22:01:13 #startmeeting 22:01:14 Meeting started Tue Apr 3 22:01:13 2012 UTC. The chair is danwent. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:01:15 Useful Commands: #action #agreed #help #info #idea #link #topic. 22:01:29 #info Agenda: http://wiki.openstack.org/Network/Meetings 22:02:11 will hopefully be pretty quick, we have a lot to do before thursday :) 22:02:24 #info essex rc2 is out: https://bugs.launchpad.net/quantum/+milestone/essex-rc2 22:02:50 there are no currently known issues with RC2, and barring any major issues, it will be the final release. 22:03:01 woohoo 22:03:18 phew… finally, someone said something. I was beginning to wonder if I was talking to no one :) 22:03:46 Ok, I just sent an email to the ML about some outstanding docs issues 22:03:53 I'm always here for you, Dan :) 22:04:00 :P 22:04:12 ok, remaining docs tasks are: 22:04:24 1) create updated version of the 1.1 API spec 22:04:42 2) review admin guide for any changes since E-4 (should be fairly minor) 22:05:06 3) make sure use of Quantum with distro packaging is documented (as this is how most people will consume quantum) 22:05:20 If you have cycles, please let me know so you can pitch in here. 22:05:32 The other main area to help out is distro testing. 22:05:42 Since all packaging for quantum is new, there are some rough edges. 22:06:18 Fedora packaging has gotten a lot of love, so I suspect that is in pretty good shape, but we need more testing on it. 22:06:40 *nod* specifically, ucs plugin is basically untested 22:07:13 and for ovs, we don't have real tunnelling support, so no bother testing that 22:07:19 cdub: i'm upgrading my F16 kernel so I can hopefully test OVS plugin later today 22:07:43 danwent: great, we've done basic sanity testing on that and the bridge plugin. not cisco, nvp or ryu 22:08:01 but overall there has been a lot of work put into integrating quantum with fedora, much appreciated 22:08:22 yeah, it's coming together pretty well (bug reports happily accepted of course ;) 22:08:46 On Ubuntu, last night we got to the point where the quantum client could talk to the quantum server using the fakeplugin 22:09:10 but there seem to be issues with the agents not being packaged as binaries. 22:09:24 the underlying python classes are there, but there are no binaries, and no init scripts. 22:09:33 hmm, perhaps fallout from the setup.py changes 22:09:37 I'm working with Dave from canonical on this. 22:10:09 but the summary is, both platforms could use a lot more eyeballs, since a lot of this packaging is brand new 22:10:39 remember that if people can't use quantum with the main distros, people will conclude that the underlying software doesn't work, so testing distro integration is important. 22:10:57 any other distros that we should be looking at testing on? I believe there are debian packages, but I don't have any details on that. 22:11:03 should we be testing ubuntu packaging with 11.10 or 12.04? 22:11:03 I can help with ubuntu 22:11:12 great 22:11:15 12.04 22:11:18 ok 22:11:28 see the email I sent to the ML with the basic descriptions of the packages 22:12:03 I believe just doing apt-get install quantum-server will install the client and all other dependencies. 22:12:38 I'll shift my discussion of remaining packaging issues onlist, so we can avoid any duplicate troubleshooting/error reporting 22:12:49 anything else on distro testing? 22:13:09 realistically, we have ONE MORE DAY before the release, so we really need to focus efforts 22:13:42 #info devstack + quantum is still broken due to config change, please review: https://review.openstack.org/#change,6178 22:13:59 I don't really like this patch, but its a simple work around that doesn't require reverting the old patch. 22:14:10 if someone else from the quantum team could review it, I'd appreciate it. 22:14:23 I took a look at it earler.. I'll +1 it 22:14:24 then I can hopefully corner some devstack people to approvide it. 22:14:32 approve 22:14:52 man, I REALLY need to sleep more :) 22:15:19 don't misunderestimate sleep deprecation 22:15:21 debo-os: update on the devstack excercise script? plan is that devstack folks will merge after essex branches? 22:15:27 yes 22:15:38 cdub: i unfortunately have deprecated sleep 22:15:40 waiting for the merge ... 22:15:53 danwent: you just need to start drinking coffee.. :) 22:16:07 bhall: i'll never give in 22:16:11 danwent: in favor of sleepv2 i hope ;) 22:16:16 :) 22:16:18 first one is free buddy 22:16:19 never! non coffee drinkers untie! 22:16:32 untie? or unite? 22:16:37 yes 22:16:41 carlp needs coffee as well 22:16:45 ok, last topic, summit 22:16:59 #info hopefully everyone has seen the etherpad by now: http://etherpad.openstack.org/quantum-folsom 22:17:21 I sent an email to the ML yesterday proposing some definition to help frame the many discussion topics related to "L3". 22:17:47 so hopefully "L3" can be banned as a phrase in our discussion in favor of more concrete terms. 22:17:54 please provide feedback via email or etherpad. 22:18:13 I'm starting to send emails to people signed up as drivers for the key topics, asking them to register summit sessions. 22:18:27 will do, I'm hoping to have a proposal tomorrow afternoon to the list 22:18:40 I'll probably send an email out tomorrow about critical areas where we have yet been able to trick (er, i mean convince) someone into running the sssion. 22:19:00 carlp: new topic, or just opinions on an existing topic? 22:19:02 danwent: i expect we'll have ours in w/in the next day 22:19:19 opinions on the existing topic 22:19:23 ok, any new topics, we definitely want to get those proposed ASAP. 22:19:48 opinions on existing topics can continue to flow in between now and summit 22:20:05 new meaning not currently on etherpad? 22:20:18 but remember, please don't let summit stuff distract you from the release work…. there will be plent of time to do stuff after the release 22:20:22 cdub: yes 22:20:48 danwent: ok, thanks 22:21:13 #topic open discussion 22:21:21 anything? 22:21:36 please think about how you can pitch in and help with testing + docs. 22:21:54 I'll see what I can do, I'm somwhat swamped with work ATM 22:22:14 I hate it when all the deadlines start converging 22:22:14 carlp: yup :) 22:22:20 quick q: there are some topics on the list for sessions that almost cover it, but nothing specific for nova network parity. 22:23:04 aabes: there won't be a specific session on it, but many of the sessions combine to approach that goal. is there a specific topic you see as missing? 22:24:24 ok, well aabes, feel free to sync with me offline. I'll call the meeting to a close so everyone else can take off :) 22:24:27 #endmeeting