16:00:16 #startmeeting containers 16:00:17 Meeting started Tue Oct 14 16:00:16 2014 UTC and is due to finish in 60 minutes. The chair is adrian_otto1. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:18 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:21 The meeting name has been set to 'containers' 16:00:24 #link https://wiki.openstack.org/wiki/Meetings/Containers#Agenda_for_2014-10-14_1600_UTC Our Agenda 16:00:28 #topic Roll Call 16:00:33 Adrian Otto 16:01:56 \o 16:02:03 \o 16:02:22 hello there kebray and nelsnelson 16:02:44 greetings, adrian_otto1 16:03:45 Hello 16:03:51 hi iqbalmohomed 16:04:07 we will begin in just a moment 16:05:32 Hello! 16:05:40 ok, let's begin. Anyone who has not chimed in yet, you may type at any time to be recorded in attendance. 16:05:44 hi dguryanov 16:05:54 #topic Announcements 16:06:41 1) Next Tuesday, we have a scheduled team meeting. I am traveling, and should return on time to chair. 16:07:03 I have taken the day off as a vacation day. 16:07:19 would anyone like to be a chair in the event I can not attend? 16:08:30 the week after that I am speaking at Zendcon 16:08:58 and the week after that we are in Paris for the OpenStack Summit 16:09:51 the week after that is Veterans day in the US (Government holiday) 16:10:11 so… how would you all like to proceed with team meetings? 16:11:07 One option is to schedule our next meeting for Nov 18th 16:11:43 that would give us time to engage with the TC on how Magnum should be run (as part of OpenStack Compute, or as a Stackforge project) 16:12:04 and then regroup once we have clarity on that point. The next TC meeting is rather far out 16:12:32 When's the next TC meeting? 16:12:43 October 21 16:12:50 Okay. 16:13:02 but on controversial topics they rarely decide in one meeting 16:13:09 it usually takes two or three 16:13:47 so we would need to have most of the discussion in advance on the mailing list if we want a decision to happen on Oct 21. 16:14:57 ok, so another option would be for another team member to step in to chair team meetings in my absense 16:15:00 Thoughts? 16:16:09 Okay, any objections to scheduling our next meeting for Nov 18 at 22:00 UTC? 16:16:21 No ... that sounds alright 16:16:49 I argee 16:17:08 #agreed Our next meeting will be 2014-11-18 at UTC 2200 16:17:20 any other announcements from team members? 16:19:31 ok 16:19:47 #topic Review Action Items 16:19:59 * adrian_otto1 adrian_otto to attempt to identify Linux kernel patches that could enable use of nested containers in non-privileged mode 16:20:19 the meeting I had with Docker was rescheduled for later this week 16:20:30 so I will carry this action item forward to next week. 16:21:22 #action adrian_otto to attempt to identify Linux kernel patches that could enable use of nested containers in non-privileged mode 16:22:05 #topic Open Discussion 16:22:19 I have a question about networking for containers. 16:22:56 Is there any consdieration for multiple IPs for containers? 16:23:26 you mean multiple vnics inside the container? 16:23:31 yeah. 16:23:48 the containers use the IP addresses from the instance they are created in 16:23:57 so if you have a Nova instance with multiple IP 16:24:15 then you can use those with the containers that are create within that instance 16:24:20 +d 16:24:35 it is definitely possible to have a container with multiple network interfaces 16:24:44 yes. 16:24:47 adrian_otto1, iqbalmohomed: thanks. 16:24:49 I was studying this project recently: https://github.com/zettio/weave 16:25:18 they add an ethwe to docker containers in addition to the regular eth0 that docker sets up inside the container 16:26:02 iqbalmohomed: thanks. will take a look. 16:26:03 yes, I have looked at weave, and it looks promising, but I have a number of concerns about how it will scale beyond a few hundred containers 16:26:38 the system architecture (at first glance) appears to suffer from the N-Squared problem. 16:26:56 Why do we need weave when Neutron can already make overlay networks? 16:27:52 I don't think we do ... was merely offering it as evidence of multiple vnics inside a container :) 16:28:17 adrian_otto1: agreed. I am not sure how the overlay would scale 16:33:26 ok, any other topics for discussion today? 16:36:09 thanks everyone for attending! 16:36:13 #endmeeting