[00:25] <nemothekid> Do I install juju on a single machine (like a CI server) or do I install juju on every developers machine?
[00:27] <hatch> nemothekid, you install the juju client on every developers machine
[00:27] <Makyo> nemothekid: The client talks to a central "bootstrap" node that controls all of the units involved in the environment.
[00:27] <Makyo> (or any number of environments)(
[00:39] <nemothekid> Makyo: okay, can I have the GUI exist on the "bootstrap" node and do everything from there?
[00:40] <nemothekid> basically all I want to have is the ability to just push to a git repo, and then for the bootstrap node to figure everything out
[00:41] <Makyo> nemothekid: you can have the GUI on the bootstrap node with `juju deploy cs:precise/juju-gui --to 0`
[00:45] <Makyo> nemothekid: as for the second part, others may have suggestions for post-receive hook type stuff wrt git repos? What one could do is have the config-changed hook pull HEAD of master, then the post-receive hook in your repo do something like, say, juju set <service> last-update-timestamp=<curr time>
[00:45] <nemothekid> thats perfect
[00:45] <nemothekid> really glad I cam across this project then
[00:46] <Makyo> nemothekid: however, the bootstrap node doesn't necessarily listen for changes, that'd be something you'd have to automate outside of juju.  Another option is to, say, up the revision number of your charm and then upgrade the charm for your service.
[00:46] <Makyo> There are a few options, I guess :)
[00:46] <Makyo> Cheers
[00:47] <Makyo> The config-changed idea would also work for CI stuff, too, I guess.

[01:44] <Makyo> hatch: https://www.exratione.com/2013/02/nodejs-and-forever-as-a-service-simple-upstart-and-init-scripts-for-ubuntu/
[07:20] <nesusvet_> Hello everyone!
[10:11] <nesusvet_> I tried to deploy the MAAS environment, and run into the issue when I see assigned server under MAAS, but juju can't recognize it. I did the "destroy-environment " command and deploy everything again via "juju bootstrap" but the same issue happens again and again
[10:12] <nesusvet_> I see only one sever of two
[11:15] <Jujufy> hey, how long should a `juju status' take?
[12:46] <nesusvet_> Jujufy, sorry for the delay, it shows information immediately!
[13:25] <Jujufy1> thanks for answering, although the webchat stopped responding
[13:26] <Jujufy> ok finally set up a proper IRC client
[13:28] <Jujufy> i am updating 13.04 -> 13.10 and i spotted some maas updates
[13:29] <Jujufy> will try to see if juju feels better afterwards
[13:46] <Jujufy> upgrade done but nothing changed with juju
[13:46] <Jujufy> i'm trying to get a time for it
[13:51] <Jujufy> got the result back: http://paste.ubuntu.com/6283048/
[13:53] <Jujufy> i already have an environment bootstrapped
[13:53] <Jujufy> it shows up in `juju env'
[13:54] <Jujufy> i'm trying `juju status' with `-e' but it looks like it will turn out the same
[13:55] <Jujufy> meanwhile, the original bootstrapping process did convert one maas node to `Allocated to root'
[13:56] <Jujufy> so i guess at least that worked
[14:47] <Jujufy> got the same error , this time it does mention my environment
[14:47] <Jujufy> the timeout seems to be 10min
[15:57] <marcoceppi> Jujufy: What version of juju did you bootstrap with?
[19:55] <danielesalatti> Hi all! I need a little help… How can I tell juju to use a t1.micro when deploying? I tried bootstrapping with --constraints "cpu-count=0 mem=512M", "cpu-power=0 mem=512M" and "instance-type=t1.micro" but none of these works for me…
[20:40] <marcoceppi> danielesalatti: cpu-power=0 mem=128M should have done it