[02:17] <flaccid> are any ubuntu images or AKIs tickless?
[02:17] <flaccid> i.e. tickless kernel
[14:24] <SpamapS> smoser: the idea I'm throwing out in the pad is that its a bit silly for us, fedora, centos, rhel, sles, etc, to maintain our own systems for a symbolic os to provider to image id mapping..
[14:25] <SpamapS> smoser: I know we'd rather everybody use Ubuntu, but the best way to do that is to influence the process by which they choose images, by providing a widespread tool and a way for others to get on board with it.
[14:25] <smoser> SpamapS, yeah....
[14:25] <smoser> understandable.
[14:26] <SpamapS> Much like we're doing w/ cobbler really.
[14:26] <smoser> cloudimg-query ec2 us-east-1 oneiric i386 daily
[14:26] <smoser> cloudimg-query rackspace lucid release
[14:26] <smoser> the like.
[14:27] <smoser> and the backend service to push to to have that data consistent.
[14:28] <SpamapS> This whole process of figuring out the AMI to use is a bit silly isn't it?
[14:28] <SpamapS> I know we're solving this on Ubuntu, but it would be even better to solve this on Mac OS and Fedora and etc. as well.
[14:33] <smoser> i agree. other than, really, why would i want to make it easy for users to bring unnecessary pain by using something other than Ubuntu?
[14:33] <SpamapS> Its a flawed idea that you can get all users on Ubuntu
[14:34] <SpamapS> but you can make it easier for people to get Ubuntu
[14:36] <smoser> i was being sarcastic, SpamapS.  Really, why would you pick *this* time to assume i was being serious.
[14:39] <SpamapS> its 7:30 here.. no coffee.. no people.. social skills don't get started until 9:00
[14:54] <SpamapS> can an ebs root be resized and the root re-size recognized without rebooting an instance?
[15:15] <smoser> SpamapS, yes! magic it just happends on natty or greater.
[15:16] <smoser> http://ubuntu-smoser.blogspot.com/2011/07/getting-larger-root-volume-on-cluster.html
[15:18] <smoser> for other than hvm on ec2, though, it magically resized i think in maverick. as you dont need a trick to get the kernel to re-read the disk partition table (as there is none)
[15:19] <smoser> my most recent images published on canonistack hit this code path too.
[15:43] <SpamapS> smoser: ok, trying to make sure we have a story for "large storage" ensemble nodes
[17:50] <koolhead17> hi all
[18:40] <SpamapS> kirkland: whats the package that gives you a command on oneiric to say "give me a lucid instance" ?
[18:40] <SpamapS> smoser: ^^ ?
[18:40] <smoser> cloud-utils
[18:41] <smoser> the help is non-existant (i believe Kirkland said he'd do that)
[18:41] <SpamapS> :)
[18:42] <smoser> --dry-run will tell you what it is going to do
[18:43] <SpamapS> cool
[18:43] <kirkland> SpamapS: you want: ubuntu-ec2-run lucid
[18:43] <kirkland> SpamapS: personally, i always want "daily"
[18:43] <kirkland> SpamapS: so for me, it's "ubuntu-ec2-run lucid daily"
[18:44] <kirkland> SpamapS: i also usually want t1.micro
[18:44] <kirkland> SpamapS: so that's "ubuntu-ec2-run lucid daily t1.micro"
[19:31] <smoser> SpamapS, i just pushed usage to upstream cloud-utils branch: https://code.launchpad.net/~cloud-utils-dev/cloud-utils/trunk
[19:31] <smoser> that is probably worth pulling back to ubuntu