want to make sure this is implementable with DVR
15:36:17 <Swami> mlavalle: ok, still we need to see how the portbinding works if targeted with DVR.
15:36:36 <Swami> mlavalle: I will take a closer look at the spec and provide my comments in the spec.
15:36:47 <mlavalle> perfect, thanks
15:37:22 <mlavalle> the other patch I wanted to give you a heads up is this one: https://review.openstack.org/#/c/453458/
15:38:51 <mlavalle> There is an approved spec for this https://specs.openstack.org/openstack/neutron-specs/specs/backlog/pike/layer-3-rate-limit.html
15:39:09 <mlavalle> I just want to make sure the team is aware of this
15:39:37 <mlavalle> in case we have something to say
15:39:44 <mlavalle> I intend to review the code
15:40:17 <haleyb> mlavalle: i'm aware now :)  i am wondering why that doesn't have a reviewer assigned, or maybe it was on Kevin's watch and it was his?
15:40:45 <mlavalle> that was probably the case
15:40:50 <haleyb> or it's in the spec
15:41:23 <haleyb> hmm, just the RFE bug
15:41:42 <mlavalle> yeap
15:41:47 <davidsha> Spec is over a year old and not merged
15:42:00 <mlavalle> the spec is merged
15:42:15 <mlavalle> that is why we have https://specs.openstack.org/openstack/neutron-specs/specs/backlog/pike/layer-3-rate-limit.html
15:42:23 <mlavalle> it's in our backlog
15:42:38 <mlavalle> haleyb: would you like to be the approver?
15:42:45 <davidsha> mlavalle: Ah the other soec must have been abandoned then
15:42:49 <davidsha> https://review.openstack.org/#/c/337969/
15:44:02 <haleyb> mlavalle: i guess i could, sure
15:44:29 <haleyb> i stepped into that one
15:44:51 <mlavalle> haleyb: ok, I intend to help reviewing it, so I won't leave you alone
15:45:14 <mlavalle> I will look to see if there is a blueprint
15:45:24 <haleyb> thanks, it's probably almost done if the patch is that old
15:45:35 <mlavalle> and if not create one and add it to the Q-2 milestone
15:46:01 <mlavalle> I will also ask slawek to weigh in
15:46:26 <mlavalle> any other topics?
15:47:02 <haleyb> i have one comment
15:47:05 <mlavalle> sure
15:47:49 <haleyb> there are number (don't know how many) reviews fixing bugs that have fallen into the backlog, i'm hoping to get to looking at them the next couple of weeks, but might need other reviewers
15:48:04 <haleyb> most are smaller, but rather get them fixed than forget about them
15:48:17 <mlavalle> yeah, I'll be glad to help
15:48:21 <mlavalle> Thanks!
15:48:32 <mlavalle> add me to those reviews, please
15:48:35 <Swami> haleyb: sure I can help as well.
15:48:41 <haleyb> sometimes a one-line change requires more thought than a larger one
15:48:59 <mlavalle> LOL, yeah
15:49:21 <davidsha> Add me as well!
15:49:44 <haleyb> i had wondered why my review board was so long and noticed some of these just waiting for attention
15:49:57 <haleyb> that's all i had
15:50:15 <mlavalle> haleyb: did we cover all that we needed to discuss from the CI meeting?
15:50:57 <haleyb> mlavalle: yes, i think the one thing was the bug Swami was working on (tempest scenario), or was there another thing?
15:51:11 <mlavalle> nope, just making sure :-)
15:52:00 <mlavalle> ihrachys is pretty strict following up with the action items ;-)
15:52:12 <Swami> haleyb: got it
15:52:18 <haleyb> yes, i've been working on them
15:52:47 <mlavalle> and I don't want the L3 Olympus to be called out for missing an action item :-)
15:53:02 <haleyb> Swami: np, we want to have that job voting, and it's so close, just two areas, trunk and dvr
15:53:51 <mlavalle> ok team
15:53:52 <Swami> haleyb: agreed
15:53:58 <mlavalle> thanks for attending
15:54:04 <mlavalle> talk to you next week
15:54:14 <mlavalle> Swami: safe travels
15:54:14 <davidsha> Cya!
15:54:17 <Swami> bye
15:54:21 <mlavalle> #endmeeting