15:10:27 <tmorin> #startmeeting bgpvpn
15:10:28 <openstack> Meeting started Tue Nov  8 15:10:27 2016 UTC and is due to finish in 60 minutes.  The chair is tmorin. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:10:30 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:10:31 <matrohon> hi
15:10:33 <openstack> The meeting name has been set to 'bgpvpn'
15:10:36 <timirnich> hi folks
15:10:39 <pcarver> tmorin: Hi, was starting to wonder if we had a DST glitch
15:11:01 <bobmel> tmorin: Hi
15:11:03 <tmorin> pcarver: it would have been a possibility, but I'm just late :)
15:11:07 <tmorin> hi bobmel!
15:11:12 <doude> hi
15:11:39 <bfernando> hello
15:12:07 <tmorin> let's start... ?
15:12:31 <tmorin> #topic what hapenned since barcelona...
15:13:29 <tmorin> essentially work on two fronts: work on the OSC by doude, and various openstack CI config tweaks
15:14:14 <bobmel> tmorin: I started work on scenario test
15:14:35 <tmorin> ah yes, this is another item!
15:15:00 <bobmel> tmorin: Then I was a bit decked out by a Barcelona bug
15:15:23 <bobmel> so less progress than I had hoped
15:15:38 <tmorin> we also merged a minor trivial fix and a contrail driver bugfix (which needs to be backported)
15:16:11 <tmorin> bobmel: it's good that its started, don't hesitate to push an unfinished change if you want feedback
15:16:29 <bobmel> tmorin: Yes I plan to push a WIP patch soon
15:16:38 <tmorin> bobmel: very cool !
15:17:21 <tmorin> one last thing: I'm pushing a fix so that the job automatically adjusting the *requirements.txt files, works
15:17:32 <tmorin> see https://review.openstack.org/394936
15:17:46 <tmorin> let's discuss the stadium scorecard...
15:17:55 <tmorin> #topic meeting neutron stadium requirements
15:18:23 <tmorin> the draft for the assessment for networking-bgpvpn is at https://review.openstack.org/#/c/383906/
15:18:36 <tmorin> nice HTML at http://docs-draft.openstack.org/06/383906/9/check/gate-neutron-specs-docs-ubuntu-xenial/d125a41//doc/build/html/specs/stadium/ocata/networking-bgpvpn.html
15:18:58 <tmorin> I'd say we are doing quite good
15:19:16 <tmorin> many things marked as work in progress are soon to be finalized
15:19:36 <tmorin> the assessment conclusion currently is: There are some gaps that need attention most notably API documentation, client mappings and functional/scenario testing.
15:19:37 <pcarver> tmorin: I've been following that, but one thing I'm not clear on is what's the pass/fail criteria?
15:19:42 <tmorin> and the three have WIP items
15:19:48 <pcarver> Do we need "Y" on 100% of the items?
15:20:23 <tmorin> I don't know, I think this will be up to the Neutron PTL
15:22:02 <tmorin> I would think, from my discussion with him, that if we are advancing on the areas where we were lacking, we should be fine
15:22:04 <tmorin> and we are
15:22:35 <tmorin> the work on OSC is mostly landed (doude)
15:22:46 <tmorin> the work on API doc also (pcarver)
15:23:11 <tmorin> and the work on scenario testing is started and we should have a pointer as a "proof" some times soon (bobmel)
15:23:17 <tmorin> so I'm optimistic
15:25:14 <tmorin> what's our next topic ?
15:25:46 <matrohon> bobmel did you notice this WIP  : https://review.openstack.org/#/c/386418/6
15:26:16 <tmorin> ^ proposal for rally testing of the API
15:26:22 <bobmel> matrohon: Oh, no I wasn't aware of it
15:26:33 <tmorin> matrohon: not directly related to tempest scenario testing is it ?
15:27:00 <matrohon> bobmel, tmorin : not sure, i'm not a specialist of testing framework
15:27:15 <matrohon> but both are performng scenario tests
15:27:51 <matrohon> AFAIU tempest is functional test, while rally is stress test
15:28:05 <bobmel> tmorin: matrohon: Yes but Rally uses tempest afaik.
15:28:27 <matrohon> bobmel, ha,  was not aware of that
15:29:10 <tmorin> matrohon, bobmel: my understanding ate least is that rally tests do not replace tempest scenario testing
15:29:13 <bobmel> matrohon: well, I thought so. "knowing" too strong a word
15:30:13 <bobmel> tmorin: No I don't think so either. But I thought Rally made use of tempest collateral
15:30:42 <matrohon> well, the patch submitted in rally doesn't use what already exist in tempest for bgpvpn
15:31:06 <matrohon> but it would be a great idea for it to use tempest client
15:31:07 <bobmel> Anyway, I can look into that patch.
15:31:41 <tmorin> ok
15:31:51 <tmorin> next topic ?
15:32:04 <tmorin> #topic problems with networking-bgpvpn gate
15:32:13 <tmorin> our gate has had problems in the past 10-15 days
15:32:25 <tmorin> I've started investigating with infra folks
15:32:31 <matrohon> bobmel, ok, it was just a remark to let you know this work exists, I didn't want to unmotivate you :)
15:32:45 <bobmel> matrohon: No worries :-)
15:32:51 <tmorin> the job seems to crash systematically on one of the VM provider
15:33:09 <matrohon> tmorin, I've tried to follow your debug session, it relates to ipv6 test VM?
15:33:20 <tmorin> it ends up working after a few attempts, because it ends up being scheduled on a provider on which the crash does not occur
15:33:45 <tmorin> matrohon: the provider on which the problems appears happens to be a provider only giving v6 access to the VMs
15:33:54 <tmorin> matrohon: but it does not mean the problem is v6 related
15:34:38 <tmorin> tmorin: anyway, it was just to keep everyone posted, I'm sure will find out what is the problem...
15:38:18 <tmorin> next topic ?
15:38:32 <tmorin> #topic status on drivers - bagpipe
15:38:39 <tmorin> just a quick word on bagpipe
15:38:57 <tmorin> the stadium assessment for bagpipe is in progress as well
15:39:10 <tmorin> we have clarified the area that need work with the PTL
15:39:29 <tmorin> like for networking-bgpvpn, many items are in progress
15:39:44 <bobmel> tmorin: Is it a both makes it or none?
15:40:06 <tmorin> bobmel: not as far as I know
15:40:53 <tmorin> bobmel: what /might/ happen is that if the status of the non-BGPVPN-related portions of networking-bagpipe were not meeting stadium criteria, then the BGPVPN parts could be kept in networking-bgpvpn
15:41:18 <tmorin> bobmel: but I'm optimistic we'll wrap up what is needed so that networking-bagpipe does meet the criteria
15:41:42 <bobmel> ok, I see. That's a good way to handle any gaps if need be
15:43:06 <tmorin> bobmel: yes, that would be a last resort, but it means its not a "both or none" perspective
15:43:19 <tmorin> #topic driver status -- contrail
15:43:24 <tmorin> doude ?
15:44:00 <doude> I'm just starting to work on a new version of the contrail driver
15:44:13 <doude> a production ready one
15:44:38 <tmorin> that include work on both sides right ? (contrail-side + bgpvpn driver on neutron's side)
15:44:48 <doude> I started to work on a contrail patch to add the bgpvpn resource into the data model
15:45:01 <doude> yes, principaly on the contrail side
15:45:17 <tmorin> ok, very nice to hear!
15:45:40 <doude> for the moment that new driver will have same limit as the PoC one
15:45:58 <doude> we will see later to add router assoc and support l2 VPN
15:46:08 <tmorin> ok
15:47:33 <tmorin> next topic...
15:47:38 <tmorin> anything else to discuss ?
15:47:56 <tmorin> timirnich: anything else on OPNFV or ODL ?
15:48:02 <tmorin> #topic open discussion
15:48:51 <timirnich> No I don't have anythign today
15:49:43 <tmorin> matrohon: anything ?
15:49:57 <matrohon> not for me
15:50:47 <tmorin> ok, we are done then...
15:50:55 <tmorin> thanks everyone!
15:51:03 <tmorin> #endmeeting