15:00:40 #startmeeting bgpvpn 15:00:40 Meeting started Tue Apr 5 15:00:40 2016 UTC and is due to finish in 60 minutes. The chair is matrohon. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:41 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 15:00:44 The meeting name has been set to 'bgpvpn' 15:00:55 hi everyone 15:01:02 hello 15:01:14 hi pcarver, vthapar, timirnich 15:01:47 I think tmorin is unavailable today, so I'll chair this meeting 15:02:18 #topic release 15:02:34 let's check what's needed to have our mitaka release 15:05:15 apart from vthapar's patch, I don't think we need anything else in the mitaka release 15:05:52 https://bugs.launchpad.net/bgpvpn/+bug/1560842 15:05:53 Launchpad bug 1560842 in bgpvpn "Add precommit validations in ODL driver" [Medium,In progress] - Assigned to Vishal Thapar (vthapar) 15:06:03 matrohon: somewhat related question, will we be backporting pre-commit to liberty? 15:06:44 vthapar, I don't think so, it's a feature addiction, not a bug fix, the stable team will probably not accept it 15:07:03 s/addiction/addition :) 15:07:25 vthapar, https://review.openstack.org/#/c/296289/3..4 15:07:37 vthapar, looks like only UTs are missing? 15:08:01 one more thing... I was testing the patch and ran into an issue, maybe coz my understanding of how precommit is wrong. 15:08:28 vthapar, what issue? 15:08:39 when I Call find_bgpvpns_for_network() it also returns the bgpvpn currently being associated. 15:08:51 so, if bgpvpns: is always true... 15:09:07 I had to do this to make it work: 15:09:09 - if bgpvpns: 15:09:09 + if bgpvpns and net_assoc['bgpvpn_id'] != bgpvpns[0]['id'] : 15:09:30 wanted to check before pushing another patch, if this is expected or some gap in my understanding. 15:10:18 vthapar, well, you just give this method a net_id, so it returns any associated bgpvpns 15:10:56 my assumption was that since it is fetching from db and this is precommit, it isn't in db yet and should return emtpy 15:11:21 vthapar, oh I see sorry! 15:12:07 vthapar, the find_bgpvpns_for_network() is called in the same session as the one used for the commit 15:13:01 vthapar, so in this session, the bgpvpn is already associated, but this session is not commited yet, so this entry is not really the DB, only in this sqlalchemy session 15:13:23 aha, got it. this will help. 15:13:24 vthapar, I'm calling pre_commit method, after the session.add 15:13:34 hey, sorry I am later 15:13:38 late 15:13:47 enikher, hi 15:14:02 I'll push this patch. take a look at rest of comments. if base code all looks fine, I'll get started on UTs also. I think create/update bgpvpn methods are good as is, so will get on the UTs for those first. 15:14:21 vthapar, fine thanks 15:14:47 with Niko back I got someone to pick brains for help with UTs :) 15:15:37 vthapar, the session.add is done here : https://github.com/openstack/networking-bgpvpn/blob/master/networking_bgpvpn/neutron/services/service_drivers/driver_api.py#L132 15:15:46 vthapar, and then I call the precommit 15:16:17 and the I exit the with statement, which trigger the session.commit 15:16:35 ok let's move on 15:16:55 #topic gate tests 15:17:38 tmorin is struggling at running bgpvpn+bagpipe tests in the gate : https://review.openstack.org/#/c/297260/ 15:18:57 enikher's past just need the approval from the neutron guys 15:19:02 s/past/patch 15:19:08 https://review.openstack.org/#/c/258522/ 15:19:49 mestery, dougwig, armax : can you help us having ^^ merge in infra? 15:21:44 matrohon: looking 15:21:47 We'll propably have to clen up what tempest tests we really want to run in our gate, but that can be done in our rc files, as net-ovn does 15:21:57 dougwig, thanks 15:23:49 #topic open discussion 15:24:06 does want anyone want to discuss about something? 15:26:45 ok let's close this meeting 15:26:51 #end-meeting 15:26:56 #endmeeting