17:00:53 <LouisF> #startmeeting service_chaining
17:00:54 <openstack> Meeting started Thu Apr 13 17:00:53 2017 UTC and is due to finish in 60 minutes.  The chair is LouisF. Information about MeetBot at http://wiki.debian.org/MeetBot.
17:00:55 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
17:00:57 <openstack> The meeting name has been set to 'service_chaining'
17:01:00 <LouisF> hi all
17:01:03 <pcarver> hi
17:01:24 <LouisF> pcarver: hi
17:01:26 <doonhammer> Hi all
17:01:34 <bcafarel> hello everybody
17:01:34 <LouisF> doonhammer: hi John
17:01:39 <LouisF> bcafarel: hi
17:02:00 <vks1> hi all
17:02:13 <igordcard> hi all
17:02:22 <LouisF> #topic agenda
17:02:26 <igordcard> unfortunately I have to leave now
17:03:08 <LouisF> igordcard: hi
17:03:47 <LouisF> https://wiki.openstack.org/wiki/Meetings/ServiceFunctionChainingMeeting#Agenda_for_the_Networking-SFC_Meeting_.284.2F13.2F2017.29
17:03:48 <igordcard> please leave any questions at the service graph reviews, I'll read them and reply, but I have to go now! have a great meeting
17:04:01 <LouisF> igordcard: ttyl
17:04:07 <bcafarel> igordcard: hi and bye then ;)
17:04:20 <igordcard> aiming to have graphs done by next week
17:04:28 <LouisF> igordcard: great
17:05:23 <LouisF> #topic CLI
17:05:35 <LouisF> CLI in python-neutronclient https://review.openstack.org/#/c/409759/
17:06:08 <LouisF> mohan is working on this
17:06:41 <LouisF> making progress..
17:06:54 <LouisF> #topic API reference
17:07:07 <LouisF> pcarver: how is it going?
17:07:38 <pcarver> I think the API reference is done, just waiting on completion of the API definition (because Armando wants to make the merge of the ref dependent on the merge of the def)
17:08:08 <bcafarel> that's what I understood too
17:08:09 <pcarver> The API def is close to done, but I'm still doing some trial and error on the unit test because I don't really understand exactly how the unit test works.
17:08:47 <LouisF> pcarver: ok thanks
17:09:25 <LouisF> #topic Pike work
17:09:54 <LouisF> #topic Tap SF
17:10:35 <vks1> there is an issue with MPLS when trying to flood on multiple VXLAN tunne;
17:11:10 <vks1> LouisF: you are also in the email chain, Farhad had faced similar issue sometime back
17:11:12 <LouisF> vks1: i've seen some emails on that
17:11:44 <LouisF> vks1: did he find the problem?
17:12:14 <vks1> LouisF:  Things work fine when packet is sent to a single tunnel. But with multiple tunnel , OVS drops packet
17:12:27 <vks1> LouisF: Nope
17:12:56 <vks1> The same doesn't happens without using MPLS, I cheked that
17:13:21 <vks1> and no reply from community either :)
17:13:48 <LouisF> vks1: so this occurs with changes you made to implement the tap functionality?
17:13:55 <vks1> yeah
17:14:34 <vks1> LouisF: without mpls even TAP seems to be fine
17:14:53 <LouisF> vks1: can you send Farhad your code
17:15:17 <vks1> LouisF: I have sent him entire OVS dump
17:15:25 <vks1> *flow dump
17:15:44 <LouisF> vks1: when you "with/without mpls" what do you mean exactly?
17:16:25 <vks1> LouisF: I was suspecting there can be issue with the way I am programming flow
17:16:57 <vks1> LouisF: so to confirm that , it is not a problem I experimented traffic without using MPLS
17:17:07 <vks1> and it worked without any issue
17:18:05 <LouisF> can you forward me the ovs dump and changes to ovs
17:18:25 <vks1> LouisF: can you go through email once and let me know your thought ?
17:18:32 <LouisF> vks1: ok
17:18:42 <vks1> LouisF: you and cathy are already part of email
17:18:51 <vks1> LouisF: I will resend it
17:18:58 <LouisF> vks1: i will have a look at it
17:19:23 <LouisF> vks1: looks like we are making progress
17:19:50 <LouisF> vks1: thanks for your work
17:20:02 <vks1> LouisF: this multiple tunnel issue has consumed lot of time otherwise it would have been ready by now
17:20:37 <LouisF> vks1: yeah that happens
17:21:20 <LouisF> #topic OVN driver
17:22:02 <LouisF> doonhammer: i have not had a chance to review your ovn patch yet
17:22:22 <LouisF> doonhammer: i will do today
17:23:05 <doonhammer> Thanks I am ready to submit an updated version - just doing a few more tests
17:23:19 <LouisF> doonhammer: ok i will review that
17:23:28 <doonhammer> I think I have covered all the feedback from the OVN/OVS team
17:23:40 <LouisF> doonhammer: great
17:24:21 <LouisF> #topic non-transparent SF
17:25:21 <LouisF> i will post an update to the spec https://review.openstack.org/#/c/448284/
17:26:36 <LouisF> #topic Recent work
17:27:01 <LouisF> bcafarel: thanks for Fix UT failures with multiple flow classifiers
17:27:15 <LouisF> https://review.openstack.org/#/c/456602/
17:27:34 <bcafarel> this had been on my list for some time :) the rdo package had unit tests disabled because of that
17:27:54 <LouisF> bcafarel: again thanks
17:28:57 <LouisF> there is also Correct dl_type for non-SFC-proxied hops
17:29:07 <LouisF> https://review.openstack.org/#/c/455732/
17:29:47 <LouisF> igor needs this for the non SFC-proxy work
17:30:51 <LouisF> #topic other items
17:31:06 <LouisF> anything else to discuss?
17:31:35 <LouisF> who will be at the Boston summit?
17:32:13 <bcafarel> a few Neutron people from Red Hat will be, but not me
17:32:25 <LouisF> bcafarel: ok
17:32:55 <LouisF> ok i'll end early thanks all
17:33:07 <LouisF> bye all
17:33:10 <vks1> thanks all
17:33:11 <bcafarel> early is nice :) bye
17:33:19 <vks1> LouisF: please go thru email
17:33:27 <LouisF> vks1: will do
17:33:33 <LouisF> #endmeeting