07:05:24 #startmeeting networking_midonet 07:05:25 Meeting started Tue Dec 8 07:05:24 2015 UTC and is due to finish in 60 minutes. The chair is yamamoto. Information about MeetBot at http://wiki.debian.org/MeetBot. 07:05:26 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 07:05:28 The meeting name has been set to 'networking_midonet' 07:05:44 #link https://wiki.openstack.org/wiki/Meetings/NetworkingMidoNet agenda 07:06:07 not many items on agenda. this will be a short meeting i guess. 07:06:21 #topic Announcements 07:06:33 do anyone have announcements? 07:07:20 we are past mitaka-1. we are not doing milestones though. 07:07:35 I wonder if we should 07:07:38 not sure it qualifies as an announcement, but: we now have gerritbot in #midonet for networking-midonet :) 07:07:58 red_trela: thank you for setting it up! 07:08:18 red_trela: thanks! 07:08:28 #info we have gerritbot in #midonet for networking-midonet 07:09:20 move on 07:09:22 no accouncement from me 07:09:23 #topic Bugs 07:09:36 #link https://bugs.launchpad.net/networking-midonet/ 07:10:11 the number of NEW bugs seems growing but still small 07:10:34 no urgent bugs afaik 07:10:46 looks completely manageable 07:11:45 ryu25: i think you are going to set up bug deputy 07:12:30 yamamoto: I set it up for midonet but I think it should be different for networking-midonet due to difficulty of doing both 07:12:39 so let's set one up for networking-midonet 07:12:52 i see 07:12:53 I would be more than happy to volunteer for the first week 07:13:31 thank you! 07:13:32 yamamoto: currently there are not too many developers that are familiar with both projects, and they are tracked rather differently 07:13:55 though now that we are inching towards adopting more openstack-ways for midonet, this problem may go away 07:14:08 i agree 07:14:15 yamahata: is this meeting for neutron midonet only or for midonet as well? 07:14:21 let's start from this week. 07:14:27 irenab: plugin only 07:14:27 irenab: networking-midonet only 07:14:39 ok 07:14:41 there will be a separate one for midonet later today 07:15:39 yamamoto: I'm not sure what the best way to track the deputies. I suppose we can just update the meetings page like in Neutron 07:15:43 what do you think? 07:16:23 ryu25: i think it's fine. 07:16:31 ryu25: +1 07:17:09 #action yamamoto will create a bug deputy schedule on the meetings wiki 07:17:32 btw, did anyone follow the current neutron deputy discussion on the openstack-dev list? 07:17:36 #info bug deputy for the first week will be ryu25 07:18:08 red_trela: how current? url? 07:18:35 last week: http://lists.openstack.org/pipermail/openstack-dev/2015-December/081132.html 07:19:27 the summary of the thread is "the process is working well" afaik. 07:19:37 alright then :) 07:20:37 yeah i think generally people seem to be happy with the bug deputy process 07:20:38 anyone has a specific bug to discuss here? 07:21:10 no i have no comment on the bugs 07:22:07 lets move on 07:22:11 #topic Open Discussion 07:22:22 i have a question 07:22:36 do we want rfe bug process similar neutron? 07:23:17 yamamoto: I am not sure. I think having blueprint is enough 07:23:55 since networking-midonet is part of Neutron, do we have a choice? 07:24:18 red_trela: what do you mean by part of neutron? 07:24:21 blueprints are not convenient for discussion. it's ok as far as a single company just decides the direction, though. 07:24:25 in terms of governance 07:25:16 I think we do have a choice 07:25:27 in both processes, we still need drivers to agree on whether we accept them or not right? 07:25:59 irenab: you prefer blueprint because it's less heavy? 07:26:10 ryu25: yes 07:26:17 less steps in the process 07:27:06 neutron proccesses still evolve 07:27:07 probably "drivers" thing is more appropriate for midonet meeting, not this meeting. 07:27:20 lets keep with waht works well for networking-midonet 07:28:36 isn't the purpose of the new RFE approach, that the discussion can happen very early in the process, while the blueprint approach requires quite some upfront work that might be lost if it's not accepted or needs considerable change? 07:29:39 Also, can't they co-exist? if you want an early opinion or look for more helping hands, go with an RFE. Otherwise, with a blueprint. 07:29:43 red_trela: spec will require more work, blueprint at launchpad is short desctiption 07:30:30 so the extra step in the RFE process is the evaluation of the initial request? 07:30:31 ah, I thought it was the other way around...sorry for the noise :) 07:30:47 I think we should have policies defined, so contributors will be aware of the procedures. Any process can be adopted. 07:31:31 lets be agile and improve the process if it does not work well 07:32:19 irenab: do you mean to document the current procedure? 07:32:42 i can go either way. Why not just adopt what neutron is doing then? (RFE) 07:32:47 yamamoto: yes, something like neutron having policies defined in the source tree 07:32:56 irenab: do you volunteer? 07:33:19 yamamoto: I do not mind. 07:33:59 please add action for me to do so 07:34:27 i tend to think it's simpler to adapt neutron policies but i'm fine with either ways as far as someone volunteer to document. 07:35:12 #action irenab document the current procedure/policy for new features 07:35:17 irenab: thank you! 07:35:48 yamamoto: :-). I want to check similar projects, so be aligned with other vendor plugins 07:37:01 we should probably revisit this in 6 months or so, see how the new process is working for Neutron 07:37:04 irenab: good idea. i've heard odl has some procedure. 07:37:29 red_trela: I would even do it earlier 07:37:47 red_trela: by the end of Mitaka 07:37:52 +1 earlier, 6 months sound too long 07:38:24 yamamoto: thanks, will check with odl 07:38:52 it seems we finished agenda items. does anyone has anything more to discuss? 07:39:08 "6 months" was more of a random term anyway...I was just thinking that they might do a retrospect after Mitaka which we could use as a base 07:39:57 yes, one thing - once MidoNet is part of OpenStack (i.e. an accepted project), should we move networking-midonet from Neutron to MidoNet, governance wise? 07:40:38 we'll probably also have to see what the Neutron folks think about this, if we want to move 07:40:53 red_trela: i'm not sure. can't we belong to the both? 07:41:31 no 07:41:31 red_trela: I'm not sure if that will make sense since it's more tightly coupled with neutron for things like release cycles 07:41:44 \o/ 07:41:57 apuimedo: hello 07:41:59 sorry that I'm late 07:42:13 where are we on getting the plugin packages on distros? 07:42:24 rdo and Ubuntu cloud archives 07:42:38 ryu25: yea, that's what I thought - just been wondering whether people maybe think that it's the same people, so it should be governed together 07:42:49 but I'm fine with leaving it as-is :) 07:43:00 i definitely wouldn't mind consolidating if it makes sense 07:43:17 but i don't think it makes sense at this moment 07:43:30 ryu25: +1 07:43:54 got it 07:44:33 red_trela: let's see how it goes running two separate projects. We may feel differently later 07:44:39 sure 07:45:59 nothing more from me 07:46:07 apuimedo: i have no idea about packaging 07:46:15 ryu25: ^^ 07:46:48 i thought apuimedo answered his own question 07:50:00 maybe apuimedo needs to clarify the question 07:51:05 I'll send to ml 07:51:08 sorry ;P 07:51:31 np 07:51:56 see you later on the midonet meeting 07:51:56 3 07:52:09 #endmeeting