[00:20] <rmiguel> hi friends, I have doubts concerning the nodes maas, the documentation does not make clear what procedure to be performed after the MAAS set the IP SERVER, NODE in the installation, after which it shuts the machine automoticatimente and does not proceed to installation of the NODE .
[00:21] <rmiguel> Grateful for help
[11:46] <xp1990> hi guys, got an issue regarding bootstrapping
[11:46] <xp1990> the system deploys an instance
[11:46] <xp1990> it sets up fine
[11:46] <xp1990> but it can't connect
[11:47] <xp1990> it hangs at attempting to connect to the instance
[11:47] <xp1990> after 10 minutes it terminates
[11:47] <xp1990> it seems that maybe there is nothing listening on the API port?
[12:14] <xp1990> Anyone?
[12:14] <marcoceppi> xp1990: can you ssh in to the instance? what provider are you using?
[12:15] <xp1990> I cannot SSH in. but I know ssh is running.
[12:15] <xp1990> I can ping the created machine, and in my openstack I can see the machine running
[12:15] <xp1990> the logs all show a correct set up
[12:15] <xp1990> but the last thing juju reports is Attempting to connect to 10.0.10.5:22
[12:15] <xp1990> Attempting to connect to 161.74.26.74:22
[12:15] <marcoceppi> xp1990: without connecting to the machine it's going to be hard to say what's wrong
[12:16] <xp1990> Yeah
[12:16] <xp1990> :\
[12:16] <marcoceppi> xp1990: if you can get to the logs, can you paste the cloud-init log?
[12:16] <xp1990> I only have what openstack tells me
[12:16] <xp1990> the cloud-init log finishes successfully.
[12:16] <xp1990> cloud-init boot finished at Wed, 19 Feb 2014 12:08:57 +0000. Up 31.23 seconds
[12:17] <xp1990> I have a feeling it could be a networking issue.....
[12:17] <xp1990> ....but maybe not?
[12:17] <xp1990> Because the instance is pingable
[13:18] <Dave-J> Hi Guys, Can anyone tell me, if JuJu is still limited to only booting Ubuntu based guests - I saw this post with patches to support CentOS, but I wondered if that had ever made it back into the trunk?
[13:18] <Dave-J> http://f.souza.cc/2012/07/using-juju-to-orchestrate-centos-based.html
[13:36] <tomixxx3> hi, with the default "admin" user, is it ok if i get "INteral Server Error" when i cklick to "Instances & Volumes", "Images & Snapshots" ?
[13:36] <tomixxx3> in openstack dashboard
[13:37] <marcoceppi> Dave-J: official support is still Ubuntu, but we're working on officially bridging other operating systems
[13:38] <marcoceppi> Dave-J: also, that fork of juju is quite old, it's based on juju 0.5 and the latest stable is 1.16.6
[13:38] <Dave-J> marcoceppi:  Thanks.  Yeah that's why I was wondering if any changes had made their way into the trunk, as I would like to try it out for a proof of concept.
[13:39] <Dave-J> marcoceppi: any idea on time frame for bridging to other OS ?
[13:40] <marcoceppi> Dave-J: it's probably right now to do, the only current limitations with other platforms are: You'll have to create your own metadata stream to use images other than Ubuntu, you'll have to make sure that image has cloud-init installed, you'll have to have a charm that isn't tied to Ubuntu (apt, etc)
[13:40] <marcoceppi> probable*
[13:43] <marcoceppi> Dave-J: so if you wanted to try, that's the steps I would look in to
[13:44] <Dave-J> marcoceppi: Thanks - so I'm already using juju on my private openstack deployment, and had to generate imagemetadata.json - so I'm comfortable with that.  I'll try to put a sample charm together and see if it works.  Thanks
[16:00] <bloodearnest> is there anyway to call juju-log outside a hook context? (without juju run, as am on 1.16)
[16:02] <mgz> bloodearnest: nope, but yo can just use rsyslog directly I suspect?
[16:04] <bloodearnest> mgz: yeah, I could. It's from a cron job installed by the charm
[17:59] <marcoceppi> bloodearnest: mgz well you /can/ but it's quite a lot of work
[22:13] <danob> hi all
[22:15] <marcoceppi> danob: hi o/
[22:16] <danob> how do i run unit-get privet-address/public-address in lxc container?
[22:18] <danob> i mean where is unit-get command?
[22:19] <marcoceppi> danob: you can only run run unit-get from a hook, you can't just run it
[22:21] <danob> marcoceppi: hmm thanks
[22:21] <lazyPower> danob: you can use debug-hooks environment during hook execution to run the command
[22:24] <danob> lazyPower: can you point me a doc? is debug-hook is not part of juju and charm-tools package ryt?
[22:24] <marcoceppi> danob: juju debug-hooks is part of juju core
[22:25] <marcoceppi> https://juju.ubuntu.com/docs/authors-hook-debug.html
[22:26] <danob> marcoceppi: thanks
[22:48] <danob> why juju is written in GO?
[23:07] <danob> found this [http://dave.cheney.net/wp-content/uploads/2012/08/august-go-meetup.pdf] it answered my question :)
[23:18] <nxvl> hi
[23:18] <nxvl> where can i see what a charm is actually installing
[23:19] <nxvl> i just spin off a juju gui and a mongodb instance, but i'm still a little lost
[23:19] <nxvl> i'm not sure if the mongodb charm i just deployed can be easily turned into a cluster later on
[23:20] <nxvl> and i want to deploy my own application behind nginx running over gunicorn and monitored by new-relic, but i'm quite clueless on where to find that information
[23:41] <marcoceppi> nxvl: mongo will scale properly
[23:42] <marcoceppi> you can view the charms here
[23:43] <marcoceppi> http://manage.jujucharms.com/charms/precise/mongodb
[23:43] <marcoceppi> Clicking on Repository will bring you to the code repo for that charm
[23:43] <marcoceppi> It also includes the README rendered which should answer a lot of your questions
[23:44] <nxvl> yup, starting by knowing i know the author :D
[23:47] <marcoceppi> nxvl: good, if it doesn't work then you know who to blame ;)
[23:48] <nxvl> marcoceppi: and who to poke an annoy :D
[23:48] <nxvl> that's one of the awesome things of being a former motu ;)
[23:54] <danob> how do i test run a charm?
[23:55] <danob> without creating any unit
[23:56] <danob> is it possible?