Friday, 2011-11-04

[2011/11/04 00:09:27] @ Log started by gepetto
[2011/11/04 00:09:27] @ Quit: hooper: Ping timeout: 253 seconds
[2011/11/04 00:16:11] @ jaytaph joined channel #puppet
[2011/11/04 00:26:45] @ Quit: jbraeuer: Quit: jbraeuer
[2011/11/04 00:28:01] @ caviwipes joined channel #puppet
[2011/11/04 00:29:51] @ Quit: rcrowley: Quit: Computer has gone to sleep.
[2011/11/04 00:32:19] @ DJClean joined channel #puppet
[2011/11/04 00:38:48] @ adimania joined channel #puppet
[2011/11/04 00:41:18] @ lkoranda joined channel #puppet
[2011/11/04 00:41:47] @ ckauhaus joined channel #puppet
[2011/11/04 00:46:24] @ jedi4ever joined channel #puppet
[2011/11/04 00:47:28] @ Quit: z3ndrag0n: Ping timeout: 255 seconds
[2011/11/04 00:48:37] @ Stenbryggen joined channel #puppet
[2011/11/04 00:51:54] @ Quit: mosen: Quit: Leaving
[2011/11/04 00:52:22] @ Quit: adimania: Ping timeout: 252 seconds
[2011/11/04 00:55:09] @ brtl joined channel #puppet
[2011/11/04 00:55:27] @ adimania joined channel #puppet
[2011/11/04 00:55:44] @ jbraeuer joined channel #puppet
[2011/11/04 00:57:28] @ Quit: jbraeuer: Client Quit
[2011/11/04 00:58:45] @ Quit: bmahe: Remote host closed the connection
[2011/11/04 00:59:22] @ ddoc961 is now known as ddoc961[afk]
[2011/11/04 01:01:49] @ jodok joined channel #puppet
[2011/11/04 01:01:50] @ Quit: jodok: Changing host
[2011/11/04 01:01:50] @ jodok joined channel #puppet
[2011/11/04 01:04:05] @ Quit: adimania: Ping timeout: 240 seconds
[2011/11/04 01:04:25] @ Quit: jhelwig_: Quit: Zoom!
[2011/11/04 01:05:25] @ zz_jhelwig joined channel #puppet
[2011/11/04 01:08:19] @ toi joined channel #puppet
[2011/11/04 01:08:43] @ Quit: jhelwig: Quit: jhelwig
[2011/11/04 01:08:54] @ mmoya joined channel #puppet
[2011/11/04 01:09:30] @ Quit: caviwipes: Quit: Computer has gone to sleep.
[2011/11/04 01:09:35] @ pmorillon joined channel #puppet
[2011/11/04 01:10:26] @ zz_jhelwig is now known as jhelwig
[2011/11/04 01:10:32] @ Quit: jhelwig: Changing host
[2011/11/04 01:10:32] @ jhelwig joined channel #puppet
[2011/11/04 01:10:49] @ Vulcan[OMNI] joined channel #puppet
[2011/11/04 01:14:35] @ Quit: mmoya: Ping timeout: 258 seconds
[2011/11/04 01:14:36] @ Quit: jsurfer: Quit: jsurfer
[2011/11/04 01:14:59] @ mmoya joined channel #puppet
[2011/11/04 01:16:33] @ Quit: carpenoctem: Ping timeout: 265 seconds
[2011/11/04 01:16:34] @ Quit: the_ramink: Ping timeout: 260 seconds
[2011/11/04 01:17:56] @ Quit: kubicek: Excess Flood
[2011/11/04 01:18:45] @ kubicek joined channel #puppet
[2011/11/04 01:19:30] @ Quit: mpdude: Remote host closed the connection
[2011/11/04 01:20:00] @ csanchez joined channel #puppet
[2011/11/04 01:22:01] @ caviwipes joined channel #puppet
[2011/11/04 01:23:57] @ raphdg joined channel #puppet
[2011/11/04 01:27:09] @ KristianDK joined channel #puppet
[2011/11/04 01:28:00] <KristianDK> Hey, I'm having troubles with checking if a specific key exists in a hash, i tried it like this, but without too much success - http://dpaste.com/647143/ - how do is it correctly done?
[2011/11/04 01:28:33] @ jaytaph_ joined channel #puppet
[2011/11/04 01:30:29] @ Quit: jedi4ever: Quit: jedi4ever
[2011/11/04 01:32:17] @ Quit: kenperkins: Quit: Computer has gone to sleep.
[2011/11/04 01:32:27] @ Quit: jaytaph: Ping timeout: 244 seconds
[2011/11/04 01:32:58] @ Quit: jaytaph_: Ping timeout: 244 seconds
[2011/11/04 01:35:44] @ Razique joined channel #puppet
[2011/11/04 01:37:21] @ trickyf joined channel #puppet
[2011/11/04 01:38:07] @ mpdude joined channel #puppet
[2011/11/04 01:38:09] @ Vasik\|work joined channel #puppet
[2011/11/04 01:43:38] @ wallblom_ joined channel #puppet
[2011/11/04 01:46:35] @ pieter_ joined channel #puppet
[2011/11/04 01:47:39] @ Quit: Vulcan[OMNI]: Quit: Vulcan[OMNI]
[2011/11/04 01:47:59] @ duerF joined channel #puppet
[2011/11/04 01:48:24] @ TheDigitalNinja joined channel #puppet
[2011/11/04 01:49:17] @ Quit: pmorillon: Quit: pmorillon
[2011/11/04 01:49:29] @ Quit: lkoranda: Quit: Ex-Chat
[2011/11/04 01:49:53] @ Quit: mpdude: Ping timeout: 276 seconds
[2011/11/04 01:50:11] @ mpdude joined channel #puppet
[2011/11/04 01:55:51] <KristianDK> Also, checked out this ticket, and even though it says "released with 2.0.0 just now" and i have 2.7.1, it still reports "Unknown function has_ley"
[2011/11/04 01:55:54] <KristianDK> http://projects.puppetlabs.com/issues/8708
[2011/11/04 01:57:46] @ Quit: mpdude: Ping timeout: 244 seconds
[2011/11/04 01:59:39] <_rc> 2.0.0 of https://github.com/puppetlabs/puppetlabs-stdlib, not of puppet
[2011/11/04 02:00:04] @ lkoranda joined channel #puppet
[2011/11/04 02:00:54] @ Quit: Moe__: Ping timeout: 260 seconds
[2011/11/04 02:01:21] @ pmorillon joined channel #puppet
[2011/11/04 02:02:29] <KristianDK> _rc, ah sorry - never came across that before - i guess ill have to implement that too then ,)
[2011/11/04 02:03:43] <blahdeblah> Hi all. I have a client who needs to migrate an existing 32-bit server to new hardware and 64-bit OS. Are there any architecture dependencies in puppet which would prevent /var/lib/puppet being copied over from the old system to the new and maintaining its existing certificate?
[2011/11/04 02:06:49] @ thenulldevice joined channel #puppet
[2011/11/04 02:07:59] @ MattyM joined channel #puppet
[2011/11/04 02:10:17] @ skm joined channel #puppet
[2011/11/04 02:10:29] <dcarley> blahdeblah: nope. but re-signing would still be prefered.
[2011/11/04 02:10:47] <blahdeblah> dcarley: Any particular reason for that preference?
[2011/11/04 02:12:36] <dcarley> blahdeblah: ability to revoke the old machine in the event that it doesn't really get wiped. the chance to remove any cruft in $vardir, $clientbucket permitting.
[2011/11/04 02:12:53] @ Quit: kinsel: Ping timeout: 245 seconds
[2011/11/04 02:14:38] <blahdeblah> dcarley: The old machine is not really an issue in this case. It's a remote office bare metal server hosting the VPN connection to the puppet server, so there's not really any way the old one could successfully connect without disrupting the connection to the new one. What sort of cruft accrues in $vardir besides $clientbucket?
[2011/11/04 02:15:04] @ ezryx joined channel #puppet
[2011/11/04 02:16:54] @ jhelwig is now known as zz_jhelwig
[2011/11/04 02:17:13] @ Holocaine joined channel #puppet
[2011/11/04 02:18:29] <dcarley> blahdeblah: theoretically speaking- the SSL cert could be taken and used elsewhere to obtain or inject information that it shouldn't.
[2011/11/04 02:18:45] <blahdeblah> good point
[2011/11/04 02:21:22] <blahdeblah> I will work a certificate re-signing into the migration procedure. Are there any good tools for automating this into the machine bootstrap?
[2011/11/04 02:22:24] @ mogitaff joined channel #puppet
[2011/11/04 02:23:52] @ derjohn_foo joined channel #puppet
[2011/11/04 02:24:51] @ Quit: lovemj: Read error: Connection reset by peer
[2011/11/04 02:24:53] @ jbraeuer joined channel #puppet
[2011/11/04 02:25:22] @ lovemj joined channel #puppet
[2011/11/04 02:27:13] @ Deesl joined channel #puppet
[2011/11/04 02:27:13] @ Quit: Deesl: Excess Flood
[2011/11/04 02:27:17] @ Quit: zipkid: Ping timeout: 240 seconds
[2011/11/04 02:27:33] @ zipkid joined channel #puppet
[2011/11/04 02:27:40] @ Deesl joined channel #puppet
[2011/11/04 02:27:51] @ xover1 joined channel #puppet
[2011/11/04 02:28:13] <xover1> once I have made my site.pp changes, how do I refresh the config file?
[2011/11/04 02:28:18] <xover1> service puppet restart?
[2011/11/04 02:29:01] <dcarley> blahdeblah: which part? revocation, request or signing?
[2011/11/04 02:29:45] <_rc> xover1: the puppetmaster should see it's changed and just recompile it
[2011/11/04 02:29:46] <dcarley> xover1: in it's default configuration the agent will periodically check the server for changes and execute them.
[2011/11/04 02:30:27] <xover1> is there a syntax tool checker?
[2011/11/04 02:31:16] <Spads> xover1: I use puppetdoc to check for simple syntax errors
[2011/11/04 02:31:29] <Spads> it'll miss a lot, but it catches the usual missing comma type stuff
[2011/11/04 02:31:48] @ jaytaph joined channel #puppet
[2011/11/04 02:32:33] <dcarley> you can use "puppet apply --parseonly manifest.pp"
[2011/11/04 02:32:50] <blahdeblah> dcarley: all of the above
[2011/11/04 02:33:47] @ jab_doa joined channel #puppet
[2011/11/04 02:34:31] @ Quit: Ownage: Quit: Leaving
[2011/11/04 02:34:48] @ Ramonster joined channel #puppet
[2011/11/04 02:36:12] @ kinsel joined channel #puppet
[2011/11/04 02:38:23] @ dle joined channel #puppet
[2011/11/04 02:39:22] @ Quit: skm: Quit: Computer has gone to sleep.
[2011/11/04 02:41:14] @ Quit: artkrz: Quit: Linkinus - http://linkinus.com
[2011/11/04 02:41:28] @ MarkD joined channel #puppet
[2011/11/04 02:41:39] @ Quit: MarkD: Client Quit
[2011/11/04 02:44:16] @ Quit: ckauhaus: Ping timeout: 244 seconds
[2011/11/04 02:44:51] <n1x0n> Hello, I understand there are no loops in puppet language (from google) - but I need to execute command foo <rpmname> for each rpm I plan to pass to my define (as an array) {no, I'm not trying to install rpm via exec!} - any suggestions how to tackle this ?
[2011/11/04 02:45:37] <robinbowes> Mr. Le Volcane, would you happen to be around for a few mins to go over a couple of hiera topics?
[2011/11/04 02:46:26] <Spads> n1x0n: you can create an exec and pass in a list of strings and it'll do them in no particular order
[2011/11/04 02:47:29] <blahdeblah> n1x0n: syntax is like this: define mydef() { exec { "foo $name": ... } } mydef { [ "rpm1", "rpm2", "rpm3", ... ]: ... }
[2011/11/04 02:48:04] @ artkrz joined channel #puppet
[2011/11/04 02:49:20] <n1x0n> thanks both, I'll give it a go
[2011/11/04 02:50:11] <xover1> I am getting this error when trying to copy a file
[2011/11/04 02:50:12] <xover1> Could not evaluate: Could not retrieve information from source(s)
[2011/11/04 02:50:13] @ adimania joined channel #puppet
[2011/11/04 02:50:18] <xover1> the file is listed under modules.
[2011/11/04 02:50:42] @ Quit: walterheck: Quit: Tribily - Monitor your servers more easily! - http://tribily.com
[2011/11/04 02:50:46] <_rc> rest of the error message please
[2011/11/04 02:51:00] @ AikiLinux joined channel #puppet
[2011/11/04 02:51:13] <_rc> you probably misformed the source line, but without seeing the message we can't be sure
[2011/11/04 02:51:52] <xover1> file {'/root/.ssh/authorized_keys':
[2011/11/04 02:51:52] <xover1> ensure => present,
[2011/11/04 02:51:52] <xover1> source => "puppet:///modules/ssh_keys_master/master"
[2011/11/04 02:51:52] <xover1> }
[2011/11/04 02:52:20] <_rc> and so the file is in modules/ssh_keys_master/files/master?
[2011/11/04 02:52:28] @ Quit: derjohn_foo: Read error: Connection reset by peer
[2011/11/04 02:52:28] <xover1> yes
[2011/11/04 02:52:36] <xover1> the file is 'master'
[2011/11/04 02:52:36] <_rc> and is readable by the puppet user?
[2011/11/04 02:53:03] <xover1> -rw-r--r-- 1 root root
[2011/11/04 02:53:04] @ derjohn_foo joined channel #puppet
[2011/11/04 02:53:10] @ Quit: TheDigitalNinja: Remote host closed the connection
[2011/11/04 02:53:13] <_rc> the directories above?
[2011/11/04 02:53:30] <xover1> drwxr-xr-x 2 root root
[2011/11/04 02:53:54] <blahdeblah> xover1: It's not likely to be your issue, but it's usually good to add a comma at the end of every line - saves going back debugging it later when you add a line below and wonder why your manifest broke.
[2011/11/04 02:54:03] @ AikiLinux left channel #puppet ()
[2011/11/04 02:54:25] <xover1> i actually just added that.
[2011/11/04 02:54:27] <_rc> all the directories above?
[2011/11/04 02:54:51] <xover1> they area ll 755 root:root
[2011/11/04 02:55:13] <_rc> as a simple test I do 'sudo -u puppet stat /etc/puppet/modules/ssh_keys_master/files/master' to double-check the puppet user can see the file
[2011/11/04 02:55:23] <xover1> I was able to change the contents of the file before, but as i dont want my keys inside site.pp i wanted to copy the file.
[2011/11/04 02:55:24] <_rc> don't paste that
[2011/11/04 02:55:28] <_rc> a yes/no will do
[2011/11/04 02:56:05] <xover1> i dont have that stat command
[2011/11/04 02:56:22] <_rc> well a simple cat will do
[2011/11/04 02:56:22] <blahdeblah> xover1: try ls -l instead
[2011/11/04 02:56:41] <xover1> unknown command ls
[2011/11/04 02:56:47] @ TheDigitalNinja joined channel #puppet
[2011/11/04 02:56:57] <blahdeblah> ??
[2011/11/04 02:57:14] <blahdeblah> what about su - puppet -c 'ls -l /etc/puppet/modules/ssh_keys_master/files/master'?
[2011/11/04 02:57:18] @ ckauhaus joined channel #puppet
[2011/11/04 02:57:49] @ Quit: derjohn_foo: Ping timeout: 240 seconds
[2011/11/04 02:57:53] <xover1> i get this
[2011/11/04 02:57:53] <blahdeblah> (minus the ? - that's just me)
[2011/11/04 02:57:54] <xover1> ambiguous option: -rw-r--r--
[2011/11/04 02:57:55] <xover1> Try 'puppet apply --help'
[2011/11/04 02:58:04] <blahdeblah> you're using the wrong quotes
[2011/11/04 02:58:11] <blahdeblah> ' not `
[2011/11/04 02:58:11] @ MrPunch joined channel #puppet
[2011/11/04 02:58:40] @ derjohn_foo joined channel #puppet
[2011/11/04 02:58:42] <xover1> Available commands are: agent, apply, cert, describe, doc, filebucket, inspect, kick, license, master, queue, resource
[2011/11/04 02:59:08] <_rc> you're running puppet rather than becoming puppet to run the command.
[2011/11/04 02:59:14] <blahdeblah> You're doing something very wrong. _rc just wants you to verify permissions from the shell. Don't add puppet on the front.
[2011/11/04 02:59:33] <iclebyte> any idea what features are planned for the dashboard?
[2011/11/04 02:59:55] <_rc> blahdeblah: I think they missed the 'su -' from your 'su - puppet -c'
[2011/11/04 03:00:02] <blahdeblah> evidently :-
[2011/11/04 03:00:04] <blahdeblah> :-)
[2011/11/04 03:00:10] <MrPunch> Morning - hopefully a quick question. I am using cobbler to autoprovision and it leaves a cobbler.repo file around that i want to delete. I then have a stack of modules for my infrastructure setup that fail due to the presence of this file. If I create a file type to remove it where would I need to put it to ensure it always fires before module installation?
[2011/11/04 03:00:25] <KaZeR_W> hi there
[2011/11/04 03:00:47] <MrPunch> I'm guessing that putting it in /etc/puppet/manifests/init.pp might work?
[2011/11/04 03:00:48] @ Quit: kinsel: Ping timeout: 244 seconds
[2011/11/04 03:01:11] <blahdeblah> MrPunch: There is not usually any file of that name - site.pp is the master manifest.
[2011/11/04 03:01:11] <KaZeR_W> MrPunch, i'm not an expert, but i'd say : 1) add a require to your module installation, 2) use stages
[2011/11/04 03:01:17] <_rc> you'd want to put it in an early runstage
[2011/11/04 03:02:05] <MrPunch> sorry, mean site.pp - would that then fire first? Stages are new to me, but i am a total puppet newb :)
[2011/11/04 03:02:21] <MrPunch> I'll try to find the docs on stages
[2011/11/04 03:02:49] <KaZeR_W> MrPunch, quick sample :
[2011/11/04 03:02:51] <KaZeR_W> stage {pre: before=>Stage["main"]}
[2011/11/04 03:02:55] <_rc> site.pp just means it'd be done globally, and wouldn't neccesarily enforce any ordering
[2011/11/04 03:02:59] <KaZeR_W> stage {'yum': require => Stage['pre'] }
[2011/11/04 03:03:41] <_rc> itym class { "yum": stage => "pre" }
[2011/11/04 03:04:55] <xover1> sorry i am not getting this
[2011/11/04 03:05:01] <MrPunch> aha, found the docs - so what is good practice for where to set up the stages?
[2011/11/04 03:05:01] <xover1> what is the problem with copyig this file?
[2011/11/04 03:05:16] @ jaymurtii joined channel #puppet
[2011/11/04 03:05:25] @ Quit: adimania: Ping timeout: 240 seconds
[2011/11/04 03:05:42] <xover1> is that puppet pro boo any good?
[2011/11/04 03:05:44] <_rc> xover1: well we didn't see the full errormessage, but it seemed to be that the puppetmaster process couldn't get the metadata from the file
[2011/11/04 03:05:58] <_rc> xover1: which is why we asked you to check permissions
[2011/11/04 03:06:04] @ Quit: TheDigitalNinja: Ping timeout: 255 seconds
[2011/11/04 03:06:19] <blahdeblah> xover1: Yes, the book is good - highly recommended to work through it from start to finish if you're new.
[2011/11/04 03:06:22] <NuclearLucifer> xover1, You mean Pro Puppet book?
[2011/11/04 03:06:28] <xover1> stat /etc/puppetlabs/puppet/modules/ssh_keys_master/master
[2011/11/04 03:06:28] <xover1> File: `/etc/puppetlabs/puppet/modules/ssh_keys_master/master'
[2011/11/04 03:06:28] <xover1> Size: 3047 Blocks: 8 IO Block: 4096 regular file
[2011/11/04 03:06:28] <xover1> Device: 801h/2049d Inode: 8209 Links: 1
[2011/11/04 03:06:28] <xover1> Access: (0644/-rw-r--r--) Uid: ( 0/ root) Gid: ( 0/ root)
[2011/11/04 03:06:31] <xover1> Access: 2011-11-04 09:26:51.000000000 +0000
[2011/11/04 03:06:31] <xover1> Modify: 2011-11-04 09:19:37.000000000 +0000
[2011/11/04 03:06:31] <xover1> Change: 2011-11-04 09:19:37.000000000 +0000
[2011/11/04 03:06:48] <_rc> xover1: stop pasting lines into the channel. use pastie.org please
[2011/11/04 03:06:48] <blahdeblah> xover1: Next time, use pastebin please!
[2011/11/04 03:06:54] <blahdeblah> lol
[2011/11/04 03:07:13] <_rc> and did you do that stat as the puppet user?
[2011/11/04 03:07:43] <NuclearLucifer> Very good book. It's good to have some knowledge base before reading it, but it's written in simple way.
[2011/11/04 03:07:47] <xover1> who is the puppet user?
[2011/11/04 03:07:54] <xover1> pe-puppet?
[2011/11/04 03:07:59] <blahdeblah> MrPunch: site.pp is fine for that, but suit yourself - as long as it's included
[2011/11/04 03:08:02] <_rc> it's the user your puppetmaster process it running as
[2011/11/04 03:08:04] <_rc> could be
[2011/11/04 03:08:22] <xover1> its run as root
[2011/11/04 03:08:28] <MrPunch> ok, thanks for the help - useful as ever - off to have a play
[2011/11/04 03:08:43] <MrPunch> :D
[2011/11/04 03:08:51] <_rc> you start it as root, it should drop permissions to a less priveledged user
[2011/11/04 03:09:03] <xover1> how can i tell then who is the user?
[2011/11/04 03:09:04] <xover1> sigh
[2011/11/04 03:09:09] <_rc> check your process table
[2011/11/04 03:09:20] @ DigitalNinja joined channel #puppet
[2011/11/04 03:09:30] <xover1> senger-spawn-server ?
[2011/11/04 03:09:54] <NuclearLucifer> xover1, Go read some doscs.
[2011/11/04 03:09:56] <_rc> no
[2011/11/04 03:10:29] <blahdeblah> MrPunch: another way to do it might be to set your resource as a dependency for all packages, e.g. Package { require => Class['myclasstodeletecobblerrepo'] }
[2011/11/04 03:11:24] <MrPunch> ok, that sounds a bit more coupled than the stages technique though
[2011/11/04 03:11:43] <_rc> xover1: the master process will be spawned by that, and is typically called 'master'
[2011/11/04 03:12:39] <blahdeblah> MrPunch: Whatever works for you. If package installation is what cobbler.repo breaks, i would probably use Package, because one day you might need to do package installations in an early stage.
[2011/11/04 03:13:24] <xover1> if i do ps -ef \| grep puppet, all the processes run as root
[2011/11/04 03:13:40] <blahdeblah> xover1: what about ps -ef\|grep master
[2011/11/04 03:14:02] <xover1> pe-puppet
[2011/11/04 03:14:37] <_rc> that's your puppetmaster process, and so that's the user you need to ensure can read the file and its metadata
[2011/11/04 03:15:13] @ EL84 joined channel #puppet
[2011/11/04 03:15:17] @ EL84 left channel #puppet ()
[2011/11/04 03:16:13] <xover1> su -c 'stat /etc/puppetlabs/puppet/modules/ssh_keys_master/master' pe-puppet
[2011/11/04 03:16:17] <xover1> returns nothim
[2011/11/04 03:16:26] <xover1> works as root
[2011/11/04 03:16:41] <_rc> and that's your problem. fix it.
[2011/11/04 03:16:49] <xover1> the folders are 755
[2011/11/04 03:16:56] <xover1> im not getting you.
[2011/11/04 03:17:17] <_rc> the pe-puppet user can't stat the file, so you get the errormessage that you get.
[2011/11/04 03:17:28] <blahdeblah> xover1: If they're all 755, you would be able to stat it
[2011/11/04 03:17:44] <_rc> check all the directories along the path
[2011/11/04 03:17:56] <xover1> I already have
[2011/11/04 03:18:14] @ Quit: Spads: Quit: rebooting for new kernel
[2011/11/04 03:18:16] <blahdeblah> selinux stopping it perhaps?
[2011/11/04 03:18:22] <_rc> /etc /etc/puppetlabs /etc/puppetlabs/puppet etc
[2011/11/04 03:19:22] <xover1> im using debian, i dont think it uses selinux or apparmour
[2011/11/04 03:21:02] <xover1> how do i restart puppet?
[2011/11/04 03:21:08] <xover1> pe-mcollective?
[2011/11/04 03:21:10] @ kelseyhightower joined channel #puppet
[2011/11/04 03:23:36] <KaZeR_W> i'm back with my "err: Could not run Puppet configuration client: Parameter name failed: Invalid host name" issue. can anyone help?
[2011/11/04 03:23:40] @ Yoe joined channel #puppet
[2011/11/04 03:23:53] <Yoe> hi -- I'm trying to set some variables depending on where a machine is physically located
[2011/11/04 03:23:57] <KaZeR_W> hostname == hostname -f == facter \| grep -i fqdn
[2011/11/04 03:24:12] <_rc> KaZeR_W: does your manifest have any host {} resources?
[2011/11/04 03:24:14] <Yoe> because there are a few mirrors for services like kerberos, dns, ldap, etc
[2011/11/04 03:24:15] <KaZeR_W> facter \| grep -i hostname is the correct shortname
[2011/11/04 03:24:20] <KaZeR_W> _rc, let me check
[2011/11/04 03:25:05] <Yoe> what I tried is have a class "loc" where a variable "$ldapsrv" is defined, and then have a class for each site that inherits from that "loc" class and sets the actual variables to useful values
[2011/11/04 03:25:40] <Yoe> so that I then could use templates to create config files, which would use the values of those variables
[2011/11/04 03:25:41] <KaZeR_W> _rc, you mean host-specific ressource declaration, instead of a regexp pattern ?
[2011/11/04 03:25:45] <_rc> Yoe: I'd use extlookup or heira
[2011/11/04 03:25:57] <_rc> KaZeR_W: yes, I mean a resource
[2011/11/04 03:26:00] <misc> or user a custom fact for that
[2011/11/04 03:26:02] <Yoe> _rc: any further pointers to what those are? I'm fairly new to puppet
[2011/11/04 03:26:23] <_rc> Yoe: http://www.devco.net/archives/2011/06/05/hiera_a_pluggable_hierarchical_data_store.php
[2011/11/04 03:26:56] <_rc> I always manage to transpose the ie, sorry
[2011/11/04 03:27:53] <KaZeR_W> _rc, my manifest has a bunch of host resources yes. some are patterns, some are the exact hostname
[2011/11/04 03:28:02] @ kelseyhightower_ joined channel #puppet
[2011/11/04 03:28:21] <_rc> KaZeR_W: huh? a pattern host resource?
[2011/11/04 03:29:05] <Yoe> _rc: that seems like a perfect fit for my needs, thanks. I'll go and read up on how to use it, now :-)
[2011/11/04 03:29:21] <_rc> KaZeR_W: I explcitly mean things like host { "foo": ip => "1.2.3.4" } http://docs.puppetlabs.com/references/stable/type.html#host
[2011/11/04 03:29:33] @ Quit: kelseyhightower: Ping timeout: 245 seconds
[2011/11/04 03:29:33] <KaZeR_W> ah! yes, indeed
[2011/11/04 03:29:40] @ kelseyhightower joined channel #puppet
[2011/11/04 03:30:39] <_rc> as the error message 'Invalid host name' only comes from the host type
[2011/11/04 03:31:37] <KaZeR_W> _rc, bingo. commenting the host ressource seems to fix the issue. good shot! thanks
[2011/11/04 03:32:21] <KaZeR_W> now it seems so obvious
[2011/11/04 03:32:33] @ Quit: kelseyhightower_: Ping timeout: 258 seconds
[2011/11/04 03:32:50] @ _plg joined channel #puppet
[2011/11/04 03:32:56] <KaZeR_W> googling led me to an irc archive where people were suspecting an issue with dns resolution, or hostname configuration on the node
[2011/11/04 03:33:10] <_rc> let's see the resource?
[2011/11/04 03:34:35] <_rc> https://github.com/puppetlabs/puppet/blob/master/lib/puppet/type/host.rb#L59 is the validation it applies
[2011/11/04 03:35:04] <_plg> hi, i have a module that copies a folder into place and that works fine, now if I try to make sure that a sub folder is exacutable in the end that sub folder is not there
[2011/11/04 03:35:18] <_plg> any guesses?
[2011/11/04 03:35:54] <bd__> _plg: missing dependeny on the parent folder?
[2011/11/04 03:36:13] <KaZeR_W> thanks _rc i'm having a look to see if i can figure out why it fails
[2011/11/04 03:36:16] <_plg> require => File [ "/usr/local/apache-cassandra-1.0.0" ]
[2011/11/04 03:36:29] @ Quit: ezmobius: Remote host closed the connection
[2011/11/04 03:36:33] <_plg> its there, let me check if it has any minor error in the path
[2011/11/04 03:36:45] <bd__> _plg: pastie, please.
[2011/11/04 03:36:50] <_plg> sry
[2011/11/04 03:37:22] <bd__> _plg: no worries, I just need a little bit more context
[2011/11/04 03:37:37] <KaZeR_W> _rc, hint : displaying the failing hostname makes it really easy to find the faulty config entry :)
[2011/11/04 03:37:43] <KaZeR_W> should i fill a ticket ?
[2011/11/04 03:38:07] <_plg> http://pastebin.com/tbMhgU9d
[2011/11/04 03:38:09] <_rc> sounds worthwhile
[2011/11/04 03:38:21] <KaZeR_W> k
[2011/11/04 03:38:40] <KaZeR_W> host fixed, now it works. thanks a lot, i've spent an incredible amount of time on this issue
[2011/11/04 03:38:52] <_plg> if i remove the 3rd file section it works fine
[2011/11/04 03:40:25] @ Spads joined channel #puppet
[2011/11/04 03:40:32] @ Quit: Spads: Changing host
[2011/11/04 03:40:32] @ Spads joined channel #puppet
[2011/11/04 03:40:39] <bd__> _plg: why the recurse on the bin directory?
[2011/11/04 03:41:37] @ Quit: duerF: Ping timeout: 255 seconds
[2011/11/04 03:42:02] <_plg> bd__: grant execution in the bin folder and all its files
[2011/11/04 03:43:53] <bd__> _plg: ic, sadly I have no real idea. maybe try and removing the require from the link and see what happens.
[2011/11/04 03:44:20] @ bencc left channel #puppet ()
[2011/11/04 03:44:39] @ Quit: artkrz: Quit: Linkinus - http://linkinus.com
[2011/11/04 03:45:15] <_plg> i based myself on a similar example I saw, but there the file sources and similar were a bit confusing... it was made to a previous puppet version I gess
[2011/11/04 03:45:29] @ kinsel joined channel #puppet
[2011/11/04 03:45:30] <_plg> ok i will try
[2011/11/04 03:45:34] <bd__> _plg: alternatively also use a source statement for the bin directory too
[2011/11/04 03:45:49] <_plg> rigth
[2011/11/04 03:45:51] <_plg> thanks
[2011/11/04 03:46:01] <bd__> you're welcome
[2011/11/04 03:48:10] @ Quit: Enf: Quit: Leaving
[2011/11/04 03:48:36] @ Quit: choonming: Quit: Leaving
[2011/11/04 03:48:40] @ Quit: derjohn_foo: Read error: Connection reset by peer
[2011/11/04 03:49:33] @ derjohn_foo joined channel #puppet
[2011/11/04 03:49:42] <_plg> on little question, is it normal to take 2 minutes for puppet to start doing something
[2011/11/04 03:49:46] <_plg> ?
[2011/11/04 03:50:36] <andrewvos> No
[2011/11/04 03:50:44] <bd__> _plg: try puppet agent --test --trace --verbose --noop
[2011/11/04 03:50:55] @ duerF joined channel #puppet
[2011/11/04 03:51:21] @ dnjaramba joined channel #puppet
[2011/11/04 03:52:06] @ derjohn_mob joined channel #puppet
[2011/11/04 03:54:01] @ Quit: derjohn_foo: Ping timeout: 244 seconds
[2011/11/04 03:55:41] @ Quit: duerF: Ping timeout: 260 seconds
[2011/11/04 03:56:16] <_plg> debugging on my machine so I using puppet apply
[2011/11/04 03:56:21] <_plg> no --test
[2011/11/04 03:57:01] @ Kazuhiro joined channel #puppet
[2011/11/04 03:57:20] <_plg> but i used the other flags, for now its notice: info: Applying configuration version '1320404117'
[2011/11/04 03:59:29] @ Yoe left channel #puppet ()
[2011/11/04 04:01:47] @ duerF joined channel #puppet
[2011/11/04 04:01:52] @ Quit: dnjaramba: Remote host closed the connection
[2011/11/04 04:05:26] @ Quit: KristianDK: Quit: KristianDK
[2011/11/04 04:06:24] @ Quit: duerF: Ping timeout: 248 seconds
[2011/11/04 04:06:33] @ skm joined channel #puppet
[2011/11/04 04:06:34] @ dnjaramba joined channel #puppet
[2011/11/04 04:07:15] @ Quit: csanchez: Quit: csanchez
[2011/11/04 04:13:10] <robinbowes> xover1: _rc have you used hiera with puppet
[2011/11/04 04:13:15] <robinbowes> Oops
[2011/11/04 04:13:36] <robinbowes> _rc: have you used hiera with puppet
[2011/11/04 04:13:45] @ blahdeblah left channel #puppet ()
[2011/11/04 04:13:55] @ blahdeblah joined channel #puppet
[2011/11/04 04:15:14] @ Quit: pmorillon: Quit: pmorillon
[2011/11/04 04:17:54] @ Quit: frostynova: Remote host closed the connection
[2011/11/04 04:18:53] <robinbowes> I'm doing some simple testing and am getting: Unknown function hiera at /root/tt.pp:1
[2011/11/04 04:19:04] <xover1> is there a package for debian 6?
[2011/11/04 04:19:20] <blahdeblah> xover1: Yes, it's in the standard Debian repos.
[2011/11/04 04:19:39] <robinbowes> tt.pp just contains: $webname = hiera("webname")
[2011/11/04 04:19:46] <_rc> robinbowes: not yet, it's on my todo
[2011/11/04 04:20:06] <robinbowes> Can't get it working
[2011/11/04 04:20:09] <_rc> isn't that the 'copy a directory from the gem into your puppetmaster'
[2011/11/04 04:20:20] <_rc> then pluginsync/bounce
[2011/11/04 04:20:21] <robinbowes> I think I've done that
[2011/11/04 04:20:36] <xover1> apt cant find it
[2011/11/04 04:21:01] <robinbowes> We've actually got several environments
[2011/11/04 04:21:22] @ Kazuhiro left channel #puppet ()
[2011/11/04 04:21:36] <blahdeblah> xover1: From my puppet master: http://pastebin.com/gs2vE7Kn
[2011/11/04 04:22:22] <xover1> i cant find it
[2011/11/04 04:22:38] <xover1> which source do i have to uncheck in sources.list?
[2011/11/04 04:22:42] <robinbowes> But I've copied the hiera stuff into <env>/plugins/puppet/parser/functions for all of them
[2011/11/04 04:23:27] <_rc> env/plugins/lib/puppet/parser/functions, no?
[2011/11/04 04:23:51] <robinbowes> Is it??? That might explain it!
[2011/11/04 04:23:54] <_rc> also I think environments don't cope so well with custom functions
[2011/11/04 04:24:10] <_rc> http://docs.puppetlabs.com/guides/plugins_in_modules.html
[2011/11/04 04:24:12] <blahdeblah> xover1: Here's mine: http://pastebin.com/cQVayQ3D
[2011/11/04 04:24:23] <blahdeblah> xover1: Gotta go - good luck sorting out your issue
[2011/11/04 04:25:19] <xover1> i have just installed debian and am getting this error/
[2011/11/04 04:25:20] <xover1> http://pastebin.com/27kCvBX0
[2011/11/04 04:26:16] <xover1> durig installation is was complaining about internet access.
[2011/11/04 04:26:28] <xover1> i have nothing in sources.list.d
[2011/11/04 04:27:04] <robinbowes> Hmm, so they have to go in a module?
[2011/11/04 04:28:34] <xover1> im going to resort to centos again i think, debian is too much of a ballbreaker with stuff like this
[2011/11/04 04:32:34] @ erase joined channel #puppet
[2011/11/04 04:32:39] <robinbowes> If you are using a version of Puppet prior to 0.24.0, or have some other compelling reason to not use plugins in modules, functions can also be loaded from .rb files in the following locations:
[2011/11/04 04:32:44] <robinbowes> $libdir/puppet/parser/functions
[2011/11/04 04:32:52] <robinbowes> Which is $libdir?
[2011/11/04 04:32:59] @ erase left channel #puppet ()
[2011/11/04 04:33:12] <Spads> robinbowes: typically /var/lib/puppet/, but it might be changed in your puppet.conf
[2011/11/04 04:33:37] <robinbowes> No, don't think so
[2011/11/04 04:33:42] <_rc> it's better to just put them in modules
[2011/11/04 04:33:46] @ Quit: gaveen: Remote host closed the connection
[2011/11/04 04:34:09] <_rc> pluginsync will put them in $libdir for you
[2011/11/04 04:34:13] <robinbowes> Won't I then have to load the module to use the functions?
[2011/11/04 04:34:16] <_rc> no
[2011/11/04 04:34:43] <_rc> same as you don't need to for custom facts, pluginsync takes care of it
[2011/11/04 04:35:14] @ mpdude joined channel #puppet
[2011/11/04 04:35:23] <robinbowes> OK, so I should put the hiera files in: <env>/modules/hiera/lib/puppet/parser/functions ?
[2011/11/04 04:36:54] * Spads ♥ pluginsync
[2011/11/04 04:37:06] <fragfutter> can i add a new key to a hash? like $my_hash["foobar"] = 2000?
[2011/11/04 04:37:39] <_rc> puppet apply smalltestcase.pp
[2011/11/04 04:38:06] <xover1> is there not a tutorial on how to install this on debian and centos?
[2011/11/04 04:38:26] @ Quit: MattyM: Remote host closed the connection
[2011/11/04 04:38:49] <xover1> i have it working in centos now, but debian is bitching about dependencies too
[2011/11/04 04:39:33] <_rc> are you adding repositories to the systems, or just ramming individual packages onto the machines? if the latter then it's no surprise
[2011/11/04 04:39:58] <xover1> where are the repositories listed?
[2011/11/04 04:40:18] <_rc> which repositories are you using?
[2011/11/04 04:40:49] @ KristianDK joined channel #puppet
[2011/11/04 04:42:23] @ erthad left channel #puppet ()
[2011/11/04 04:42:41] @ Quit: ezryx: Read error: Connection reset by peer
[2011/11/04 04:42:50] @ ezryx joined channel #puppet
[2011/11/04 04:43:31] <walkeran> If you are looking for the centos repos... check yum.puppetlabs.com. IMHO, they do a pretty horrid job of advertising/publishing that URL (maybe it's on purpose... I guess I don't know)
[2011/11/04 04:43:33] @ erthad joined channel #puppet
[2011/11/04 04:43:45] <walkeran> s/centos/yum/
[2011/11/04 04:43:57] <xover1> i have centos sorted
[2011/11/04 04:44:05] <xover1> i just had to add epel repo and that sorts it
[2011/11/04 04:44:29] <xover1> on debian however, i have a load of dependcies that are not met using the main repo even though the puppet-master is listed
[2011/11/04 04:44:32] <walkeran> Oh... meh. Poo poo on the epel repo :) They still have 0.25 in the stable one?
[2011/11/04 04:44:46] <xover1> i thought the point of apt is that they were no dependency issues
[2011/11/04 04:44:51] <robinbowes> epel-testing ftw
[2011/11/04 04:44:58] <walkeran> epel testing has 2.6, right?
[2011/11/04 04:45:10] <robinbowes> epel has too, I think?
[2011/11/04 04:45:22] <walkeran> Oh.. that could be. It's been a while since I looked
[2011/11/04 04:45:24] <robinbowes> testing has a later version though
[2011/11/04 04:45:35] @ axisys joined channel #puppet
[2011/11/04 04:47:04] <dcarley> epel5/epel6 both have 2.6.6 with some backported security fixes now.
[2011/11/04 04:47:53] <xover1> im running debian though mate
[2011/11/04 04:47:57] <xover1> which repos do i need?
[2011/11/04 04:54:44] @ jajajajaj joined channel #puppet
[2011/11/04 04:55:18] @ pmorillon joined channel #puppet
[2011/11/04 04:55:35] @ xover1 left channel #puppet ()
[2011/11/04 05:06:35] @ Quit: BadKarma: Quit: leaving
[2011/11/04 05:12:16] @ Quit: erodriguez: Quit: erodriguez
[2011/11/04 05:13:50] @ Quit: Holocaine:
[2011/11/04 05:15:45] @ beaknit joined channel #puppet
[2011/11/04 05:21:46] @ rgsteele joined channel #puppet
[2011/11/04 05:23:45] @ xover1 joined channel #puppet
[2011/11/04 05:23:56] <xover1> i thought installing this on centos would make things easier
[2011/11/04 05:24:07] <walkeran> Thought you had centos figgered out!
[2011/11/04 05:24:09] <xover1> all the master server folders are missing now, do i have to run a script?
[2011/11/04 05:24:17] <xover1> that was as an agent
[2011/11/04 05:24:20] <walkeran> Ah
[2011/11/04 05:24:28] <xover1> i need to replace this debian master, so i intsalled master on centos
[2011/11/04 05:24:35] @ andrewvos left channel #puppet ()
[2011/11/04 05:24:38] <walkeran> What master folders? /etc/puppet/modules and /etc/puppet/manifests?
[2011/11/04 05:24:56] <xover1> I only have manifests folder and three config giles
[2011/11/04 05:25:04] <xover1> do i need to create these manually?
[2011/11/04 05:25:05] <walkeran> That's about all you need
[2011/11/04 05:25:11] <xover1> oh ok
[2011/11/04 05:26:24] @ Quit: ezryx: Remote host closed the connection
[2011/11/04 05:26:30] <walkeran> When you drop your modules in, you'll need /etc/puppet/modules, but other than that, your site.pp and anything it imports go in manifests, and you don't need much for config files aside from puppet.conf. Maybe a fileserver.conf and auth.conf... perhaps autosign.conf. But the defaults are usually good enough to get rolling, I think
[2011/11/04 05:26:39] @ kolla joined channel #puppet
[2011/11/04 05:32:57] @ Quit: semiosis: Quit: Time40
[2011/11/04 05:33:39] @ uphillian joined channel #puppet
[2011/11/04 05:36:08] @ semiosis joined channel #puppet
[2011/11/04 05:39:51] <greenmang0> hey friends, i have 2 modules, both of them require a package "openjdk-6-jdk"... now i have , these 2 modules may be installed separately or together, but when i install them together, I get error "Error 400 on SERVER: Duplicate definition: Package[openjdk-6-jdk] is already defined"
[2011/11/04 05:39:55] <xover1> omg
[2011/11/04 05:40:05] <greenmang0> what can be done to solve this?
[2011/11/04 05:40:35] <fragfutter> greenmang0: you might want to read documentation. and exactly this problem was brought up yesterday while you were present...
[2011/11/04 05:40:38] <nevyn> greenmang0: make the openjdk package virtual and realise it in both clases
[2011/11/04 05:41:11] <fragfutter> or make a common and include it in both.
[2011/11/04 05:42:22] <greenmang0> fragfutter, i already have a common, but then i inherit this "common" in other 10 modules as well, which don't need "openjdk" at all ...
[2011/11/04 05:42:52] <greenmang0> fragfutter, can you point me to the exact docs link please?
[2011/11/04 05:43:00] <greenmang0> nevyn, ok .. will try that ...
[2011/11/04 05:43:04] <fragfutter> i'm sure you will figure it out... it's not the _one_ ring.
[2011/11/04 05:43:58] @ lpaul7 joined channel #puppet
[2011/11/04 05:44:02] @ ezryx joined channel #puppet
[2011/11/04 05:44:04] <nevyn> fragfutter: won't including it in both cause the same problem (ie it appears twice in the graph)
[2011/11/04 05:45:51] @ astrostl joined channel #puppet
[2011/11/04 05:46:59] <walkeran> What the.... just make a simple-as-heck openjdk module, and include the module from both of the modules that need it.
[2011/11/04 05:47:21] <dcarley> what walkeran said.
[2011/11/04 05:48:08] <nevyn> hrm I must be showing my age. I really should get work to send me on puppet training.
[2011/11/04 05:49:14] <walkeran> If that doesn't work, get them to purchase Pro Puppet (book) for you. Awesome resource, I'd have to say
[2011/11/04 05:49:28] <walkeran> Even if you don't have any prior puppet knowledge, really
[2011/11/04 05:50:51] @ sduchesneau joined channel #puppet
[2011/11/04 05:52:57] @ Quit: ezryx: Remote host closed the connection
[2011/11/04 05:53:05] <xover1> it seems toe me that the centos package has all the config files missing
[2011/11/04 05:53:43] <xover1> there is no puppet.conf file
[2011/11/04 05:54:17] <dcarley> xover1: where's the package from and what version?
[2011/11/04 05:55:13] <nevyn> walkeran: I didn't think including a module in 2 places would solve the problem
[2011/11/04 05:55:24] <nevyn> in that the resource graph still has the package resource twice
[2011/11/04 05:55:30] <nevyn> which is a problem for graph resolution
[2011/11/04 05:55:35] @ Quit: NuclearLucifer: Ping timeout: 276 seconds
[2011/11/04 05:55:53] <walkeran> nevyn: You can include a module from as many places as you want, and all it does is what it sounds like... it just "includes" it in the catalog
[2011/11/04 05:55:59] <xover1> (1/2): puppet-server-2.6.6-3.el5.noarch.rpm \| 20 kB 00:00
[2011/11/04 05:55:59] <xover1> (2/2): puppet-2.6.6-3.el5.noarch.rpm \| 830 kB 00:00
[2011/11/04 05:56:02] <walkeran> If it's there... it's already included.
[2011/11/04 05:56:29] <nevyn> so modules are virtual then?
[2011/11/04 05:56:33] <nevyn> effectively?
[2011/11/04 05:56:39] <dcarley> nevyn: from the language tutorial: "Classes, on the other hand, are guaranteed to be singletons — you can include them as many times as you want and you’ll only ever get one copy of the resources."
[2011/11/04 05:56:53] <nevyn> ok.
[2011/11/04 05:57:04] <walkeran> Sorry, yeah, my terminology was a tad wrong. Classes, not modules. dcarley++ :)
[2011/11/04 05:57:43] <dcarley> i looked it up because i figured i'd screw the programming terminology ;)
[2011/11/04 05:57:52] <xover1> jesus this is painful
[2011/11/04 05:58:01] <xover1> i had this working from source, now i use rpm and its not working
[2011/11/04 05:58:03] <xover1> ffs
[2011/11/04 05:58:11] <walkeran> xover1: So just copy the files you had from your working install.....
[2011/11/04 05:58:22] <xover1> err: Could not retrieve catalog from remote server: certificate verify failed
[2011/11/04 05:58:22] <xover1> warning: Not using cache on failed catalog
[2011/11/04 05:58:38] <xover1> i have deleted it now all becaise that ssh file wouldnt copy
[2011/11/04 05:58:46] <walkeran> xover1: I assume you generated new certificates? That'll happen if so
[2011/11/04 05:58:50] <xover1> im trying to setup the most basic of implementations
[2011/11/04 05:59:09] <dcarley> xover1: the configs are definitely in that package.
[2011/11/04 05:59:39] <xover1> how do i do that?
[2011/11/04 06:01:01] <xover1> i have instaleld puppet-server
[2011/11/04 06:01:09] <xover1> what do i do next?
[2011/11/04 06:01:12] <walkeran> xover1: So, if you are trying to set up the most basic of implementations, I'm assuming you are doing the following: 1) Installed a puppet master with default/no configs, 2) Installed an agent on another server that's never talked to another puppetmaster, 3) Started the master on the server to generate the puppetmaster configs, 4) Set up at least a basic site.pp with a default node
[2011/11/04 06:01:52] <bd__> xover1: http://www.howtoforge.com/installing_puppet_on_ubuntu
[2011/11/04 06:02:09] <walkeran> Thought you were using centos...
[2011/11/04 06:02:39] @ zwi joined channel #puppet
[2011/11/04 06:02:40] <walkeran> well.. ok... distro aside, that should be about the same, after installing the packages.
[2011/11/04 06:03:49] @ Quit: DigitalNinja: Ping timeout: 255 seconds
[2011/11/04 06:04:01] @ tremble joined channel #puppet
[2011/11/04 06:04:02] <walkeran> So, if you have the master installed on your server, and the agent installed on your client, start at step 6 in that tutorial.
[2011/11/04 06:04:35] @ zorg1 joined channel #puppet
[2011/11/04 06:04:52] @ Quit: zorg1: Remote host closed the connection
[2011/11/04 06:07:41] @ hal-j joined channel #puppet
[2011/11/04 06:08:52] @ cliff-hm joined channel #puppet
[2011/11/04 06:09:19] @ gniks joined channel #puppet
[2011/11/04 06:09:39] @ Quit: Deesl: Quit: Bye!
[2011/11/04 06:10:02] <kjetilho> is there a way to get the parser to validate a manifest on Puppet 2.6.2 ?
[2011/11/04 06:10:28] @ Xanxess joined channel #puppet
[2011/11/04 06:11:29] <walkeran> kjetilho: I could be wrong on this one, but try: puppet --parseonly foo.pp
[2011/11/04 06:12:04] <kjetilho> aha! it is the --ignoreimport which breaks it
[2011/11/04 06:12:23] <kjetilho> this is for a commit hook
[2011/11/04 06:12:31] <walkeran> Ah, gotcha
[2011/11/04 06:12:38] <kjetilho> I guess I'll just use Perl to remove the import statements, then :-p
[2011/11/04 06:12:51] <_plg> is there a way of having a for cycle in a manifest? Imagine I have to give a number id to each machine
[2011/11/04 06:12:55] <walkeran> why would there be import statements?
[2011/11/04 06:13:07] <kjetilho> import 'nodes/*.pp' basically
[2011/11/04 06:13:28] @ MattyM joined channel #puppet
[2011/11/04 06:13:30] <walkeran> Ah ok... that makes sense, then :) I was going to get on your case for not using modules correctly :D
[2011/11/04 06:13:39] <kjetilho> :-)
[2011/11/04 06:13:43] <kjetilho> probably just 2-3 occurences
[2011/11/04 06:13:48] <walkeran> gotcha
[2011/11/04 06:15:08] @ choonming joined channel #puppet
[2011/11/04 06:15:13] @ rsnhyperion joined channel #puppet
[2011/11/04 06:15:28] @ Quit: kermit: Quit: Leaving.
[2011/11/04 06:15:43] @ kermit joined channel #puppet
[2011/11/04 06:15:58] @ frostynova joined channel #puppet
[2011/11/04 06:18:44] @ ohadlevy left channel #puppet ("Leaving")
[2011/11/04 06:18:57] @ ohadlevy joined channel #puppet
[2011/11/04 06:19:00] @ ezryx joined channel #puppet
[2011/11/04 06:19:40] @ mcarpent joined channel #puppet
[2011/11/04 06:24:02] @ Quit: gniks: Quit: Leaving.
[2011/11/04 06:24:48] @ st33v joined channel #puppet
[2011/11/04 06:30:23] @ Chesther joined channel #puppet
[2011/11/04 06:33:00] @ Quit: sduchesneau: Ping timeout: 260 seconds
[2011/11/04 06:33:41] @ Quit: bobra: Ping timeout: 240 seconds
[2011/11/04 06:34:45] @ Quit: jajajajaj: Ping timeout: 252 seconds
[2011/11/04 06:37:17] @ Quit: JodieC: Ping timeout: 240 seconds
[2011/11/04 06:37:47] <xover1> is the problem that my client already contacted a previous puppet master?
[2011/11/04 06:38:02] <xover1> do i need to delete the conig?
[2011/11/04 06:40:23] @ Quit: rsnhyperion: Ping timeout: 245 seconds
[2011/11/04 06:40:27] @ BLZbubba joined channel #puppet
[2011/11/04 06:40:41] <walkeran> xover1: You'll have to clean the generated ssl certs from the client (rm -rf /var/lib/puppet/ssl)
[2011/11/04 06:40:51] @ ldlework joined channel #puppet
[2011/11/04 06:42:11] @ Quit: kubicek: Excess Flood
[2011/11/04 06:42:27] @ jedi4ever joined channel #puppet
[2011/11/04 06:42:48] @ kubicek joined channel #puppet
[2011/11/04 06:44:49] <xover1> there are no files there
[2011/11/04 06:44:52] <xover1> i still get this
[2011/11/04 06:44:53] <xover1> err: /File[/var/opt/lib/pe-puppet/lib]: Failed to generate additional resourc
[2011/11/04 06:45:04] <xover1> using 'eval_generate': Connection refused - connect(2
[2011/11/04 06:45:13] @ Quit: MrPunch: Ping timeout: 265 seconds
[2011/11/04 06:45:26] @ Quit: Xanxess: Quit: Page closed
[2011/11/04 06:46:03] @ IndyMichaelB joined channel #puppet
[2011/11/04 06:46:45] <walkeran> xover1: Hrmm.. what agent package did you install on the client?
[2011/11/04 06:47:01] <walkeran> Using puppet enterprise??
[2011/11/04 06:47:07] @ Quit: marrusl: Quit: Ex-Chat
[2011/11/04 06:47:20] <xover1> err: Could not send report: SSL_connect returned=1 errno=0 state=SSLv3 read serv
[2011/11/04 06:47:45] <xover1> compiled on the client
[2011/11/04 06:47:50] <xover1> on the server i used the rpm
[2011/11/04 06:47:56] <_rc> "Connection refused" - the master isn't listening/accepting connections
[2011/11/04 06:48:05] <xover1> _rc: thats been sorted
[2011/11/04 06:48:10] <xover1> i get the ssl error
[2011/11/04 06:48:20] <walkeran> Oy vey... beats me where your stuff is, then. Check in your puppet.conf file to find your libdir
[2011/11/04 06:48:26] <walkeran> or ssldir
[2011/11/04 06:49:25] @ Quit: skm: Quit: Computer has gone to sleep.
[2011/11/04 06:49:28] <walkeran> Or, even better, run: puppet config print ssldir
[2011/11/04 06:49:57] <walkeran> I'm guessing it somewhere under that goofy-ass /var/opt/lib/pe-puppet/lib path you have going on
[2011/11/04 06:50:31] <xover1> where should it be installed then?
[2011/11/04 06:50:49] <xover1> that is where optional software goes, no?
[2011/11/04 06:51:39] <_rc> many of us are used to the users/paths used by the opensource version as packaged by RH/Debian, so the PE stuff looks weird
[2011/11/04 06:51:40] <walkeran> I guess, if you installed from source, that's probably the best place. Still seems like an overly confusing path. I'd opt for /opt/puppet/var/lib, /opt/puppet/bin, etc
[2011/11/04 06:51:55] @ olibuijr joined channel #puppet
[2011/11/04 06:52:14] <walkeran> But yeah... like _rc said, I'm used to /etc for the configs, /var/lib/puppet as vardir, and the binaries in /usr/bin
[2011/11/04 06:52:46] <olibuijr> I have a question, can i somehow change the string that puppet uses to tell aptitude to install packages, to allow untrusted packages?
[2011/11/04 06:54:32] @ mawser joined channel #puppet
[2011/11/04 06:54:44] <walkeran> olibuijr: In /etc/apt/apt.conf.d/gpg-checker.conf you can set "GPG::Check false;" to allow it to install unsigned packages, but AFAIK, you can't pass an additional argument in the package install resource. There might be another way to do it for a specific repo, too, but I'm not sure as I don't use Apt
[2011/11/04 06:57:48] @ Tobu joined channel #puppet
[2011/11/04 06:57:49] <olibuijr> alright, thanks, ill give it a shot!
[2011/11/04 06:58:27] @ benji_ joined channel #puppet
[2011/11/04 06:58:54] @ Quit: themadcanudist: Quit: Leaving.
[2011/11/04 07:00:14] @ sduchesneau joined channel #puppet
[2011/11/04 07:02:49] <hacim> i'm trying to upgrade to 2.7.6, but its not finding basic facter variables for templates
[2011/11/04 07:02:57] <hacim> eg. Failed to parse template apache/munin-stats: Could not find value for 'hostname'
[2011/11/04 07:03:08] <hacim> does anyone know anything about this?
[2011/11/04 07:05:00] @ jim80net joined channel #puppet
[2011/11/04 07:05:58] @ Quit: caviwipes: Quit: . . . n/a . . .
[2011/11/04 07:06:14] @ Quit: benji_: Quit: leaving
[2011/11/04 07:08:18] @ aj__ joined channel #puppet
[2011/11/04 07:08:25] @ lborda joined channel #puppet
[2011/11/04 07:10:56] <xover1> where is refers to installing modules?
[2011/11/04 07:10:58] @ derjohn_foo joined channel #puppet
[2011/11/04 07:11:07] <xover1> do i just copy my files into a dir that i create under modules?
[2011/11/04 07:11:08] @ Quit: derjohn_mob: Read error: Connection reset by peer
[2011/11/04 07:12:05] <_rc> that's usually what it means, but without knowing what the it you're referring to means
[2011/11/04 07:13:21] @ Quit: aj__: Ping timeout: 258 seconds
[2011/11/04 07:17:10] @ lwhalen joined channel #puppet
[2011/11/04 07:17:51] @ skm joined channel #puppet
[2011/11/04 07:17:52] <xover1> http://pastebin.com/28kxkU3K
[2011/11/04 07:19:05] @ matchaw_ joined channel #puppet
[2011/11/04 07:19:06] @ Quit: matchaw_: Read error: Connection reset by peer
[2011/11/04 07:19:12] <xover1> http://pastebin.com/AqfSLghT
[2011/11/04 07:19:17] @ Quit: matchaw___: Ping timeout: 240 seconds
[2011/11/04 07:19:25] <_rc> that's a part from a manifest. if you put that in modules/ssh/init.pp wrapped with class ssh { } it'll probably do what you intend
[2011/11/04 07:19:52] @ illsci joined channel #puppet
[2011/11/04 07:20:10] <_rc> the file needs to have the path modules/ssh/files/keys
[2011/11/04 07:20:23] <illsci> Is there anything special you have to do to get templates to work when you're using environments?
[2011/11/04 07:20:45] <illsci> I'm seeing that the "production" environment's modules templates are used instead of the ones inside the modules in my "testing" environment....
[2011/11/04 07:21:09] <illsci> ronments/test/modules/ovpn/manifests/config.pp: content => template("ovpn/server.erb"),
[2011/11/04 07:21:14] @ asteve joined channel #puppet
[2011/11/04 07:21:15] @ Quit: asteve: Changing host
[2011/11/04 07:21:15] @ asteve joined channel #puppet
[2011/11/04 07:21:28] <illsci> Is there a different path I should put in the template() ?
[2011/11/04 07:21:38] @ Quit: derjohn_foo: Ping timeout: 245 seconds
[2011/11/04 07:22:35] <illsci> this was driving me nuts last night... I couldn't figure out why it kept rendering the same values on different hosts... and it was because I had a hard coded value in my "production" environment but my other ones had a lookup call
[2011/11/04 07:23:01] <illsci> So I need to figure out how to use templates with environments so they pick the template inside that environments directory... and not the "production" one...
[2011/11/04 07:24:39] @ Quit: skm: Quit: Computer has gone to sleep.
[2011/11/04 07:26:00] @ derjohn_foo joined channel #puppet
[2011/11/04 07:27:41] @ labrown joined channel #puppet
[2011/11/04 07:30:42] @ rsnhyperion joined channel #puppet
[2011/11/04 07:32:49] <xover1> how do i specify the destination then?
[2011/11/04 07:33:01] <xover1> the point being that i want to copy to /root/.ssh
[2011/11/04 07:33:21] @ neonflux joined channel #puppet
[2011/11/04 07:33:26] <xover1> i was using this website example
[2011/11/04 07:33:26] <xover1> http://askubuntu.com/questions/61080/how-to-copy-file-to-all-clients-using-puppet
[2011/11/04 07:34:42] @ skm joined channel #puppet
[2011/11/04 07:34:54] <xover1> jesus
[2011/11/04 07:34:56] <xover1> this is crazy
[2011/11/04 07:34:56] <ewoud> xover1: from that example: file { "/etc/sudoers": ... }
[2011/11/04 07:35:04] <ewoud> /etc/sudoers is the destination
[2011/11/04 07:35:19] <xover1> so i have used /root/.ssh/authorized_keys
[2011/11/04 07:35:27] <xover1> and my source is modules/ssh/keys
[2011/11/04 07:35:29] <xover1> what is the issue?
[2011/11/04 07:36:14] <ewoud> xover1: what's the error?
[2011/11/04 07:36:26] <ewoud> and can you paste the module on something like pastebin?
[2011/11/04 07:36:41] @ Quit: sduchesneau: Ping timeout: 252 seconds
[2011/11/04 07:37:15] <ewoud> xover1: oh nm, I see you pasted that before on pastebin
[2011/11/04 07:38:08] <ewoud> I have no experience using puppet apply + fileserver; I heard it works, but only used the puppet server myself
[2011/11/04 07:38:12] <xover1> http://pastebin.com/AqfSLghT
[2011/11/04 07:38:48] <xover1> i dont know wat this module stuff is
[2011/11/04 07:38:54] <xover1> its just a text file
[2011/11/04 07:39:00] <xover1> that i put under modules
[2011/11/04 07:39:49] <illsci> ok so this is issue 7497...
[2011/11/04 07:40:11] @ Quit: ckauhaus: Quit: Leaving.
[2011/11/04 07:40:22] <illsci> so I have to set templatedir in the environments config...
[2011/11/04 07:40:30] <ewoud> xover1: hmm, and your puppet.conf?
[2011/11/04 07:43:23] @ Quit: dnjaramba: Read error: Connection reset by peer
[2011/11/04 07:43:25] @ Quit: ezryx: Remote host closed the connection
[2011/11/04 07:43:42] @ dnjaramba joined channel #puppet
[2011/11/04 07:44:34] @ Quit: Stenbryggen: Quit: Leaving.
[2011/11/04 07:45:48] @ ezryx joined channel #puppet
[2011/11/04 07:46:01] @ Quit: xover1: Read error: Connection reset by peer
[2011/11/04 07:47:02] <illsci> yeah... I'm not understanding this... http://docs.puppetlabs.com/guides/environment.html How do you configure environments to use template's inside modules inside an environment?
[2011/11/04 07:48:13] <_Spike_> very carefully!
[2011/11/04 07:48:23] <asteve> and slowly with caution
[2011/11/04 07:49:04] @ Quit: skm: Quit: Computer has gone to sleep.
[2011/11/04 07:49:30] @ skm joined channel #puppet
[2011/11/04 07:49:59] @ Quit: j^2: Ping timeout: 276 seconds
[2011/11/04 07:50:02] @ Quit: skm: Remote host closed the connection
[2011/11/04 07:50:27] @ xover1 joined channel #puppet
[2011/11/04 07:50:44] @ j^2 joined channel #puppet
[2011/11/04 07:50:54] @ Quit: jodok: Quit: jodok
[2011/11/04 07:52:22] @ ghg joined channel #puppet
[2011/11/04 07:53:11] @ zorg1 joined channel #puppet
[2011/11/04 07:53:33] @ Quit: pmorillon: Quit: pmorillon
[2011/11/04 07:54:19] @ jodok joined channel #puppet
[2011/11/04 07:54:20] @ Quit: jodok: Changing host
[2011/11/04 07:54:20] @ jodok joined channel #puppet
[2011/11/04 07:56:02] @ Quit: thenulldevice: Quit: Leaving.
[2011/11/04 07:56:16] <Volcane> illsci: no config needed, it just does it
[2011/11/04 07:56:36] @ Quit: MattyM: Ping timeout: 256 seconds
[2011/11/04 07:57:34] @ themadcanudist joined channel #puppet
[2011/11/04 07:57:38] @ Quit: Vasik\|work: Remote host closed the connection
[2011/11/04 08:00:34] <illsci> hmm it's totally not working
[2011/11/04 08:01:01] @ ckauhaus joined channel #puppet
[2011/11/04 08:01:28] <illsci> once I updated the "production" environments template it started working...
[2011/11/04 08:01:41] <Volcane> how do you set the environment to use?
[2011/11/04 08:01:58] <illsci> puppet agent --test --environment test
[2011/11/04 08:02:17] <xover1> http://pastebin.com/mUYmZr8j
[2011/11/04 08:03:20] @ Quit: ghg: Quit: Computer has gone to sleep.
[2011/11/04 08:03:24] <illsci> I verified that the catalogs being compiled on the master showed they were using the "test" environment
[2011/11/04 08:03:41] @ pmorillon joined channel #puppet
[2011/11/04 08:03:49] <illsci> and no matter what I did it would keep using hte hardcoded value in the template in the "production" env...
[2011/11/04 08:04:18] <illsci> the master is puppet 2.7.1 and the agents are 2.6.2
[2011/11/04 08:04:22] <illsci> maybe that has something to do with it..
[2011/11/04 08:06:30] <illsci> Another question I had is will puppet notice when facts change and rerender templates?
[2011/11/04 08:06:41] <illsci> templates doing fact lookups
[2011/11/04 08:07:54] @ csanchez joined channel #puppet
[2011/11/04 08:08:43] @ Quit: dbarcelo: Quit: Leaving
[2011/11/04 08:08:45] @ gniks joined channel #puppet
[2011/11/04 08:09:06] @ MattyM joined channel #puppet
[2011/11/04 08:10:56] @ hobbeswalsh joined channel #puppet
[2011/11/04 08:12:21] @ benjiedmund joined channel #puppet
[2011/11/04 08:12:56] @ Quit: st33v: Ping timeout: 260 seconds
[2011/11/04 08:13:05] @ rcrowley joined channel #puppet
[2011/11/04 08:13:46] @ Quit: jaydg: Quit: Verlassend
[2011/11/04 08:13:56] @ Quit: lpaul7: Quit: Leaving
[2011/11/04 08:14:25] @ hyde joined channel #puppet
[2011/11/04 08:14:52] @ Quit: kermit: Ping timeout: 252 seconds
[2011/11/04 08:16:40] @ ghg joined channel #puppet
[2011/11/04 08:16:40] @ soh joined channel #puppet
[2011/11/04 08:16:49] <aranc23> illsci: yes
[2011/11/04 08:19:36] @ kermit joined channel #puppet
[2011/11/04 08:21:08] @ indigo__ joined channel #puppet
[2011/11/04 08:22:12] @ benkevan joined channel #puppet
[2011/11/04 08:23:38] @ pjsz_ joined channel #puppet
[2011/11/04 08:26:03] @ TRV joined channel #puppet
[2011/11/04 08:26:50] @ mawser_ joined channel #puppet
[2011/11/04 08:26:51] @ Quit: mawser: Read error: Connection reset by peer
[2011/11/04 08:26:51] @ mawser_ is now known as mawser
[2011/11/04 08:27:36] @ Quit: jab_doa: Quit: Verlassend
[2011/11/04 08:28:31] @ st33v_ joined channel #puppet
[2011/11/04 08:29:01] @ Quit: dnjaramba: Ping timeout: 240 seconds
[2011/11/04 08:29:14] @ cruzfox joined channel #puppet
[2011/11/04 08:29:42] @ dnjaramba joined channel #puppet
[2011/11/04 08:29:59] <astrostl> can i access the shell's $HOSTNAME var from an exec/unless? if so, how?
[2011/11/04 08:30:15] <indigo__> when i set some resource defaults like "Package { ... }" inside a class, what is the scope of those defaults?
[2011/11/04 08:30:31] @ st33v_ is now known as st33v
[2011/11/04 08:31:12] <indigo__> i had thought they would apply only to packages i declare inside the class where i declared the defaults, but it's looking like also it's applying to packages in classes included by my class that defines the defaults.
[2011/11/04 08:32:05] @ Quit: kermit: Ping timeout: 252 seconds
[2011/11/04 08:33:48] @ st33v_ joined channel #puppet
[2011/11/04 08:34:05] @ tizzo-afk is now known as tizzo
[2011/11/04 08:34:14] <Volcane> yeah unfortunatly it applies to included classes too
[2011/11/04 08:34:18] @ Quit: toi: Ping timeout: 252 seconds
[2011/11/04 08:34:42] @ riha left channel #puppet ()
[2011/11/04 08:35:09] @ TRV left channel #puppet ()
[2011/11/04 08:35:11] @ Quit: kubicek: Excess Flood
[2011/11/04 08:35:30] @ kubicek joined channel #puppet
[2011/11/04 08:36:18] @ saysjonathan joined channel #puppet
[2011/11/04 08:36:38] @ Quit: st33v: Ping timeout: 245 seconds
[2011/11/04 08:36:38] @ st33v_ is now known as st33v
[2011/11/04 08:36:39] <kjetilho> astrostl: '/bin/bash -c "echo $HOSTNAME"'
[2011/11/04 08:36:45] <indigo__> i'm reading http://docs.puppetlabs.com/guides/scope_and_puppet.html and i guess that's going away. which is good.
[2011/11/04 08:36:52] @ kenperkins joined channel #puppet
[2011/11/04 08:37:03] @ Quit: rsnhyperion: Quit: rsnhyperion
[2011/11/04 08:37:09] <astrostl> kjetilho: thx
[2011/11/04 08:37:36] <indigo__> if i'm reading it correctly, the defaults that resources from say the "apache" class find depend on where the first "include apache" happens to be.
[2011/11/04 08:38:15] <kjetilho> astrostl: actually I don't know if it works
[2011/11/04 08:38:35] @ Quit: sumsuch: Quit: sumsuch
[2011/11/04 08:38:40] <astrostl> i used $(hostname) instead
[2011/11/04 08:39:50] <kjetilho> aha. then your problem was that HOSTNAME wasn't set in the environment
[2011/11/04 08:40:45] <kjetilho> I thought you wanted to test for the case where ${::hostname} (in Puppet) != $HOSTNAME (in environment)
[2011/11/04 08:40:54] @ Quit: lkoranda: Quit: Ex-Chat
[2011/11/04 08:40:57] @ Arash joined channel #puppet
[2011/11/04 08:41:12] <indigo__> so let's say in the top scope i do Package{require=>Exec['apt-update']} and then in some class I want to make packages require something else, also. Is there a way I can express "add these requirements, plus whatever other requirements might have been defined in parent scopes"?
[2011/11/04 08:41:34] <kjetilho> +>
[2011/11/04 08:41:53] <Arash> hello
[2011/11/04 08:41:55] <kjetilho> but it doesn't work for defaults
[2011/11/04 08:41:58] <kjetilho> I think
[2011/11/04 08:42:17] <Arash> is there any chance to configure ipfw with puppet
[2011/11/04 08:42:37] <kjetilho> indigo__: sorry, I'm confused. if you specify anything concrete in a resource, the default is ignored
[2011/11/04 08:42:43] <kjetilho> so the answer is "no"
[2011/11/04 08:43:23] @ Quit: kenperkins: Quit: Computer has gone to sleep.
[2011/11/04 08:45:32] <indigo__> here's the real situation: i'm managing apt sources, so I need a dependency that makes sure anything that changes /etc/apt/sources.d and then apt-get update runs before attempting to install any packages. So I accomplished that with a Package{require=>Exec['apt-update']} in site.pp at the top. But I also have one class that requires some packages from backports, so I have to install a file in /etc/apt/preferences.d to make them available
[2011/11/04 08:45:38] @ Demosthenes joined channel #puppet
[2011/11/04 08:46:16] <indigo__> maybe +> is the thing to use. I'm testing it now.
[2011/11/04 08:46:45] <xover1> what is this modules business?
[2011/11/04 08:46:50] @ Atwa1337 joined channel #puppet
[2011/11/04 08:47:01] <xover1> do i have to somewhere declare that i have a file in modules that i want to copy to my clients?
[2011/11/04 08:47:07] <xover1> this is painfuil
[2011/11/04 08:47:44] @ Quit: saysjonathan: Quit: leaving
[2011/11/04 08:47:46] <indigo__> xover1: yes, if you put a file in a module, you also must have a manifest that says what to do with it.
[2011/11/04 08:47:58] @ Vulcan[OMNI] joined channel #puppet
[2011/11/04 08:48:00] <indigo__> xover1: well, you can put files in modules and not do anything with them, but I'm not sure why that's useful.
[2011/11/04 08:48:12] @ bretm joined channel #puppet
[2011/11/04 08:48:40] <indigo__> xover1: try v
[2011/11/04 08:48:42] @ Quit: olibuijr: Remote host closed the connection
[2011/11/04 08:48:44] <indigo__> xover1: try http://docs.puppetlabs.com/learning/modules1.html
[2011/11/04 08:49:59] <xover1> indigo__: all i want to do is copy a file from master to client lol
[2011/11/04 08:50:06] <xover1> apparently its really difficult
[2011/11/04 08:50:35] @ Quit: ckauhaus: Ping timeout: 244 seconds
[2011/11/04 08:50:53] <indigo__> xover1: if that's all you want to do, then use scp.
[2011/11/04 08:51:23] <indigo__> you don't need puppet to copy files.
[2011/11/04 08:52:48] @ Quit: dle: Quit: Quit
[2011/11/04 08:52:53] <xover1> to all my servers?
[2011/11/04 08:52:55] <xover1> hmm
[2011/11/04 08:52:56] @ ckauhaus joined channel #puppet
[2011/11/04 08:53:18] @ Quit: lofidellity: Ping timeout: 245 seconds
[2011/11/04 08:54:00] <indigo__> for s in EACH OF YOUR SERVERS; do scp file "$s:"; done
[2011/11/04 08:54:25] <bd__> also, clusterssh and/or parallell-rsync
[2011/11/04 08:55:02] <indigo__> or use NFS or similar
[2011/11/04 08:55:34] <xover1> urgh no because if you dont have ssh access then you cannot actually scp can you
[2011/11/04 08:55:48] <indigo__> what are you doing running servers without ssh access?
[2011/11/04 08:55:57] <xover1> i want to copy the keys
[2011/11/04 08:55:58] @ bretm_ joined channel #puppet
[2011/11/04 08:56:10] @ Quit: Meier: Quit: Leaving.
[2011/11/04 08:56:49] @ Quit: raphdg: Remote host closed the connection
[2011/11/04 08:58:10] @ Quit: jaytaph: Quit: jaytaph
[2011/11/04 08:58:29] @ Quit: ghanima: Ping timeout: 252 seconds
[2011/11/04 08:59:22] @ Quit: bretm: Ping timeout: 260 seconds
[2011/11/04 08:59:24] @ Quit: greenmang0: Quit: Leaving
[2011/11/04 09:00:23] @ Quit: ezryx: Remote host closed the connection
[2011/11/04 09:00:34] <joe-mac> if you need a key on a server in a chicken and egg scenario use wget in your kickstart %post or your post-install cmd if you're using preseed
[2011/11/04 09:01:47] @ maes_ joined channel #puppet
[2011/11/04 09:02:00] @ Quit: Vulcan[OMNI]: Quit: Vulcan[OMNI]
[2011/11/04 09:03:16] @ dan__t joined channel #puppet
[2011/11/04 09:03:44] <dan__t> Hi.
[2011/11/04 09:03:55] @ MattyM76 joined channel #puppet
[2011/11/04 09:04:08] @ Quit: maes_`: Ping timeout: 245 seconds
[2011/11/04 09:04:17] @ Quit: MattyM: Read error: No route to host
[2011/11/04 09:04:43] @ yann2 joined channel #puppet
[2011/11/04 09:05:15] @ krish joined channel #puppet
[2011/11/04 09:05:20] <krish> hi
[2011/11/04 09:05:31] <yann2> hello! I want to create a custom fact, let's say "firewall_roles", on all my server nodes. Can I iterate through the values of that fact for all the nodes, to generate a file on every node?
[2011/11/04 09:05:36] <krish> i have this package ensure => present
[2011/11/04 09:05:37] <yann2> not sure if my question is clear enough
[2011/11/04 09:05:56] <krish> but i want to exclude some depends of that package from install
[2011/11/04 09:06:02] <krish> howd be a best way to do it?
[2011/11/04 09:06:28] <fhd> I want to manage the puppet master itself via puppet, how do you guys normally do that? Apply manifests locally or does it contact itself using puppetd?
[2011/11/04 09:06:59] <indigo__> fhd: i have mine contacting itself
[2011/11/04 09:07:07] @ Quit: ckauhaus: Ping timeout: 244 seconds
[2011/11/04 09:07:27] @ Quit: csanchez: Quit: csanchez
[2011/11/04 09:07:47] @ xover1 left channel #puppet ()
[2011/11/04 09:07:58] <bd__> fhd: contacting itself, as it is actually a cluster and i want all nodes to be up to date :)
[2011/11/04 09:08:19] @ lofidellity joined channel #puppet
[2011/11/04 09:08:23] <fhd> bd__, indigo__: Thanks, was just wondering what's the common way. Contacting itself seems to be it :)
[2011/11/04 09:10:12] @ skippy joined channel #puppet
[2011/11/04 09:13:11] @ Quit: MattyM76: Ping timeout: 276 seconds
[2011/11/04 09:13:22] @ Stenbryggen joined channel #puppet
[2011/11/04 09:13:24] <unixdaemon> masterzen: Volcane any idea where in the file type it checks if the file needs to have modifications or not?
[2011/11/04 09:13:32] @ Quit: zorg1: Quit: Leaving
[2011/11/04 09:13:42] @ sduchesneau joined channel #puppet
[2011/11/04 09:15:35] <mawser> Can I define the libdir in an environment? Puppet throws error when I try to use a defined environment that is not production
[2011/11/04 09:15:36] <mawser> err: /File[/var/lib/puppet/lib]: Could not evaluate: Could not retrieve information from environment dev source(s) puppet://puppet/plugins
[2011/11/04 09:15:36] <mawser> notice: /File[/var/lib/puppet/lib/puppet]: Dependency File[/var/lib/puppet/lib] has failures: true
[2011/11/04 09:16:37] @ shadoi joined channel #puppet
[2011/11/04 09:18:05] @ Quit: sduchesneau: Ping timeout: 240 seconds
[2011/11/04 09:19:27] @ sumsuch joined channel #puppet
[2011/11/04 09:19:42] <_plg> hi, is there a tool where I can say something like node N1, N2, N3 and classes C1,C2 and it would create me a manifest for that nodes ?
[2011/11/04 09:20:22] @ rsnhyperion joined channel #puppet
[2011/11/04 09:20:58] @ Quit: pieter_: Quit: Ex-Chat
[2011/11/04 09:22:26] <_plg> before that, is there a way of knowing all the modules that I have in my master?
[2011/11/04 09:22:52] @ Quit: Arash: Quit: Page closed
[2011/11/04 09:23:09] @ p3rror joined channel #puppet
[2011/11/04 09:23:53] @ Quit: benkevan: Quit: leaving
[2011/11/04 09:24:09] @ Quit: st33v: Quit: st33v
[2011/11/04 09:24:13] @ Quit: derjohn_foo: Ping timeout: 240 seconds
[2011/11/04 09:24:32] @ kenperkins joined channel #puppet
[2011/11/04 09:25:11] <skippy> the Pro Puppet book states "For networks of one to two thousand Puppet managed nodes, a single Puppet master system
[2011/11/04 09:25:14] <skippy> running inside of Apache with Passenger is often sufficient
[2011/11/04 09:25:17] <skippy> "
[2011/11/04 09:25:22] @ Quit: rcrowley: Quit: Computer has gone to sleep.
[2011/11/04 09:25:32] <skippy> what kind of hardware is necessary on the Puppet Master to support that?
[2011/11/04 09:25:41] <joe-mac> idk, is that without stored configs?
[2011/11/04 09:26:47] @ johnston joined channel #puppet
[2011/11/04 09:27:15] @ Quit: lborda: Quit: Ex-Chat
[2011/11/04 09:27:46] @ lak joined channel #puppet
[2011/11/04 09:27:47] @ Quit: lak: Changing host
[2011/11/04 09:27:47] @ lak joined channel #puppet
[2011/11/04 09:29:38] @ st33v joined channel #puppet
[2011/11/04 09:30:43] <kjetilho> skippy: a modern server. 16G RAM, eight cores, fast disks
[2011/11/04 09:31:17] <skippy> kjetilho: thanks. We'll be supporting ~2K servers when we're all done, so want to build for the right scale.
[2011/11/04 09:31:29] <skippy> I guess that rules out running our master as a VMware guest for now.
[2011/11/04 09:32:16] @ Mladen joined channel #puppet
[2011/11/04 09:34:28] <Mladen> hi
[2011/11/04 09:34:38] <Mladen> anybody could give me some info about puppet?
[2011/11/04 09:35:02] @ Quit: helindbe: Quit: helindbe
[2011/11/04 09:35:45] <bd__> o_O
[2011/11/04 09:37:39] <Mladen> anybody
[2011/11/04 09:37:44] <Mladen> ?
[2011/11/04 09:37:52] <skippy> Mladen: what specifically would you like to know?
[2011/11/04 09:38:48] @ ezryx joined channel #puppet
[2011/11/04 09:39:01] <Mladen> o thanx God there is people here..:) I was thinking web forums killed IRC :)
[2011/11/04 09:39:22] <skippy> Mladen: there's lots of people here, but it helps to ask a specific question
[2011/11/04 09:39:40] @ mpdude_ joined channel #puppet
[2011/11/04 09:39:45] @ Quit: mogitaff: Remote host closed the connection
[2011/11/04 09:39:55] <Mladen> well.. I would like to receive some input from somebody that using puppet in production.. I have lot of Linux server that has same/similar configuration...I developed bash scripts and debian packages to automate installation...
[2011/11/04 09:40:03] @ Quit: ezryx: Remote host closed the connection
[2011/11/04 09:40:05] <Mladen> I believe some tool like puppet would be next step..
[2011/11/04 09:40:17] <skippy> Mladen: yes. Using Puppet is the next step.
[2011/11/04 09:40:28] @ Quit: mpdude: Read error: Connection reset by peer
[2011/11/04 09:40:43] <Mladen> since I do not know too much about puppet, I wonder how hard is it to configure/use...and is there some WebGUI or something like that?
[2011/11/04 09:40:52] @ mpdude joined channel #puppet
[2011/11/04 09:41:01] <skippy> Mladen: no, there is no gui for configuring Puppet yet.
[2011/11/04 09:41:12] * indigo__ hopes "not ever"
[2011/11/04 09:41:22] <skippy> Mladen: have you read the information on the Puppet Labs website? It's a really good place to start
[2011/11/04 09:41:54] <skippy> Mladen: this is a good starting point: http://docs.puppetlabs.com/learning/
[2011/11/04 09:42:01] @ peridot joined channel #puppet
[2011/11/04 09:42:01] <Mladen> I downloaded some presentation.. inside it I found this IRC channel and wanted to see if comunity is acctually alive..
[2011/11/04 09:42:18] @ ranulf joined channel #puppet
[2011/11/04 09:42:23] <Mladen> OK..
[2011/11/04 09:42:24] <fenrus02> quite so. see how many people are here?
[2011/11/04 09:42:42] @ ckauhaus joined channel #puppet
[2011/11/04 09:42:44] <Mladen> I will check official docs..
[2011/11/04 09:43:02] @ ranulf left channel #puppet ()
[2011/11/04 09:43:48] @ Quit: mpdude_: Ping timeout: 244 seconds
[2011/11/04 09:43:51] <skippy> I'm not trying to ignore you with an RTFM, but if you want to get familiar with Puppet in general, that link is likely more helpful than us trying to answer vague questions.
[2011/11/04 09:44:07] <skippy> suffice to say, the Puppet community is very active.
[2011/11/04 09:44:26] <skippy> If you have specific questions after getting familiar with Puppet, then this is the place to ask.
[2011/11/04 09:44:41] @ Quit: Razique: Quit: Razique
[2011/11/04 09:44:47] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 09:44:48] <Mladen> I understand. And I appreciate your efforts to help!
[2011/11/04 09:45:19] <Mladen> just..I would like to hear real users experience..like..could you provide me some example how you use puppet, and what you like the most about it?
[2011/11/04 09:45:49] <skippy> we're still planning our Puppet installation, actually. but we intend to use it to configure hosts after we Kickstart them.
[2011/11/04 09:45:56] <_rc> it's a fine way to make all of your broken systems consistently broken
[2011/11/04 09:46:05] <bd__> :)
[2011/11/04 09:46:08] <peridot> I'm switching from an older version of cfengine to puppet, and I'm finding it very powerful. It lets me do things that I couldn't do before.
[2011/11/04 09:46:16] <skippy> we currently use a collection of RPMs and scripts to do that configuration, but the scripts are messy and overly complex and hard to maintain.
[2011/11/04 09:46:17] @ helindbe joined channel #puppet
[2011/11/04 09:46:44] <bd__> Mladen: wikimedia recently opened ther puppet config repository. while it is nothing youi'd want read now, it can be helpful later.
[2011/11/04 09:47:52] <peridot> Like one of the tasks it's doing is: I need to populate a dozen different systems with a file which sets an environmental variable on a specific user's login which incorporates the mac address of that system.
[2011/11/04 09:49:00] @ matchaw_ joined channel #puppet
[2011/11/04 09:49:45] @ trickyf left channel #puppet ()
[2011/11/04 09:49:56] <matchaw_> hi all, anyone know how to look through an array when passed to a define?
[2011/11/04 09:50:31] @ tightwork joined channel #puppet
[2011/11/04 09:51:28] @ soh joined channel #puppet
[2011/11/04 09:52:38] @ Quit: mpdude: Remote host closed the connection
[2011/11/04 09:52:50] <Mladen> do you know what is difference between pupet open source and enterprise versions?
[2011/11/04 09:53:22] <skippy> if you want to use the Puppet Dashboard, the Enterprise version makes is much, much easier to install.
[2011/11/04 09:53:39] <skippy> Enterprise also gets you professional support, if that's important to you.
[2011/11/04 09:53:44] <magicaltrout> mladen i use it to control databases abduction bi systems :-)
[2011/11/04 09:53:56] <magicaltrout> er
[2011/11/04 09:54:01] <magicaltrout> and
[2011/11/04 09:54:09] <boklm> matchaw_: unfortunatly, I think it's not possible to iterate on the array, but if using it as name of an other resource, multiple resources are created (one for each element of the array)
[2011/11/04 09:54:16] <magicaltrout> silly predictive text
[2011/11/04 09:54:23] @ williemach joined channel #puppet
[2011/11/04 09:54:49] <matchaw_> boklm: Thanks, I thought this but when trying it with an augeas resource it doesn't seem to run augeas twice?
[2011/11/04 09:55:27] @ williemach left channel #puppet ()
[2011/11/04 09:55:54] <magicaltrout> much much easier ? i think that's a bit steep ;-)
[2011/11/04 09:56:19] <tightwork> say I want to build an interface that accepts variables that are then inserted into a manifest. What would I be looking at?
[2011/11/04 09:56:20] <skippy> not having to muck around with Passenger and gems is a big win in my book. :)
[2011/11/04 09:56:50] <boklm> matchaw_: for instance, to create a file for each element of the array: file { "something-$array":
[2011/11/04 09:57:32] * magicaltrout installs the deb and runs rake ;-)
[2011/11/04 09:57:41] @ Vulcan[OMNI] joined channel #puppet
[2011/11/04 09:57:42] <matchaw_> boklm, this this seems to work fine, but using augeas resource it doesn't work?
[2011/11/04 09:58:26] @ kaptk2 joined channel #puppet
[2011/11/04 09:58:43] @ shintaku joined channel #puppet
[2011/11/04 09:58:54] <tightwork> simple example, say I want a rails interface that asks a user 'what file name?' .. and it inserts the name of the file into the manifest ..?
[2011/11/04 09:59:03] <boklm> matchaw_: I don't see why it shouldn't work, but didn't try it
[2011/11/04 09:59:12] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 09:59:54] <shintaku> i'm not familiar with puppet yet, but having a feeling if i were using it, it would make tracking what's on vm host easier. if puppet is appropriate here, can someone reply with how they use it on http://serverfault.com/questions/327877/how-to-keep-track-of-vm-hosts-and-guests-and-ip-addresses. Thanks
[2011/11/04 10:00:02] <boklm> matchaw_: but it only works on name, not other parameters
[2011/11/04 10:00:04] <mawser> I'm running in to a problem when I set --environment to dev
[2011/11/04 10:00:07] <mawser> Could not retrieve information from environment dev source(s) puppet://puppet/plugins
[2011/11/04 10:00:35] <mawser> Any thoughts on where I might be getting this error from? --environment production has no errors
[2011/11/04 10:00:38] @ benkevan joined channel #puppet
[2011/11/04 10:01:01] @ Quit: alanevans: Remote host closed the connection
[2011/11/04 10:01:09] <matchaw_> boklm, ok. any ideas on how to iterate over an array when the variable is not $name?
[2011/11/04 10:01:17] <tightwork> GUI -> form field -> insert into puppet ?
[2011/11/04 10:02:06] <peridot> Can anyone point me to a good resource for setting up passenger for puppet on rhel? I finally went through rpm hell and got it installed, but the instructions for setting it up are all off...different directories, the manifest.pp I'm supposed to use is for debian systems, etc.
[2011/11/04 10:02:11] <kjetilho> matchaw_: Ruby DSL
[2011/11/04 10:02:24] <kjetilho> or use a helper define
[2011/11/04 10:02:47] @ rharrison joined channel #puppet
[2011/11/04 10:02:47] <matchaw_> kjetilho: what do you mean a helper define?
[2011/11/04 10:02:47] <boklm> matchaw_: creating a new define taking this argument as name
[2011/11/04 10:02:59] <matchaw_> boklm, ah ok.
[2011/11/04 10:03:10] <boklm> ah yes, maybe with ruby dsl
[2011/11/04 10:03:13] <mawser> peridot: Was this helpful? http://projects.puppetlabs.com/projects/1/wiki/Using_Passenger
[2011/11/04 10:03:15] <peridot> mawser: You have your puppet.conf setup correctly to point to your different environments?
[2011/11/04 10:03:25] <matchaw_> kjetilho: any examples with ruby DSL
[2011/11/04 10:03:40] @ duelafn joined channel #puppet
[2011/11/04 10:04:21] @ Quit: rharrison: Client Quit
[2011/11/04 10:04:23] <mawser> peridot: yes in my puppet.conf I have two subsections [dev] and [production] and each has it's own modulepath and a common site.pp
[2011/11/04 10:04:43] <kjetilho> scope.lookupvar("something").each do \|var\| ... end
[2011/11/04 10:04:45] <unixdaemon> shintaku: mcollective and either puppetdashboard or foreman would be the big win
[2011/11/04 10:04:46] <mawser> [dev]
[2011/11/04 10:04:46] <mawser> modulepath = /etc/puppet/env/dev:/etc/puppet/env/common
[2011/11/04 10:04:46] <mawser> manifest = /etc/puppet/manifests/site.pp
[2011/11/04 10:05:09] <shintaku> unixdaemon: thanks
[2011/11/04 10:05:11] <kjetilho> matchaw_: seriously, look at the docs at puppetlabs
[2011/11/04 10:05:14] <shintaku> i'll check that out
[2011/11/04 10:05:22] @ Quit: duelafn: Read error: Connection reset by peer
[2011/11/04 10:05:24] @ rharrison joined channel #puppet
[2011/11/04 10:05:25] <unixdaemon> shintaku: http://puppetlabs.com/blog/puppet-mcollective-make-for-quick-inventory-queries-part-2-of-2/
[2011/11/04 10:05:35] <unixdaemon> And you can write your own little reporting scripts
[2011/11/04 10:05:40] @ duelafn joined channel #puppet
[2011/11/04 10:05:48] <unixdaemon> all of which can use the facter facts
[2011/11/04 10:05:56] @ Quit: astrostl: Read error: Operation timed out
[2011/11/04 10:06:22] @ ianbrandt joined channel #puppet
[2011/11/04 10:06:25] <clustermagnet> guys, if a file changes, i would like to run a command
[2011/11/04 10:06:30] @ lborda joined channel #puppet
[2011/11/04 10:06:44] <clustermagnet> cant i notify a command from file?
[2011/11/04 10:07:02] <clustermagnet> can i do this:?
[2011/11/04 10:07:06] @ Quit: jodok: Quit: jodok
[2011/11/04 10:07:07] <tightwork> welcome to the Internet
[2011/11/04 10:07:15] @ Quit: Mladen: Ping timeout: 265 seconds
[2011/11/04 10:07:18] <tightwork> on ur own
[2011/11/04 10:07:35] @ Quit: shintaku: Quit: shintaku
[2011/11/04 10:07:59] <gniks> clustermagnet: file { "title": notify => Service[ 'service' ] }
[2011/11/04 10:08:05] <clustermagnet> http://pastie.org/2810897
[2011/11/04 10:09:07] @ Quit: ckauhaus: Ping timeout: 244 seconds
[2011/11/04 10:10:25] @ Quit: st33v: Quit: st33v
[2011/11/04 10:11:08] <peridot> I think maybe use notify => Exec [ 'blah' ]
[2011/11/04 10:11:17] <clustermagnet> peridot: lol yes, fixed :)
[2011/11/04 10:11:26] <clustermagnet> thanks! i think i maybe haz it...
[2011/11/04 10:11:27] @ Quit: bretm_: Ping timeout: 252 seconds
[2011/11/04 10:11:31] @ Quit: Ramonster: Quit: So Long and thanks for all the fish
[2011/11/04 10:11:34] <clustermagnet> stupid cold, cant think
[2011/11/04 10:11:46] @ astrostl joined channel #puppet
[2011/11/04 10:13:10] @ st33v joined channel #puppet
[2011/11/04 10:13:19] @ ckauhaus joined channel #puppet
[2011/11/04 10:13:19] @ MrRothstein joined channel #puppet
[2011/11/04 10:13:24] @ Quit: ghg: Quit: Computer has gone to sleep.
[2011/11/04 10:14:00] @ bretm joined channel #puppet
[2011/11/04 10:15:08] <hacim> i've got a define that is using the $name for a parameter, but its being set to the class name
[2011/11/04 10:15:14] <MrRothstein> hi I'm trying to create a module for webspheremq, i'm working on the server class. the class installs the required packages and a resource creates a queue manager. I'm getting a circular dependency... I'm not sure why the class depends on the resource
[2011/11/04 10:15:20] <hacim> am I not supposed to use the $namevar for that?
[2011/11/04 10:15:22] <Volcane> hacim: pastie
[2011/11/04 10:15:36] @ ckauhaus1 joined channel #puppet
[2011/11/04 10:15:47] @ soh joined channel #puppet
[2011/11/04 10:15:48] @ Quit: setient: Ping timeout: 245 seconds
[2011/11/04 10:15:59] <MrRothstein> here is my class http://pastie.org/2810921 and here is the error http://pastie.org/2810929
[2011/11/04 10:16:00] @ Quit: soh: Client Quit
[2011/11/04 10:16:04] @ Quit: robinbowes: Quit: leaving
[2011/11/04 10:16:06] @ Quit: ckauhaus1: Client Quit
[2011/11/04 10:16:26] <tightwork> internets
[2011/11/04 10:16:31] @ robinbowes joined channel #puppet
[2011/11/04 10:16:43] <MrRothstein> i'd like to understand why the class depends on the resource
[2011/11/04 10:17:04] <hacim> Volcane: https://paste.debian.net/142701/
[2011/11/04 10:17:49] <Volcane> hacim: version?
[2011/11/04 10:18:01] <hacim> Volcane: i've tried 2.6.12 and 2.7.6
[2011/11/04 10:18:01] @ mitchellh joined channel #puppet
[2011/11/04 10:18:11] <Volcane> in 2.7 that should work
[2011/11/04 10:18:28] <Volcane> and 2.6.12 actually
[2011/11/04 10:18:35] <hacim> it was working in 2.6.12 before
[2011/11/04 10:18:43] <hacim> then I tried to install 2.7 and things got wonky
[2011/11/04 10:18:58] @ peridot left channel #puppet ()
[2011/11/04 10:18:58] <hacim> what is ending up happening is $check_health_mode = $name,
[2011/11/04 10:19:05] <Volcane> i dont run 2.7 so cant confirm it works - should though
[2011/11/04 10:19:08] <hacim> gets set to the class name (site-mysql::server)
[2011/11/04 10:20:03] @ peridot joined channel #puppet
[2011/11/04 10:20:12] @ setient joined channel #puppet
[2011/11/04 10:20:45] @ Quit: pmorillon: Quit: pmorillon
[2011/11/04 10:21:18] @ Quit: andreaturli: Quit: http://quassel-irc.org - Chat comfortably. Anywhere.
[2011/11/04 10:22:07] <peridot> In the passenger docs, they refer to a puppetmaster.conf file that you need to move to /etc/httpd/conf.d/, but they don't tell me where this file comes from.
[2011/11/04 10:22:54] <Volcane> hacim: puppet apply seems to suggest it works fine - master might behave diff i guess
[2011/11/04 10:22:58] @ ezryx joined channel #puppet
[2011/11/04 10:22:59] <peridot> Is that the apache vhost config?
[2011/11/04 10:23:08] <agaffney> peridot: yes
[2011/11/04 10:23:18] @ skippy left channel #puppet ()
[2011/11/04 10:23:20] <agaffney> it's likely in the ext/ directory from the tarball
[2011/11/04 10:23:29] <agaffney> in the EPEL RPMs, that's installed at /usr/share/puppet/ext/
[2011/11/04 10:23:43] <peridot> thanks
[2011/11/04 10:23:44] <hacim> hm
[2011/11/04 10:24:23] <Volcane> hacim: http://www.pastie.org/2810969
[2011/11/04 10:24:51] <hacim> yeah, thats what I thought
[2011/11/04 10:25:53] <tightwork> Imagine I had a graphical user interface that accepted user input. How could I insert that into a manifest? I assume the easiest way is to use a ruby rails app, and there is a gem.. is there a puppet api, can anyone show me examples.. does this exist?
[2011/11/04 10:26:19] <Volcane> tightwork: http://docs.puppetlabs.com/guides/external_nodes.html
[2011/11/04 10:26:20] <Sylnai> tightwork: what kind of information are you inserting?
[2011/11/04 10:28:36] <duelafn> Hello, I have a directory { recursive => true }, but only one file in that dir needs to trigger a server restart. Of course, exec { blah: subscribe => File['/dir/the_file.conf'] } won't work ("Could not find dependency File['/dir/the_file.conf']"). Is it possible to specify the dependency without spelling out all files in the directory?
[2011/11/04 10:29:03] <tightwork> Sylnai: virtual host info
[2011/11/04 10:29:11] <tightwork> Sylnai: apache virtual host
[2011/11/04 10:29:20] @ Quit: kaptk2: Quit: Leaving.
[2011/11/04 10:29:33] <Sylnai> I store that kind of stuff in LDAP and use an external node classifier to polish it up a bit
[2011/11/04 10:29:41] @ kelseyhightower_ joined channel #puppet
[2011/11/04 10:29:42] @ djm68 joined channel #puppet
[2011/11/04 10:29:54] <tightwork> Sylnai: hopefully, interface should ask domain? and then insert variable into ServerName variable for vhost.conf of that vhost
[2011/11/04 10:30:13] @ Quit: kelseyhightower: Ping timeout: 255 seconds
[2011/11/04 10:30:13] @ kelseyhightower_ is now known as kelseyhightower
[2011/11/04 10:30:31] <tightwork> Volcane: thanks.. im reading now
[2011/11/04 10:31:01] <indigo__> so, i guess a puppet:/// URI isn't valid for a package source on os x? :(
[2011/11/04 10:31:40] <tightwork> Volcane: so basically I just need to get my GUI to talk yaml as an external_node to puppet?
[2011/11/04 10:31:56] <clustermagnet> guys, latest puppet 2.7.6… agent, not really working :)
[2011/11/04 10:31:56] <clustermagnet> debug: file_metadata supports formats: b64_zlib_yaml marshal pson raw yaml; using pson
[2011/11/04 10:32:03] <clustermagnet> kinda sitting there......
[2011/11/04 10:33:30] @ claussen joined channel #puppet
[2011/11/04 10:33:33] @ Quit: frostynova: Remote host closed the connection
[2011/11/04 10:34:22] @ jaytaph joined channel #puppet
[2011/11/04 10:35:28] <MrRothstein> it looks like using a "notify" inside a resource for a service defined in the class that also contains the resource definition causes a circular dependency. can someone explain why it's like that?
[2011/11/04 10:35:54] <MrRothstein> specifically, why is there a dependency from the class to the resource type
[2011/11/04 10:35:57] <Volcane> MrRothstein: show the code
[2011/11/04 10:36:09] <MrRothstein> http://pastie.org/2810921
[2011/11/04 10:36:16] <gepetto> ::redmine:: Wiki edit: Downloading_Puppet (#114) @ http://projects.puppetlabs.com/projects/puppet/wiki/Downloading_Puppet?version=114 (by Michael Stahnke)
[2011/11/04 10:37:07] <MrRothstein> it works fine when i take out the notify from exec "create listener", but i'd like to get the service restarted at that point
[2011/11/04 10:37:26] @ Quit: KristianDK: Quit: KristianDK
[2011/11/04 10:38:28] @ jsurfer joined channel #puppet
[2011/11/04 10:39:40] @ alanevans joined channel #puppet
[2011/11/04 10:40:10] @ jsm joined channel #puppet
[2011/11/04 10:40:17] @ the_ramink joined channel #puppet
[2011/11/04 10:40:18] @ Quit: hal-j: Quit: Leaving.
[2011/11/04 10:40:56] <MrRothstein> Volcate, i get the following error when i try to run it http://pastie.org/2810929
[2011/11/04 10:41:24] <MrRothstein> i dont get the "Class[Webspheremq::Server] => Webspheremq::Server::Create_qmgr[TEST])" part
[2011/11/04 10:42:19] <Volcane> line 27 tells it to require the class server
[2011/11/04 10:42:29] <Volcane> so create_qmgr can only happen after server
[2011/11/04 10:42:44] <Volcane> but then in create_qmgt you notify the service and you cannot notify things thats already happened
[2011/11/04 10:42:47] <Volcane> so its cirular
[2011/11/04 10:43:22] <peridot> Are these versions of rack and passenger in the wiki the ones I should be using? 1.1.0 and 2.2.15?
[2011/11/04 10:43:52] @ hal-j joined channel #puppet
[2011/11/04 10:44:12] @ mpdude joined channel #puppet
[2011/11/04 10:44:40] <Volcane> wiki noun, a website that's always out of date
[2011/11/04 10:45:03] <peridot> That's why I ask :P
[2011/11/04 10:45:08] <Volcane> look at the info on docs.puppetlabs.com
[2011/11/04 10:45:29] <the_ramink> peridot: which distro are you running?
[2011/11/04 10:45:44] <peridot> rhel 5.6
[2011/11/04 10:46:29] <MrRothstein> volcane, is there a way to restart a service after a config change? A server definition includes that service. I'd like the ability to add multiple qmgrs with create_qmgr and have the service restarted after each one.
[2011/11/04 10:46:48] <MrRothstein> and i need the server packages in order to execute the steps to create a new qmgr
[2011/11/04 10:46:52] <Volcane> MrRothstein: yes there is - but by adding the require you are requiring that the service is completed before the define.
[2011/11/04 10:46:56] @ Quit: j^2: Ping timeout: 248 seconds
[2011/11/04 10:46:56] <Volcane> MrRothstein: line 27
[2011/11/04 10:47:16] <Volcane> MrRothstein: so the service is done being managed - you cant then ask it to restart it because you have 2 competing statements
[2011/11/04 10:47:34] <Volcane> MrRothstein: 'only manage it before the define' and 'restart it after the define'
[2011/11/04 10:47:47] <Volcane> you cant have both without inventing a new concept of time.
[2011/11/04 10:47:54] @ j^2 joined channel #puppet
[2011/11/04 10:48:02] <clustermagnet> guys, is htere a known issue with puppet killing CPU running reports on the agent?
[2011/11/04 10:48:22] <clustermagnet> just upgraded facter and puppet agents, and its bareeeely running
[2011/11/04 10:48:22] <MrRothstein> so, by the time i define the service resource, i'm supposed to be done with any config it depends on?
[2011/11/04 10:48:24] @ Razique joined channel #puppet
[2011/11/04 10:48:24] <clustermagnet> killing cpu, etc
[2011/11/04 10:48:38] <the_ramink> peridot: those should work. If you're running 2.6.6 or better Puppet you can install passenger 3.0.9 from the Phusion Passenger rpm repo which has some nice features.
[2011/11/04 10:48:42] @ Quit: pjsz_: Quit: pjsz_
[2011/11/04 10:48:42] <clustermagnet> in debug mode, i see debug: file_metadata supports formats: b64_zlib_yaml marshal pson raw yaml; using pson
[2011/11/04 10:48:46] <Volcane> MrRothstein: line 27
[2011/11/04 10:48:50] <clustermagnet> and then it finally finishes....
[2011/11/04 10:48:55] <tightwork> ZOMG
[2011/11/04 10:49:01] @ duelafn left channel #puppet ()
[2011/11/04 10:49:31] <Volcane> MrRothstein: you _require_ the server class - so everything in the class needs to be completed prior to the define. you cannot then notify something thats completed
[2011/11/04 10:49:32] <clustermagnet> notice: Finished catalog run in 338.95 seconds
[2011/11/04 10:50:15] @ Quit: claussen: Quit: claussen
[2011/11/04 10:51:31] @ kermit joined channel #puppet
[2011/11/04 10:52:05] <astrostl> is http://projects.puppetlabs.com/projects/1/wiki/Using_Stored_Configuration the definitive stored configuration doc?
[2011/11/04 10:52:31] @ Quit: ckauhaus: Ping timeout: 244 seconds
[2011/11/04 10:52:45] @ newtMcKerr joined channel #puppet
[2011/11/04 10:53:13] @ Quit: jbraeuer: Quit: jbraeuer
[2011/11/04 10:53:17] @ JodieC joined channel #puppet
[2011/11/04 10:53:50] @ tvincent joined channel #puppet
[2011/11/04 10:56:35] @ brtl left channel #puppet ("PING 1320429421")
[2011/11/04 10:57:30] @ alop joined channel #puppet
[2011/11/04 10:59:00] @ lindenle joined channel #puppet
[2011/11/04 10:59:25] @ Ashutto joined channel #puppet
[2011/11/04 10:59:38] @ Quit: lak: Quit: Leaving.
[2011/11/04 10:59:47] @ Quit: n0ts: Ping timeout: 276 seconds
[2011/11/04 11:00:33] @ Quit: mitchellh: Quit: Computer has gone to sleep
[2011/11/04 11:01:01] <Volcane> astrostl: best look on docs.puppetlabs.com
[2011/11/04 11:01:14] <MrRothstein> volcane, i change require to include and everything worked as expected, but i'm not sure if it worked because the packages were installed on the previous runs...
[2011/11/04 11:01:19] @ mitchellh joined channel #puppet
[2011/11/04 11:02:34] @ cwebber joined channel #puppet
[2011/11/04 11:03:03] @ Quit: cruzfox: Quit: cruzfox
[2011/11/04 11:03:05] @ derjohn_foo joined channel #puppet
[2011/11/04 11:03:40] @ n0ts joined channel #puppet
[2011/11/04 11:04:05] <astrostl> Volcane: i did check there first. just grabbed the pdf of the entire thing, the only ref in it is a link to http://projects.puppetlabs.com/projects/1/wiki/Using_Stored_Configuration, so it's definitive for now
[2011/11/04 11:04:16] @ cruzfox_ joined channel #puppet
[2011/11/04 11:04:23] <astrostl> surprising to me that it hasn't been rolled into the main site yet
[2011/11/04 11:05:16] <illsci> how do you do subscribe with a symbol... is it ~
[2011/11/04 11:05:48] <walkeran> illsci: Are you talking about resource chaining?
[2011/11/04 11:05:53] <illsci> yeah
[2011/11/04 11:06:01] <Volcane> astrostl: http://docs.puppetlabs.com/guides/exported_resources.html
[2011/11/04 11:06:04] @ Quit: choonming: Read error: Connection timed out
[2011/11/04 11:06:09] <walkeran> -> for a require relationship, and ~> for notify
[2011/11/04 11:06:22] <Volcane> astrostl: so yeah i guess thats it then
[2011/11/04 11:06:28] <astrostl> Volcane: ahh, not referred to as "stored configuration"
[2011/11/04 11:06:31] <astrostl> tyvm
[2011/11/04 11:06:38] @ Quit: tlocalhos: Remote host closed the connection
[2011/11/04 11:06:49] <Volcane> astrostl: yeah - that page though link to the wiki page you found so probably still good
[2011/11/04 11:06:58] <illsci> -> is for require ~> is for notify... is there one for subscribe
[2011/11/04 11:07:04] <astrostl> what is that wiki anyway?
[2011/11/04 11:07:14] <Volcane> astrostl: user contributed docs
[2011/11/04 11:07:22] <astrostl> gotcha, thx
[2011/11/04 11:07:48] @ Quit: mawser: Quit: mawser
[2011/11/04 11:07:49] @ tlocalhos joined channel #puppet
[2011/11/04 11:07:52] <indigo__> lindenle: subscribe is the same as notify, in the other direction.
[2011/11/04 11:08:08] <illsci> ok..
[2011/11/04 11:08:11] @ imroot joined channel #puppet
[2011/11/04 11:08:35] <illsci> so the file notifies the service
[2011/11/04 11:09:10] <alop> So, I've got a question that I can't find good examples for...
[2011/11/04 11:09:21] @ los324 joined channel #puppet
[2011/11/04 11:09:32] <alop> I understand with file source, I can specify a list, first match wins
[2011/11/04 11:09:37] @ schewara joined channel #puppet
[2011/11/04 11:09:46] <alop> and with templates, I can specify a list for concatenation
[2011/11/04 11:09:52] @ Quit: _plg: Quit: _plg
[2011/11/04 11:09:52] <alop> I'm wondering
[2011/11/04 11:09:54] <imroot> hello, I have a few more questions. I was able to set up a manifest and apply it to the main server. How do I schedule it and also run it on another node running the agent?
[2011/11/04 11:10:17] <walkeran> alop: To answer your question before you ask it... yes, with a custom function
[2011/11/04 11:10:46] <alop> I was going down the path of file existence...
[2011/11/04 11:11:17] <alop> like content => template ['user-stub','user-$host'] - where 'user-$host' may or maynot exist
[2011/11/04 11:12:10] <alop> I may be able to do this with the concat module I'm already using, that might be a winner
[2011/11/04 11:12:32] <walkeran> alop: Yeah. I'm trying to find the function now. It's one that Volcane worked on at one point, and there's a pastie and a gist out there somewhere, I believe. multisource_template or something like that.. (at least I think that's what you are looking for)
[2011/11/04 11:12:33] <the_ramink> illsci: notify is almost always simpler to use than subscribe. In the case of adding vhosts you can notify the Apache service, but it's impossible to subscribe to all the vhosts you could every have.
[2011/11/04 11:12:51] <alop> Yeah, I was thinking Volcane already did it
[2011/11/04 11:12:56] @ Quit: astrostl:
[2011/11/04 11:12:58] <Volcane> https://gist.github.com/388123/dbf3170868ddddb38ae2f52484f34f30564191ba
[2011/11/04 11:12:58] <indigo__> what's the <\| \|> thing called and where is it documented?
[2011/11/04 11:13:25] <walkeran> indigo__: It's for "realizing", and it's in with exported or virtual resources
[2011/11/04 11:13:25] <alop> Thanks again!
[2011/11/04 11:14:03] <indigo__> walkeran: i've read some stuff that calls it a "collection operator" which seems to indicate it may have a use beyond virtual resources
[2011/11/04 11:14:09] <walkeran> alop: Here's one of them http://pastie.org/666728 ... I believe there's another version somewhere that handles multiple files differently.
[2011/11/04 11:14:26] <walkeran> indigo__: It does. You can also use it for resource chaining
[2011/11/04 11:14:59] <walkeran> indigo__: I guess "collection" is what it's called, and "realizing" is what it's for. You win :)
[2011/11/04 11:15:15] <indigo__> walkeran: yeah...that's where i'm interested with it. What I'm wondering specifically is when i say Foo <\| \|>, that gives me all the Foos, but what exactly does "all" mean?
[2011/11/04 11:15:33] <indigo__> is it all the Foos, ever, regardless of where in my manifests i declare them?
[2011/11/04 11:15:42] <indigo__> or, all the foos in some scope?
[2011/11/04 11:15:46] <Volcane> indigo__: ones declared in classes you included in this host
[2011/11/04 11:16:17] <illsci> word
[2011/11/04 11:16:45] <indigo__> so if i put "Foo <\| \|> -> Exec['something']", and then later "foo { ... }", that foo that i declared later will still run before Exec['something'] ?
[2011/11/04 11:17:31] <illsci> spaceship operator
[2011/11/04 11:17:35] <indigo__> so, for example, if i wanted to guarantee that apt-get update ran before installing packages, then putting something like "Exec['apt-update'] -> Package <\| provider='apt' \|>" at the top of site.pp would be a good idea?
[2011/11/04 11:17:38] <Volcane> indigo__: create a small test.pp file with just the resources you want and run puppet apply test.pp
[2011/11/04 11:17:43] <Volcane> indigo__: simple way to test it for yourself
[2011/11/04 11:18:02] <walkeran> Oh yeah! Spaceship operator :P How could I ever forget something silly like that :P
[2011/11/04 11:18:33] @ Quit: markatto: Read error: Connection reset by peer
[2011/11/04 11:18:41] @ ghg joined channel #puppet
[2011/11/04 11:18:47] <illsci> indigo__: I do something like that for apt... but I run the exec apt-get update whenver the sources.list.d files change...
[2011/11/04 11:18:51] <imroot> Do I have to use a spaceship operator to connect my agents to the master?
[2011/11/04 11:19:10] <illsci> hmm maybe I should do it before package installs..
[2011/11/04 11:19:54] <indigo__> illsci: yeah, the problem i have is a class which requires a package which requires a change to sources.list.d (and then update) before it can be installed.
[2011/11/04 11:19:56] @ Quit: Andorka: Read error: Connection reset by peer
[2011/11/04 11:20:07] @ Andorka joined channel #puppet
[2011/11/04 11:20:21] <indigo__> ie, squeeze-backports
[2011/11/04 11:21:30] <imroot> the documentation fails to explain how to set up new server as an agent and how to schedule the manifesting. can someone give me some insigjht on how to do this?
[2011/11/04 11:21:32] <walkeran> imroot: No, you have to use a spaceship operator to connect your space shuttle crew to ground control. But in all seriousness.. was that a real question?
[2011/11/04 11:22:01] <imroot> sorry, I'm new to this.
[2011/11/04 11:22:34] <walkeran> imroot: Oh, sorry :) Well, as far as scheduling the agent runs, if you are running the puppet agent in daemonized mode, it will run every 30 minutes.
[2011/11/04 11:22:44] @ mawser joined channel #puppet
[2011/11/04 11:22:47] <MrRothstein> volcane, looks like include works (removed the packages to test), thx
[2011/11/04 11:23:01] <walkeran> imroot: If you aren't daemonizing it, then it only runs as often as you choose to run it... be it by hand, or via cron, etc
[2011/11/04 11:23:41] <imroot> it looks like its daemonized.
[2011/11/04 11:24:25] <mawser> Having an issue where a host defined as dev is not able to retrieve certain plugins and functions as compared to the same host defined as production. I believe I have the dev and production environment defined correctly in my puppet.conf for client and master
[2011/11/04 11:24:26] @ Quit: derjohn_foo: Ping timeout: 258 seconds
[2011/11/04 11:24:26] <mawser> Could not evaluate: Could not retrieve information from environment dev source(s) puppet://puppet/plugins
[2011/11/04 11:24:31] <mawser> Any insight?
[2011/11/04 11:24:36] <imroot> when I look at the dashboard, it's reporting that the master is reporting but the additional server I setup as an agent is not.
[2011/11/04 11:25:01] <illsci> hmm So I have a has_role() function that I use to decide which modules to include... but I'm thinking about using that inside the classes too...
[2011/11/04 11:25:03] @ Quit: vkdir: Remote host closed the connection
[2011/11/04 11:25:14] @ vkdir joined channel #puppet
[2011/11/04 11:25:15] <walkeran> imroot: Well, you might want to check the agent's logs, wherever you chose to log to.. which would be syslog by default
[2011/11/04 11:25:21] <illsci> I have an rsyslog setup that sets up different logging for files and stuff based on what kind of system it is...
[2011/11/04 11:25:40] <illsci> I was going to do has_role(blah).. use this template or this template...
[2011/11/04 11:25:55] <illsci> would it be better to just have flat files and use different namespaced classes...
[2011/11/04 11:26:21] <imroot> OK, it's complaining abbout a certificate.
[2011/11/04 11:26:40] <illsci> like include rsyslog::weblogsetup or rsyslog::dblogsetup... vs template('rsyslog/weblogs.conf'), or template('rsyslog/dblogs.erb'),
[2011/11/04 11:26:46] <imroot> it's spewing puppet-agent[19726]: Did not receive certificate
[2011/11/04 11:26:57] @ Quit: ezryx: Remote host closed the connection
[2011/11/04 11:28:11] <Volcane> imroot: did you sign it on the master?
[2011/11/04 11:28:15] <walkeran> imroot: Then you have to sign the cert on the master
[2011/11/04 11:28:25] <imroot> no, how do I do that?
[2011/11/04 11:28:35] <Volcane> imroot: what version puppet?
[2011/11/04 11:28:54] @ Quit: Razique: Quit: Razique
[2011/11/04 11:28:57] <imroot> the latest as of yesterday, enterprise.
[2011/11/04 11:29:02] @ tobert joined channel #puppet
[2011/11/04 11:29:28] <the_ramink> illsci: rsyslog supports a config dir so you should use both a sub class and a specifc template
[2011/11/04 11:29:56] <Volcane> imroot: the pe documentation has a section on "Signing Agent Certificates" ?
[2011/11/04 11:30:02] @ soh joined channel #puppet
[2011/11/04 11:30:27] <tobert> anybody have strong preferences for ec2 instance types to run a large-ish puppetmaster?
[2011/11/04 11:30:55] <tobert> about 800 clients
[2011/11/04 11:30:57] <rocket> how do I depend on a file/directory I am not explicitly managing with puppet?
[2011/11/04 11:31:10] <Volcane> rocket: u cant
[2011/11/04 11:31:13] @ Quit: los324: Quit: Leaving.
[2011/11/04 11:31:46] <alop> Volcane: Love the "circular logic" comment in issue #1818
[2011/11/04 11:31:46] <gepetto> alop: Volcane: #1818 is http://projects.puppetlabs.com/issues/show/1818 "Puppet - Feature #1818: Allow multiple sources for a template and only use the first template if it exists. It has a status of Duplicate and is assigned to Nigel Kersten"
[2011/11/04 11:32:45] <rocket> ok I guess I am going to have to have createrepo always run on every puppet run :/
[2011/11/04 11:33:15] @ csanchez joined channel #puppet
[2011/11/04 11:33:19] @ nuncanada joined channel #puppet
[2011/11/04 11:33:50] <Volcane> rocket: if its an exec then look at the creates, onlyif, unless properties of the exec
[2011/11/04 11:34:00] <imroot> OK, thanks, I've signed the cert. How do I force it to run an updat instead of waiting 30 mins?
[2011/11/04 11:34:12] <rocket> Volcane: but the folder it creates isnt managed by puppet
[2011/11/04 11:34:31] <Volcane> rocket: OK, dont read the docs then and keep suffering :)
[2011/11/04 11:34:35] <walkeran> rocket: If you have stuff managing that folder outside of puppet, chances are, puppet isn't the best choice of tools for running the createrepo
[2011/11/04 11:35:14] <rocket> walkeran: I am managing everything going into the repo as far as rpms are concerned with puppet
[2011/11/04 11:35:25] <rocket> I am recursively copying them into the repo
[2011/11/04 11:35:39] <walkeran> rocket: So you ARE managing the folder with puppet...
[2011/11/04 11:35:55] @ Quit: Ashutto: Ping timeout: 244 seconds
[2011/11/04 11:36:04] @ Quit: hobbeswalsh: Quit: hobbeswalsh
[2011/11/04 11:36:15] <rocket> the repo folder yes, the folder inside the repo folder that createrepo generates I do not have a puppet definition for
[2011/11/04 11:36:32] @ sp33t left channel #puppet ("Leaving")
[2011/11/04 11:36:49] @ Quit: benkevan: Ping timeout: 255 seconds
[2011/11/04 11:36:56] @ sp33t joined channel #puppet
[2011/11/04 11:37:19] <rocket> but I have been burned by this process I have it setup wrong, because if there is an error and the rpms get put in place and then the puppet run is interrupted before the create repo happens
[2011/11/04 11:37:37] <rocket> then it never happens because the notify trigger doesnt fire
[2011/11/04 11:37:42] <rocket> on the second run
[2011/11/04 11:37:52] <rocket> but I will look at the docs again for the file resource stuff
[2011/11/04 11:38:01] <rocket> I am probably missing something stupid
[2011/11/04 11:38:18] @ derjohn_foo joined channel #puppet
[2011/11/04 11:39:08] @ toobulkeh` is now known as toobulkeh
[2011/11/04 11:39:23] <walkeran> rocket: Ah, I guess I see where you are coming from... but if your puppet run fails mid-way through, there is some sort of other underlying problem that should be solved. A run erroring out in the middle could cause all sorts of inconsistencies in all aspects. ie: a config is updated, but the service doesn't get restarted, etc.
[2011/11/04 11:40:13] <indigo__> so, the laguage guide gives an example: Yumrepo <\| \|> -> Package <\| provider == yum \|>
[2011/11/04 11:40:20] @ rutger_ joined channel #puppet
[2011/11/04 11:40:39] <rocket> walkeran: I know in many cases its me being impatient with my testing etc as I am developing
[2011/11/04 11:40:57] <walkeran> rocket: Ahha... YOU are the underlying problem :D heh
[2011/11/04 11:41:09] <indigo__> i guess the intent there is to say to install a package by yum, you have to manage the yum repo first. But, when I try the same thing with provider == apt, it seems to not create any dependencies. Package <\| \|> works, Package <\| provider == apt \|> does not.
[2011/11/04 11:41:18] @ bmahe joined channel #puppet
[2011/11/04 11:41:18] @ Quit: bmahe: Changing host
[2011/11/04 11:41:18] @ bmahe joined channel #puppet
[2011/11/04 11:41:27] <Volcane> indigo__: indeed, the docs are wrong.
[2011/11/04 11:41:45] @ Quit: nated: Remote host closed the connection
[2011/11/04 11:41:46] <rocket> in some cases .. but it did lead me to this edge case .. for some reason yum.repo got updated before the repo was created with anything in it .. and then it just wouldnt work etc
[2011/11/04 11:41:46] <Volcane> indigo__: it only determines if its apt/yum at run time but it builds the execution order at compile time.
[2011/11/04 11:42:08] <walkeran> rocket: You might be able to do some magic with onlyif and a "find" command that compares the timestamp of your repo file with *.rpm
[2011/11/04 11:42:09] @ Quit: nuncanada: Quit: Leaving
[2011/11/04 11:42:18] @ los324 joined channel #puppet
[2011/11/04 11:42:33] <rocket> walkeran: genius .. why didnt I think of that :p
[2011/11/04 11:42:44] <rocket> thanks thats just what I needed
[2011/11/04 11:42:57] * Volcane did mention the onlyif command :)
[2011/11/04 11:43:01] <Volcane> option even
[2011/11/04 11:43:13] <rocket> Volcane: you did .. I was stuck on the file resource piece
[2011/11/04 11:43:32] <rocket> I was not widening my area of focus .. I was looking at the one little rock in the forest
[2011/11/04 11:43:38] <rocket> not the whole forest :p
[2011/11/04 11:43:41] <Volcane> yeah
[2011/11/04 11:43:48] <imroot> it looks like my agent node did its catalog scan and is talking to the master. thanks.
[2011/11/04 11:44:25] @ Quit: los324: Client Quit
[2011/11/04 11:44:36] <imroot> last question, I have the manifest set up on tyhe master, where do I tell it to run on all the agents?
[2011/11/04 11:45:17] <matchaw_> Hi, is it possible to require an augease resource in another define: i.e require another::define::here["my-augeas-resource"]
[2011/11/04 11:45:28] <matchaw_> as it seems to be failing to find it when i try it this way
[2011/11/04 11:46:12] @ trickyf joined channel #puppet
[2011/11/04 11:47:36] <alop> So here's what I'm working on... I've already created a 'likewise' module to install the latest version of likewise, set it up, join the domain. Unfortunately, this would grant EVERY employee access to our servers. So...
[2011/11/04 11:47:56] <alop> I made a 'aaa' module, which will take the pasties you guys provided
[2011/11/04 11:48:04] @ Quit: mitchellh: Quit: Computer has gone to sleep
[2011/11/04 11:48:06] <walkeran> imroot: That logic all relates to the "node" definitions you have set up, and what classes are included inside those nodes.
[2011/11/04 11:48:35] <alop> and concatenate a /etc/user.allow file based on a stub, a (may or may not exist) hostgroup file, and a (may or may not exist) host file
[2011/11/04 11:48:49] <alop> and hostgroups managed via foreman
[2011/11/04 11:48:58] @ Deesl joined channel #puppet
[2011/11/04 11:48:59] @ Quit: Deesl: Excess Flood
[2011/11/04 11:49:14] <Dominic> matchaw_: sure, just reference it as Augeas["foo"]
[2011/11/04 11:49:14] <alop> This would be so much easier if I worked at a small company!
[2011/11/04 11:49:17] <Volcane> imroot: generally you dont want to have all nodes get a change at the same time - it will effectivvely break them all at once if you made a mistake but also cause bottlenecks everywhere
[2011/11/04 11:49:23] @ Quit: MrRothstein: Quit: Page closed
[2011/11/04 11:49:31] @ Deesl joined channel #puppet
[2011/11/04 11:49:32] @ Quit: Deesl: Excess Flood
[2011/11/04 11:49:49] <Volcane> imroot: best to just let it run at its schedule or look at something like mcollective which lets you trigger runs in a way thats less damaging
[2011/11/04 11:50:06] @ Deesl joined channel #puppet
[2011/11/04 11:50:06] @ Quit: Deesl: Excess Flood
[2011/11/04 11:50:27] @ brandon_dacrib joined channel #puppet
[2011/11/04 11:50:38] @ Deesl joined channel #puppet
[2011/11/04 11:50:42] @ mitchellh joined channel #puppet
[2011/11/04 11:50:51] <imroot> I'm just testing things out right now in an isolated environment . However breaking all my servers at once does sounfd like fun. :)
[2011/11/04 11:50:52] <matchaw_> Dominic: Thanks, I thought perhaps I would have to reference the define containing the augeas resource
[2011/11/04 11:51:13] <Volcane> imroot: to just do an interactive test do puppet agent --test
[2011/11/04 11:51:29] <Volcane> imroot: that does a run now and shows you the changes etc it makes on stdout
[2011/11/04 11:51:41] <imroot> i want to schedule things not interactive.
[2011/11/04 11:51:53] <Volcane> then just leave the daemons running
[2011/11/04 11:51:58] <Volcane> and they'll check in every 30 mins
[2011/11/04 11:52:22] <indigo__> is there a way to inject a noop into the dependency graph for the sake of making the .dot output more readable?
[2011/11/04 11:52:36] <imroot> OK, then where can I find out when the next run will take place?
[2011/11/04 11:52:52] <shadoi> indigo__: probably easier to parse the .dot file and manipulate it directly if needed.
[2011/11/04 11:53:01] <Dominic> matchaw_: either, it doesn't matter. If you've got other things in the define you also depend on, that makes more sense.
[2011/11/04 11:53:16] @ Quit: schewara: Quit: Ex-Chat
[2011/11/04 11:53:23] <Volcane> imroot: dont think u can figure out how long it has to wait before next run
[2011/11/04 11:53:33] <imroot> is the 30 min configurable? if so where can I do that?
[2011/11/04 11:53:43] <indigo__> for example, all packages require any apt repositories and any apt preferences, which results in N*M edges. If there were a noop that was before all packages, and required the repos and prefs, then it would be much easier to read
[2011/11/04 11:53:57] <Volcane> imroot: yeah in puppet.conf set runinterval i think - see the config reference on docs.puppetlabs.com
[2011/11/04 11:53:59] <matchaw_> When passing an array to a resource how do you ensure the order at which entries in the array are executed?
[2011/11/04 11:54:34] <indigo__> i was thinking maybe an exec{'/bin/true': updateonly=>true} but...
[2011/11/04 11:55:37] <walkeran> matchaw_: No can do... at least not in relation to one another
[2011/11/04 11:56:13] @ justin_donaldson joined channel #puppet
[2011/11/04 11:56:37] <matchaw_> walkeran, thanks. I have an augeas resource being ran from $name which is an array, but the problem is that there is no way to order the array execution.
[2011/11/04 11:57:37] @ Quit: pugnacity: Ping timeout: 244 seconds
[2011/11/04 11:59:19] @ langworthy joined channel #puppet
[2011/11/04 12:01:49] <imroot> If I look at syslog I can get an idea about when the next catalog update will take place. It ran but did not apply my manifest though.
[2011/11/04 12:01:50] @ pugnacity joined channel #puppet
[2011/11/04 12:02:26] <Volcane> imroot: if it ran and didnt apply it then you probably didnt assign the class with what u want to do to the node name
[2011/11/04 12:02:45] <brandon_dacrib> Is there a resource type for managing network interfaces on a linux host ? I checked here but came up empty: http://docs.puppetlabs.com/references/stable/type.html
[2011/11/04 12:02:52] <brandon_dacrib> I feel like I must have missed it
[2011/11/04 12:03:26] <asteve> that can be tricky, you could knock out your box!
[2011/11/04 12:03:39] <rocket> another question is it possible to do this
[2011/11/04 12:04:02] <rocket> never mind its stupid .. :p
[2011/11/04 12:04:43] @ hooper joined channel #puppet
[2011/11/04 12:04:52] @ marrusl joined channel #puppet
[2011/11/04 12:05:06] @ Spads_ joined channel #puppet
[2011/11/04 12:05:34] @ langworthy_ joined channel #puppet
[2011/11/04 12:05:44] <imroot> Volcane: I didnt assign a class, I just created a simple manifest to create a file. I was assuming it will run it everywhere, even on the master
[2011/11/04 12:05:45] <brandon_dacrib> I really just want to set the route on a group of hosts, namely the default gateway
[2011/11/04 12:06:07] <brandon_dacrib> I had it being set to one value in cobbler: but now I have a case where half of them need a different default gateway
[2011/11/04 12:06:10] <walkeran> imroot: Ah. that's where my answer comes in :) Look into node definitions
[2011/11/04 12:06:24] <tightwork> OMG computers!
[2011/11/04 12:06:54] <walkeran> imroot: Make your manifest stuff into a module, and include the classes in the nodes you want. If you want something to run on all nodes that you don't have an explicit definition for, look into the "default" node
[2011/11/04 12:07:17] @ Quit: langworthy: Ping timeout: 240 seconds
[2011/11/04 12:07:18] @ langworthy_ is now known as langworthy
[2011/11/04 12:08:10] @ Quit: alop: Quit: alop
[2011/11/04 12:08:41] @ Quit: Spads: Ping timeout: 276 seconds
[2011/11/04 12:09:00] <peridot> So I've finally gotten passenger up and running with puppetmaster. Now when I try to connect with my clients, I get an error message: "err: Could not request certificate: Retrieved certificate does not match private key; please remove certificate from server and regenerate it with the current key " Do I need to regenerate them, or is this some error in my config do you think?
[2011/11/04 12:09:19] @ jnewland joined channel #puppet
[2011/11/04 12:09:36] <asteve> peridot: node config plus master config on pastie.org
[2011/11/04 12:10:08] <asteve> also, did you have it running with the previous web engine or is this a new install?
[2011/11/04 12:10:09] <indigo__> so if i require => Class['someclass'], what exactly does that mean? Does it require everything in the class? Or just require that the class is declared?
[2011/11/04 12:10:16] @ Quit: marrusl: Quit: Ex-Chat
[2011/11/04 12:10:24] <gniks> requires everything in the class
[2011/11/04 12:10:26] <peridot> I had it running with webrick and the certs worked just fine.
[2011/11/04 12:10:26] <asteve> indigo__: it will require that everything in the class passes
[2011/11/04 12:10:28] <imroot> node definitions? I want it to run everywhere. Do I have to define all my servers?
[2011/11/04 12:10:50] @ lilmatt joined channel #puppet
[2011/11/04 12:10:57] @ Quit: lilmatt: Client Quit
[2011/11/04 12:11:02] <walkeran> imroot: Read my whole answer... if you don't want to define nodes, use the default node definition
[2011/11/04 12:11:59] @ ocherno joined channel #puppet
[2011/11/04 12:12:01] @ Quit: rgsteele: Quit: ChatZilla 0.9.86.1 [Firefox 3.0.7/2009030423]
[2011/11/04 12:12:41] @ Quit: devdas: Ping timeout: 260 seconds
[2011/11/04 12:12:49] <Volcane> imroot: yeah thats the puppet model - if you have a class full of 'stuff to do everywhere' you can define a base node and inherit or just include the class everywhere. the pe dashboard makes this easier by doing groups and stuff
[2011/11/04 12:13:20] @ Quit: mitchellh: Quit: Computer has gone to sleep
[2011/11/04 12:14:26] @ Quit: derjohn_foo: Ping timeout: 256 seconds
[2011/11/04 12:15:10] @ Quit: st33v: Quit: st33v
[2011/11/04 12:16:30] <peridot> asteve: master puppet.conf and apache config on http://pastebin.com/YLY90Awf client puppet.conf on http://pastebin.com/2hUBUJWX
[2011/11/04 12:17:32] @ los324 joined channel #puppet
[2011/11/04 12:18:49] <indigo__> what's the difference between expanded_relationships.dot and relationships.dot? Oddly, the latter seems more verbose, but I can't say exactly how.
[2011/11/04 12:18:58] <imroot> my manifest has no class. wI add the default node definition and it complains that its missing a hostname.
[2011/11/04 12:19:06] @ ocherno left channel #puppet ()
[2011/11/04 12:19:30] @ tilgovi joined channel #puppet
[2011/11/04 12:19:31] @ Quit: tilgovi: Changing host
[2011/11/04 12:19:31] @ tilgovi joined channel #puppet
[2011/11/04 12:19:49] @ st33v joined channel #puppet
[2011/11/04 12:21:09] @ los324 left channel #puppet ()
[2011/11/04 12:21:13] <Volcane> imroot: if you're going to use puppet in a way thats non standard we are not likely to be able to help you unless you provide a lot of information on pastie.org, file locations, file contents what you're trying etc
[2011/11/04 12:21:18] <imroot> after correcting the syntax, It now returns Could not parse for environment production.
[2011/11/04 12:21:28] <peridot> When I remove certs and try to get a new one, I get the error: "err: Could not retrieve catalog from remote server: Error 403 on SERVER: Forbidden request: 172.30.0.80(172.30.0.80) access to /certificate_revocation_list/ca [find] at line 93"
[2011/11/04 12:21:56] <imroot> what the hell is pastie.org ? I never heard of it.
[2011/11/04 12:22:14] <Volcane> it lets you paste file contents and give us a short link to the contents of the file
[2011/11/04 12:22:19] <Volcane> rather than just paste a bunch of stuff here
[2011/11/04 12:22:38] <imroot> All I want to do is run a simple manifest against all the servers I have (2).
[2011/11/04 12:22:47] <Volcane> generally saying 'i have this weird thing cos i tried to do this' is about 10% as useful as just showing the code
[2011/11/04 12:22:50] @ mitchellh joined channel #puppet
[2011/11/04 12:23:17] @ Quit: FlexGuy: Ping timeout: 240 seconds
[2011/11/04 12:24:07] <Volcane> (and any associated errors)
[2011/11/04 12:24:59] <imroot> I can't believe how complicated it is to do a simple task.
[2011/11/04 12:25:43] @ david____ joined channel #puppet
[2011/11/04 12:25:53] <Volcane> you probably have a small error somewhere. and you are not helping us help you.
[2011/11/04 12:26:31] <imroot> I thimk the error is gaping holes in the documentation.
[2011/11/04 12:26:52] <Volcane> fair enough, I've now spent more time asking you to enable me to help you then I'd have spent helping you
[2011/11/04 12:27:08] @ Spads joined channel #puppet
[2011/11/04 12:27:14] <Volcane> so now you'll have to rely on someone else
[2011/11/04 12:27:38] <imroot> my question was how do I run a manifest against all my servers? I'm not asking for debugging.
[2011/11/04 12:27:52] @ sdog joined channel #puppet
[2011/11/04 12:27:57] <walkeran> imroot: To be fair... I learned everything I know about puppet from the docs, from personal experience (playing with what I learned from the docs), and most recently a bit from the Pro Puppet book.
[2011/11/04 12:28:13] @ Ashutto joined channel #puppet
[2011/11/04 12:28:43] <peridot> imroot: The docs are actually quite good. To run a manifest aginst your servers, you have to run on the client something like: puppet agent --test
[2011/11/04 12:29:01] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 12:29:09] <Volcane> peridot: first he has to put the manifest in the right place - but he refuse to show us the content of the manifest or tell us where he put it.
[2011/11/04 12:29:36] <imroot> mainly because nobidy has asked.
[2011/11/04 12:29:37] @ Quit: sdog: Changing host
[2011/11/04 12:29:37] @ sdog joined channel #puppet
[2011/11/04 12:29:45] <walkeran> omg Volcane asked you like a bajillion times
[2011/11/04 12:29:52] <imroot> where should I put the manifest?
[2011/11/04 12:29:54] @ Quit: Spads_: Ping timeout: 260 seconds
[2011/11/04 12:29:59] @ Quit: krish: Read error: Connection reset by peer
[2011/11/04 12:30:02] <Volcane> " <Volcane> imroot: if you're going to use puppet in a way thats non standard we are not likely to be able to help you unless you provide a lot of information on pastie.org, file locations, file contents what you're trying etc"
[2011/11/04 12:30:15] <Volcane> and then i asked you like several times over
[2011/11/04 12:30:18] <imroot> Volcane: What is standard then?
[2011/11/04 12:30:19] <Volcane> and explain what pastie is
[2011/11/04 12:30:23] <Volcane> etc
[2011/11/04 12:30:39] @ Quit: Deesl: Quit: Bye!
[2011/11/04 12:30:49] <Volcane> we are here DONATING our free time to you. Please dont take the piss and waste it by not following simple requests
[2011/11/04 12:30:50] @ Quit: Ashutto: Client Quit
[2011/11/04 12:30:51] <imroot> my manifest sits on /root on my master machine.
[2011/11/04 12:30:58] <peridot> pastie.org
[2011/11/04 12:31:14] <Volcane> but anyway, over to someone else
[2011/11/04 12:32:05] @ _plg joined channel #puppet
[2011/11/04 12:32:16] @ Quit: _plg: Client Quit
[2011/11/04 12:32:28] <imroot> I kind of thought that Volcane doesn't know a basic answer. Soryy if I am too new at this.
[2011/11/04 12:33:49] <imroot> all I want to know is how do I run a manifest against all my machines?
[2011/11/04 12:33:55] <walkeran> imroot: It's fine that you are new at it... but honestly, you WILL HAVE TO invest a fair amount of "playing around" time in order to learn the ropes. There's no definitive way that you must set up your environment, but you have to understand how puppet works. You might not want to jump in right away with the idea that you are going to make modifications to all of your servers on the first day
[2011/11/04 12:34:07] <peridot> Anyways...setting up passenger, and getting the error: "err: Could not retrieve catalog from remote server: Error 403 on SERVER: Forbidden request: 172.30.0.80(172.30.0.80) access to /certificate_revocation_list/ca [find] at line 93" when I try to connect to the puppetmaster.
[2011/11/04 12:34:12] <peridot> master puppet.conf and apache config on http://pastebin.com/YLY90Awf client puppet.conf on http://pastebin.com/2hUBUJWX
[2011/11/04 12:34:29] <imroot> I'm just running a basic test to see how this functions.
[2011/11/04 12:34:49] @ Quit: st33v: Quit: st33v
[2011/11/04 12:35:33] <walkeran> imroot: That said, on your master, by default, the starting-point manifest is at /etc/puppet/manifests/site.pp ... in there you can either "import nodes.pp" and create a nodes.pp along-side it with your node definitions, or you can just dump everything into that file. However, I would NOT dump everything into that file, as it will hurt you later on
[2011/11/04 12:35:49] @ Quit: cruzfox_: Quit: cruzfox_
[2011/11/04 12:35:51] @ Quit: newtMcKerr: Quit: Leaving.
[2011/11/04 12:36:07] @ Quit: david____: Ping timeout: 265 seconds
[2011/11/04 12:36:28] <walkeran> imroot: Then, you'll want to create a simple module to test out and play with... you might want to read the guide on creating a module before you do that, but everything related to that first module you create is going to be located in /etc/puppet/modules/<yourmodulename>
[2011/11/04 12:36:43] @ verwilst joined channel #puppet
[2011/11/04 12:37:23] <Volcane> walkeran: unfortunately he's also using puppet enterprise which has other locations - so hence i am asking more questions than usual to try and extract relevant information
[2011/11/04 12:37:43] @ trickyf left channel #puppet ()
[2011/11/04 12:37:55] <peridot> Doesn't puppet enterprise provide support?
[2011/11/04 12:37:58] <walkeran> Ahh, I was unaware about the PE part of it... that actually has some good jump-start documentation, doesn't it?
[2011/11/04 12:38:19] <Volcane> generally if it was /etc/puppet/modules it would be /etc/puppetlabs/puppet/modules on pe
[2011/11/04 12:38:38] <the_ramink> peridot: my eyes! Alphabetize that nonsense or just remove 90% of it. In any case you're missing ssl_client_header and ssl_client_verify_header which are needed when using Passenger.
[2011/11/04 12:38:48] <imroot> my manifests are in /etc/puppetlabs/puppet/manifests. So is this the dfault directory to put all my manifests in so the will automatically run?
[2011/11/04 12:39:01] <walkeran> Ugh
[2011/11/04 12:39:20] <peridot> the_ramink: Sorry, I cut out all the comments, will remember to sort things next time!
[2011/11/04 12:39:25] <imroot> the pe document is only half there.
[2011/11/04 12:39:36] @ Quit: jsm: Ping timeout: 258 seconds
[2011/11/04 12:40:07] <imroot> it tells you how to isntall and really not much practical info on how to use it.
[2011/11/04 12:40:21] <Volcane> if you're brand new, dont want to get the book, dont want to attend training etc. I'd recommend http://docs.puppetlabs.com/learning/
[2011/11/04 12:40:27] <Volcane> downlaod the virtual machine and follow that
[2011/11/04 12:40:40] @ derjohn_mob joined channel #puppet
[2011/11/04 12:41:07] <walkeran> Volcane: Out of curiosity, do you know if PE also has the vardir at a different location? Or does it end up at /var/lib/puppet by default, too?
[2011/11/04 12:41:11] <peridot> the_ramink: Thank you! I'd gotten that part on my test system, but forgot when reapplying it!
[2011/11/04 12:41:22] <Volcane> walkeran: /opt/puppetlabs/var....
[2011/11/04 12:41:35] <walkeran> Gotcha
[2011/11/04 12:41:58] <imroot> wrong. /opt/puppet/var
[2011/11/04 12:42:18] <imroot> is there something I need to do there?
[2011/11/04 12:42:24] <walkeran> Nah. I was just curious
[2011/11/04 12:42:41] <Volcane> yeah sorry /opt/puppet - so much for consistency
[2011/11/04 12:42:45] <walkeran> Hehe :)
[2011/11/04 12:43:16] <imroot> and /etc/puppetlabs/puppet
[2011/11/04 12:43:53] @ cruzfox joined channel #puppet
[2011/11/04 12:45:44] @ Quit: derjohn_mob: Ping timeout: 258 seconds
[2011/11/04 12:47:20] <imroot> so I copied my test.pp to the /etc/puppetlabs/puppet/manifests directory. will iit run on a regular basis now?
[2011/11/04 12:47:45] <imroot> all it does is ctreate a file in /tmp. it works fine if I run it standalone
[2011/11/04 12:48:15] @ Quit: pugnacity: Ping timeout: 244 seconds
[2011/11/04 12:48:30] @ hobbeswalsh joined channel #puppet
[2011/11/04 12:49:33] @ Quit: lindenle: Ping timeout: 245 seconds
[2011/11/04 12:49:54] @ st33v joined channel #puppet
[2011/11/04 12:50:15] @ newtMcKerr joined channel #puppet
[2011/11/04 12:51:46] @ lindenle joined channel #puppet
[2011/11/04 12:53:08] @ jsm joined channel #puppet
[2011/11/04 12:53:24] <jorhett> unless site.pp imports it, it'll never be seen
[2011/11/04 12:53:32] <jorhett> unlike pp files inside modules that you load
[2011/11/04 12:53:52] @ peridot left channel #puppet ()
[2011/11/04 12:53:53] <imroot> how would I get site.pp to import my manifest?
[2011/11/04 12:54:09] @ claussen joined channel #puppet
[2011/11/04 12:54:12] <imroot> sorry, I'm new at this
[2011/11/04 12:57:26] @ Quit: newtMcKerr: Quit: Leaving.
[2011/11/04 12:58:38] @ nycko is now known as cereal_bars
[2011/11/04 12:59:24] @ cereal_bars is now known as nycko
[2011/11/04 12:59:59] @ dwalu joined channel #puppet
[2011/11/04 13:01:00] @ erodriguez joined channel #puppet
[2011/11/04 13:01:03] @ Quit: cruzfox: Quit: cruzfox
[2011/11/04 13:01:22] @ soh joined channel #puppet
[2011/11/04 13:01:25] @ Quit: tilgovi: Ping timeout: 255 seconds
[2011/11/04 13:01:52] <walkeran> imroot: You can do "import test.pp" in your site.pp...but keep in mind, that's not actually learning how to use puppet... that's just seeing it in action. Placing all of your classes and stuff (ie: modules) in your manifests dir is NOT recommended, by any means.
[2011/11/04 13:02:08] @ pugnacity joined channel #puppet
[2011/11/04 13:02:45] <imroot> So where would best practices tell me to put them?
[2011/11/04 13:03:32] <imroot> do I have to import all of the manifests that I want to use?
[2011/11/04 13:04:18] @ lak joined channel #puppet
[2011/11/04 13:04:19] @ Quit: lak: Changing host
[2011/11/04 13:04:19] @ lak joined channel #puppet
[2011/11/04 13:04:34] <walkeran> imroot: Well, let's see here... Best practices... uhm.. Oh! Look! There's a guide for that http://docs.puppetlabs.com/guides/best_practices.html ... looks like it links to the "Puppet modules" guide, and modules are what I have been trying to tell you about at least 3 times now.
[2011/11/04 13:06:27] @ Quit: lak: Client Quit
[2011/11/04 13:06:31] <mawser> I'm having trouble switching over to a multi environment setup, and I believe I have my dev environment correct in puppet.conf but when running against it I'm erroring
[2011/11/04 13:06:32] <mawser> Could not evaluate: Could not retrieve information from environment dev source(s) puppet://puppet/plugins
[2011/11/04 13:06:36] <mawser> Any thoughts?
[2011/11/04 13:06:49] @ newtMcKerr joined channel #puppet
[2011/11/04 13:07:20] <imroot> thanks for the link to the best practices. it doesn't show up on the puppetlabs website
[2011/11/04 13:08:39] <dan__t> Error 400 on SERVER: Generators can only contain alphanumerics, file separators, and dashes
[2011/11/04 13:08:41] <dan__t> damn you.
[2011/11/04 13:08:45] <imroot> i guess because it says "This document is a stub."
[2011/11/04 13:09:06] <walkeran> This URL: http://docs.puppetlabs.com/ has been given to you a couple times
[2011/11/04 13:09:10] <walkeran> It's linked directly on that page
[2011/11/04 13:09:15] <walkeran> As is ALL of the other useful docs
[2011/11/04 13:09:15] <adrient> whois imroot
[2011/11/04 13:09:28] <adrient> Well, that was awkward
[2011/11/04 13:09:33] <walkeran> :P
[2011/11/04 13:10:47] @ Quit: Chesther: Quit: Leaving.
[2011/11/04 13:11:09] <zleslie> our docs are awesome
[2011/11/04 13:11:18] <zleslie> but it does involve reading
[2011/11/04 13:11:21] <zleslie> which is a downside
[2011/11/04 13:11:22] <walkeran> imroot: And I know how much it sucks to hear this, but honestly, you should almost read through the entire doc before even starting to play around
[2011/11/04 13:11:24] @ Quit: dwalu: Quit: dwalu
[2011/11/04 13:11:40] @ illsci left channel #puppet ()
[2011/11/04 13:11:40] <tvincent> the docs are 99% better then most.
[2011/11/04 13:11:55] <walkeran> zleslie, tvincent: ++
[2011/11/04 13:11:58] <imroot> the docs have huge gaps in them. Otherwise they are truely awsome.
[2011/11/04 13:12:11] <walkeran> The gap is that you haven't even looked at them yet...
[2011/11/04 13:12:18] @ Quit: lindenle: Read error: Connection reset by peer
[2011/11/04 13:12:20] <tvincent> imroot: Yes there is always room for improvement.
[2011/11/04 13:12:33] <tvincent> imroot: but I don't feel there are "huge" gaps.
[2011/11/04 13:12:59] <imroot> if they are so truely awsome, please point me to the section where it tells you how to set up your manifests to run on a regulary scheduled time.
[2011/11/04 13:13:15] @ lindenle joined channel #puppet
[2011/11/04 13:13:41] <zleslie> imroot: module layout: http://docs.puppetlabs.com/guides/modules.html
[2011/11/04 13:13:48] <imroot> I do see a lot of good info on the language syntax, but the docs are horrible when it comes to actually using the product.
[2011/11/04 13:13:52] <dan__t> manifests don't run on a regularly scheduled time.
[2011/11/04 13:14:08] <tvincent> You could manually run puppet as a cron job on each individual machine.
[2011/11/04 13:14:23] <dan__t> manifest : A configuration file written in the Puppet language. These files should have the .pp extension.
[2011/11/04 13:14:30] <dan__t> yeah but that's not running a manifest.
[2011/11/04 13:14:49] <dan__t> You could tell an "apache config" to run every hour by telling apache to restart via cron every hour, too.
[2011/11/04 13:14:53] <zleslie> imroot: try: http://docs.puppetlabs.com/learning/
[2011/11/04 13:14:54] <mawser> I've been poring over the docs to try to get a handle on multiple environments. I have my puppet.conf set up the way indicated, both on the client and the master, I can run just fine against the production environment but when I run against the dev environment it's ilke there is a variable that is pointing to the wrong place, but I can't figure out which
[2011/11/04 13:14:57] <tvincent> imroot: Puppet community as a learn by example kind of thing going on IMHO.
[2011/11/04 13:15:21] <zleslie> mawser: show us the puppet.conf
[2011/11/04 13:15:26] @ jodok joined channel #puppet
[2011/11/04 13:15:26] <tvincent> Pro Puppet is another good one as long as you check the erratta before using the code.
[2011/11/04 13:15:33] @ Quit: jodok: Changing host
[2011/11/04 13:15:33] @ jodok joined channel #puppet
[2011/11/04 13:15:38] <mawser> imroot: When we set up new hosts we install a cron that runs at different times based on the last octet of the ip, once every 10 minutes
[2011/11/04 13:15:56] <zleslie> there is also --splay
[2011/11/04 13:16:12] @ Quit: st33v: Quit: st33v
[2011/11/04 13:16:23] <mawser> zleslie: Here is a snippet specific to the environments, I can show other parts if desired
[2011/11/04 13:16:28] <mawser> [dev]
[2011/11/04 13:16:28] <mawser> modulepath = /etc/puppet/env/dev:/etc/puppet/env/common
[2011/11/04 13:16:28] <mawser> manifest = /etc/puppet/manifests/site.pp
[2011/11/04 13:16:29] <mawser> [production]
[2011/11/04 13:16:29] <mawser> modulepath = /etc/puppet/env/production:/etc/puppet/env/common
[2011/11/04 13:16:29] <mawser> manifest = /etc/puppet/manifests/site.pp
[2011/11/04 13:16:41] <jsm> pastie.org
[2011/11/04 13:16:50] <zleslie> imroot: the other bit you asked about: http://docs.puppetlabs.com/references/stable/configuration.html#runinterval
[2011/11/04 13:16:52] <indigo__> mawser: here's how i do it http://pastebin.com/5Krkzspk
[2011/11/04 13:16:54] <zleslie> all at docs.pupptelabs.com
[2011/11/04 13:16:58] <mawser> jsm: Thanks I'll use that
[2011/11/04 13:17:30] <walkeran> mawser: You might also need "manifestdir" to be defined
[2011/11/04 13:17:56] <zleslie> mawser: we do just about the same thing as indigo__
[2011/11/04 13:17:57] <mawser> http://pastie.org/2811859
[2011/11/04 13:18:05] <dan__t> Hm, so I can't pass a variable to generate() ?
[2011/11/04 13:18:20] @ lindenle_ joined channel #puppet
[2011/11/04 13:18:54] <walkeran> mawser: http://pastie.org/2811861 <-- this is what I used to use. You could easily tweak it for your folder structure... but I think perhaps all you need is to add manifestdir
[2011/11/04 13:19:05] @ Quit: lindenle: Read error: Connection reset by peer
[2011/11/04 13:19:58] <imroot> Thank you everyone for your insight.. If I want to use cron, I surely don't need to waste my time with this.
[2011/11/04 13:20:18] @ Quit: claussen: Quit: claussen
[2011/11/04 13:20:21] <mawser> Although to be honest about manifest and templates, we don't have a specific manifest. site.pp is virtual empty, http://pastebin.com/kxQLxXba and most of the manifesting is done at the module level
[2011/11/04 13:20:28] <mawser> same with templates
[2011/11/04 13:21:14] <dan__t> $check_user_exists = generate("/usr/local/bin/check_user_exists.sh $username")
[2011/11/04 13:21:17] <dan__t> So that'll never work, right
[2011/11/04 13:21:39] <whack> dan__t: you can do that yes
[2011/11/04 13:21:49] <whack> dan__t: but keep in mind that generate() like all functions are executed on the master
[2011/11/04 13:21:54] <whack> not the agent.
[2011/11/04 13:22:15] <dan__t> yep. that'e xactly what I want.
[2011/11/04 13:22:27] @ Quit: imroot: Quit: Page closed
[2011/11/04 13:22:27] <walkeran> mawser: Yeah, don't get me wrong, I've got basically nothing in my manifests/site.pp, as well.. but you still need that point for it to jump in to
[2011/11/04 13:22:33] @ IndyMichaelB_ joined channel #puppet
[2011/11/04 13:22:38] @ Quit: IndyMichaelB_: Client Quit
[2011/11/04 13:22:46] <dan__t> Though when I run it, I get: Error 400 on SERVER: Generators can only contain alphanumerics, file separators, and dashes
[2011/11/04 13:23:04] <whack> dan__t: generate() probably only takes a path to a script
[2011/11/04 13:23:15] <dan__t> I need to give it an argument.
[2011/11/04 13:23:35] @ Quit: kermit: Ping timeout: 252 seconds
[2011/11/04 13:23:36] @ Quit: lindenle_: Read error: Connection reset by peer
[2011/11/04 13:23:37] <dan__t> guess i can use some bastardized implementation of exec
[2011/11/04 13:24:09] <walkeran> mawser: I could be completely wrong, too, but I seem to remember having trouble without actually setting manifestdir, when I originally started using environments
[2011/11/04 13:24:10] <dan__t> that sucks.
[2011/11/04 13:25:04] <dan__t> oh. duh.
[2011/11/04 13:25:18] @ Quit: jsm: Ping timeout: 260 seconds
[2011/11/04 13:25:21] @ lindenle joined channel #puppet
[2011/11/04 13:25:26] <dan__t> = generate("/usr/local/bin/check_user_exists",$username)
[2011/11/04 13:25:36] <dan__t> generate takes arguments, not arguments to the application being ran
[2011/11/04 13:25:39] @ zz_jhelwig is now known as jhelwig
[2011/11/04 13:26:05] @ Quit: IndyMichaelB: Ping timeout: 240 seconds
[2011/11/04 13:26:27] <mawser> walker an, all: I've changed my puppet.conf to basically this for that env, http://pastebin.com/ftNBtN1P but still getting this: http://pastebin.com/QqCbGdQK when running a command like puppetd --test --environment dev --evaltrace --debug --verbose
[2011/11/04 13:26:34] @ jsm joined channel #puppet
[2011/11/04 13:26:58] @ Quit: jim80net: Quit: Leaving.
[2011/11/04 13:27:27] @ Quit: lwhalen: Quit: Computer has gone to sleep
[2011/11/04 13:27:32] <walkeran> mawser: Are you using pluginsync?
[2011/11/04 13:27:56] <mawser> walkeran: pluginsync = true in my puppet.conf
[2011/11/04 13:28:08] <walkeran> Oh, I guess I could have looked at that pastie of yours :P Derp!
[2011/11/04 13:28:10] <mawser> walkeran: In the [main] sectoin
[2011/11/04 13:28:12] <walkeran> Look at #2244
[2011/11/04 13:28:12] <gepetto> walkeran: #2244 is http://projects.puppetlabs.com/issues/show/2244 "Puppet - Bug #2244: pluginsync fails when no source is available. It has a status of Accepted and is assigned to -"
[2011/11/04 13:28:46] @ st33v joined channel #puppet
[2011/11/04 13:28:54] <walkeran> fyi: I found that on this page http://bitcube.co.uk/content/puppet-errors-explained Which has your specific error on it. I haven't run into this before, but it looks like it could be what's going on
[2011/11/04 13:28:56] @ kermit joined channel #puppet
[2011/11/04 13:29:14] @ BManojlovic joined channel #puppet
[2011/11/04 13:29:17] @ Quit: BManojlovic: Changing host
[2011/11/04 13:29:17] @ BManojlovic joined channel #puppet
[2011/11/04 13:29:31] <mawser> walkeran: yeah I was looking at that, trying (somewhat unsuccessfully) to apply that to my env. I'll bang away at it some more
[2011/11/04 13:29:31] @ Quit: lindenle: Read error: Connection reset by peer
[2011/11/04 13:29:43] <mawser> walkeran: Looking at both those links that is
[2011/11/04 13:29:59] @ lindenle joined channel #puppet
[2011/11/04 13:30:51] @ Quit: labrown: Quit: Leaving
[2011/11/04 13:31:20] @ cruzfox joined channel #puppet
[2011/11/04 13:31:41] @ Quit: jsurfer: Read error: Operation timed out
[2011/11/04 13:35:29] @ Quit: lborda: Quit: Ex-Chat
[2011/11/04 13:35:29] @ Quit: uphillian: Quit: Leaving.
[2011/11/04 13:35:45] @ lborda joined channel #puppet
[2011/11/04 13:36:34] <walkeran> mawser: Do you use the modules/<name>/lib/puppet/{type,provider,parser/functions} locations, or do you have some in modules/<name>/plugins?
[2011/11/04 13:37:30] <mawser> walkeran: former not the latter
[2011/11/04 13:37:35] <walkeran> interesting..
[2011/11/04 13:37:56] <walkeran> mawser: as a test, if you turn off pluginsync, does it work, or at least give you a different error?
[2011/11/04 13:38:15] @ newtMcKerr1 joined channel #puppet
[2011/11/04 13:38:38] @ Quit: newtMcKerr: Ping timeout: 258 seconds
[2011/11/04 13:38:57] <mawser> walkeran: we use the functions like datacetner.rb and ip to cron etc… and yeah I just tried with pluginsync false on master and client and it worked. The bug indicates that we should not be syncing w/o a plugin source, although I'm curious as to why it works on env = production
[2011/11/04 13:39:29] <vkdir> does puppet use apt or aptitude when it installs debian packages?
[2011/11/04 13:40:26] @ Quit: p3rror: Ping timeout: 244 seconds
[2011/11/04 13:40:44] <mawser> vkdir: it does
[2011/11/04 13:44:20] <vkdir> ohm, no I meant which one does it use
[2011/11/04 13:44:29] <vkdir> s/ohm/uhm/
[2011/11/04 13:45:37] @ Quit: tvsutton: Quit: Lost terminal
[2011/11/04 13:46:39] <whack> vkdir: depends on what it detects.
[2011/11/04 13:47:48] <mawser> vkdir: It can use both, good ref: http://docs.puppetlabs.com/references/latest/type.html#package
[2011/11/04 13:48:39] @ Quit: erodriguez: Quit: erodriguez
[2011/11/04 13:49:10] @ Quit: newtMcKerr1: Quit: Leaving.
[2011/11/04 13:49:47] @ erodriguez joined channel #puppet
[2011/11/04 13:55:14] <mawser> walkeran: By way of update, it works if I specify pluginsource in the client puppet.conf. Still strange to me that it was environment constrained, not sure I understand why production doesn't have this same problem
[2011/11/04 13:55:15] <vkdir> mawser: ah, thx sir
[2011/11/04 13:55:17] @ tvsutton joined channel #puppet
[2011/11/04 13:55:18] @ Quit: shadoi: Quit: Leaving.
[2011/11/04 13:55:18] @ jbraeuer joined channel #puppet
[2011/11/04 13:55:19] @ IndyMichaelB joined channel #puppet
[2011/11/04 13:55:49] @ Quit: mpdude: Remote host closed the connection
[2011/11/04 13:56:33] @ Quit: beaknit: Read error: Connection reset by peer
[2011/11/04 13:57:02] <cwebber> does anyone happen to know if you can use source to specify what repo on debian for a package?
[2011/11/04 13:57:30] @ beaknit joined channel #puppet
[2011/11/04 13:57:34] @ Quit: mitchellh: Quit: Computer has gone to sleep
[2011/11/04 14:01:37] @ Quit: sumsuch: Quit: sumsuch
[2011/11/04 14:02:08] @ krish joined channel #puppet
[2011/11/04 14:04:21] @ Quit: indigo__: Quit: leaving
[2011/11/04 14:06:37] @ Quit: paroneayea: Read error: Connection reset by peer
[2011/11/04 14:06:40] @ Quit: bretm: Quit: Leaving
[2011/11/04 14:07:01] @ jsurfer joined channel #puppet
[2011/11/04 14:07:05] @ Quit: brandon_dacrib: Quit: brandon_dacrib
[2011/11/04 14:07:29] @ imroot joined channel #puppet
[2011/11/04 14:07:55] @ failure joined channel #puppet
[2011/11/04 14:08:20] @ Quit: imroot: Client Quit
[2011/11/04 14:09:00] @ Quit: st33v: Quit: st33v
[2011/11/04 14:10:59] @ newtMcKerr joined channel #puppet
[2011/11/04 14:11:48] @ Quit: newtMcKerr: Client Quit
[2011/11/04 14:12:21] @ tilgovi joined channel #puppet
[2011/11/04 14:12:22] @ Quit: tilgovi: Changing host
[2011/11/04 14:12:22] @ tilgovi joined channel #puppet
[2011/11/04 14:12:31] <Radex> hello, i'm calling function with two array $array1 and $array2, in function i'm getting $array1 element over $name variable. How to get $array2 element?
[2011/11/04 14:12:52] @ Quit: jsurfer: Read error: Connection reset by peer
[2011/11/04 14:13:08] @ Quit: cruzfox: Quit: cruzfox
[2011/11/04 14:13:14] @ paroneayea joined channel #puppet
[2011/11/04 14:13:19] @ Quit: gniks: Quit: Leaving.
[2011/11/04 14:13:22] @ jsurfer joined channel #puppet
[2011/11/04 14:14:31] @ Quit: rutger_: Remote host closed the connection
[2011/11/04 14:16:06] @ jongrocho joined channel #puppet
[2011/11/04 14:18:22] @ newtMcKerr joined channel #puppet
[2011/11/04 14:20:42] @ Spads_ joined channel #puppet
[2011/11/04 14:23:38] @ Quit: Spads: Ping timeout: 260 seconds
[2011/11/04 14:25:08] <whack> Radex: I don't really understand your question
[2011/11/04 14:27:31] <dan__t> meh. so i have a define. it has an exec. that exec depends on a file being put there by Puppet. Not sure how to make this work. I'm putting the file resource for that file everywhere - but I can't make sure it's put on the puppetmaster before it runs.
[2011/11/04 14:28:02] <whack> dan__t: the puppet master doesn'ta ct on any resources, only the agent does
[2011/11/04 14:29:01] <thedonvaughn> dan__t: if the exec depends on a file, why not have the exec inside the file resource?
[2011/11/04 14:29:03] @ Quit: mcarpent: Quit: Thanks for the fish.
[2011/11/04 14:29:06] <dan__t> sorry - the agent runs on a machine designated as master. the define only gets ran on the master, according to my logic.
[2011/11/04 14:29:09] @ cruzfox joined channel #puppet
[2011/11/04 14:29:11] <whack> dan__t: gotcha
[2011/11/04 14:29:27] <whack> dan__t: mycustomdefine { "foo": require => File["blah"]; }
[2011/11/04 14:29:36] <thedonvaughn> dan__t: er same class wit hteh file resource. or just require the file
[2011/11/04 14:29:45] <dan__t> oh. no shit.
[2011/11/04 14:29:46] <thedonvaughn> ok what whack said :)
[2011/11/04 14:29:47] @ pjsz_ joined channel #puppet
[2011/11/04 14:29:54] <whack> or if it's just your exec, exec { "foo": require => File["blah"]; }
[2011/11/04 14:29:58] <dan__t> where do I define the file resource then? can I do it inside that function?
[2011/11/04 14:30:03] <dan__t> duh.
[2011/11/04 14:30:04] @ Quit: lborda: Ping timeout: 255 seconds
[2011/11/04 14:30:14] <dan__t> wait no. i use 'generate'. I can't specify a 'require' for 'generate'
[2011/11/04 14:30:21] <dan__t> though I could wrap generate in a class...?
[2011/11/04 14:30:26] <whack> no
[2011/11/04 14:30:37] <dan__t> no.
[2011/11/04 14:30:39] <whack> function execute before the catalog is compiled, and dependencies are handled after that.
[2011/11/04 14:30:55] <whack> why are you using generate() ?
[2011/11/04 14:31:12] <whack> if you want your puppet agent to run a script, exec {} is the way to do that
[2011/11/04 14:31:17] <dan__t> oh, right
[2011/11/04 14:31:27] <dan__t> I want to determine true/false from the script
[2011/11/04 14:31:39] <dan__t> using && true \|\| false inside the script
[2011/11/04 14:31:58] @ Quit: hyde: Quit: Leaving
[2011/11/04 14:32:52] @ Quit: jhelwig: Quit: Zoom!
[2011/11/04 14:33:20] <dan__t> ok.
[2011/11/04 14:33:56] <dan__t> I don't want to specify a require when I use the define. that's the whole point of mkaing a define; so only one piece of data is required.
[2011/11/04 14:34:04] <dan__t> i'll toy around with exec.
[2011/11/04 14:34:32] @ dwalu joined channel #puppet
[2011/11/04 14:34:56] <dan__t> alright. i can use an exec. and use 'returns'
[2011/11/04 14:35:15] @ zz_jhelwig joined channel #puppet
[2011/11/04 14:35:22] @ zz_jhelwig is now known as jhelwig
[2011/11/04 14:35:37] <dan__t> No. damnit. Because I want to use the output of the script to satisfy: if ($if_some_condition == 'true') {
[2011/11/04 14:35:41] @ Quit: jhelwig: Changing host
[2011/11/04 14:35:41] @ jhelwig joined channel #puppet
[2011/11/04 14:36:24] <whack> I'm confused
[2011/11/04 14:36:27] <whack> what are you trying to do?
[2011/11/04 14:36:49] <dan__t> I'm trying to search ldap for a value. If the value is present, then I want an if{} block to run.
[2011/11/04 14:36:52] @ kainos joined channel #puppet
[2011/11/04 14:37:07] <whack> that's what facts are for, really
[2011/11/04 14:37:19] <dan__t> Alright, I'll give you that.
[2011/11/04 14:37:40] <dan__t> i'm better at writing bastardized ruby inside of puppet than I am enough ruby for a fact.
[2011/11/04 14:37:59] <whack> since generate() runs before resources do, you can't make generate() depend on the exec. There might be some hacks you can do work around it, but I'd just use a fact.
[2011/11/04 14:38:07] <dan__t> right
[2011/11/04 14:38:10] <whack> there are puppet modules that let you run shell scripts for facts
[2011/11/04 14:38:28] <whack> so whatever paranoia you have about writing working ruby-puppet-fact code can be solved by using that module
[2011/11/04 14:38:41] <dan__t> serious?
[2011/11/04 14:38:45] <dan__t> drop some names for me please
[2011/11/04 14:38:48] @ Quit: hobbeswalsh: Quit: hobbeswalsh
[2011/11/04 14:39:12] @ kainos is now known as ichristo
[2011/11/04 14:39:40] <dan__t> External Facts
[2011/11/04 14:39:52] @ Quit: cruzfox: Quit: cruzfox
[2011/11/04 14:40:13] @ Quit: pugnacity: Ping timeout: 240 seconds
[2011/11/04 14:40:26] @ Quit: beaknit: Quit: beaknit
[2011/11/04 14:41:16] <dan__t> don't know if that would work either. i need to pass a variable to this. The data i'm looking for is regarding a specific user in ldap. basically, if the user exists in ldap, then run this if{} block.
[2011/11/04 14:41:45] @ Quit: asteve: Quit: asteve
[2011/11/04 14:42:09] @ hobbeswalsh joined channel #puppet
[2011/11/04 14:43:00] @ kjalil joined channel #puppet
[2011/11/04 14:43:28] <dan__t> yea, ya see
[2011/11/04 14:43:56] <dan__t> fubar.
[2011/11/04 14:44:12] <dan__t> afaik i can't pass data to a fact.
[2011/11/04 14:44:32] @ Quit: kjalil: Client Quit
[2011/11/04 14:44:56] <torrancew> dan__t: sounds like you need a custom function...
[2011/11/04 14:44:57] @ hal-j1 joined channel #puppet
[2011/11/04 14:45:09] @ Quit: hal-j: Read error: No route to host
[2011/11/04 14:45:22] <dan__t> i wrote a define. obv not a function.
[2011/11/04 14:46:11] @ Quit: countskm: Quit: leaving
[2011/11/04 14:46:16] <gepetto> ::redmine:: Wiki edit: Downloading_Puppet (#115) @ http://projects.puppetlabs.com/projects/puppet/wiki/Downloading_Puppet?version=115 (by Matthaus Litteken)
[2011/11/04 14:46:39] <torrancew> dan__t: if you need external information and need to pass a variable to it, I think you'll want a custom puppet function
[2011/11/04 14:46:42] @ countskm joined channel #puppet
[2011/11/04 14:46:56] <dan__t> right
[2011/11/04 14:47:16] <dan__t> like i said, i don't know ruby well - at all.
[2011/11/04 14:47:31] <torrancew> dan__t: I feel your pain, sir
[2011/11/04 14:48:16] <torrancew> I'll paste a tiny function I wrote, you're welcome to try to draw from it
[2011/11/04 14:48:28] <dan__t> oh cool, thanks.
[2011/11/04 14:49:12] <whack> dan__t: in that case you you'll need a function that queries ldap for you
[2011/11/04 14:49:25] <dan__t> yea
[2011/11/04 14:49:27] <dan__t> well shit.
[2011/11/04 14:50:54] @ rsnhyperion_ joined channel #puppet
[2011/11/04 14:52:16] @ cruzfox joined channel #puppet
[2011/11/04 14:52:23] @ no-name- joined channel #puppet
[2011/11/04 14:52:26] @ kjalil joined channel #puppet
[2011/11/04 14:53:50] <dan__t> http://holyhandgrenade.org/blog/2011/08/runbooks-are-stupid-and-youre-doing-them-wrong/
[2011/11/04 14:53:51] <dan__t> wow.
[2011/11/04 14:54:10] @ Quit: kjalil: Client Quit
[2011/11/04 14:54:29] @ Quit: rsnhyperion: Ping timeout: 240 seconds
[2011/11/04 14:54:33] @ kjalil joined channel #puppet
[2011/11/04 14:55:09] <whack> wow at what?
[2011/11/04 14:55:16] @ Quit: rsnhyperion_: Ping timeout: 255 seconds
[2011/11/04 14:55:26] <torrancew> dan__t: https://gist.github.com/1340585 <--- the 'kmods' is a custom fact in our environment, but it shows you more or less how to format the function itself
[2011/11/04 14:55:32] @ Quit: kjalil: Client Quit
[2011/11/04 14:55:34] <dan__t> at the article.
[2011/11/04 14:55:45] @ Quit: mawser: Quit: mawser
[2011/11/04 14:56:05] @ Quit: ldlework: Ping timeout: 240 seconds
[2011/11/04 14:56:36] <dan__t> kmods spits out data in a format of foo,bar,one,two,three,fourtytwo
[2011/11/04 14:56:37] <dan__t> etc
[2011/11/04 14:57:10] <torrancew> ya
[2011/11/04 14:57:22] <torrancew> csv of all loaded kernel modules
[2011/11/04 14:57:26] @ Quit: spiette: Quit: Leaving
[2011/11/04 14:57:55] <dan__t> got it.
[2011/11/04 14:58:06] <torrancew> obviously, I could just as easily have done included = false; if modules.include?(mod); included = true ...
[2011/11/04 14:58:11] @ kjalil joined channel #puppet
[2011/11/04 14:58:15] <torrancew> and honestly, that would be much more clear
[2011/11/04 14:58:21] <dan__t> hrm, i'd need that script of mine to be present, or make a fact to return all ldap users
[2011/11/04 14:58:25] <dan__t> the latter seems kind of messy
[2011/11/04 14:59:20] <torrancew> dan__t: you could just use a ruby ldap library to query, I imagine
[2011/11/04 14:59:50] @ no-name- left channel #puppet ("Leaving")
[2011/11/04 15:00:27] <dan__t> supposing one knew ruby well enough :)
[2011/11/04 15:00:33] <dan__t> gotta head to a meeting. Thanks for the ino.
[2011/11/04 15:01:27] <dan__t> info, too.
[2011/11/04 15:02:20] <torrancew> no worries
[2011/11/04 15:02:24] <torrancew> happy hacking
[2011/11/04 15:02:25] @ shadoi joined channel #puppet
[2011/11/04 15:05:33] @ peridot joined channel #puppet
[2011/11/04 15:07:06] @ st33v joined channel #puppet
[2011/11/04 15:07:46] <peridot> Hey, I just wanted to say thanks to all of you for the help. I've got our cfengine migrated to puppet with passenger going on 10 systems in a staging phase. :)
[2011/11/04 15:08:12] @ gniks joined channel #puppet
[2011/11/04 15:08:26] @ tlocalhos is now known as localhost
[2011/11/04 15:09:32] <haus> PSA: Puppet Dashboard 1.2.3rc1 available in the usual locations
[2011/11/04 15:13:19] @ Quit: localhost: Quit: Ex-Chat
[2011/11/04 15:14:02] @ localhost joined channel #puppet
[2011/11/04 15:14:42] <CrazyGir> what is the vision for the dashboard? what sorts of things is it not meant to do?
[2011/11/04 15:15:49] @ Quit: BManojlovic: Ping timeout: 240 seconds
[2011/11/04 15:16:49] @ Quit: themadcanudist: Quit: Leaving.
[2011/11/04 15:16:53] @ Quit: tvincent:
[2011/11/04 15:18:10] @ Quit: flakrat: Quit: Leaving
[2011/11/04 15:18:11] @ Quit: cruzfox: Quit: cruzfox
[2011/11/04 15:18:19] @ Quit: krish: Quit: Leaving.
[2011/11/04 15:18:40] @ Quit: sdog: Quit: Leaving.
[2011/11/04 15:18:54] @ choonming joined channel #puppet
[2011/11/04 15:20:12] @ Quit: st33v: Quit: st33v
[2011/11/04 15:22:52] @ sumsuch joined channel #puppet
[2011/11/04 15:26:45] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 15:28:08] @ diq joined channel #puppet
[2011/11/04 15:29:13] @ soh joined channel #puppet
[2011/11/04 15:30:34] @ st33v joined channel #puppet
[2011/11/04 15:33:30] @ Quit: soh: Client Quit
[2011/11/04 15:34:56] @ Quit: maes_: Ping timeout: 248 seconds
[2011/11/04 15:35:21] @ Quit: alanevans: Remote host closed the connection
[2011/11/04 15:36:31] @ claussen joined channel #puppet
[2011/11/04 15:37:49] @ BMDan joined channel #puppet
[2011/11/04 15:38:40] <CrazyGir> what does this suggest is wrong? err: Could not create resources for managing Puppet's files and directories in sections [:main, :agent, :ssl]: Could not find a default provider for user
[2011/11/04 15:39:07] <BMDan> @maintainers: Regarding http://puppetlabs.com/security/cve/cve-2011-3872/faq/#q2, one other way to verify that you aren't vulnerable: openssl asn1parse -dump -in /var/lib/puppet/ssl/certs/$(hostname -f).pem \| grep -c Alternate
[2011/11/04 15:39:14] <BMDan> 0 = not vulnerable on this client
[2011/11/04 15:39:20] <BMDan> 1 or more = vulnerable
[2011/11/04 15:39:32] @ cruzfox joined channel #puppet
[2011/11/04 15:40:08] @ Quit: BMDan: Client Quit
[2011/11/04 15:40:11] @ soh joined channel #puppet
[2011/11/04 15:40:28] @ Quit: soh: Client Quit
[2011/11/04 15:43:30] <CrazyGir> has puppet lost support for openbsd?
[2011/11/04 15:43:42] <dwalu> do you have to store a password in puppet.conf's [master] for a mysql/storeconfigs setup or is there a way to send a hash instead?
[2011/11/04 15:44:00] <CrazyGir> http://projects.puppetlabs.com/issues/6693 <--- this suggest my error noted above is related to not having a way to creat/manage users/groups
[2011/11/04 15:44:30] <Volcane> CrazyGir: add sbin dirs to your path
[2011/11/04 15:44:38] <Volcane> thats one possible cause
[2011/11/04 15:45:32] <CrazyGir> Volcane: $PATH: /sbin:/usr/sbin:/bin:/usr/bin:/usr/X11R6/bin:/usr/local/sbin:/usr/local/bin
[2011/11/04 15:45:44] <CrazyGir> maybe /usr/local/sbin?
[2011/11/04 15:45:44] <Volcane> ok
[2011/11/04 15:45:54] <Volcane> nah should be /usr/sbin and /sbin usually
[2011/11/04 15:46:09] <CrazyGir> hmmm
[2011/11/04 15:46:18] <Volcane> thats the only times i saw it before, could no doubt be other cuases too
[2011/11/04 15:46:26] <CrazyGir> I'm seeing the same behaviour on more than one openbsd system
[2011/11/04 15:46:29] <Volcane> are you on openbsd?
[2011/11/04 15:46:32] <CrazyGir> 'yep
[2011/11/04 15:47:16] <Volcane> cant say i recall any openbsd users :)
[2011/11/04 15:47:27] <CrazyGir> you should read about mtier then
[2011/11/04 15:47:30] <Volcane> does openbsd use pw or useradd?
[2011/11/04 15:47:44] <Volcane> oh i know openbsd, i mean i dont recall puppet users who use openbsd
[2011/11/04 15:47:47] <CrazyGir> they're huge puppet users, building stuff for fortune 500 companies
[2011/11/04 15:47:52] <clustermagnet> guys, cant seem to get a new cert generated for client…. wtf ive removed /etc/puppet/ssl…e tc
[2011/11/04 15:47:59] <Volcane> CrazyGir: ah nice
[2011/11/04 15:47:59] <clustermagnet> can someone help plz? i get the infamous: Could not run: Retrieved certificate does not match private key; please remove certificate from server and regenerate it with the current key
[2011/11/04 15:48:12] <CrazyGir> Volcane: openbsd uses useradd, honestly, I haven't had a problem with puppet on openbsd before
[2011/11/04 15:48:31] <CrazyGir> I just deployed a whole bunch of new systems I need to get up to speed
[2011/11/04 15:48:49] <clustermagnet> puppetca —list -a \| grep blah does not find it...
[2011/11/04 15:48:57] <CrazyGir> Volcane: http://undeadly.org/cgi?action=article&sid=20110420080633
[2011/11/04 15:49:06] <torrancew> clustermagnet: delete /var/lib/puppet/ssl on the client (NOT the master)
[2011/11/04 15:49:12] <Volcane> yeah and useradd provider seems to be default so that should be fine
[2011/11/04 15:49:21] <torrancew> clustermagnet: or /etc/puppet/ssl
[2011/11/04 15:49:31] <clustermagnet> torrancew: :) no kidding :)
[2011/11/04 15:49:31] <torrancew> and also, revoke the cert on the master
[2011/11/04 15:49:33] <clustermagnet> yep did that too
[2011/11/04 15:49:40] <clustermagnet> rm -rf /var/lib/puppet actually, wiped the whole thing
[2011/11/04 15:49:41] <clustermagnet> :)
[2011/11/04 15:49:52] <clustermagnet> torrancew: cant find the cert on master :)
[2011/11/04 15:50:02] <torrancew> clustermagnet: puppet cert --revoke $nodename
[2011/11/04 15:50:09] <Volcane> CrazyGir: you can try run it with --debug --verbose --test and see - it should show how it tries to figure out how to manage users
[2011/11/04 15:50:15] <torrancew> (or puppetca, if you're on an old version)
[2011/11/04 15:50:15] <clustermagnet> yes, not there
[2011/11/04 15:50:16] <torrancew> ah
[2011/11/04 15:50:20] <torrancew> sorry, misunderstood you
[2011/11/04 15:50:21] <peridot> clustermagnet: is it waiting to be signed?
[2011/11/04 15:50:22] @ soh joined channel #puppet
[2011/11/04 15:50:25] <clustermagnet> nope
[2011/11/04 15:52:28] <clustermagnet> oh, foudn it
[2011/11/04 15:52:37] <clustermagnet> nvm, thanks guys :)
[2011/11/04 15:52:42] <clustermagnet> old image
[2011/11/04 15:53:35] <CrazyGir> ouch..
[2011/11/04 15:53:37] <CrazyGir> posting
[2011/11/04 15:54:48] <CrazyGir> Volcane: http://dpaste.com/647470/
[2011/11/04 15:54:53] <tobert> ugh, is debian backports really the best way to get 2.7 on ubuntu lucid?
[2011/11/04 15:55:12] <CrazyGir> install from source?
[2011/11/04 15:55:37] <tobert> I was hoping to avoid that mess
[2011/11/04 15:55:41] <CrazyGir> I have a script, I don't think twice about it anymore :0
[2011/11/04 15:55:43] <CrazyGir> :P
[2011/11/04 15:55:58] <CrazyGir> it isn't nearly as messy as backports, from my experience
[2011/11/04 15:56:09] @ Quit: dwalu: Quit: dwalu
[2011/11/04 15:56:23] <Volcane> CrazyGir: ah, openbsd doesnt have 'chage'?
[2011/11/04 15:56:24] @ Quit: st33v: Quit: st33v
[2011/11/04 15:56:43] <CrazyGir> wth is chage?
[2011/11/04 15:56:48] <Volcane> chage (1) - change user password expiry information
[2011/11/04 15:57:00] <CrazyGir> I don't believe so
[2011/11/04 15:57:10] <Volcane> bugger
[2011/11/04 15:57:11] <CrazyGir> never heard of it, honestly
[2011/11/04 15:57:12] <tobert> eh, it works, just a little gross for my bootstrap stuff
[2011/11/04 15:57:21] <CrazyGir> script it?
[2011/11/04 15:57:27] <tobert> of course
[2011/11/04 15:57:27] <CrazyGir> even with vms, that is easy
[2011/11/04 15:57:31] <CrazyGir> all you need is wget and ruby
[2011/11/04 15:57:38] <Volcane> then a specific openbsd provider probably is needed :( do you have different versions on openbsd that works and openbsd that doesnt?
[2011/11/04 15:58:02] <CrazyGir> hmmm
[2011/11/04 15:58:05] <CrazyGir> maybe 2.6.x
[2011/11/04 15:58:19] <CrazyGir> these must have been recent changed?
[2011/11/04 15:58:22] <CrazyGir> *changes
[2011/11/04 15:58:24] <the_ramink> CrazyGir: the puppetlabs apt repo is looking newly up to date though I haven't tried them yet, http://apt.puppetlabs.com/ubuntu/pool/main/p/puppet/
[2011/11/04 15:58:37] <CrazyGir> Volcane: what exactly needs to happen for puppet to be happy here?
[2011/11/04 15:58:38] <jbaldridge> tobert: I've had good luck with grabbing the deb sources from apt.puppetlabs.com
[2011/11/04 15:58:55] <Volcane> CrazyGir: see in lib/puppet/provider/user/useradd.rb see if the ones that work has chage near the top where it lists the commands
[2011/11/04 15:58:57] <CrazyGir> the_ramink: tobert would like that :)
[2011/11/04 15:59:17] <the_ramink> oops, my bad.
[2011/11/04 15:59:23] <tobert> exactly what I was looking for
[2011/11/04 15:59:53] @ Quit: neonflux: Quit: Leaving.
[2011/11/04 15:59:57] <CrazyGir> Volcane: is puppet supposed to install stuff in /lib/puppet/ ?
[2011/11/04 16:00:12] <Volcane> CrazyGir: no - it would be in the ruby libdir
[2011/11/04 16:00:31] <Volcane> sorry, so for me its /usr/lib/ruby/site_ruby/1.8/puppet/provider/user/useradd.rb
[2011/11/04 16:00:42] <Volcane> probably on bsd its like usr/local/lib blah blah
[2011/11/04 16:01:38] <CrazyGir> yep
[2011/11/04 16:02:25] <CrazyGir> what's the best way to remove puppet if you installed by hand?
[2011/11/04 16:02:41] <Volcane> hehe
[2011/11/04 16:02:44] <CrazyGir> :P
[2011/11/04 16:03:06] <Volcane> it should be mostly in ruby site lib and bin dirs
[2011/11/04 16:03:10] @ st33v joined channel #puppet
[2011/11/04 16:03:32] <CrazyGir> yea, looks like I can just rm a few things
[2011/11/04 16:03:42] <CrazyGir> I love simplicity :)
[2011/11/04 16:03:57] @ Quit: idn: Ping timeout: 260 seconds
[2011/11/04 16:04:40] <hostdream> hello, can anyone tell me how to get ride of this error ? Error 400 on SERVER: stack level too deep
[2011/11/04 16:05:58] <CrazyGir> oh so wonderful
[2011/11/04 16:06:36] <CrazyGir> openbsd ships with a patched 2.6.3 as a package
[2011/11/04 16:06:40] <CrazyGir> it works find
[2011/11/04 16:06:42] <CrazyGir> *fine
[2011/11/04 16:06:48] <Volcane> CrazyGir: lol
[2011/11/04 16:07:06] <Volcane> CrazyGir: nice, wish they pushed that upstream cos this is bloody tedious :)
[2011/11/04 16:07:09] <CrazyGir> I think I would just need to update chage for passwd
[2011/11/04 16:07:24] <CrazyGir> how do you mean?
[2011/11/04 16:07:36] <Volcane> if they took their patches and send it to the puppet project
[2011/11/04 16:07:42] <Volcane> to properly support openbsd
[2011/11/04 16:07:48] @ lak joined channel #puppet
[2011/11/04 16:07:48] @ Quit: lak: Changing host
[2011/11/04 16:07:49] @ lak joined channel #puppet
[2011/11/04 16:08:39] <CrazyGir> I'd be a little surprised if he hadn't
[2011/11/04 16:08:44] <CrazyGir> I will poke the maintainer
[2011/11/04 16:09:01] @ idn joined channel #puppet
[2011/11/04 16:09:43] <the_ramink> hostdream: IIRC that's kinda hard to pin down. I'd back out of the last change or two and see if you can isolate what introduced it.
[2011/11/04 16:09:48] <countskm> CrazyGir: used to be an openbsd'er i feel for you ;-)
[2011/11/04 16:10:18] <torrancew> hostdream: I had that issue when using activerecord 3.1 with puppet once, fwiw, but yeah, it is a vague message that can be caused by a number of things
[2011/11/04 16:10:20] <peridot> Is there a way to do notify something if an rpm install fails?
[2011/11/04 16:10:21] @ tizzo is now known as tizzo-afk
[2011/11/04 16:10:22] <hostdream> the_ramink : I've updated the puppet server to 2.7
[2011/11/04 16:10:49] <hostdream> torrancew : I've it on 1 of my 40 hosts that's weird
[2011/11/04 16:12:48] <CrazyGir> countskm: how so?
[2011/11/04 16:12:48] @ los324 joined channel #puppet
[2011/11/04 16:12:56] @ Quit: kermit: Quit: Leaving.
[2011/11/04 16:13:10] @ kermit joined channel #puppet
[2011/11/04 16:13:25] <peridot> I want to download an rpm and install it if it fails to install from the repo. Can I do that in puppet?
[2011/11/04 16:13:45] @ Razique joined channel #puppet
[2011/11/04 16:13:50] @ Quit: jsm: Quit: Leaving.
[2011/11/04 16:14:30] <Volcane> peridot: not really
[2011/11/04 16:14:46] <countskm> CrazyGir: anything "out of the norm" takes more effort in puppet.. i am managing aix and find myself writing custom providers to get things done
[2011/11/04 16:15:25] <countskm> CrazyGir: "out of the norm" w/ respect to os / distribution
[2011/11/04 16:16:16] @ Quit: shadoi: Quit: Leaving.
[2011/11/04 16:16:16] <torrancew> hostdream: I'm referring to #9290, hope that helps
[2011/11/04 16:16:16] <gepetto> torrancew: hostdream: #9290 is http://projects.puppetlabs.com/issues/show/9290 "Puppet - Bug #9290: Puppet master fails with 'stack level too deep' error when storeconfigs = true with rails stack 3.1.0. It has a status of Accepted and is assigned to -"
[2011/11/04 16:16:39] <hostdream> ok I will check that, thanks
[2011/11/04 16:17:22] <CrazyGir> mmm
[2011/11/04 16:17:40] <CrazyGir> I usually have prety good luck with openbsd in that regard, but yea, you are still correct
[2011/11/04 16:17:53] @ los324 is now known as cLos
[2011/11/04 16:20:45] <CrazyGir> this is frustrating: /Stage[main]/Base/Package[nmap]/ensure: change from absent to present failed: You must specify a package source for BSD packages
[2011/11/04 16:20:59] <CrazyGir> in openbsd you set PKG_PATH, eg: export PKG_PATH="ftp://ftp3.usa.openbsd.org/pub/OpenBSD/4.9/packages/i386"
[2011/11/04 16:22:13] <torrancew> CrazyGir: so make sure your puppetmaster gets that in its environment, and that should work, I imagine
[2011/11/04 16:22:28] @ Quit: hobbeswalsh: Quit: hobbeswalsh
[2011/11/04 16:22:53] @ Quit: pjsz_: Quit: pjsz_
[2011/11/04 16:22:58] <CrazyGir> isn't that executed on the client side?
[2011/11/04 16:23:26] <CrazyGir> eg, functionally, is puppet not running aptitude install foo or pkg_add foo
[2011/11/04 16:23:39] <torrancew> you're correct, but the solution should still work
[2011/11/04 16:23:43] <CrazyGir> and in this case, pkg_add is the one that is relying on PKG_PATH, not puppet
[2011/11/04 16:23:46] <torrancew> ie invoke it as PKG_PATH=$foo puppet
[2011/11/04 16:24:33] <CrazyGir> oh yes, puppet would run as another user internally, correct?
[2011/11/04 16:24:40] <CrazyGir> bah, same thing
[2011/11/04 16:25:18] <torrancew> :(
[2011/11/04 16:25:54] <CrazyGir> yea, a little like wth, this was not such a pain in the past
[2011/11/04 16:25:59] <CrazyGir> (puppet on openbsd)
[2011/11/04 16:26:13] @ Quit: cruzfox: Quit: cruzfox
[2011/11/04 16:27:49] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 16:28:41] @ Quit: dan__t: Ping timeout: 276 seconds
[2011/11/04 16:28:52] @ Quit: the_ramink: Ping timeout: 255 seconds
[2011/11/04 16:31:10] @ Quit: lak: Quit: Leaving.
[2011/11/04 16:31:56] @ Quit: newtMcKerr: Quit: Leaving.
[2011/11/04 16:32:05] @ Quit: tizzo-afk: Ping timeout: 240 seconds
[2011/11/04 16:33:10] @ tizzo joined channel #puppet
[2011/11/04 16:33:25] @ Quit: jongrocho: Remote host closed the connection
[2011/11/04 16:33:56] @ cLos is now known as foolme1s
[2011/11/04 16:35:40] @ jongrocho joined channel #puppet
[2011/11/04 16:36:18] @ soh joined channel #puppet
[2011/11/04 16:36:49] @ Quit: jongrocho: Remote host closed the connection
[2011/11/04 16:39:13] @ Quit: davewongillies: Ping timeout: 255 seconds
[2011/11/04 16:39:51] @ Quit: jaminja: Ping timeout: 260 seconds
[2011/11/04 16:40:22] @ jaminja joined channel #puppet
[2011/11/04 16:40:32] @ Quit: jaminja: Changing host
[2011/11/04 16:40:32] @ jaminja joined channel #puppet
[2011/11/04 16:41:49] <jiboumans> Hi, I hope there's someone around that can help me figure out a puppet issue. My puppet clients on newly launched instances are reporting 'err: Could not retrieve catalog from remote server: SSL_connect returned=1 errno=0 state=SSLv3 read server certificate B: certificate verify failed'. The cert signs cleanly on the master, but even removing the cert on the master (puppet cert --clean) AND removing the /var/lib/puppet/ssl and rerunning the whole thing
[2011/11/04 16:42:05] <jiboumans> is not leading to a successful run; I sign the cert again on the server, but a subsequent run on the client gives the same error again.
[2011/11/04 16:42:26] <jiboumans> This was working until at least oct 28th (last launched instance), and is no longer working today. I can't find anythign specific that changed in the puppet installation in that time frame.
[2011/11/04 16:43:28] <peridot> systems are complex things...what other changes have there been? I'd run it in --verbose mode too if you haven't already.
[2011/11/04 16:43:42] @ cosman2001 joined channel #puppet
[2011/11/04 16:44:14] @ Quit: benjiedmund: Remote host closed the connection
[2011/11/04 16:44:20] <jiboumans> peridot: i'm running the client with --test --debug
[2011/11/04 16:44:26] <jiboumans> it's not giving any useful diagnostic other than that
[2011/11/04 16:44:53] <cosman2001> I just upgrade to 2.7.6 and I get this error on all clients undefined method `Pathname' for #<Puppet::Type::File:0x2a975f5648>
[2011/11/04 16:44:54] <jiboumans> peridot: i've verified time settings as well
[2011/11/04 16:45:11] <cosman2001> did something change in 2.7.6?
[2011/11/04 16:45:11] <jiboumans> there's no noticable drift
[2011/11/04 16:45:36] <peridot> how about your ruby? have you upgraded that?
[2011/11/04 16:45:47] <Volcane> clocks in sync?
[2011/11/04 16:45:57] <jiboumans> no, we're running 10.04 lts and using its ruby package
[2011/11/04 16:45:59] <jiboumans> Volcane: yes
[2011/11/04 16:46:24] <jiboumans> i'd say the first step is to get why the certs don't match, and i can't figure out how to do that
[2011/11/04 16:46:52] <jiboumans> our method of launching instances also hasn't changed
[2011/11/04 16:46:53] <CrazyGir> what is the sensible way to make changes to your manifests dependent on the node's OS?
[2011/11/04 16:47:09] <CrazyGir> i primarily support 2 OS (debian/ubuntu and OpenBSD)
[2011/11/04 16:47:14] <whack> CrazyGir: there's all kinds of facts available for that
[2011/11/04 16:47:31] <whack> like $operatingsystem, etc
[2011/11/04 16:47:54] * countskm forsees a puppet "design patterns" book in the future :-)
[2011/11/04 16:48:28] <CrazyGir> whack: this is less of a question of what is available to the dev, and more like "what is a good way to go about this", as countskm is pointing out
[2011/11/04 16:48:54] <whack> case $operatingsystem { "debian" => ... or whatevers }
[2011/11/04 16:48:58] <peridot> openssl x509 -text -noout -in /var/lib/puppet/ssl/certs/hostname.tld.pem \| grep -A2 Validity (replacing hostname.tld) from http://projects.puppetlabs.com/projects/1/wiki/Certificates_And_Security
[2011/11/04 16:49:10] <Volcane> CrazyGir: include "apache::${operatingsystem}"
[2011/11/04 16:49:13] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 16:49:13] <Volcane> is one possibility
[2011/11/04 16:49:20] @ jorhett left channel #puppet ()
[2011/11/04 16:49:29] <CrazyGir> is there a doc that has some examples?
[2011/11/04 16:50:57] <CrazyGir> http://docs.puppetlabs.com/guides/parameterized_classes.html <--- maybe this is what I ought to be wrapping my head around?
[2011/11/04 16:51:21] @ Quit: erodriguez: Quit: erodriguez
[2011/11/04 16:51:35] <rodjek> CrazyGir: no, this one. http://docs.puppetlabs.com/guides/language_guide.html#conditionals
[2011/11/04 16:52:03] <jiboumans> peridot: Validity
[2011/11/04 16:52:03] <jiboumans> Not Before: Nov 3 23:15:18 2011 GMT
[2011/11/04 16:52:04] <jiboumans> Not After : Nov 1 23:15:18 2016 GMT
[2011/11/04 16:52:07] <jiboumans> which seems reasonable
[2011/11/04 16:52:52] @ davewongillies joined channel #puppet
[2011/11/04 16:52:59] <jiboumans> peridot: comparing that to another (working) host, it looks similar
[2011/11/04 16:53:02] <CrazyGir> nice!
[2011/11/04 16:54:35] <peridot> and it's the same image that your last successful client launched on?
[2011/11/04 16:55:02] <Volcane> jiboumans: whats the time on the master?
[2011/11/04 16:55:30] <jiboumans> $ date
[2011/11/04 16:55:30] <jiboumans> Fri Nov 4 23:55:23 UTC 2011
[2011/11/04 16:55:37] @ foolme1s is now known as los324
[2011/11/04 16:55:38] <Volcane> weird
[2011/11/04 16:55:39] <jiboumans> Volcane: ^
[2011/11/04 16:57:01] <jiboumans> Volcane: i have no idea where to look beyond this, and it's ground our ops to a halt, so i'm highly motivated to get to the bottom of this
[2011/11/04 16:57:50] @ Quit: ghg: Quit: Computer has gone to sleep.
[2011/11/04 16:59:23] <peridot> jiboumans: what versions of ruby/puppet do you have?
[2011/11/04 16:59:39] @ Quit: countskm: Quit: leaving
[2011/11/04 16:59:53] @ countskm joined channel #puppet
[2011/11/04 16:59:54] <jiboumans> peridot: puppet 2.6.3, ruby 1.8.7.249-2
[2011/11/04 17:00:34] @ tizzo is now known as tizzo-afk
[2011/11/04 17:01:04] @ Quit: rharrison: Quit: Leaving
[2011/11/04 17:01:15] <peridot> Is that the same as the versions on your working clients?
[2011/11/04 17:01:40] <jiboumans> peridot: http://pastebin.com/y5Eg4hf4
[2011/11/04 17:01:40] <jiboumans> yes
[2011/11/04 17:01:56] <jiboumans> peridot: that's the --test --debug output on the client
[2011/11/04 17:02:42] <jiboumans> peridot: we have 100+ nodes under puppet control, and we're very strict on how they're launched
[2011/11/04 17:02:49] @ jongrocho joined channel #puppet
[2011/11/04 17:03:46] @ Quit: jaminja: Ping timeout: 260 seconds
[2011/11/04 17:05:32] @ Quit: verwilst: Read error: Operation timed out
[2011/11/04 17:05:33] @ jaminja joined channel #puppet
[2011/11/04 17:07:31] @ mitchellh joined channel #puppet
[2011/11/04 17:07:55] @ Quit: mitchellh: Client Quit
[2011/11/04 17:08:26] @ devdas joined channel #puppet
[2011/11/04 17:08:34] @ mitchellh joined channel #puppet
[2011/11/04 17:09:50] <peridot> jiboumans: All I can suggest is that you look at the changes that were made since your last successful launch, assuming you are launching from the same exact image? Ubuntu 10.04 doesn't describe an exact state, as packages are constantly upgraded.
[2011/11/04 17:09:54] @ Quit: sumsuch: Quit: sumsuch
[2011/11/04 17:10:23] <jiboumans> peridot: i have, and i am.. but what i need to know is what i can do to debug puppet
[2011/11/04 17:10:27] <jiboumans> it just says 'doesnt work'
[2011/11/04 17:10:46] <Volcane> unfortunately in this case it just tells you what the underlying ssl libraries has to say
[2011/11/04 17:10:53] @ Quit: mitchellh: Client Quit
[2011/11/04 17:10:58] <jiboumans> volcane: can i call those and get more information?
[2011/11/04 17:11:03] <Volcane> and those dont provide more info - all references i can find to that combo though suggests time related issues
[2011/11/04 17:11:04] <jiboumans> i'm not very fluent on ssl
[2011/11/04 17:11:11] <peridot> You could strace it :P
[2011/11/04 17:11:26] <Volcane> I'd delete the ssl directory o the client
[2011/11/04 17:11:26] <jiboumans> .. i did already, the output is lengthy, so i need to filter
[2011/11/04 17:11:30] <Volcane> do a --clean client on the master
[2011/11/04 17:11:32] <jiboumans> volcane: already done
[2011/11/04 17:11:33] <jiboumans> and done
[2011/11/04 17:11:35] <whack> you can try debugging it with openssl s_client
[2011/11/04 17:11:37] <peridot> and restart the master
[2011/11/04 17:11:57] @ sumsuch joined channel #puppet
[2011/11/04 17:12:01] <jiboumans> whack: pointer woudl be great.. i've managed to avoid ssl like the plague.. it may be biting me now
[2011/11/04 17:12:02] <whack> from the client, use openssl s_client with the client's cert and master CA/cert against the master
[2011/11/04 17:12:06] <Volcane> peridot: whats the time on the client - show the same date command as earlier
[2011/11/04 17:12:17] <whack> jiboumans: man s_client should get you started
[2011/11/04 17:12:23] <jiboumans> whack: fair enough
[2011/11/04 17:12:23] <Volcane> errr, jiboumans even
[2011/11/04 17:12:27] <whack> in general, openssl s_client -connect puppet:8143 or whatever
[2011/11/04 17:12:34] <whack> other flags will let you set the certs
[2011/11/04 17:12:37] <whack> and verify levels
[2011/11/04 17:12:54] <eric0> jiboumans: also--just to pile on here-- make sure the $ssldir/crl.pem on the client doesnt contain the serial number of the server's cert
[2011/11/04 17:13:19] @ mitchellh joined channel #puppet
[2011/11/04 17:13:19] @ Quit: mitchellh: Client Quit
[2011/11/04 17:13:27] <jiboumans> eric0: i've simply moved the ssl dir aside
[2011/11/04 17:13:42] <peridot> or bad permissions
[2011/11/04 17:13:46] <Volcane> jiboumans: show output of date on the client
[2011/11/04 17:13:55] <jiboumans> coming
[2011/11/04 17:13:56] <eric0> jiboumans: the client will download a new crl first time it runs, it lives on the server at $ssldir/ca/ca_crl.pem
[2011/11/04 17:14:17] <jiboumans> date
[2011/11/04 17:14:18] <jiboumans> Sat Nov 5 00:14:03 UTC 2011
[2011/11/04 17:14:35] <Volcane> yeah seems fine
[2011/11/04 17:14:56] <eric0> jiboumans: i made this diagram that may help you correlate messages on the client to ssl activity http://dl.dropbox.com/u/18472980/puppet-ssl-timeseq.png
[2011/11/04 17:15:16] <Volcane> eric0: nice
[2011/11/04 17:15:42] <peridot> eric0: I'm saving that :P
[2011/11/04 17:15:54] <jiboumans> well spin my nipple nuts and send me to alaska
[2011/11/04 17:16:10] <peridot> that's a new one
[2011/11/04 17:16:12] <jiboumans> peridot: 'have you tried turning it off and on again' was the right answer
[2011/11/04 17:16:17] <eric0> heres another for agent activity http://dl.dropbox.com/u/18472980/puppet-agent-timeseq.png
[2011/11/04 17:16:17] <jiboumans> peridot: kryten from red dwarf
[2011/11/04 17:16:31] <jiboumans> peridot: restarting apache made it unfuck
[2011/11/04 17:16:35] <Volcane> heh
[2011/11/04 17:16:37] <peridot> heh, glad to hear it :)
[2011/11/04 17:16:43] <jiboumans> .. that is.. disturbing
[2011/11/04 17:16:59] @ Quit: DJClean: Read error: Operation timed out
[2011/11/04 17:17:02] * eric0 needs to fuck off away from work
[2011/11/04 17:17:30] <jiboumans> eric0: i'm saving those diagrams, awesome, thanks
[2011/11/04 17:17:53] @ Quit: Vulcan[OMNI]: Quit: Vulcan[OMNI]
[2011/11/04 17:18:00] <jiboumans> i'm taking guesses now as to why the fuck an apache restart would make a difference
[2011/11/04 17:18:03] <jiboumans> and there will be rewards
[2011/11/04 17:18:36] <peridot> config changes or cache issue would be my guess
[2011/11/04 17:18:52] <Volcane> yeah some form of negative cache or something like that
[2011/11/04 17:19:18] <jiboumans> nothing changed to the puppet install on the master, so.. cache then?
[2011/11/04 17:19:26] <peridot> "Using cached certificate"
[2011/11/04 17:19:37] * jiboumans adds 'restart apache' to his things to do when puppet is making him upset
[2011/11/04 17:19:42] @ Quit: ianbrandt: Quit: ianbrandt
[2011/11/04 17:20:17] <peridot> At least you didn't get to the bottom of the list: kick the server a few times to see if it makes a difference.
[2011/11/04 17:22:12] <jiboumans> peridot: ….
[2011/11/04 17:22:13] <jiboumans> oh mhy
[2011/11/04 17:22:41] <countskm> peridot: sometimes physical violence is appropriate
[2011/11/04 17:22:52] <peridot> It gets the neurons flowing
[2011/11/04 17:23:04] <countskm> :-)
[2011/11/04 17:23:16] <peridot> ;-)
[2011/11/04 17:23:23] @ Quit: jaminja: Ping timeout: 252 seconds
[2011/11/04 17:23:31] <peridot> Well, have a good weekend all
[2011/11/04 17:23:35] @ Quit: peridot: Quit: ~ Trillian Astra - www.trillian.im ~
[2011/11/04 17:23:43] <jiboumans> thanks for your help all, much appreciated
[2011/11/04 17:24:08] <jiboumans> if you make it to hadoop world, remind me i owe you a drink :)
[2011/11/04 17:24:11] @ jaminja joined channel #puppet
[2011/11/04 17:24:11] @ Quit: jaminja: Changing host
[2011/11/04 17:24:12] @ jaminja joined channel #puppet
[2011/11/04 17:24:25] @ Quit: jbraeuer: Quit: jbraeuer
[2011/11/04 17:27:34] @ Quit: johnston: Ping timeout: 265 seconds
[2011/11/04 17:27:34] @ Holocaine joined channel #puppet
[2011/11/04 17:28:33] <tobert> hmm is anybody using ldap and having puppet write them to /etc/passwd, etc.?
[2011/11/04 17:28:55] <tobert> we currently generate a huge class and I want to get puppet to pull uids/gids directly if I can
[2011/11/04 17:30:17] <rodjek> i'm not, but it's possible enough
[2011/11/04 17:31:00] @ Quit: claussen: Quit: claussen
[2011/11/04 17:31:43] @ los324 is now known as foolme1s
[2011/11/04 17:37:55] @ Quit: djm68: Quit: leaving
[2011/11/04 17:39:00] @ dwalu joined channel #puppet
[2011/11/04 17:39:47] @ erodriguez joined channel #puppet
[2011/11/04 17:40:57] @ tdo joined channel #puppet
[2011/11/04 17:41:01] @ tdo left channel #puppet ()
[2011/11/04 17:41:21] @ Quit: langworthy: Quit: langworthy
[2011/11/04 17:41:49] @ Quit: csanchez: Quit: csanchez
[2011/11/04 17:44:37] @ socomm joined channel #puppet
[2011/11/04 17:45:09] @ Quit: Holocaine:
[2011/11/04 17:45:12] @ los324_ joined channel #puppet
[2011/11/04 17:45:46] <foolme1s> msg
[2011/11/04 17:46:16] @ skm joined channel #puppet
[2011/11/04 17:52:32] @ Quit: hooper: Ping timeout: 276 seconds
[2011/11/04 17:53:06] @ Quit: jongrocho: Remote host closed the connection
[2011/11/04 17:58:00] @ thenulldevice joined channel #puppet
[2011/11/04 17:58:47] @ Quit: nottings: Remote host closed the connection
[2011/11/04 18:00:07] @ Quit: erodriguez: Quit: erodriguez
[2011/11/04 18:01:34] @ tizzo-afk is now known as tizzo
[2011/11/04 18:02:11] @ Quit: los324_: Quit: leaving
[2011/11/04 18:02:56] @ tizzo is now known as tizzo-afk
[2011/11/04 18:05:49] @ Quit: agagag: Ping timeout: 258 seconds
[2011/11/04 18:07:03] @ Quit: jaminja: Ping timeout: 245 seconds
[2011/11/04 18:07:14] @ jazo joined channel #puppet
[2011/11/04 18:08:35] @ Quit: jazo:
[2011/11/04 18:09:33] @ Quit: mmoya: Ping timeout: 245 seconds
[2011/11/04 18:09:46] @ jazo joined channel #puppet
[2011/11/04 18:09:56] <jazo> !lista
[2011/11/04 18:10:34] @ ken_barber joined channel #puppet
[2011/11/04 18:10:40] <jazo> i soliti idioti
[2011/11/04 18:11:22] @ rcrowley joined channel #puppet
[2011/11/04 18:11:26] @ tizzo-afk is now known as tizzo
[2011/11/04 18:11:40] <mig5> sono subito da lei
[2011/11/04 18:12:43] @ agagag joined channel #puppet
[2011/11/04 18:13:23] @ Quit: kenperkins: Quit: Computer has gone to sleep.
[2011/11/04 18:16:55] <CrazyGir> ssl woes: http://dpaste.com/647511/
[2011/11/04 18:17:33] <CrazyGir> I just cleared the ssl path and allowed puppet to recreate it, signed a new cert, and still this error
[2011/11/04 18:17:59] <CrazyGir> I am all about being an informed user, but puppet has some terrible guidance when it comes to error messages and error handling
[2011/11/04 18:20:21] <Volcane> CrazyGir: if you use --server=some.weird.name you need to be sure that the actual puppet master was configured to be called by that name
[2011/11/04 18:20:31] <rodjek> CrazyGir: is the system time in sync on your puppetmaster and client?
[2011/11/04 18:20:44] @ soh joined channel #puppet
[2011/11/04 18:20:47] <Volcane> CrazyGir: generally just using 'puppet' works fine without tweaks
[2011/11/04 18:21:05] @ Quit: st33v: Quit: st33v
[2011/11/04 18:22:22] @ jaminja joined channel #puppet
[2011/11/04 18:22:22] @ Quit: jaminja: Changing host
[2011/11/04 18:22:23] @ jaminja joined channel #puppet
[2011/11/04 18:23:40] <CrazyGir> Volcane: that was wrong, and so I cleared the ssl directories on both sides and reran after correcting the fqdn
[2011/11/04 18:23:47] <CrazyGir> rodjek: good question
[2011/11/04 18:24:25] @ tizzo is now known as tizzo-afk
[2011/11/04 18:26:08] @ newtMcKerr joined channel #puppet
[2011/11/04 18:27:51] @ Quit: jaminja: Ping timeout: 256 seconds
[2011/11/04 18:29:01] @ Quit: tilgovi: Remote host closed the connection
[2011/11/04 18:29:42] @ johnston joined channel #puppet
[2011/11/04 18:29:59] @ Quit: jazo:
[2011/11/04 18:31:02] @ tizzo-afk is now known as tizzo
[2011/11/04 18:31:07] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 18:32:15] @ soh joined channel #puppet
[2011/11/04 18:33:10] @ Spads joined channel #puppet
[2011/11/04 18:33:20] @ st33v joined channel #puppet
[2011/11/04 18:33:27] @ st33v left channel #puppet ()
[2011/11/04 18:36:28] @ jaminja joined channel #puppet
[2011/11/04 18:36:29] @ Quit: jaminja: Changing host
[2011/11/04 18:36:29] @ jaminja joined channel #puppet
[2011/11/04 18:36:44] @ Quit: Spads_: Ping timeout: 276 seconds
[2011/11/04 18:39:08] @ tizzo is now known as tizzo-afk
[2011/11/04 18:40:19] @ Ownage joined channel #puppet
[2011/11/04 18:40:30] <Ownage> can someone explain to me if this is a bug or if I'm crazy or something
[2011/11/04 18:40:38] @ claussen joined channel #puppet
[2011/11/04 18:41:01] <CrazyGir> ?
[2011/11/04 18:41:47] @ z3ndrag0n joined channel #puppet
[2011/11/04 18:42:24] <Ownage> http://pastebin.com/hEpMnXWJ
[2011/11/04 18:42:45] <Ownage> seems insane to have an order required for attributes
[2011/11/04 18:43:01] <Ownage> especially for unrelated attributes
[2011/11/04 18:43:02] @ Quit: rcrowley: Quit: Computer has gone to sleep.
[2011/11/04 18:43:20] <Ownage> oh man
[2011/11/04 18:43:25] <Ownage> I know I forgot the comma
[2011/11/04 18:43:26] <Ownage> I hate you
[2011/11/04 18:43:29] <dpittman> Ownage: You don't have a ',' after the '}' at the end of the fstype thing. :)
[2011/11/04 18:44:16] @ Quit: jaminja: Ping timeout: 248 seconds
[2011/11/04 18:45:21] @ Quit: ken_barber: Quit: Leaving.
[2011/11/04 18:45:45] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 18:47:06] @ Quit: jsurfer: Read error: Operation timed out
[2011/11/04 18:47:13] @ Quit: ezekiel: Ping timeout: 258 seconds
[2011/11/04 18:49:14] @ pjsz_ joined channel #puppet
[2011/11/04 18:50:25] <CrazyGir> I'm getting err: Could not run Puppet configuration client: Could not find user puppet
[2011/11/04 18:50:35] <CrazyGir> despite having _puppet:_puppet
[2011/11/04 18:51:08] @ Holocaine joined channel #puppet
[2011/11/04 18:52:14] <CrazyGir> this is silly..
[2011/11/04 18:52:38] <Volcane> the unix user is _puppet?
[2011/11/04 18:52:53] @ Quit: thenulldevice: Quit: Leaving.
[2011/11/04 18:53:03] <CrazyGir> yep
[2011/11/04 18:53:13] <Volcane> why? :P
[2011/11/04 18:53:33] <CrazyGir> openbsd convention.. users for system services are prefixed with a _
[2011/11/04 18:53:38] <Volcane> wow
[2011/11/04 18:53:56] <Volcane> did you configure it with user= and group= ?
[2011/11/04 18:53:56] <CrazyGir> why? everyone does things a little differently.. that's why we have puppet!
[2011/11/04 18:54:08] <Volcane> in the puppet.conf
[2011/11/04 18:54:10] <CrazyGir> should that be in [main] ?
[2011/11/04 18:54:26] <Volcane> probably in [master]
[2011/11/04 18:55:07] <jbaldridge> errr... wouldn't the local agent attempt to run as "puppet" then?
[2011/11/04 18:55:08] <CrazyGir> does the master dameon respect pkill -HUP?
[2011/11/04 18:55:21] <CrazyGir> jbaldridge: I was going to ask the same
[2011/11/04 18:55:31] <CrazyGir> should be in both [master] / [agent] then?
[2011/11/04 18:55:38] <CrazyGir> putting it in [main] failed
[2011/11/04 18:55:46] <CrazyGir> unless my restarting puppet master failed
[2011/11/04 18:56:41] <jbaldridge> I don't think a HUP would work if you're changing the user.
[2011/11/04 18:57:05] <CrazyGir> it's already running as _puppet
[2011/11/04 18:57:08] @ jaminja joined channel #puppet
[2011/11/04 18:57:09] @ Quit: jaminja: Changing host
[2011/11/04 18:57:09] @ jaminja joined channel #puppet
[2011/11/04 18:57:09] <CrazyGir> it's the client that is failing
[2011/11/04 18:58:31] <CrazyGir> same when trying clients on other systems (I had been focused on running puppet agent on the master, against puppet master)
[2011/11/04 18:59:19] <tobert> ugh, ssh_authorized_key doesn't match on the key itself ...
[2011/11/04 19:00:02] @ ShakataGaNai joined channel #puppet
[2011/11/04 19:01:18] <Volcane> matches on the key name
[2011/11/04 19:01:26] <tobert> that's ... awful
[2011/11/04 19:02:00] <tobert> oh well I'm sure somebody had a reason for that choice, whether I disagree with it or not
[2011/11/04 19:04:13] <Volcane> only a bad choice cos the type also doesnt support purging
[2011/11/04 19:04:29] <tobert> there's ensure => absent unless that doesn't work
[2011/11/04 19:04:35] <Volcane> thats not purge
[2011/11/04 19:04:49] <Volcane> purge is remove keys from the file that puppet didnt specifically put there
[2011/11/04 19:05:15] <tobert> ahh yeah that would be nice ... also tricky to detect unless it keyed off the key itself and wrote a custom key name
[2011/11/04 19:05:22] <tobert> sshd pretty much ignores those anyways
[2011/11/04 19:05:41] @ Quit: johnston: Ping timeout: 265 seconds
[2011/11/04 19:06:10] <tobert> sshd-rsa ... tobert@puppet-#{catalog} or something like that
[2011/11/04 19:06:11] @ Quit: sumsuch: Quit: sumsuch
[2011/11/04 19:07:08] <Volcane> i dont see the problem
[2011/11/04 19:09:57] @ ShakataGaNai_ joined channel #puppet
[2011/11/04 19:10:57] @ alanevans joined channel #puppet
[2011/11/04 19:11:06] @ Quit: Razique: Quit: Razique
[2011/11/04 19:11:34] @ ken_barber joined channel #puppet
[2011/11/04 19:12:06] @ Quit: zwi: Ping timeout: 260 seconds
[2011/11/04 19:13:15] @ Quit: ShakataGaNai: Ping timeout: 258 seconds
[2011/11/04 19:14:46] @ Quit: ShakataGaNai_: Client Quit
[2011/11/04 19:20:23] @ jsurfer joined channel #puppet
[2011/11/04 19:24:04] @ erodriguez joined channel #puppet
[2011/11/04 19:24:16] @ soh joined channel #puppet
[2011/11/04 19:32:25] @ Theory joined channel #puppet
[2011/11/04 19:33:52] <Theory> Is there a recommended way of representing actions which should be performed once and only once per resource?
[2011/11/04 19:34:00] @ kolla_ joined channel #puppet
[2011/11/04 19:34:06] <Theory> My particular case is setting usernames and passwords in a database to an initial value
[2011/11/04 19:36:00] @ Quit: ken_barber: Quit: Leaving.
[2011/11/04 19:36:50] @ st33v joined channel #puppet
[2011/11/04 19:40:38] @ Quit: newtMcKerr: Quit: Leaving.
[2011/11/04 19:44:45] @ zwi joined channel #puppet
[2011/11/04 19:44:59] @ Quit: Ownage: Ping timeout: 276 seconds
[2011/11/04 19:45:44] @ Ownage joined channel #puppet
[2011/11/04 19:47:24] @ Quit: foolme1s: Ping timeout: 258 seconds
[2011/11/04 19:47:53] @ jongrocho joined channel #puppet
[2011/11/04 19:52:07] @ kenperkins joined channel #puppet
[2011/11/04 19:52:56] @ st33v left channel #puppet ()
[2011/11/04 19:53:57] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 19:55:19] @ Quit: jongrocho: Remote host closed the connection
[2011/11/04 19:58:32] @ jhelwig is now known as zz_jhelwig
[2011/11/04 20:00:31] @ Quit: pjsz_: Quit: pjsz_
[2011/11/04 20:09:10] <tobert> hmm we want to have a no cross-module dependency rule, but the damn jvm seems like it should be an exception
[2011/11/04 20:09:52] @ Quit: Holocaine:
[2011/11/04 20:09:58] <tobert> class appthatlurvesjava::install { include java::jdk6 }
[2011/11/04 20:10:34] @ Holocaine joined channel #puppet
[2011/11/04 20:11:06] @ Quit: Holocaine: Client Quit
[2011/11/04 20:13:24] @ Quit: bmahe: Remote host closed the connection
[2011/11/04 20:15:35] <tobert> whack, you around? how are you handling the jdk problem in a world of no cross-module deps?
[2011/11/04 20:16:25] @ nomyself joined channel #puppet
[2011/11/04 20:17:56] @ Quit: Ownage: Ping timeout: 260 seconds
[2011/11/04 20:18:41] @ Ownage joined channel #puppet
[2011/11/04 20:21:16] <eric0> IMO thats the correct way to express it, i don't think its helpful to totally eschew cross module dependencies
[2011/11/04 20:21:54] <rocket> how do you guys manage the certs if you are rebuilding a vm lots for testing purposes?
[2011/11/04 20:22:04] <eric0> for instance i have a module that manages mrepo configs, which needs a webserver, so i include software::httpd ... thats a clean modular encapsulation IMO
[2011/11/04 20:22:19] <rocket> is there a way to have a set of hostnames always accept new certs etc?
[2011/11/04 20:22:34] <rocket> even if it has been created previously?
[2011/11/04 20:24:08] @ Quit: kenperkins: Quit: Computer has gone to sleep.
[2011/11/04 20:24:56] @ Holocaine joined channel #puppet
[2011/11/04 20:25:25] <tobert> eric0: yeah, you're absolutely right, I was just trying to think through how it wouldn't be stupid if I was fanatical about it
[2011/11/04 20:25:47] @ zz_jhelwig is now known as jhelwig
[2011/11/04 20:28:12] @ st33v joined channel #puppet
[2011/11/04 20:28:44] @ Quit: ichristo: Quit: Textual IRC Client: http://www.textualapp.com/
[2011/11/04 20:29:11] @ jsurfer_ joined channel #puppet
[2011/11/04 20:30:04] @ Quit: st33v: Client Quit
[2011/11/04 20:32:24] @ Quit: Holocaine:
[2011/11/04 20:32:28] @ soh joined channel #puppet
[2011/11/04 20:32:38] @ Quit: jsurfer: Ping timeout: 258 seconds
[2011/11/04 20:32:38] @ jsurfer_ is now known as jsurfer
[2011/11/04 20:35:28] @ Quit: jaymurtii: Ping timeout: 255 seconds
[2011/11/04 20:37:53] @ Holocaine joined channel #puppet
[2011/11/04 20:38:17] @ Quit: jsurfer: Quit: jsurfer
[2011/11/04 20:40:03] @ frostynova joined channel #puppet
[2011/11/04 20:40:14] @ Quit: lofidellity: Quit: ZNC - http://znc.sourceforge.net
[2011/11/04 20:43:38] @ Quit: kolla_: Remote host closed the connection
[2011/11/04 20:45:00] @ Quit: cliff-hm: Ping timeout: 260 seconds
[2011/11/04 20:45:14] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 20:51:31] @ jsurfer joined channel #puppet
[2011/11/04 20:52:55] @ bmahe joined channel #puppet
[2011/11/04 20:52:56] @ Quit: bmahe: Changing host
[2011/11/04 20:52:56] @ bmahe joined channel #puppet
[2011/11/04 20:54:27] @ Quit: IndyMichaelB: Remote host closed the connection
[2011/11/04 20:54:44] @ IndyMichaelB joined channel #puppet
[2011/11/04 20:54:58] @ Quit: rektide: Ping timeout: 245 seconds
[2011/11/04 21:00:08] @ kenperkins joined channel #puppet
[2011/11/04 21:00:46] @ los3241 joined channel #puppet
[2011/11/04 21:02:48] @ los3241 left channel #puppet ()
[2011/11/04 21:02:57] <CrazyGir> fun.. my "puppet can't find the user" issue is related to my configs
[2011/11/04 21:03:14] <CrazyGir> using a config from puppet --genconfig works fine, fails mine
[2011/11/04 21:03:43] @ Quit: lbt: Ping timeout: 245 seconds
[2011/11/04 21:03:43] <CrazyGir> I have a feeling it is related to the directories specified, but I have no real idea
[2011/11/04 21:04:41] @ tttflexguy joined channel #puppet
[2011/11/04 21:06:50] @ Quit: Holocaine:
[2011/11/04 21:08:24] <CrazyGir> unfortunately, genconfig has _so_ much in it, I am not really able to identify what is missing..
[2011/11/04 21:09:38] @ rcrowley joined channel #puppet
[2011/11/04 21:10:34] @ jongrocho joined channel #puppet
[2011/11/04 21:13:43] @ lwhalen joined channel #puppet
[2011/11/04 21:13:58] <tttflexguy> anybody knows about likewise?
[2011/11/04 21:14:51] @ Quit: jongrocho: Remote host closed the connection
[2011/11/04 21:16:21] @ thenulldevice joined channel #puppet
[2011/11/04 21:19:26] @ soh joined channel #puppet
[2011/11/04 21:28:30] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 21:28:34] @ Quit: frostynova: Ping timeout: 258 seconds
[2011/11/04 21:31:26] @ Quit: rcrowley: Quit: Computer has gone to sleep.
[2011/11/04 21:34:06] @ Quit: kelseyhightower: Quit: kelseyhightower
[2011/11/04 21:34:18] @ frostynova joined channel #puppet
[2011/11/04 21:35:30] @ rektide joined channel #puppet
[2011/11/04 21:35:45] @ Quit: frostynova: Remote host closed the connection
[2011/11/04 21:39:47] @ soh joined channel #puppet
[2011/11/04 21:42:24] @ adimania joined channel #puppet
[2011/11/04 21:43:06] @ IndyMichaelB_ joined channel #puppet
[2011/11/04 21:45:49] @ Quit: z3ndrag0n: Quit: Colloquy for iPhone - http://colloquy.mobi
[2011/11/04 21:46:02] @ Quit: cosman2001: Quit: cosman2001
[2011/11/04 21:46:34] @ Quit: jodok: Quit: jodok
[2011/11/04 21:47:12] @ Quit: IndyMichaelB: Ping timeout: 248 seconds
[2011/11/04 21:47:12] @ IndyMichaelB_ is now known as IndyMichaelB
[2011/11/04 21:54:38] @ thenulldevice1 joined channel #puppet
[2011/11/04 21:55:55] @ Quit: claussen: Remote host closed the connection
[2011/11/04 21:56:36] @ Quit: thenulldevice: Ping timeout: 260 seconds
[2011/11/04 22:10:08] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/11/04 22:23:02] @ rcrowley joined channel #puppet
[2011/11/04 22:23:55] @ Quit: adimania: Ping timeout: 255 seconds
[2011/11/04 22:31:38] @ DigitalNinja joined channel #puppet
[2011/11/04 22:36:48] @ Quit: jaminja: Ping timeout: 248 seconds
[2011/11/04 22:44:22] @ Quit: dwalu: Quit: dwalu
[2011/11/04 22:50:06] @ Quit: rcrowley: Quit: Computer has gone to sleep.
[2011/11/04 22:50:21] @ Quit: gniks: Quit: Leaving.
[2011/11/04 22:57:48] @ the_ramink joined channel #puppet
[2011/11/04 23:07:11] @ Quit: the_ramink: Ping timeout: 260 seconds
[2011/11/04 23:12:55] @ Quit: Ownage: Read error: Connection reset by peer
[2011/11/04 23:13:16] @ Ownage joined channel #puppet
[2011/11/04 23:14:13] @ Quit: ks_Marian: Ping timeout: 240 seconds
[2011/11/04 23:15:43] @ Quit: jamison: Ping timeout: 252 seconds
[2011/11/04 23:17:22] @ Quit: xizhendu: Ping timeout: 252 seconds
[2011/11/04 23:17:45] @ Quit: jhulten: Quit: ZNC - http://znc.sourceforge.net
[2011/11/04 23:18:38] @ jamison joined channel #puppet
[2011/11/04 23:18:45] @ jhulten joined channel #puppet
[2011/11/04 23:31:06] @ Quit: lwhalen: Quit: Computer has gone to sleep
[2011/11/04 23:39:31] @ Quit: tremble: Ping timeout: 260 seconds
[2011/11/04 23:43:15] @ Quit: n0ts: Ping timeout: 252 seconds
[2011/11/04 23:43:28] @ tilgovi joined channel #puppet
[2011/11/04 23:43:29] @ Quit: tilgovi: Changing host
[2011/11/04 23:43:29] @ tilgovi joined channel #puppet
[2011/11/04 23:45:14] @ n0ts joined channel #puppet
[2011/11/04 23:55:00] @ thenulldevice joined channel #puppet
[2011/11/04 23:56:49] @ Quit: thenulldevice1: Ping timeout: 252 seconds
[2011/11/04 23:57:17] @ Quit: abetterlie: Ping timeout: 240 seconds

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!