15:11:11 <tmorin> #startmeeting bgpvpn
15:11:12 <openstack> Meeting started Tue Mar  7 15:11:11 2017 UTC and is due to finish in 60 minutes.  The chair is tmorin. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:11:13 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
15:11:15 <openstack> The meeting name has been set to 'bgpvpn'
15:11:17 <tmorin> hi everyone
15:12:21 <tmorin> let's see who is around ...
15:13:06 <tmorin> hi doude, bobmel, lhinds, pcarver, timirnich...?   (bgpvpn weekly IRC)
15:13:23 <weshay> Gabriele Cerami's move to Ireland
15:13:29 <weshay> sry
15:13:33 <lhinds> hi tmorin
15:13:47 <tmorin> hi luke
15:13:57 <lhinds> i will ping ricardo, once sec.
15:14:32 <bobmel> hi
15:14:51 <tmorin> lhinds: nothing "mandatory", but you haden't really been invited to come to this weekly meeting before, and I thought we might discuss/mention the puppet/tripleo work in progress
15:14:54 <tmorin> hi bobmel
15:16:17 <lhinds> makes sense tmorin , I think rnoriega should be here as well (although he has a meeting in parallel)
15:16:41 <rnoriega> tmorin, lhinds, here... but not 100%
15:17:36 <tmorin> hi rnoriega / np I'm just taking the opportunity of the weekly meeting to mention your puppet/tripleo work
15:17:55 <rnoriega> tmorin, awesome. Thank you very much.
15:18:13 <tmorin> we'll wait a bit to see who else would come, but we aren't always many
15:18:15 <rnoriega> tmorin, any question so far??
15:18:24 <rnoriega> tmorin, nice.
15:18:26 <tmorin> nope, we're just starting
15:20:41 <tmorin> ok, let's start
15:20:58 <tmorin> we're just me, bobmel, lhinds, rnoriega
15:21:17 <tmorin> here is a proposed "agenda" (light, this week)
15:22:10 <tmorin> a) API evolutions for Pike
15:22:28 <tmorin> b) work in progress on puppet+tripleo packaging
15:22:42 <tmorin> c) bagpipe driver work in progress
15:22:50 <tmorin> any suggestion ?
15:22:59 <tmorin> #topic API evolution for Pike
15:23:55 <tmorin> not much has been discussed in the past week
15:24:09 <tmorin> we'll need more feedback so that it can advance efficiently
15:24:36 <tmorin> see in particular:
15:24:37 <tmorin> https://etherpad.openstack.org/p/bgpvpn_port_routes
15:24:39 <tmorin> and
15:24:47 <tmorin> https://etherpad.openstack.org/p/bgpvpn_advanced_features
15:24:53 <tmorin> next topic ?
15:25:01 <tmorin> #topic work in progress on puppet+tripleo packaging
15:25:32 <tmorin> I wanted to mention here that rnoriega has been working on puppet and tripleo packaging of networking-bgpvpn
15:26:11 <tmorin> #link https://review.openstack.org/#/q/owner:rnoriega%2540redhat.com+and+message:bgpvpn
15:26:31 <tmorin> this is expected to land for Pike, right, rnoriega ?
15:26:32 <rnoriega> as an update to the community, the networking-bgpvpn is well in progress.
15:26:48 <bobmel> nice!
15:26:58 <tmorin> yes, this is really nice to see this support arrive !
15:27:12 <rnoriega> bagpipe driver is coming as the reference architecture, so we need to integrate it as well in tripleo in order to get the blueprint implemented
15:27:21 <rnoriega> tmorin, yes. It's planned for Pike
15:27:38 <tmorin> rnoriega: ok!
15:27:48 <rnoriega> I have two main concernes
15:28:30 <rnoriega> from the perspective of TripleO, integrating bagpipe would require installing the proper package (networking-bagpipe) and do all the configuration via puppet
15:29:12 <rnoriega> do you think is it feaseable to start that integration eventhough the bagpipe code is still not ported to the right repo???
15:29:50 <rnoriega> my second concern is about TripleO CI testing. But that could be discussed in a future meeting
15:30:47 <tmorin> rnoriega: I'm actively working on merging bagpipe-bgp into networking-bagpipe, so hopefully that will help
15:31:22 <tmorin> rnoriega: the location/format of configuration files will *not* change, so file templates could be started now
15:31:43 <lhinds> rnoriega: you can bring tmorin onto the reviews, and he can check vars / paths etc, i think that should help
15:31:56 <rnoriega> tmorin, awesome. So next week I'll start with bagpipe TripleO integration
15:31:59 <tmorin> rnoriega: no new python package will be added to networking_bagpipe, bagpipe_bgp will be included there as a submodule
15:32:16 <rnoriega> lhinds, yes! that's a good idea. I'll be adding tmorin to every patch as reviewer
15:32:38 <tmorin> rnoriega: cool! I'll have a look and raise warning whenever I find things that would be impacted by the migraion of bagpipe-bgp into networking-bagpipe
15:32:46 <tmorin> tmorin: works for me
15:33:12 <rnoriega> tmorin, lhinds, it's all set then!
15:34:02 <tmorin> #topic bagpipe driver
15:34:12 <tmorin> well, the key thing has been said above
15:34:28 <tmorin> work is in progress to merge bagpipe-bgp into networking-bagpipe
15:35:08 <bobmel> have you figured out how to get the git history?
15:35:15 <tmorin> yes
15:35:22 <tmorin> this has been done nicely
15:35:39 <tmorin> we have a feature/bagpipe_bgp_merge branch in networking-bagpipe repo
15:36:09 <tmorin> in this branch all the code from bagpipe-bgp github has is included, with the history, under bagpipe_bgp directory
15:36:29 <tmorin> I already have a few change in flight to move this code and other files in the right places
15:36:50 <tmorin> one unseen piece of work has appeared: now that exabgp will be a clean dependency of networking-bagpipe (instead of an embedded fork), it is supposed to be python3 friendly
15:37:21 <tmorin> and this is actually significant work, even to get only the unit tests working, because exabgp is not currently python3-friendly at all
15:37:36 <bobmel> oh, i see
15:37:41 <tmorin> but the authors would be happy to receive python3 fixes
15:38:11 <tmorin> so I'm working on improving exabgp python3 support at least up to a point where networking-bagpipe unit test pass
15:38:23 <tmorin> and improving python3 support in bagpipe-bgp at the same time
15:39:10 <tmorin> rnoriega, lhinds: I hadn't anticipated this, but hopefully, this will advance fast enough as well
15:39:49 <rnoriega> tmorin, ack
15:40:31 <tmorin> #topic open discussion
15:40:37 <tmorin> anything else to discuss today ?
15:41:50 <lhinds> not from me
15:42:02 <bobmel> not from either
15:42:11 <tmorin> ok, we're done for today then!
15:42:16 <tmorin> by everyone!
15:42:19 <tmorin> #endmeeting