22:00:51 <adrian_otto> #startmeeting containers
22:00:52 <openstack> Meeting started Tue May  5 22:00:51 2015 UTC and is due to finish in 60 minutes.  The chair is adrian_otto. Information about MeetBot at http://wiki.debian.org/MeetBot.
22:00:54 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
22:00:56 <openstack> The meeting name has been set to 'containers'
22:01:08 <adrian_otto> #link https://wiki.openstack.org/wiki/Meetings/Containers#Agenda_for_2015-05-05_2200_UTC Our Agenda
22:01:13 <adrian_otto> #topic Roll Call
22:01:20 <adrian_otto> Adrian Otto
22:01:23 <tcammann_> o/
22:01:26 <Tango> Ton Ngo
22:01:45 <juggler> o/
22:01:45 <jay-lau-513> jay-lau-513
22:01:52 <hongbin> o/
22:01:53 <rpothier> o/
22:02:10 <apmelton> o/
22:02:12 <thomasem> o/
22:03:06 <adrian_otto> hello tcammann, Tango, juggler, hongbin, rpothier, apmelton, and thomasem
22:03:14 <juggler> hey everyone
22:03:26 <thomasem> howdy
22:03:27 <adrian_otto> hi jay-lau-513
22:03:38 <juggler> no sdake today?
22:03:43 <adrian_otto> he's around
22:03:50 <juggler> :)
22:04:06 <adrian_otto> but his client is dicsonnected
22:04:26 <adrian_otto> he mentioned he had something he was heading out to do
22:04:40 <adrian_otto> ok, let's begin
22:04:58 <adrian_otto> #topic Announcements
22:05:00 <adrian_otto> 1) Reminder: Our IRC Meeting will be skipped on 2015-05-19 because we will be at the Vancouver Design Summit
22:05:06 <adrian_otto> 2) We now have a kilo-stable branch for magnum and python-magnumclient:
22:05:12 <adrian_otto> #link https://review.openstack.org/#/admin/projects/openstack/magnum,branches magnum branches
22:05:18 <adrian_otto> #link https://review.openstack.org/#/admin/projects/openstack/python-magnumclient,branches python-magnumclient branches
22:05:30 <adrian_otto> the purpose of this branch is to have a place to backport bugs
22:06:00 <adrian_otto> if you try to put new features in that branch, or fix tech debt there, your patch shall be marked by a core reviewer with a -2
22:06:17 <adrian_otto> all new features belong only in our master (Liberty) branch
22:06:20 <adrian_otto> any questions?
22:06:27 <juggler> i have a q
22:06:32 <adrian_otto> proceed juggler
22:06:54 <adrian_otto> I will also offer one more restriction
22:06:56 <juggler> so when I implement filed bug: https://bugs.launchpad.net/magnum/+bug/1451678, should it go into branches?
22:06:56 <openstack> Launchpad bug 1451678 in Magnum "Add link to dev-manual-devstack.rst into document dev-quickstart.rst" [Undecided,New] - Assigned to P Rivera (juggler)
22:07:26 <adrian_otto> your bug must be triaged by a bug manager prior to merging any code in our kilo-stable branch
22:07:36 <adrian_otto> this is contrary to our policy for master
22:07:49 <juggler> cool. let me know where it needs to go!
22:07:50 <adrian_otto> where the bug can be opened, patched, and committed with no action by anyone but you
22:08:02 <adrian_otto> let me read the bug, one moment
22:08:20 <adrian_otto> a documentation bug like that belongs in *both* branches
22:08:28 <adrian_otto> so it will actually be two commits
22:08:39 <juggler> got it
22:08:47 <adrian_otto> as we have no intent to ever merge kilo-stable back into master (liberty)
22:08:52 <adrian_otto> ok, any more questions?
22:09:22 <juggler> thanks
22:09:41 <adrian_otto> ok, rule of thumb, if it's classified as a CRITICAL bug, it will get backported
22:09:54 <adrian_otto> the lower priority levels are subject to discression
22:10:11 <adrian_otto> *discretion
22:11:00 <adrian_otto> 3) We have tagged a 2015.1.0 release of magnum and python-magnumclient
22:11:13 <adrian_otto> #link https://github.com/openstack/magnum/releases/tag/2015.1.0 today's release
22:11:29 <adrian_otto> I plan to make an email announcement for this as well.
22:12:01 <adrian_otto> that concludes prepared announcements. Any announcements from team members?
22:12:32 <adrian_otto> ok, advancing topics
22:12:38 <adrian_otto> #topic Blueprint/Task Review
22:12:47 <adrian_otto> any work items requiring team discussion today?
22:13:40 <adrian_otto> ok, advancing topics
22:13:45 <adrian_otto> #topic Open Discussion
22:14:02 <adrian_otto> I want to take a moment to point out how awesome our Kilo release was
22:14:09 <adrian_otto> 6 big features
22:14:21 <adrian_otto> 133 improvements and bug fixes
22:14:33 <adrian_otto> that's more than any open source release I have ever cut.
22:14:47 <juggler> excellent
22:14:48 <adrian_otto> well done team. The velocity here is really impressive
22:15:23 <adrian_otto> we now have 40 engineers who have contributed (commits) from 18 affiliations
22:15:26 <juggler> how's the progress on the room topics for the summit?
22:15:46 <adrian_otto> we have an etherpad we touched on last week:
22:16:24 <juggler> 40 engineers!
22:16:57 <adrian_otto> darn where did I put that URL, one sec, looking in logs
22:17:14 <juggler> etherpad.openstack.org/p/liberty-magnum-topics
22:17:20 <juggler> is that it?
22:17:37 <adrian_otto> #link https://etherpad.openstack.org/p/liberty-magnum-topics Liberty Design Summit Topics
22:17:43 <adrian_otto> yes, thanks juggler
22:17:53 <juggler> np
22:18:35 <juggler> offhand will anyone be recording one or more of those sessions?
22:18:36 <jay-lau-513> juggler i cannot open this link
22:18:58 <adrian_otto> I know the conference sessions will be recorded. I am not sure about the design sessions.
22:19:05 <jay-lau-513> juggler an error occured while loading the page
22:19:13 <adrian_otto> possibly the fishbowls, but I doubt the workrooms
22:19:14 <juggler> jay: try adrian_otto's link
22:19:52 <jay-lau-513> juggler same error
22:19:54 <adrian_otto> let's land some votes next to the proposed topics
22:20:14 <jay-lau-513> does anyone else can open the L topic link?
22:20:20 <juggler> jay: dunno...seems to work for adrian and I
22:20:30 <tcammann_> works for me too
22:20:48 <jay-lau-513> ok, let me switch a browser
22:21:00 <adrian_otto> https://etherpad.openstack.org/p/liberty-magnum-topics is working for me
22:21:12 <jay-lau-513> chrome working, thx
22:21:14 <adrian_otto> I see 10 people connected
22:21:37 <juggler> <--voting
22:21:57 <adrian_otto> juggler got creative with names!
22:22:06 * adrian_otto chuckles
22:23:20 <apmelton> can lines 56 and 58 be combined?
22:23:28 <apmelton> they seem like the same talk worded differently
22:24:36 <shhfeng> works for me, by  chrome browser
22:25:01 <adrian_otto> I consolidated networking into the other topic
22:25:16 <adrian_otto> so I am going to move the other one to our workroom section
22:25:49 <apmelton> perhaps "Design for container-to-container networking" and "Networking / tenant network integration" could simply be a networking breakout session
22:26:10 <apmelton> for reflecting as a team on what is learned in a fishbowl on networking?
22:27:04 <adrian_otto> so we cover "next features" in a fishbowl
22:27:16 <adrian_otto> possibly breaking out networking into it's own… not sure yet
22:27:26 <juggler> adrian_otto lol, thanks! :)
22:27:29 <adrian_otto> then we take the feedback from that, and explore designs to meet those
22:27:54 <adrian_otto> so we cover the topic twice, in the large and small room formats respectively
22:28:13 <adrian_otto> apmelton: make sense?
22:28:22 <apmelton> yup
22:28:46 <Tango> Any thought about storage yet?
22:29:01 <adrian_otto> Tango, let's put that up on the etherpad for sure
22:29:21 <juggler> +1 Tango
22:31:02 <adrian_otto> Tango, see "Storage Features". Any more detail that belongs in there?
22:31:19 <jay-lau-513> any thoughts for scheduler? I know that we want to use swarm to manage native docker cluster, do we need more discussion for this?
22:31:30 <jay-lau-513> such as mesos as magnum scheduler?
22:31:47 <jay-lau-513> or Gantt, seems Gantt is dead with slow progress
22:32:21 <adrian_otto> we definitely need a workroom for that one, jay-lau-513
22:32:29 <tcammann_> +1
22:33:14 <jay-lau-513> another I want to mention is that shall we have some discussion for how to make magnum as a generic container-as-a-service just like nova does for IaaS.
22:33:49 <jay-lau-513> end user do not need to care the backend but only submit request, the backend will provide container for user
22:33:51 <juggler> please observe the use of Metric system in the topic title "Magnum at 15 240 meters" :)
22:33:55 <jay-lau-513> just like nova: user do not care what hypervisor they are using, just submit request to get vm/pm
22:34:25 <adrian_otto> jay-lau-513: I added "External Scheduler Support" for your consideration
22:34:52 <jay-lau-513> adrian_otto cool
22:35:08 <tcammann_> adrian_otto: Line 69 has "Design for Magnum Scheduler", could we combine those?
22:35:18 <tcammann_> or are they signifcantly different enough
22:36:53 <adrian_otto> tcammann: good catch. combined those
22:36:58 <juggler> Does the tenant network integration look like it go under some other main topic?
22:37:15 <juggler> hmm
22:37:23 <tcammann_> A session on container snapshotting or imaging?
22:37:32 <thomasem> It will likely fall under the overlay network topic
22:37:35 <thomasem> since it's a similar concept
22:37:37 <tcammann_> i.e. integration with glance
22:38:04 <adrian_otto> tcammann: Added to line 37
22:38:05 <juggler> cool
22:38:18 <thomasem> I'd say that's the main concern with using overlay networks like Weave/Flannel
22:38:35 <thomasem> Those wouldn't work well with neutron tenant networks
22:38:42 <thomasem> afaict
22:39:26 <thomasem> The concern from our last discussion being how confusing it'd be to have two completely separate isolated network solutions available in OpenStack when using Magnum
22:39:29 <adrian_otto> ok, I sorted the fishbowls by botes
22:39:34 <adrian_otto> *votes
22:39:36 <juggler> how's about racking and stacking the top +1s in the Workrooms, and subtopicing the lower top +1s as needed?
22:39:59 <adrian_otto> good idea. doing that now
22:41:28 <juggler> just a thought: maybe put the contributing and Everything You Know stuff into the Plenary
22:42:08 <juggler> figure that people who want to learn about Magnum will attend the Plenary as an overview...
22:43:05 <juggler> that also whittles the # of topics in the fishbowls down to 4
22:47:00 <tcammann_> Another note, we really need to fix the doc link in the main magnum README, it still 404s.
22:47:13 <tcammann_> https://bugs.launchpad.net/magnum/+bug/1412380
22:47:13 <openstack> Launchpad bug 1412380 in Magnum "Magnum Documentation link report 404 error " [Medium,In progress]
22:49:02 <jay-lau-513> tcammann_ yes, it is the first link in Magnum README
22:51:41 <adrian_otto> tcammann: good catch on bug 1412380
22:51:42 <openstack> bug 1412380 in Magnum "Magnum Documentation link report 404 error " [High,In progress] https://launchpad.net/bugs/1412380
22:52:54 <jay-lau-513> another is that can sdake share something for magnum&kolla integration, seems sdake is doing something for containzing magnum service
22:52:54 <adrian_otto> ok, so on the summit topics I think I have enough input from you to proceed with filling out the program
22:53:17 <adrian_otto> if you feel strongly that something has been missed, please contact me ASAP
22:53:47 <adrian_otto> jay-lau-513: it might be more powerful to do that as a video and post it on our wiki page
22:53:53 <adrian_otto> prior to the summit
22:54:12 <jay-lau-513> adrian_otto sure, this can make magnum installation more easier
22:54:40 <adrian_otto> well, as long as you have the right number of network ports and routable ip address ranges ;-)
22:54:57 <jay-lau-513> exactly
22:55:29 <tcammann_> does sound like a tricky setup, a video would be great
22:55:33 <adrian_otto> ok, let's wrap up for today
22:55:39 <adrian_otto> but first
22:55:52 <adrian_otto> how many of you have skills to make a really good screencast?
22:56:07 <adrian_otto> a good radio voice, and can record a demo?
22:56:30 <adrian_otto> if that sounds like you, let's coordinate to get a Magnum demo on video.
22:57:02 <Tango> what do you mean by screencast?
22:57:15 <adrian_otto> you record your screen giving a demo, with a voice-over.
22:57:21 <adrian_otto> but you do not appear on camera
22:57:40 <adrian_otto> make sense Tango?
22:57:54 <Tango> ah ok, I can help with that, beendoing that a few times.
22:58:15 <adrian_otto> cool, I will PRVMSG you
22:58:20 <adrian_otto> thanks everyone for attending today
22:58:21 <Tango> someone can record the voice and I can dub over
22:58:32 <adrian_otto> our next meeting is 2015-05-12 at 1600 UTC
22:58:52 <juggler> good seeing you all again
22:58:55 <adrian_otto> #endmeeting