Sunday, 2011-10-23

[2011/10/23 00:10:26] @ Log started by gepetto
[2011/10/23 00:10:26] @ Joined channel #puppet
[2011/10/23 00:10:26] @ Topic is "Thanks for coming to PuppetConf! Videos available http://www.youtube.com/user/puppetlabsinc | Puppet 2.7.5 http://tinyurl.com/puppet275 | Puppet 2.6.11 http://tinyurl.com/puppet2611 | http://docs.puppetlabs.com | Bugs & Feature Requests http://bit.ly/dKVa0A."
[2011/10/23 00:10:26] @ Topic set by stahnma!~stahnma@fedora/stahnma on Fri Oct 21 11:39:43 -0700 2011
[2011/10/23 00:10:35] @ Mode +Ccnt by card.freenode.net
[2011/10/23 00:12:53] @ Quit: jamesturnbull: Ping timeout: 245 seconds
[2011/10/23 00:13:41] @ BManojlovic joined channel #puppet
[2011/10/23 00:14:40] @ Quit: benjiedmund: Remote host closed the connection
[2011/10/23 00:15:13] @ Quit: Holocaine:
[2011/10/23 00:27:55] @ benjiedmund joined channel #puppet
[2011/10/23 00:36:40] @ Quit: iFire: Ping timeout: 260 seconds
[2011/10/23 00:37:48] @ Quit: benjiedmund: Remote host closed the connection
[2011/10/23 00:38:39] @ iFire joined channel #puppet
[2011/10/23 00:39:02] <kesor> yeah, found it in the end.
[2011/10/23 00:46:24] @ maes_` joined channel #puppet
[2011/10/23 00:47:03] @ Quit: dnjaramba: Ping timeout: 245 seconds
[2011/10/23 00:47:16] @ Quit: bmahe: Quit: Leaving
[2011/10/23 00:48:25] @ Quit: maes_: Ping timeout: 258 seconds
[2011/10/23 00:54:17] @ Quit: neonflux: Quit: Leaving.
[2011/10/23 00:55:14] @ dnjaramba joined channel #puppet
[2011/10/23 00:56:24] @ Quit: sumsuch: Quit: sumsuch
[2011/10/23 01:23:58] @ jodok joined channel #puppet
[2011/10/23 01:39:40] @ hooper joined channel #puppet
[2011/10/23 01:46:17] @ Quit: joelio: Ping timeout: 252 seconds
[2011/10/23 01:46:37] @ Quit: apollo13: Ping timeout: 240 seconds
[2011/10/23 01:46:41] @ Quit: schimmi: Ping timeout: 258 seconds
[2011/10/23 01:47:08] @ jaminja joined channel #puppet
[2011/10/23 01:50:27] @ schimmi joined channel #puppet
[2011/10/23 01:59:40] @ nimrod10 joined channel #puppet
[2011/10/23 02:02:10] @ dissipate joined channel #puppet
[2011/10/23 02:02:13] @ Quit: nimrod10`: Ping timeout: 258 seconds
[2011/10/23 02:02:24] @ Quit: dissipate: Read error: Connection reset by peer
[2011/10/23 02:07:27] @ Quit: Psi-Jack: Quit: ZNC - http://znc.in
[2011/10/23 02:09:39] @ cosman2001 joined channel #puppet
[2011/10/23 02:11:35] @ Holocaine joined channel #puppet
[2011/10/23 02:14:05] @ Psi-Jack joined channel #puppet
[2011/10/23 02:14:59] @ apollo13 joined channel #puppet
[2011/10/23 02:15:39] @ Quit: mrsprinkles: Quit: mrsprinkles
[2011/10/23 02:17:18] @ blentz joined channel #puppet
[2011/10/23 02:23:53] @ ironwolf joined channel #puppet
[2011/10/23 02:25:45] @ Quit: blentz: Ping timeout: 244 seconds
[2011/10/23 02:37:09] @ Quit: cosman2001: Quit: cosman2001
[2011/10/23 02:42:22] @ Quit: dnjaramba: Read error: Connection reset by peer
[2011/10/23 02:44:48] @ tr41n_ joined channel #puppet
[2011/10/23 02:46:13] @ Quit: i0n: Ping timeout: 240 seconds
[2011/10/23 02:49:12] @ jonhattan joined channel #puppet
[2011/10/23 02:56:37] @ dubey joined channel #puppet
[2011/10/23 02:56:41] <dubey> hello
[2011/10/23 02:57:07] <dubey> i very new to puppet. start to learn
[2011/10/23 02:57:45] <dubey> i have installed puppet* on centos6.1. started puppetmasterd and puppet service
[2011/10/23 02:58:06] <dubey> On second centos6.1 box i have installed puppet only and started puppet service
[2011/10/23 02:58:36] <dubey> then i tried to test connectivity using "puppetd --test" which give me error :
[2011/10/23 02:58:42] <dubey> puppetd --test
[2011/10/23 02:58:42] <dubey> err: Could not request certificate: Connection timed out - connect(2)
[2011/10/23 02:58:42] <dubey> Exiting; failed to retrieve certificate and waitforcert is disabled
[2011/10/23 02:59:00] <dubey> On server puppetca --list shows nothing
[2011/10/23 02:59:23] <Dominic> looks more like a firewall or something, assuming you've either set the server name correctly or pointed the host "puppet" to your server
[2011/10/23 02:59:58] <Dominic> check that you can telnet to port 8140 on the server from the client
[2011/10/23 03:03:01] <dubey> Dominic: iptables stopped, selinux disabled,
[2011/10/23 03:03:16] <dubey> telnet to 8140 show : Escape character is '^]'.
[2011/10/23 03:03:16] <dubey> Connection closed by foreign host.
[2011/10/23 03:03:19] <dubey> last two lines
[2011/10/23 03:04:18] @ d3c joined channel #puppet
[2011/10/23 03:06:09] <d3c> I'm having a strange problem where my EC2 instance loses its FQDN everytime I apply site.pp. has anyone experienced this too?
[2011/10/23 03:10:13] <Dominic> dubey: what does `puppetd --configprint server` on the client say? Is it the address of your master?
[2011/10/23 03:10:31] @ Quit: shadoi: Quit: Leaving.
[2011/10/23 03:11:44] @ Quit: tr41n_: Ping timeout: 244 seconds
[2011/10/23 03:11:46] <dubey> Dominic: puppetd --configprint server ?
[2011/10/23 03:11:55] <Dominic> yes
[2011/10/23 03:12:24] <dubey> Result is : puppet
[2011/10/23 03:12:36] <Dominic> and is that the name of your master?
[2011/10/23 03:12:44] <dubey> No
[2011/10/23 03:12:56] <Dominic> then you'll need to reconfigure the client, or fix DNS
[2011/10/23 03:13:18] @ biomass joined channel #puppet
[2011/10/23 03:13:20] <dubey> if i put the name of master instead of server it says : invalid parameter: <nameof my server>
[2011/10/23 03:13:51] <dubey> both are able to ping. I have made an hosts entry for both
[2011/10/23 03:13:53] <Dominic> that command doesn't set the server, it's just printing out what Puppet thinks the master is
[2011/10/23 03:14:27] <Dominic> so you can either set "server = <your server>" on the client's config file, use puppetd --test --server=yourserver, or add a "puppet" record to DNS
[2011/10/23 03:14:35] <dubey> If i run : puppetd --waitforcert 30 --server vpc3.domain.in -v
[2011/10/23 03:14:49] <dubey> it says : Could not prepare for execution: Could not create PID file: /var/run/puppet/agent.pid
[2011/10/23 03:15:21] <Dominic> are you root? Does that directory exist? A package for your distribution should create directories where necessary.
[2011/10/23 03:15:27] <dubey> No DNS setup. I just made an entry in /etc/hosts file
[2011/10/23 03:15:30] <dubey> yes
[2011/10/23 03:16:33] <Dominic> ok, then you should be able to connect without --server. The PID issue should be something fairly simple to solve.
[2011/10/23 03:16:40] <dubey> Dominic: Yes, directory exists and able to cat agent.pid
[2011/10/23 03:17:31] @ Quit: wilmoore: Remote host closed the connection
[2011/10/23 03:17:54] <dubey> let me remove puppet* from my system and try again
[2011/10/23 03:18:20] @ i0n joined channel #puppet
[2011/10/23 03:21:38] <dubey> Dominic: Does puppet.conf require any chanes before starting puppet service
[2011/10/23 03:22:25] <Dominic> I guess that depends on what's in it. Most distro packages will be set up sensibly.
[2011/10/23 03:22:58] <dubey> so a frest puppet install can be started without making any change in conf file
[2011/10/23 03:23:02] <dubey> right ?
[2011/10/23 03:23:34] <Dominic> as above, that depends. But quite possibly.
[2011/10/23 03:24:39] <dubey> Ok. so if i start puppet service on client then waht would be next step
[2011/10/23 03:25:14] <Dominic> sign its certificate, then let it run
[2011/10/23 03:27:17] <dubey> what is the meanig of this : Could not prepare for execution: Could not create PID file: /var/run/puppet/agent.pid
[2011/10/23 03:29:52] @ Quit: biomass: Quit: biomass
[2011/10/23 03:31:39] @ los324 joined channel #puppet
[2011/10/23 03:31:45] @ los324 left channel #puppet ()
[2011/10/23 03:32:01] @ biomass joined channel #puppet
[2011/10/23 03:33:05] @ Quit: Psi-Jack: Quit: ZNC - http://znc.in
[2011/10/23 03:38:05] <dubey> Dominic: puppetd --server vpc3.infomedia18.in --waitforcert 60 --test
[2011/10/23 03:39:38] <dubey> worked
[2011/10/23 03:44:09] @ Psi-Jack joined channel #puppet
[2011/10/23 03:54:30] @ Quit: lbt: Quit: Konversation terminated!
[2011/10/23 04:00:09] @ lbt joined channel #puppet
[2011/10/23 04:13:19] @ Quit: hooper: Ping timeout: 258 seconds
[2011/10/23 04:14:53] @ Quit: dubey: Quit: Leaving
[2011/10/23 04:25:51] @ dubey joined channel #puppet
[2011/10/23 04:25:56] <dubey> hello
[2011/10/23 04:26:19] <dubey> i am getting this error when i run puppetd --test on both server / client
[2011/10/23 04:26:30] <dubey> err: Could not retrieve catalog from remote server: Connection timed out - connect(2)
[2011/10/23 04:26:30] <dubey> warning: Not using cache on failed catalog
[2011/10/23 04:26:30] <dubey> err: Could not retrieve catalog; skipping run
[2011/10/23 04:33:37] <omry> standalone puppetmaster?
[2011/10/23 04:33:43] <omry> try to restart it
[2011/10/23 04:33:53] @ Quit: trams: Ping timeout: 244 seconds
[2011/10/23 04:39:23] <dubey> omry: restarted it but getting same error
[2011/10/23 04:41:18] @ trams joined channel #puppet
[2011/10/23 04:41:55] <omry> try to telnet to it : telnet puppet 8140
[2011/10/23 04:44:57] @ Quit: KaiserBeto1: Quit: Leaving.
[2011/10/23 04:45:45] <dubey> omry: Connected to localhost.
[2011/10/23 04:45:46] <dubey> Escape character is '^]'.
[2011/10/23 04:45:46] <dubey> Connection closed by foreign host.
[2011/10/23 04:46:12] <omry> connection gets closed immediately or after a while/
[2011/10/23 04:46:13] <dubey> why closed
[2011/10/23 04:46:14] <omry> ?
[2011/10/23 04:46:21] <dubey> immediately
[2011/10/23 04:46:25] <omry> so something is up
[2011/10/23 04:46:28] <omry> with your server
[2011/10/23 04:46:31] <omry> anything in its log?
[2011/10/23 04:47:18] <dubey> [2011-10-23 22:48:43] ERROR RuntimeError: Client disconnected before connection could be established
[2011/10/23 04:47:33] <omry> this is from the same machine?
[2011/10/23 04:47:50] <dubey> this is on server
[2011/10/23 04:47:59] <omry> but your client is on the server as well?
[2011/10/23 04:48:15] <omry> something interferes with the connection.
[2011/10/23 04:48:16] <dubey> yes, but have on netwrok client too
[2011/10/23 04:48:24] <omry> maybe firewall, maybe some smartass security shit
[2011/10/23 04:48:57] <dubey> /usr/lib/ruby/site_ruby/1.8/puppet/network/http/webrick.rb:43:in `listen'
[2011/10/23 04:49:03] @ Quit: jaminja: Remote host closed the connection
[2011/10/23 04:49:08] <dubey> firewall is stopper, selinux is disabled
[2011/10/23 04:49:48] <dubey> does it require any changes in puppet.conf file on master server or auth.conf, fileserver.conf
[2011/10/23 04:50:28] <omry> as far as I know, just opening a connection should work on an unmodified installation
[2011/10/23 04:50:41] <omry> your connection gets closed, client things its the server, server things its the client
[2011/10/23 04:50:47] <omry> thinks
[2011/10/23 04:50:49] <omry> x2
[2011/10/23 04:51:05] <omry> can you open a tcp connection to other ports?
[2011/10/23 04:51:11] <omry> telnet puppet 80 / 22 / 24
[2011/10/23 04:51:12] <omry> 5
[2011/10/23 04:51:15] <omry> or whatever?
[2011/10/23 04:52:01] <dubey> yes, 80 is open and client is able to telnet port 80
[2011/10/23 04:52:11] <omry> I really dont know, sorry
[2011/10/23 04:52:53] <dubey> omry: np
[2011/10/23 04:53:00] <omry> you can try to switch to passenger
[2011/10/23 04:53:06] <omry> webrick is a joke anyway
[2011/10/23 04:53:35] <omry> if you plan to use puppet for production, you should do it eventually in any case.
[2011/10/23 04:55:51] <dubey> http://pastebin.com/PK2MJL2D
[2011/10/23 04:56:06] @ Quit: djmitche: Ping timeout: 244 seconds
[2011/10/23 04:56:18] @ jsurfer joined channel #puppet
[2011/10/23 04:57:48] @ Quit: jsurfer: Client Quit
[2011/10/23 04:58:31] @ djmitche joined channel #puppet
[2011/10/23 04:58:34] <dubey> ^ error log
[2011/10/23 05:01:14] @ Quit: Holocaine:
[2011/10/23 05:05:07] @ trek1s joined channel #puppet
[2011/10/23 05:12:18] <dubey> how can i change default webrick ?
[2011/10/23 05:50:41] @ Quit: dubey: Quit: Leaving
[2011/10/23 06:10:21] @ samrose joined channel #puppet
[2011/10/23 06:16:05] @ Quit: biomass: Quit: biomass
[2011/10/23 06:18:31] @ Quit: jodok: Quit: jodok
[2011/10/23 06:18:34] @ devzero is now known as devzero|afk
[2011/10/23 06:20:54] @ jodok joined channel #puppet
[2011/10/23 06:20:54] @ Quit: jodok: Changing host
[2011/10/23 06:20:54] @ jodok joined channel #puppet
[2011/10/23 06:21:32] @ Quit: thoraxe: Remote host closed the connection
[2011/10/23 06:23:24] @ biomass joined channel #puppet
[2011/10/23 06:30:08] @ Quit: oatsport1_: Ping timeout: 244 seconds
[2011/10/23 06:31:31] @ oatsport1 joined channel #puppet
[2011/10/23 06:34:40] @ Quit: kermit: Read error: Connection reset by peer
[2011/10/23 06:36:44] @ kermit joined channel #puppet
[2011/10/23 06:38:22] @ Quit: jamison: Excess Flood
[2011/10/23 06:39:39] @ jamison joined channel #puppet
[2011/10/23 06:43:06] @ biomass_ joined channel #puppet
[2011/10/23 06:43:43] @ Quit: biomass: Read error: Connection reset by peer
[2011/10/23 06:43:43] @ biomass_ is now known as biomass
[2011/10/23 07:00:52] @ Quit: justdave: Read error: Connection reset by peer
[2011/10/23 07:01:16] @ justdave joined channel #puppet
[2011/10/23 07:02:17] @ JodieC joined channel #puppet
[2011/10/23 07:15:10] @ zorg1 joined channel #puppet
[2011/10/23 07:18:02] @ neonflux joined channel #puppet
[2011/10/23 07:25:53] @ Quit: zorg1: Remote host closed the connection
[2011/10/23 07:30:24] @ Quit: biomass: Quit: biomass
[2011/10/23 07:31:40] @ rcrowley joined channel #puppet
[2011/10/23 07:42:27] @ Quit: jmccune: Read error: Operation timed out
[2011/10/23 07:42:33] @ jmccune joined channel #puppet
[2011/10/23 07:49:44] @ blentz joined channel #puppet
[2011/10/23 07:55:15] @ Quit: Espen-_o: Quit: Connection closed for inactivity
[2011/10/23 07:56:48] @ Callek joined channel #puppet
[2011/10/23 07:57:40] <Callek> so... is it possible, with no connection to a puppet master, to "dump" the manifest (including ordering of dependancies) based on a particular node, in any means?
[2011/10/23 07:58:06] <Callek> I found a diff helper ruby library, but that seems to presume a fully setup puppet instance client and master
[2011/10/23 07:59:24] @ Quit: nigelk: Ping timeout: 252 seconds
[2011/10/23 07:59:26] <Callek> I'm in a state, where I'm a windows user, installed a Linux VM, since I need to figure out what a puppet setup is doing, basically and modify it to fit new needs. The sad part is there is currently no VM/OS space left in the VLAN that houses the puppet master, and IT Sec Manager won't allow my personal VM to connect :(
[2011/10/23 07:59:40] <Callek> I have the puppet manifests though, so any suggestion welcome.
[2011/10/23 08:00:05] @ nigelk joined channel #puppet
[2011/10/23 08:00:12] <omry> what about a vpn tunnel?
[2011/10/23 08:00:15] <omry> openvpn for example
[2011/10/23 08:00:15] @ Quit: jaytaph: Quit: jaytaph
[2011/10/23 08:01:11] <Callek> well I'm also not wanting to add a node, since the Hardware stack here is different than the Puppets already setup, if that makes sense (they do have Xorg.conf stuff setup that initiates stuff for a particular nVidea chipset this machine does not have)
[2011/10/23 08:01:54] <Callek> so I was basically looking to dump the existing manifest stuff (which I have locally, from a repository checkout) if possible, without having to create a separate local VM to act as a master to this one
[2011/10/23 08:02:10] <omry> I have no experience with this
[2011/10/23 08:02:18] <omry> if this is what you want to do, I cant really help
[2011/10/23 08:02:36] <Callek> and even if I do have a VM acting as a puppet master, I don't know how to dump the actual order of operations, as puppet sees them (dependency list, etc.)
[2011/10/23 08:02:49] <omry> but I am not sure why you want to do it in the first place. it sounds to me like your problem can be solved with a vpn
[2011/10/23 08:03:06] <Callek> I admit I am relatively new to puppet, not *great* at Networking and prior to now my only Linux experience has been over ssh
[2011/10/23 08:03:25] @ Elise001 joined channel #puppet
[2011/10/23 08:03:46] <omry> can you connect to the puppet master with ssh?
[2011/10/23 08:04:05] <Callek> omry: no, the puppet master is explicitly limited to the VMs that already use it
[2011/10/23 08:04:27] <omry> you are not allowed to use it?
[2011/10/23 08:04:42] <Callek> and in this environ perf of those machines is extremely strict (I'm not technically allowed to log in as long as they are in "production", they check if they should be in prod once every 24 hours, and only run puppet on boot)
[2011/10/23 08:05:11] <omry> if you have full access to the puppetmaster configuration, just replicate it
[2011/10/23 08:05:13] <Callek> basically the setup is archaic and intended to be moved
[2011/10/23 08:05:15] <omry> run your own puppetmaster
[2011/10/23 08:05:49] @ abetterlie joined channel #puppet
[2011/10/23 08:06:07] <Callek> I don't have the details of the puppetmaster VM at this point, so it would be trying to build a mirror of it from scratch
[2011/10/23 08:06:16] <omry> nah
[2011/10/23 08:06:18] <Callek> though I have the manifests, and the other details, just not the Host OS details
[2011/10/23 08:06:19] <Callek> :(
[2011/10/23 08:06:22] <omry> installing puppetmaster is pretty easy
[2011/10/23 08:06:37] <omry> you need the manifests, the modules and the files it serves
[2011/10/23 08:06:38] <Callek> but my real goal is to dump the manifest details for a particular node
[2011/10/23 08:06:49] <omry> good luck
[2011/10/23 08:06:52] <Callek> including dependancy chain.
[2011/10/23 08:07:21] <Volcane> the graph is written to the machine but there arent tools to dump it in that way
[2011/10/23 08:07:44] <omry> the client can probably be used to dump soemthing
[2011/10/23 08:08:08] <Callek> so, basically a member of an entriely different team, wants to expose most of these details of our internal setup to community so they can setup their own instances, and contribute back to us, but needs to do it outside puppet, and needs to vet the data being exposed
[2011/10/23 08:09:32] <omry> I think the right way to do this is to provide class files / modules that does the job
[2011/10/23 08:09:37] <Callek> ...well I did find: https://github.com/ripienaar/puppet-catalog-diff which is sort of what I was looking for
[2011/10/23 08:10:01] <Callek> yea, sadly the way the previous people setup the puppet manifests doesn't do it that way (its a convoluted mess)
[2011/10/23 08:10:10] <Callek> which is also why eyeballing it will SUCK
[2011/10/23 08:10:16] @ AphisOne joined channel #puppet
[2011/10/23 08:10:25] <Callek> (I could spend 4 days unwrapping the manifests by hand for a given node, but it will suck)
[2011/10/23 08:10:45] <omry> supporting users running partial dumps of that mess will be just as painful, or more
[2011/10/23 08:12:21] @ AphisOne left channel #puppet ()
[2011/10/23 08:12:45] <Callek> well my goal is to refactor this "mess" and learning how it actually ties together will likely help me be able to do that, while still allowing that 3rd party team person to begin looking at the details
[2011/10/23 08:12:59] @ devzero|afk is now known as devzero
[2011/10/23 08:13:18] <Callek> (it uses a _lot_ of implicit requirement crap too)
[2011/10/23 08:13:47] <Callek> just would be easier to wrap around my head if it was written with more forethought
[2011/10/23 08:14:09] <Callek> and was _really_ hoping there was some way to dump this stuff without having to setup a master
[2011/10/23 08:14:20] <Callek> since I can certainly setup another Linux VM on this w7 machine I'm currently on
[2011/10/23 08:14:43] <Callek> I just have absolutely no idea how to tie VM's together in terms of networking (I'm horrid at that side of the coin)
[2011/10/23 08:15:10] * Callek hopes he is at least making sense
[2011/10/23 08:15:24] <omry> sounds like you have a legacy system people are afraid to touch
[2011/10/23 08:15:30] <omry> and you want to start using it for more things
[2011/10/23 08:15:31] <Callek> basically
[2011/10/23 08:15:41] <omry> I suggest you just create a new master, which is yours
[2011/10/23 08:15:47] <Callek> I want to start using it for more things, including Q1 2011 is to upgrade it
[2011/10/23 08:15:54] <omry> and start messing with it
[2011/10/23 08:15:57] <Callek> to newer puppet version
[2011/10/23 08:15:57] <omry> get to know it
[2011/10/23 08:16:01] <omry> feel comfortable with it
[2011/10/23 08:16:16] <omry> you can use this opportunity to upgrade puppet
[2011/10/23 08:16:40] <omry> it will take time to clean things up, but you will need to do it anyway at some point
[2011/10/23 08:16:57] <Callek> well part of the requirement from "management" for allowing me to refactor these manifests is to _not_ upgrade the underlying puppet until things are more clear to everyone
[2011/10/23 08:16:58] <Callek> :/
[2011/10/23 08:17:31] <omry> thats fine, the important step is that you get control over the puppetmaster (or a copy of it)
[2011/10/23 08:17:32] <Callek> I initially did suggest dropping it ALL on the floor and starting from scratch with puppet 2.7
[2011/10/23 08:17:37] <omry> and get comforable with changing it
[2011/10/23 08:17:59] <omry> you dont need to drop it all to switch, you can certainly reuse things.
[2011/10/23 08:18:09] @ dnjaramba joined channel #puppet
[2011/10/23 08:18:40] <Callek> omry: I don't suppose you can get me an idea of how to configure, two CentOS5.0 machines, in a VirtualBox VM on a w7 to be able to talk to each other, so one can be a master the other a slave, for puppet
[2011/10/23 08:18:41] <Callek> ?
[2011/10/23 08:19:30] <omry> well, I dont know what w7 is. but for starters you can run your client on the master
[2011/10/23 08:19:30] <Callek> (p.s. I only chose Cent 5.0 due to that being what the current node VM I'm planning to upgrade legacy stuff off of)
[2011/10/23 08:19:36] <omry> so you only need one machine
[2011/10/23 08:19:51] @ gniks joined channel #puppet
[2011/10/23 08:19:51] <Callek> yea lots of legacy.
[2011/10/23 08:19:54] <omry> besides that - virtualbox supports vm networking
[2011/10/23 08:20:00] <omry> should be easy
[2011/10/23 08:21:02] <Callek> sure, should be, happy to know how (as I said, networking info is my weakest link, as in consider me a dumb user)
[2011/10/23 08:21:02] <Callek> w7 == windows 7
[2011/10/23 08:21:27] <omry> there is a gui for this in the vm settings for vbox
[2011/10/23 08:21:41] <omry> I think they should be able to communicate by default
[2011/10/23 08:21:42] <Callek> also would _rather_ keep client and master separate as I want the client VM at least to be as bare bones as necessary (since the puppet is run from a blank OS install + puppet installed/configured) in the first place.
[2011/10/23 08:21:45] <omry> if not play with the gui
[2011/10/23 08:22:35] <omry> if you get stuck you can ask at #vbox
[2011/10/23 08:22:35] <Callek> :/
[2011/10/23 08:22:47] @ duerF joined channel #puppet
[2011/10/23 08:23:02] <omry> one hint: dont rely on ping to test communication between vms.
[2011/10/23 08:23:03] <Callek> sounds like even if I get master setup I can't dump the info I want though
[2011/10/23 08:23:34] <Callek> so sounds more likely I'll be unraveling this by hand
[2011/10/23 08:23:34] <Callek> yea, ping is by-default blocked, iirc
[2011/10/23 08:23:52] <omry> last I checked icmp was not implemented
[2011/10/23 08:23:59] <omry> in any case, this is not related to puppet.
[2011/10/23 08:24:15] <omry> I sugges #vbox if you get stuck
[2011/10/23 08:24:24] @ Quit: gniks: Client Quit
[2011/10/23 08:24:39] @ MartVisser joined channel #puppet
[2011/10/23 08:24:40] <Callek> so is dumping the info I am looking for impossible even with a master setup?
[2011/10/23 08:25:11] <Callek> at least without some really deft programming in ruby
[2011/10/23 08:25:37] <omry> dunno
[2011/10/23 08:25:49] @ cr3 joined channel #puppet
[2011/10/23 08:25:57] <omry> and it sounds like a waste of time to me in any case
[2011/10/23 08:26:28] <Callek> I'm more looking for requirements I may not notice easily, or bogus ones
[2011/10/23 08:26:59] <Callek> to trim down the current manifests a bit before refactoring
[2011/10/23 08:27:03] @ jwhitlock_zzz is now known as jwhitlock
[2011/10/23 08:27:36] <Callek> since I got lucky a few times, and saw stupid stuff like git requiring hg
[2011/10/23 08:27:36] <Callek> (to use that as an example_
[2011/10/23 08:28:08] <Callek> and of course, git isn't being built/grabbed for any Hg repo
[2011/10/23 08:28:29] <cr3> if I have Class['foo'] -> Class['bar'] and class foo notifies service baz to be restarted, it looks like baz is only retarted at the end after class bar is called. so, how can I express dependencies so that baz is restarted before bar is called?
[2011/10/23 08:30:28] <Callek> do a Service['baz'] -> Class['bar'] should work fine, iirc
[2011/10/23 08:30:45] @ jaytaph joined channel #puppet
[2011/10/23 08:31:16] * Callek defers to regular puppet people if I am wrong though
[2011/10/23 08:31:43] <cr3> Callek: I'll give it a try unless a regular puppet person says otherwise :)
[2011/10/23 08:32:06] @ Quit: duerF: Quit: Leaving
[2011/10/23 08:36:48] <cr3> Callek: that seems to work and the dependency graph generated seems to look good based on my limited knowledge on the subject, thanks!
[2011/10/23 08:37:19] <Callek> np
[2011/10/23 08:37:47] <cr3> is there a convention for using single or double quotes in puppet?
[2011/10/23 08:37:53] @ devzero is now known as devzero|afk
[2011/10/23 08:43:32] @ Quit: pabelanger: Ping timeout: 276 seconds
[2011/10/23 08:46:12] <cr3> nevermind, found it: http://docs.puppetlabs.com/guides/style_guide
[2011/10/23 08:52:48] @ Quit: bodepd: Ping timeout: 240 seconds
[2011/10/23 08:53:30] @ bodepd joined channel #puppet
[2011/10/23 08:53:46] @ newtMcKerr joined channel #puppet
[2011/10/23 08:53:46] @ Quit: trams: Ping timeout: 244 seconds
[2011/10/23 08:56:52] @ trams joined channel #puppet
[2011/10/23 08:58:04] @ Quit: fivefish: Quit: ZNC - http://znc.in
[2011/10/23 08:58:08] @ fivefish joined channel #puppet
[2011/10/23 09:01:27] @ Quit: Callek: Quit: Page closed
[2011/10/23 09:04:01] @ mrwacky42 joined channel #puppet
[2011/10/23 09:09:17] @ Quit: Elise001:
[2011/10/23 09:13:08] @ Quit: jaytaph: Quit: jaytaph
[2011/10/23 09:19:59] @ biomass joined channel #puppet
[2011/10/23 09:23:28] @ Quit: marrusl: Quit: Ex-Chat
[2011/10/23 09:31:48] @ p3rror joined channel #puppet
[2011/10/23 09:36:57] @ hooper joined channel #puppet
[2011/10/23 09:41:07] @ soh joined channel #puppet
[2011/10/23 09:41:36] @ los324 joined channel #puppet
[2011/10/23 09:41:42] @ mogwai joined channel #puppet
[2011/10/23 09:43:22] @ los324 left channel #puppet ()
[2011/10/23 09:44:28] @ kenperkins joined channel #puppet
[2011/10/23 09:46:16] @ Quit: kenperkins: Client Quit
[2011/10/23 09:49:04] @ Quit: hooper: Read error: Connection reset by peer
[2011/10/23 09:50:50] @ Quit: drnexus__: Ping timeout: 258 seconds
[2011/10/23 09:52:09] @ Quit: biomass: Ping timeout: 244 seconds
[2011/10/23 09:56:31] @ sdog joined channel #puppet
[2011/10/23 09:56:49] @ Quit: sdog: Changing host
[2011/10/23 09:56:49] @ sdog joined channel #puppet
[2011/10/23 09:57:06] @ drnexus joined channel #puppet
[2011/10/23 09:57:09] @ Quit: gdbc: Read error: Connection reset by peer
[2011/10/23 10:01:00] @ Quit: mrwacky42: Quit: Leaving.
[2011/10/23 10:02:49] @ Quit: cr3: Quit: leaving
[2011/10/23 10:07:16] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/10/23 10:08:00] @ Quit: blentz: Ping timeout: 248 seconds
[2011/10/23 10:10:01] @ Quit: Jeff_S: Quit: brb
[2011/10/23 10:11:42] @ ccaum joined channel #puppet
[2011/10/23 10:20:00] @ Quit: neonflux: Read error: Connection reset by peer
[2011/10/23 10:20:14] @ neonflux joined channel #puppet
[2011/10/23 10:35:00] @ joe-mac joined channel #puppet
[2011/10/23 10:36:38] @ Quit: oholiab: Ping timeout: 245 seconds
[2011/10/23 10:43:25] @ oholiab joined channel #puppet
[2011/10/23 10:44:28] @ Quit: Maliuta: Read error: Operation timed out
[2011/10/23 10:44:47] @ Maliuta joined channel #puppet
[2011/10/23 10:45:53] <xeor_> Is there any way to have a different puppet-enc script per environment? Or any way for the puppet-enc script to somehow detect which environment is asked for... Environment variables or something?
[2011/10/23 10:49:29] @ dissipate joined channel #puppet
[2011/10/23 10:49:45] <Dominic> xeor_: I think the client's environment setting might be reported in the facts sent to the master. You can read from the facts yaml files in the master's $vardir/yaml/facts/${fqdn}.yaml.
[2011/10/23 10:49:50] @ Quit: dissipate: Read error: Connection reset by peer
[2011/10/23 10:50:40] <xeor_> Dominic: hmm, will look into that, thanks
[2011/10/23 10:50:50] @ Quit: abetterlie: Ping timeout: 258 seconds
[2011/10/23 10:52:15] @ Quit: trek1s: Ping timeout: 276 seconds
[2011/10/23 10:53:14] @ Quit: trams: Ping timeout: 252 seconds
[2011/10/23 10:58:37] @ Quit: samrose: Remote host closed the connection
[2011/10/23 10:58:48] @ Quit: pugnacity: Ping timeout: 244 seconds
[2011/10/23 11:00:51] @ kenperkins joined channel #puppet
[2011/10/23 11:02:40] @ pugnacity joined channel #puppet
[2011/10/23 11:02:40] @ devzero|afk is now known as devzero
[2011/10/23 11:03:30] @ kelseyhightower joined channel #puppet
[2011/10/23 11:06:27] @ Quit: dnjaramba: Ping timeout: 260 seconds
[2011/10/23 11:08:33] @ Quit: kenperkins: Quit: Computer has gone to sleep.
[2011/10/23 11:15:49] @ jab_doa joined channel #puppet
[2011/10/23 11:15:49] @ Quit: jab_doa: Changing host
[2011/10/23 11:15:49] @ jab_doa joined channel #puppet
[2011/10/23 11:15:50] @ Quit: sdog: Read error: Connection reset by peer
[2011/10/23 11:18:45] @ sdog joined channel #puppet
[2011/10/23 11:22:30] @ Quit: evelyette: Remote host closed the connection
[2011/10/23 11:28:25] @ jaytaph joined channel #puppet
[2011/10/23 11:29:16] @ nul0op joined channel #puppet
[2011/10/23 11:29:22] @ kenperkins joined channel #puppet
[2011/10/23 11:32:30] @ Quit: the_ramink: Ping timeout: 255 seconds
[2011/10/23 11:32:32] @ Quit: kenperkins: Client Quit
[2011/10/23 11:33:41] @ Quit: jaytaph: Quit: jaytaph
[2011/10/23 11:36:21] @ jaytaph joined channel #puppet
[2011/10/23 11:41:19] @ Quit: jaytaph: Quit: jaytaph
[2011/10/23 11:41:31] @ cosman2001 joined channel #puppet
[2011/10/23 11:42:51] @ Quit: cosman2001: Client Quit
[2011/10/23 11:44:56] @ abetterlie joined channel #puppet
[2011/10/23 11:52:08] @ jaytaph joined channel #puppet
[2011/10/23 11:55:06] @ Quit: tmz: Ping timeout: 260 seconds
[2011/10/23 11:55:06] @ the_ramink joined channel #puppet
[2011/10/23 12:04:04] @ Quit: jaytaph: Quit: jaytaph
[2011/10/23 12:06:23] @ lak joined channel #puppet
[2011/10/23 12:06:23] @ Quit: lak: Changing host
[2011/10/23 12:06:23] @ lak joined channel #puppet
[2011/10/23 12:11:56] @ kenperkins joined channel #puppet
[2011/10/23 12:15:33] @ Quit: kenperkins: Client Quit
[2011/10/23 12:17:33] @ patcon joined channel #puppet
[2011/10/23 12:17:51] @ Quit: Xeross: Ping timeout: 244 seconds
[2011/10/23 12:22:02] @ Xeross joined channel #puppet
[2011/10/23 12:26:54] @ Quit: ccaum: Quit: ccaum
[2011/10/23 12:27:24] @ jaytaph joined channel #puppet
[2011/10/23 12:27:25] @ Quit: newtMcKerr: Quit: Leaving.
[2011/10/23 12:29:22] @ Quit: Psi-Jack: Read error: Connection reset by peer
[2011/10/23 12:30:12] @ Psi-Jack joined channel #puppet
[2011/10/23 12:31:09] @ Quit: jaytaph: Client Quit
[2011/10/23 12:40:52] @ devzero is now known as devzero|afk
[2011/10/23 12:47:26] @ hooper joined channel #puppet
[2011/10/23 12:52:55] @ Quit: jonhattan: Ping timeout: 258 seconds
[2011/10/23 12:56:38] @ Quit: MartVisser: Quit: This computer has gone to sleep
[2011/10/23 12:57:38] @ Quit: pugnacity: Ping timeout: 244 seconds
[2011/10/23 13:00:37] @ pugnacity joined channel #puppet
[2011/10/23 13:01:57] @ jaytaph joined channel #puppet
[2011/10/23 13:04:51] @ Quit: hooper: Ping timeout: 260 seconds
[2011/10/23 13:08:15] @ jonhattan joined channel #puppet
[2011/10/23 13:09:37] @ steki-BLAH joined channel #puppet
[2011/10/23 13:09:38] @ Quit: steki-BLAH: Changing host
[2011/10/23 13:09:38] @ steki-BLAH joined channel #puppet
[2011/10/23 13:11:59] @ kenperkins joined channel #puppet
[2011/10/23 13:12:41] @ Quit: steki-BLAH: Remote host closed the connection
[2011/10/23 13:13:18] @ Quit: BManojlovic: Ping timeout: 255 seconds
[2011/10/23 13:18:03] @ rmckee joined channel #puppet
[2011/10/23 13:22:51] @ manlymatt83 joined channel #puppet
[2011/10/23 13:24:03] <manlymatt83> I have a problem - I want to use ssh_authorized_key, but that seems to require the user to exist on the system. Puppet is in charge of enabling LDAP auth on my server. But even if I do Require => exec["enable-auth"], that doesn't seem to work... almost like the users have to exist before puppet starts running?
[2011/10/23 13:26:01] @ trek1s joined channel #puppet
[2011/10/23 13:27:36] <kelseyhightower> manlymatt83: so you are enabling auth for every run?
[2011/10/23 13:34:57] @ tilgovi joined channel #puppet
[2011/10/23 13:34:57] @ Quit: tilgovi: Changing host
[2011/10/23 13:34:57] @ tilgovi joined channel #puppet
[2011/10/23 13:37:38] @ newtMcKerr joined channel #puppet
[2011/10/23 13:37:48] <manlymatt83> kelseyhightower: no
[2011/10/23 13:38:01] <manlymatt83> kelseyhightower: that exec has a notifyonly
[2011/10/23 13:41:36] @ Quit: jaytaph: Read error: Connection reset by peer
[2011/10/23 13:42:09] @ cyflea joined channel #puppet
[2011/10/23 13:43:09] <kelseyhightower> manlymatt83: ok
[2011/10/23 13:43:24] <kelseyhightower> manlymatt83: so you manage your users in LDAP
[2011/10/23 13:43:36] <kelseyhightower> and you want to user puppet to manage their keys
[2011/10/23 13:44:04] <kelseyhightower> user = use
[2011/10/23 13:46:46] @ Quit: lak: Quit: Leaving.
[2011/10/23 13:47:52] @ bmahe joined channel #puppet
[2011/10/23 13:47:52] @ Quit: bmahe: Changing host
[2011/10/23 13:47:53] @ bmahe joined channel #puppet
[2011/10/23 13:49:06] @ Moe__ joined channel #puppet
[2011/10/23 13:50:28] <manlymatt83> yes kelseyhightower
[2011/10/23 13:50:35] @ jaytaph joined channel #puppet
[2011/10/23 13:50:49] <manlymatt83> I really wish there was a way to tell puppet "ignore the fact that this user doesn't exist, just push the key anyway"
[2011/10/23 13:50:52] <manlymatt83> but I can't do that.
[2011/10/23 13:50:58] <kelseyhightower> manlymatt83: your are right
[2011/10/23 13:51:14] <manlymatt83> it dies out with: err: Could not apply complete catalog: User 'manlymatt83' does not exist
[2011/10/23 13:51:32] <kelseyhightower> manlymatt83: there was some work to change that behavior, but I would recommend using a file type
[2011/10/23 13:51:56] <manlymatt83> the only way I'm thinking of is doing a Require[ Exec["check-users-enabled"] ] and have that do a test -f
[2011/10/23 13:52:03] <manlymatt83> what do you mean work to change the behavior? Is it in the new version?
[2011/10/23 13:52:08] <kelseyhightower> manlymatt83: the current ssh_authorized_type auto requires a user
[2011/10/23 13:52:18] <Moe__> I drew a blank...what is the command to register a new puppet client
[2011/10/23 13:52:21] <kelseyhightower> manlymatt83: let me find the ticket
[2011/10/23 13:52:31] <Moe__> puppet ... --test
[2011/10/23 13:52:36] <kelseyhightower> Moe__: puppet agent --test
[2011/10/23 13:52:50] <Moe__> ah
[2011/10/23 13:52:52] <Moe__> thanks
[2011/10/23 13:52:54] <kelseyhightower> Moe__: you will still need to sign the cert on the master
[2011/10/23 13:53:03] <kelseyhightower> puppet cert sign list --all
[2011/10/23 13:53:09] <kelseyhightower> puppet cert list --all
[2011/10/23 13:53:19] <kelseyhightower> puppet cert sign <hostname>
[2011/10/23 13:53:48] <Moe__> kelseyhightower: I got that thanks I just could not remember that first command thanks for the memory boost :0
[2011/10/23 13:53:59] <kelseyhightower> Moe__: cool
[2011/10/23 13:54:41] <henk> There is no command to register a new client from the client perspective. It will just request a cert if it doesn’t have one.
[2011/10/23 13:55:40] <manlymatt83> kelseyhightower: OK, thanks. So you're saying use File and just make the contents the key?
[2011/10/23 13:55:50] <kelseyhightower> manlymatt83: yep
[2011/10/23 13:55:58] <kelseyhightower> #8986
[2011/10/23 13:56:12] <kelseyhightower> manlymatt83: http://projects.puppetlabs.com/issues/8986
[2011/10/23 13:57:14] <kelseyhightower> ^^ in that ticket you can see the discussion around the ssh_authorized_key type. So in your case it may be easier to use a file resource (and the required directories)
[2011/10/23 14:01:21] <manlymatt83> kelseyhightower: Yup, you're right. So I'll put everyone's keys in /etc/ssh/authorized_keys/$user/authorized_keys, and just create those files using File[] types?
[2011/10/23 14:01:24] <manlymatt83> that makes sense, ya?
[2011/10/23 14:02:23] <kelseyhightower> manlymatt83: yep, just be sure to update sshd_config to look there
[2011/10/23 14:02:30] @ Quit: nimrod10: Ping timeout: 258 seconds
[2011/10/23 14:02:51] <kelseyhightower> manlymatt83: Test that and see if that solves your problem, I will be around here for a bit.
[2011/10/23 14:05:18] <manlymatt83> Yeah, I will. I'm actually wondering - will user => and group => work with LDAP users?
[2011/10/23 14:05:40] <manlymatt83> So if I do file .... and then do user => "manlymatt83", group => "users", and those are both ldap users/groups, will puppet barf?
[2011/10/23 14:07:22] @ Jbstlmo joined channel #puppet
[2011/10/23 14:07:55] <kelseyhightower> manlymatt83: My guess is that should work as I have used puppet in that configuration before
[2011/10/23 14:08:50] @ sumsuch joined channel #puppet
[2011/10/23 14:09:29] @ nimrod10 joined channel #puppet
[2011/10/23 14:10:24] <manlymatt83> kelseyhightower: OK, I'll take a look. My hope is that puppet just uses a chown command behind the scenes or something, which of course the system would accept since those users exist.
[2011/10/23 14:10:30] <manlymatt83> will test this.
[2011/10/23 14:10:42] <kelseyhightower> manlymatt83: yep
[2011/10/23 14:11:13] <henk> Now, the ssh_authorized_key type appears to try to write the file as the user.
[2011/10/23 14:11:54] <kelseyhightower> henk: you are correct, there is an auto require on the user going on
[2011/10/23 14:12:18] <kelseyhightower> henk: this ticket should shed some light for you: http://projects.puppetlabs.com/issues/8986
[2011/10/23 14:13:13] <kelseyhightower> henk: it maybe easier to use a file resource instead.
[2011/10/23 14:13:32] <henk> I know, I’m reading it und wondered where manlymatt83 hope came from, but perhaps I still have to get to that part :)
[2011/10/23 14:14:09] <kelseyhightower> henk: oh sorry
[2011/10/23 14:14:15] @ kolla joined channel #puppet
[2011/10/23 14:14:58] <henk> kelseyhightower: No reason to be, you are very helpful AFAICT and linked to a bug which might be good to know about
[2011/10/23 14:16:08] @ Quit: kenperkins: Quit: Computer has gone to sleep.
[2011/10/23 14:16:30] @ Quit: jaymurtii: Remote host closed the connection
[2011/10/23 14:23:52] @ ghaskins joined channel #puppet
[2011/10/23 14:24:29] @ Quit: ghaskins: Client Quit
[2011/10/23 14:25:25] @ Quit: Xeross: Ping timeout: 240 seconds
[2011/10/23 14:27:44] @ Holocaine joined channel #puppet
[2011/10/23 14:28:22] @ blentz joined channel #puppet
[2011/10/23 14:28:34] @ Quit: jaytaph: Ping timeout: 244 seconds
[2011/10/23 14:28:50] @ Quit: blentz: Read error: Connection reset by peer
[2011/10/23 14:29:09] @ blentz joined channel #puppet
[2011/10/23 14:33:26] @ Quit: jonhattan: Quit: Ex-Chat
[2011/10/23 14:33:54] @ Quit: Jbstlmo: Ping timeout: 265 seconds
[2011/10/23 14:34:08] @ Quit: sumsuch: Quit: sumsuch
[2011/10/23 14:34:53] @ j00bar joined channel #puppet
[2011/10/23 14:40:32] @ Quit: Spads: Ping timeout: 248 seconds
[2011/10/23 14:40:50] <jpalmer> Does Ken Barber hang out in here?
[2011/10/23 14:40:56] @ Xeross joined channel #puppet
[2011/10/23 14:49:37] <stahnma> jpalmer: I think so
[2011/10/23 14:51:32] <jpalmer> I've been working with him on a ticket (I've contributed a patch, and he wanted some additional unit tests and such) and I thought it'd be easier to talk to him here, but.. I don't see an obvious nick
[2011/10/23 14:54:57] <ashp> I wonder if the AWS instance would be better to use at work than the official RHEL instance, hmm
[2011/10/23 14:58:30] @ jbstlmo joined channel #puppet
[2011/10/23 14:58:42] @ sumsuch joined channel #puppet
[2011/10/23 15:00:27] <manlymatt83> Is there a group equivalent to id?
[2011/10/23 15:02:50] @ themadcanudist joined channel #puppet
[2011/10/23 15:05:40] @ Psi-Jack_ joined channel #puppet
[2011/10/23 15:08:11] <stahnma> jpalmer: it's normally ken_barber
[2011/10/23 15:08:12] <stahnma> ;)
[2011/10/23 15:08:20] @ Quit: Psi-Jack: Ping timeout: 276 seconds
[2011/10/23 15:08:20] @ Psi-Jack_ is now known as Psi-Jack
[2011/10/23 15:08:21] <jpalmer> ahh, danke.
[2011/10/23 15:09:03] @ dgillies joined channel #puppet
[2011/10/23 15:09:36] @ sticky is now known as sticky[8]
[2011/10/23 15:11:59] @ sticky[8] is now known as sticky
[2011/10/23 15:12:38] <manlymatt83> OK, so here's another question. Is there a way to pass exec{} a variable that it can use dynamically in it's command? I wish I could do something like: file { "blah": owner => manlymatt83, group => manlymatt83, onlyif => "id manlymatt83" }
[2011/10/23 15:12:45] <manlymatt83> but there's no way to do that. I can't do a dynamic onlyif like that.
[2011/10/23 15:13:01] <manlymatt83> the problem I'm having now is that LDAP may not have all the same users that I have keys for
[2011/10/23 15:13:19] <manlymatt83> And if they don't, puppet freaks out that it can't chown the key file to that user/group (this is a different problem)
[2011/10/23 15:16:56] @ johnston joined channel #puppet
[2011/10/23 15:17:25] <kjetilho> manlymatt83: I don't understand...
[2011/10/23 15:17:51] <kjetilho> oh, now I see
[2011/10/23 15:17:56] @ Peanut left channel #puppet ()
[2011/10/23 15:17:58] @ Quit: d3c: Quit: leaving
[2011/10/23 15:18:48] <ashp> http://puppetlabs.com/blog/rapid-scaling-with-auto-generated-amis-using-puppet/ holy shit this rules
[2011/10/23 15:19:06] <kjetilho> manlymatt83: you can do an exec of chown/chgrp like that?
[2011/10/23 15:19:47] <kjetilho> but Puppet really don't want to have state which varies with what the local state is already
[2011/10/23 15:21:17] @ Quit: devdas: Remote host closed the connection
[2011/10/23 15:24:02] @ MartVisser joined channel #puppet
[2011/10/23 15:25:37] @ jaymurtii joined channel #puppet
[2011/10/23 15:27:43] @ soh joined channel #puppet
[2011/10/23 15:27:48] @ mosen joined channel #puppet
[2011/10/23 15:29:23] <manlymatt83> kjetilho: No, just an exec that verifies the user exists
[2011/10/23 15:29:30] <manlymatt83> I need puppet to only run chown if the user exists on the box :)
[2011/10/23 15:29:49] <manlymatt83> maybe I should just remove the ssh key stuff from puppet all together and just make a cron that runs as root, that way I can store my keys in LDAP instead of making them part of the repository
[2011/10/23 15:30:24] @ trek1s is now known as trek1s_zZzZ
[2011/10/23 15:32:49] <manlymatt83> How big can a puppetVar be? I wonder if instead, I should just set all the authorized_key info inside a puppetVar, so puppet can take it in, de-code it (perhaps from base 64), and then install it into files.
[2011/10/23 15:32:54] <manlymatt83> But I still think a cron might be best for this stuff....
[2011/10/23 15:38:06] @ jaytaph joined channel #puppet
[2011/10/23 15:38:16] @ jaytaph left channel #puppet ()
[2011/10/23 15:40:37] <kjetilho> manlymatt83: exec { "/bin/chown foo foo": onlyif => "/bin/id foo" } should work
[2011/10/23 15:40:47] <kjetilho> it will run chown every time, though
[2011/10/23 15:41:26] <kjetilho> so you need to make the onlyif more complex if you care about that (chown(1) changes ctime even if the ownership doesn't really change, so it may be a bit of a pain)
[2011/10/23 15:41:46] @ Quit: j00bar: Quit: j00bar
[2011/10/23 15:45:18] <nigelk> I wonder how this got through... http://groups.google.com/group/puppet-users/browse_frm/thread/509fdab71b5d061b
[2011/10/23 15:46:22] @ Quit: trek1s_zZzZ: Read error: Operation timed out
[2011/10/23 15:49:58] @ Quit: jnewland: Quit: Connection closed for inactivity
[2011/10/23 15:50:51] @ Kazuhiro joined channel #puppet
[2011/10/23 15:51:02] <whack> nigelk: haha
[2011/10/23 15:53:50] <nigelk> :) going to go ban it...
[2011/10/23 15:54:02] <jbstlmo> anyone else unable to get to the puppet website?
[2011/10/23 15:55:02] <mosen> yeah wsod :)
[2011/10/23 15:55:07] <mig5> yeah.
[2011/10/23 15:55:25] <nigelk> jbstlmo: puppetlabs.com ?
[2011/10/23 15:55:35] <nigelk> huh
[2011/10/23 15:55:47] @ Quit: sumsuch: Quit: sumsuch
[2011/10/23 15:56:22] <jbstlmo> nigelk: yes, puppetlabs.com is blank
[2011/10/23 15:57:04] <nigelk> ok, alerting web folks
[2011/10/23 15:57:06] <nigelk> thanks guys!
[2011/10/23 15:57:42] <nigelk> jbstlmo: looks like it's just the front page...
[2011/10/23 15:58:04] @ Quit: themadcanudist: Ping timeout: 260 seconds
[2011/10/23 16:00:54] @ silentg joined channel #puppet
[2011/10/23 16:03:15] <whack> nigelk: 301 redirect from there with no Location: header
[2011/10/23 16:03:41] <nigelk> whack: fixed
[2011/10/23 16:03:51] <rektide> is there any way i can have a symlink created via reading the fs?
[2011/10/23 16:04:11] <rektide> i think puppet is trying to resolve my symlinks if i point a "source" at them
[2011/10/23 16:04:18] <whack> nigelk: woo
[2011/10/23 16:04:30] <whack> rektide: I don't follow
[2011/10/23 16:04:34] <whack> what is the end result you want?
[2011/10/23 16:04:44] <rektide> i'd really like to define symlinks on the FS, in the modules files/ directory, and have them copied from there verbatim onto my target system
[2011/10/23 16:05:13] <whack> why not just do file { "path...": ensure => link, target => /symlink/path/target; } ?
[2011/10/23 16:05:27] @ sumsuch joined channel #puppet
[2011/10/23 16:05:38] <nigelk> rektide: links => manage *should* work
[2011/10/23 16:05:39] <nigelk> http://docs.puppetlabs.com/references/stable/type.html#file
[2011/10/23 16:05:43] <rektide> whack: in that form i can not just browse the files/ directory to understand the structure
[2011/10/23 16:05:45] <nigelk> if you want to copy the links as-is from the master
[2011/10/23 16:06:15] <rektide> whack: i have to check both my files/ and all my manifests to inventory & know what's going to get created, which is not ideal for my particular tastes.
[2011/10/23 16:06:23] <rektide> nigelk: i will investigate, that looks super solid, many thanks
[2011/10/23 16:06:29] @ Quit: sumsuch: Client Quit
[2011/10/23 16:06:33] <nigelk> there may be a bug with it, I have a vague memory of that
[2011/10/23 16:06:46] <nigelk> but I would add that if you find yourself doing lots of filesystem/symlink management in puppet...
[2011/10/23 16:06:54] <nigelk> it's worth taking a step back to think about other approaches
[2011/10/23 16:07:13] <whack> rektide: I suppose that's a good point, but I don't have stuff in files/ mirror the exact location on the target servers anyway
[2011/10/23 16:08:13] <rektide> k guys, super thanks for the speedy replies. i'll see how it goes. :)
[2011/10/23 16:08:25] @ Quit: amoe: Ping timeout: 258 seconds
[2011/10/23 16:11:54] @ Quit: Xeross: Ping timeout: 244 seconds
[2011/10/23 16:12:36] @ Quit: MartVisser: Quit: Leaving
[2011/10/23 16:13:39] <manlymatt83> kjetilho: What do you think of my idea of making a cron push the keys to /etc/ssh/authorized_keys/<user>/authorized_keys? That way I can let puppet off the hook for dealing with something that is really run by LDAP.
[2011/10/23 16:13:55] @ Xeross joined channel #puppet
[2011/10/23 16:14:26] @ Quit: Kazuhiro: Quit: Leaving.
[2011/10/23 16:15:06] @ Kazuhiro joined channel #puppet
[2011/10/23 16:17:29] @ Quit: DJClean: Read error: Operation timed out
[2011/10/23 16:17:52] @ Quit: jodok: Quit: jodok
[2011/10/23 16:18:46] @ cwebber-mobile joined channel #puppet
[2011/10/23 16:18:48] @ jaytaph joined channel #puppet
[2011/10/23 16:19:30] <cwebber-mobile> anyone have a manifest that needs to cause a reboot?
[2011/10/23 16:19:54] <cwebber-mobile> in this case I am installing the openvz kernel
[2011/10/23 16:20:26] @ Quit: bmahe: Remote host closed the connection
[2011/10/23 16:20:54] @ amoe joined channel #puppet
[2011/10/23 16:21:26] <whack> cwebber-mobile: hmm, I suppose you could notify an exec that reboots
[2011/10/23 16:21:59] <whack> though probably want to schedule the reboot instead of doing it immediately, mostly so it happens after the puppet run completes
[2011/10/23 16:22:12] <cwebber-mobile> whack: exactly
[2011/10/23 16:22:15] @ cwebber-mobile is now known as cwebber
[2011/10/23 16:22:24] @ Quit: blast_hardcheese: Quit: ZNC - http://znc.sourceforge.net
[2011/10/23 16:23:07] <cwebber> the real problem is how do you illustrate that requirement in the graph
[2011/10/23 16:23:22] <cwebber> maybe via stages...
[2011/10/23 16:24:24] <whack> cwebber: what requirement?
[2011/10/23 16:24:32] <whack> notify => Exec["please reboot kthx"]
[2011/10/23 16:24:37] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/10/23 16:25:07] <cwebber> err
[2011/10/23 16:25:13] <whack> exec { "please reboot kthx": command => "shutdown -r 3m", refreshonly => true; }
[2011/10/23 16:26:08] <cwebber> more so, resource { require => Package[openvz-kernel] } but there needs to be a reboot to make that make sense
[2011/10/23 16:26:16] @ Kazuhiro left channel #puppet ()
[2011/10/23 16:27:43] <whack> cwebber: I see now
[2011/10/23 16:27:57] <whack> you could make an exec that checks "Are we on the right kernel?"
[2011/10/23 16:28:12] <cwebber> hmmmm
[2011/10/23 16:28:16] <cwebber> or even a fact
[2011/10/23 16:28:31] <whack> and things that need 'good kernel' require that exec, or conditionalize things based on facts like $kernelrelease
[2011/10/23 16:29:43] <cwebber> whack: thanks
[2011/10/23 16:32:50] <manlymatt83> Hi folks. So I'm having a bit of trouble with my client's request for ssh key support. I use LDAP authentication and puppet for configuration file management. Everything works great, except puppet doesn't have great support for LDAP users, so the public key via puppet method isn't working too well. I could either: a) hack puppet, b) patch openssh to look in LDAP for ssh keys, or c) write a cron that simply searches ldap for public keys and places th
[2011/10/23 16:36:07] @ los324 joined channel #puppet
[2011/10/23 16:36:09] <manlymatt83> I'd like to have my keys in LDAP. But if I do, there's no easy way to use puppet to push them. And I'd have to patch openssh to get it to look at LDAP for the keys. So instead, I'm thinking of using puppet to launch a cron on each server that runs every 5 minutes and installs public keys.
[2011/10/23 16:36:13] @ los324 left channel #puppet ()
[2011/10/23 16:40:16] <fpletz> manlymatt83: Hmm, what about a script that queries ldap and emulates authorized_keys files with fifos it's listening on? :)
[2011/10/23 16:43:25] <whack> manlymatt83: you could write a custom function that pulls ssh keys
[2011/10/23 16:44:02] <whack> ssh_authorized_key { "someuser": key => ldap_ssh_key("someuser"); }
[2011/10/23 16:44:55] <nigelk> manlymatt83: if you can't get ssh_authorized_keys to work with local home directories for LDAP accounts, please bug repot it
[2011/10/23 16:45:15] <nigelk> that at least should work.
[2011/10/23 16:45:24] <nigelk> but if you need to pull the keys from LDAP, I would either:
[2011/10/23 16:45:32] <nigelk> a) write a cron job on puppet masters to pull them all down to files regularly.
[2011/10/23 16:45:42] <nigelk> b) write a function for each one to pull it as per whack's suggestion
[2011/10/23 16:45:48] @ Quit: Moe__: Remote host closed the connection
[2011/10/23 16:46:23] <whack> in past jobs we sync'd them with cron
[2011/10/23 16:46:41] <whack> if you want to be puppety about it, you could just sync them to your puppet master and have puppet pull them like normal
[2011/10/23 16:47:04] @ Quit: tilgovi: Ping timeout: 260 seconds
[2011/10/23 16:47:42] <nigelk> I reckon sync with cron will scale better
[2011/10/23 16:47:54] <nigelk> I like separating out potential problems with ldap load from potential problems with puppet load
[2011/10/23 16:51:03] @ Quit: nul0op: Quit: Leaving
[2011/10/23 16:53:19] <whack> nod
[2011/10/23 16:56:45] <cwebber> are there any good "you been using archaic puppet for too long" presos that would bring me up to speed? looking at http://bit.ly/nedMLy and the Stages stuff, I cant tell exactly where that class { foo: stage => 'first' } stuff would go
[2011/10/23 16:57:23] <cwebber> i am thinking in the node definition
[2011/10/23 16:58:20] <whack> cwebber: depends where you want to specify it. The 'class { foo: ...} ' is the 'parameterized classes' feature
[2011/10/23 16:58:32] <cwebber> ahhh
[2011/10/23 16:59:01] @ Quit: jaytaph: Ping timeout: 260 seconds
[2011/10/23 16:59:19] <cwebber> and it looks like the ENC stuff doesnt support parameterized classes until 2.6.5
[2011/10/23 17:00:03] @ j^2 joined channel #puppet
[2011/10/23 17:02:02] @ Moe__ joined channel #puppet
[2011/10/23 17:02:36] <Moe__> hello all
[2011/10/23 17:02:51] <Moe__> has the address yum repo for puppet changed?
[2011/10/23 17:05:25] @ jaytaph joined channel #puppet
[2011/10/23 17:05:33] @ tmz joined channel #puppet
[2011/10/23 17:05:34] @ Quit: tmz: Changing host
[2011/10/23 17:05:34] @ tmz joined channel #puppet
[2011/10/23 17:05:35] @ jaytaph left channel #puppet ()
[2011/10/23 17:07:33] @ noakes joined channel #puppet
[2011/10/23 17:08:43] @ blast_hardcheese joined channel #puppet
[2011/10/23 17:15:11] @ Quit: jab_doa: Quit: Verlassend
[2011/10/23 17:15:17] <Moe__> has the address yum repo for puppet changed?
[2011/10/23 17:17:19] @ jwhitlock is now known as jwhitlock_zzz
[2011/10/23 17:24:41] @ mhiggs joined channel #puppet
[2011/10/23 17:27:21] @ ScrotusMaximus90 joined channel #puppet
[2011/10/23 17:27:28] <ScrotusMaximus90> holy puppet!
[2011/10/23 17:27:53] <whack> you've been out too long in the midnight sea
[2011/10/23 17:28:23] <ScrotusMaximus90> Didn't realize there was such a huge puppet cannel out here on free node. amazin
[2011/10/23 17:31:20] * mosen waits to see if he's talking about puppetry or puppet
[2011/10/23 17:32:00] <Moe__> lol ^
[2011/10/23 17:32:25] <ScrotusMaximus90> Oh mostly ventriloquism
[2011/10/23 17:32:35] <ScrotusMaximus90> No I am talking about PUPPET written in RUBIES
[2011/10/23 17:32:38] <ScrotusMaximus90> sheesh.
[2011/10/23 17:33:44] @ Quit: Av: Ping timeout: 260 seconds
[2011/10/23 17:35:06] @ Quit: Moe__: Remote host closed the connection
[2011/10/23 17:36:10] @ Quit: newtMcKerr: Quit: Leaving.
[2011/10/23 17:36:41] @ Quit: ScrotusMaximus90: Remote host closed the connection
[2011/10/23 17:43:16] @ tnguyen joined channel #puppet
[2011/10/23 17:43:41] @ teatimer joined channel #puppet
[2011/10/23 17:43:50] @ Quit: varadero:
[2011/10/23 17:47:44] @ Quit: nokia3510: Ping timeout: 248 seconds
[2011/10/23 17:48:49] <manlymatt83> fpletz, whack, nigelk: thanks. I think I'll go the cron on each server route. Unless there'd be an easy way to just put the cron on the puppet master or write the custom function. Could I have a custom function call a shell script or something and return data? Still, I don't want to hard-code users in my puppet repo. My SSH script would fetch them and put them in place without having hard coded users.
[2011/10/23 17:53:22] @ KaiserBeto1 joined channel #puppet
[2011/10/23 17:54:01] @ Quit: johnston: Ping timeout: 265 seconds
[2011/10/23 17:55:08] @ Quit: neoice: Read error: Operation timed out
[2011/10/23 17:58:14] @ cosman2001 joined channel #puppet
[2011/10/23 17:59:15] @ hooper joined channel #puppet
[2011/10/23 18:03:21] @ newtMcKerr joined channel #puppet
[2011/10/23 18:04:21] @ Quit: newtMcKerr: Client Quit
[2011/10/23 18:07:18] @ manlymatt83 left channel #puppet ()
[2011/10/23 18:09:30] @ soh joined channel #puppet
[2011/10/23 18:11:25] @ Quit: soh: Client Quit
[2011/10/23 18:13:04] @ soh joined channel #puppet
[2011/10/23 18:15:21] @ Quit: soh: Client Quit
[2011/10/23 18:15:28] @ Quit: tnguyen: Remote host closed the connection
[2011/10/23 18:17:13] @ Kazuhiro joined channel #puppet
[2011/10/23 18:39:17] @ Enf joined channel #puppet
[2011/10/23 18:41:57] @ nokia3510 joined channel #puppet
[2011/10/23 18:47:24] @ Quit: p3rror: Remote host closed the connection
[2011/10/23 18:49:22] @ newtMcKerr joined channel #puppet
[2011/10/23 18:50:13] @ edward_ joined channel #puppet
[2011/10/23 18:50:25] <edward_> is there a variable I can use to split output of a file?
[2011/10/23 18:50:58] @ Quit: rcrowley: Quit: Computer has gone to sleep.
[2011/10/23 18:51:03] @ Kazuhiro left channel #puppet ()
[2011/10/23 18:51:03] <edward_> so say <% if filename =~ /40-conf/ then -%> <% elsif filename =~ /50-conf/ then -%>
[2011/10/23 18:59:07] @ kelseyhightower left channel #puppet ()
[2011/10/23 18:59:23] @ robbyt joined channel #puppet
[2011/10/23 19:03:13] @ gniks joined channel #puppet
[2011/10/23 19:03:31] @ lovemj joined channel #puppet
[2011/10/23 19:08:09] @ Quit: dgillies: Remote host closed the connection
[2011/10/23 19:08:14] @ Quit: nb: Quit: ZNC - http://znc.in
[2011/10/23 19:09:00] @ realityloop joined channel #puppet
[2011/10/23 19:13:10] @ nb joined channel #puppet
[2011/10/23 19:14:05] @ ed209 left channel #puppet ()
[2011/10/23 19:16:39] @ Quit: jbstlmo: Quit: Page closed
[2011/10/23 19:30:40] @ Scottydsntknw joined channel #puppet
[2011/10/23 19:31:57] <nigelk> edward_: if you're in Ruby already, you have everything Ruby can do :)
[2011/10/23 19:32:25] <edward_> does ruby know what file it's currently in based on the manifest definition is the question I guess?
[2011/10/23 19:32:32] <nigelk> ah.
[2011/10/23 19:32:41] <edward_> our init.pp directs the file to be called different things depending on the hostgroup
[2011/10/23 19:33:06] <edward_> I'm working on a better way of doing it though, as the requirement changed on me :P
[2011/10/23 19:33:12] <nigelk> yeah, this feels backwards...
[2011/10/23 19:33:21] @ soh joined channel #puppet
[2011/10/23 19:33:31] <edward_> best to do it right first time :)
[2011/10/23 19:33:33] <edward_> thanks though
[2011/10/23 19:34:00] <nigelk> so it sounds like logic in init.pp modifies the name of the file in a resource that you're also delivering as a template? and you want inside that template to work out what the name of the file is?
[2011/10/23 19:34:36] <nigelk> (where "the file" is the path of the resource, not the manifest file this all comes from)
[2011/10/23 19:35:44] @ tilgovi joined channel #puppet
[2011/10/23 19:35:44] @ Quit: tilgovi: Changing host
[2011/10/23 19:35:44] @ tilgovi joined channel #puppet
[2011/10/23 19:36:20] @ Quit: choonming: Quit: Leaving
[2011/10/23 19:41:03] <robbyt> edward_: why not deploy files like file{'/etc/filename': content => template('$hostgroup'),} and then do file{"/etc/$hostgroup": ensure => ''/etc/filename,}
[2011/10/23 19:41:44] <robbyt> edward_: if I understand your question...
[2011/10/23 19:43:17] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/10/23 19:44:16] <edward_> robbyt: doesn't quite get what I need. I've changed the way the files are being read though, which is a much cleaner solution
[2011/10/23 19:44:19] <edward_> thanks though :)
[2011/10/23 19:56:35] <nigelk> edward_: remember in templates you have access to every variable from puppet scope...
[2011/10/23 19:56:40] @ Quit: blentz: Quit: Ex-Chat
[2011/10/23 20:01:54] @ ccaum joined channel #puppet
[2011/10/23 20:07:32] @ choonming joined channel #puppet
[2011/10/23 20:09:20] @ soh joined channel #puppet
[2011/10/23 20:11:31] @ bobra joined channel #puppet
[2011/10/23 20:12:48] @ Quit: tilgovi: Read error: Connection reset by peer
[2011/10/23 20:14:38] @ Kevin_Ding joined channel #puppet
[2011/10/23 20:17:19] @ dgillies joined channel #puppet
[2011/10/23 20:18:48] @ Quit: noakes: Quit: noakes
[2011/10/23 20:20:42] @ ghaskins joined channel #puppet
[2011/10/23 20:23:08] @ Quit: Kevin_Ding: Ping timeout: 253 seconds
[2011/10/23 20:26:51] @ Quit: Scottydsntknw: Quit: HydraIRC -> http://www.hydrairc.com <- \o/
[2011/10/23 20:31:36] @ Quit: jaymurtii: Ping timeout: 255 seconds
[2011/10/23 20:34:07] @ dissipate joined channel #puppet
[2011/10/23 20:35:07] <dissipate> anyone have a link to a good tutorial on getting puppet set up on amazon web services? (e.g. what security groups to create)
[2011/10/23 20:36:57] @ warbjoh joined channel #puppet
[2011/10/23 20:38:44] @ Quit: Holocaine:
[2011/10/23 20:41:14] @ Holocaine joined channel #puppet
[2011/10/23 20:51:01] @ druggo joined channel #puppet
[2011/10/23 20:55:43] @ Quit: DSpair: Quit: Leaving
[2011/10/23 20:57:44] @ Quit: newtMcKerr: Quit: Leaving.
[2011/10/23 21:00:45] @ cwebber is now known as cwebber-1
[2011/10/23 21:01:05] @ cwebber-1 is now known as cwebber
[2011/10/23 21:01:44] @ cwebber is now known as cwebber-1
[2011/10/23 21:01:56] @ cwebber-1 is now known as cwebber-mon
[2011/10/23 21:02:18] @ cwebber-mon is now known as cwebber-1
[2011/10/23 21:02:35] @ cwebber-1 is now known as cwebber
[2011/10/23 21:02:54] @ AshishH joined channel #puppet
[2011/10/23 21:06:49] @ bmahe joined channel #puppet
[2011/10/23 21:06:50] @ Quit: bmahe: Changing host
[2011/10/23 21:06:50] @ bmahe joined channel #puppet
[2011/10/23 21:06:59] @ mukulm joined channel #puppet
[2011/10/23 21:08:56] <mukulm> hi, please have a look at http://pastebin.com/jApHzRAm as my script is running again as the client get the update from the server ?
[2011/10/23 21:09:09] @ larstobi1 joined channel #puppet
[2011/10/23 21:09:40] <mukulm> i had applied onlyif but it seems to be not working as a conditional statement ?
[2011/10/23 21:10:37] @ Quit: neonflux: Ping timeout: 252 seconds
[2011/10/23 21:10:37] @ Quit: lohapuk: Read error: Operation timed out
[2011/10/23 21:10:38] @ Quit: ichilton: Write error: Broken pipe
[2011/10/23 21:10:38] @ Quit: gebi: Read error: Operation timed out
[2011/10/23 21:10:39] @ Quit: jY: Ping timeout: 252 seconds
[2011/10/23 21:10:40] @ Quit: tmz: Read error: Operation timed out
[2011/10/23 21:10:41] @ Quit: mogwai: Read error: Operation timed out
[2011/10/23 21:10:42] @ Quit: marcoecc: Read error: Connection reset by peer
[2011/10/23 21:10:47] @ jY- joined channel #puppet
[2011/10/23 21:10:50] @ tmz joined channel #puppet
[2011/10/23 21:10:50] @ Quit: tmz: Changing host
[2011/10/23 21:10:50] @ tmz joined channel #puppet
[2011/10/23 21:10:54] @ Quit: larstobi: Read error: Connection reset by peer
[2011/10/23 21:10:56] @ mogwai joined channel #puppet
[2011/10/23 21:11:07] @ neonflux joined channel #puppet
[2011/10/23 21:11:26] @ dissipate_ joined channel #puppet
[2011/10/23 21:11:28] @ Quit: dissipate: Read error: Connection reset by peer
[2011/10/23 21:11:29] @ Quit: semiosis: Read error: Connection reset by peer
[2011/10/23 21:11:33] @ Quit: rocket: Ping timeout: 252 seconds
[2011/10/23 21:11:34] @ Quit: dcsobralWork: Ping timeout: 252 seconds
[2011/10/23 21:11:36] @ semiosis joined channel #puppet
[2011/10/23 21:11:37] @ Quit: garethr: Max SendQ exceeded
[2011/10/23 21:11:47] @ rocket joined channel #puppet
[2011/10/23 21:11:49] @ dcsobralWork joined channel #puppet
[2011/10/23 21:11:54] @ garethr joined channel #puppet
[2011/10/23 21:12:47] <mukulm> please have a look at http://pastebin.com/jApHzRAm
[2011/10/23 21:12:49] @ ichilton joined channel #puppet
[2011/10/23 21:12:49] @ lohapuk joined channel #puppet
[2011/10/23 21:12:58] @ gebi joined channel #puppet
[2011/10/23 21:13:01] @ newtMcKerr joined channel #puppet
[2011/10/23 21:14:49] @ Quit: dissipate_: Client Quit
[2011/10/23 21:14:52] @ Quit: newtMcKerr: Client Quit
[2011/10/23 21:14:55] @ dissipate_ joined channel #puppet
[2011/10/23 21:17:44] @ Quit: neonflux: Quit: Leaving.
[2011/10/23 21:18:21] @ Quit: dissipate_: Client Quit
[2011/10/23 21:19:07] @ Quit: mosen: Ping timeout: 258 seconds
[2011/10/23 21:19:37] <mukulm> can any one help me in correcting my code as mentioned ?above
[2011/10/23 21:29:58] <rodjek> mukulm: if you run the command in onlyif manually, does it return 0?
[2011/10/23 21:33:11] @ newtMcKerr joined channel #puppet
[2011/10/23 21:33:44] @ Quit: newtMcKerr: Client Quit
[2011/10/23 21:35:23] @ Quit: limed: Ping timeout: 252 seconds
[2011/10/23 21:37:04] <mukulm> yes its returninig 0 but still then my script is running again
[2011/10/23 21:38:05] <rodjek> that's because the onlyif parameter tells the exec to run if it returns 0
[2011/10/23 21:38:21] <rodjek> you probably want unless
[2011/10/23 21:38:50] <rodjek> http://docs.puppetlabs.com/references/2.7.0/type.html#exec
[2011/10/23 21:43:05] @ Quit: robbyt: Quit: robbyt
[2011/10/23 21:44:47] <mukulm> ok thank you
[2011/10/23 21:48:06] @ Quit: hooper: Ping timeout: 255 seconds
[2011/10/23 21:50:12] @ Quit: rmckee: Quit: Leaving
[2011/10/23 21:51:15] <mukulm> but instead of returning a value to 0, the script should run once ?
[2011/10/23 21:53:28] @ Quit: mhiggs: Quit: Leaving.
[2011/10/23 21:54:54] @ limed joined channel #puppet
[2011/10/23 21:55:20] @ Quit: sdog: Ping timeout: 258 seconds
[2011/10/23 21:58:13] @ Quit: ccaum: Quit: ccaum
[2011/10/23 21:59:00] @ limed_ joined channel #puppet
[2011/10/23 22:03:24] @ Quit: mukulm: Ping timeout: 265 seconds
[2011/10/23 22:07:32] <Rince> Good morning
[2011/10/23 22:20:00] @ mukulm joined channel #puppet
[2011/10/23 22:20:29] @ Quit: shr3kst3r: Read error: Operation timed out
[2011/10/23 22:26:15] @ Quit: limed_: Ping timeout: 260 seconds
[2011/10/23 22:28:11] @ Quit: gniks: Quit: Leaving.
[2011/10/23 22:33:30] <tessier> root 20851 10.9 18.8 1032732 791948 ? Ssl Oct20 610:31 /usr/bin/python -E /usr/sbin/setroubleshootd
[2011/10/23 22:33:34] <tessier> er...
[2011/10/23 22:33:42] <tessier> root 19271 22.6 24.6 1135152 1035992 ? Ssl 16:00 89:07 /usr/bin/ruby /usr/sbin/puppetd
[2011/10/23 22:33:47] @ Quit: cwebber: Remote host closed the connection
[2011/10/23 22:33:52] <tessier> But setroubleshootd is pretty egregious in memory usage too.
[2011/10/23 22:37:17] @ Quit: soh: Quit: Computer has gone to sleep.
[2011/10/23 22:42:12] @ Meier joined channel #puppet
[2011/10/23 22:42:33] @ sdog joined channel #puppet
[2011/10/23 22:44:35] @ p3rror joined channel #puppet
[2011/10/23 22:46:01] @ jaydg joined channel #puppet
[2011/10/23 22:47:05] @ Quit: sdog: Ping timeout: 258 seconds
[2011/10/23 22:47:12] @ shr3kst3r joined channel #puppet
[2011/10/23 22:47:56] @ tal joined channel #puppet
[2011/10/23 22:49:53] @ Quit: Holocaine:
[2011/10/23 22:50:51] @ limed_ joined channel #puppet
[2011/10/23 22:51:48] @ riha joined channel #puppet
[2011/10/23 22:52:43] <tal> is it possible to use a conditional to determine node inheritance? e.g.: node '/my-regex-node/' if $somefacter == 'foo' inherits foo-node else inherits bar-node {} - pls forgive my incorrect syntax
[2011/10/23 22:53:41] @ tal_ joined channel #puppet
[2011/10/23 22:56:55] @ saiju joined channel #puppet
[2011/10/23 22:57:32] @ Quit: tal: Ping timeout: 265 seconds
[2011/10/23 22:58:30] @ Quit: mukulm: Ping timeout: 265 seconds
[2011/10/23 23:00:57] @ artkrz joined channel #puppet
[2011/10/23 23:04:51] @ Quit: realityloop: Remote host closed the connection
[2011/10/23 23:09:29] @ Quit: pdtpatrick1: Remote host closed the connection
[2011/10/23 23:10:43] @ Quit: saiju: Quit: Leaving
[2011/10/23 23:11:08] @ Quit: semiosis: Remote host closed the connection
[2011/10/23 23:11:13] @ rcrowley joined channel #puppet
[2011/10/23 23:11:36] @ Quit: warbjoh: Quit: ChatZilla 0.9.87 [Firefox 3.6.12/20101026210630]
[2011/10/23 23:11:37] @ devdas joined channel #puppet
[2011/10/23 23:14:38] @ mfladischer joined channel #puppet
[2011/10/23 23:19:28] @ toi joined channel #puppet
[2011/10/23 23:20:08] @ shug joined channel #puppet
[2011/10/23 23:20:16] @ Quit: tal_: Ping timeout: 265 seconds
[2011/10/23 23:22:44] @ swati joined channel #puppet
[2011/10/23 23:23:47] <swati> Hi, I am running puppet master and client on suse 10.1 and puppet version 2.7.5. When I run the puppet client daemon it gives me error
[2011/10/23 23:24:02] <swati> "err: Could not request certificate: Cannot save <hostname>.tspinternal; parent directory /etc/puppet/ssl/private_keys + does not exist"
[2011/10/23 23:24:08] <swati> Please suggest what to do....
[2011/10/23 23:24:13] @ Quit: tim|mac: Ping timeout: 240 seconds
[2011/10/23 23:24:17] @ EL84 joined channel #puppet
[2011/10/23 23:24:54] <beddari> swati: has it been working, ever?
[2011/10/23 23:25:09] @ ckauhaus joined channel #puppet
[2011/10/23 23:25:48] @ fladi_ joined channel #puppet
[2011/10/23 23:25:56] <swati> No, this is a fresh install and the first time itself it gives me error
[2011/10/23 23:26:23] @ Quit: mfladischer: Read error: Operation timed out
[2011/10/23 23:26:25] <beddari> swati: that dir exists and is writable? installed from gems or packages?
[2011/10/23 23:27:25] <swati> I installed from source tar gz, that directory exists and is owned by puppet user and root group and is writable by both
[2011/10/23 23:30:02] @ Quit: Azrial: Ping timeout: 244 seconds
[2011/10/23 23:30:36] @ Azrial joined channel #puppet
[2011/10/23 23:30:39] @ tim|mac joined channel #puppet
[2011/10/23 23:31:38] @ DJClean joined channel #puppet
[2011/10/23 23:31:51] @ Quit: kolla: Remote host closed the connection
[2011/10/23 23:31:59] @ saiju joined channel #puppet
[2011/10/23 23:34:12] @ gniks joined channel #puppet
[2011/10/23 23:35:28] @ Quit: gniks: Client Quit
[2011/10/23 23:37:52] @ gniks joined channel #puppet
[2011/10/23 23:42:07] @ jodok joined channel #puppet
[2011/10/23 23:42:08] @ Quit: jodok: Changing host
[2011/10/23 23:42:08] @ jodok joined channel #puppet
[2011/10/23 23:46:30] @ Quit: gniks: Quit: Leaving.
[2011/10/23 23:47:01] @ lovemj1 joined channel #puppet
[2011/10/23 23:47:14] @ Quit: lovemj: Read error: Connection reset by peer
[2011/10/23 23:49:06] @ marcus_ joined channel #puppet
[2011/10/23 23:49:13] @ bronto joined channel #puppet
[2011/10/23 23:50:42] <marcus_> my dashboard spool is some kind of flooding: /usr/share/puppet-dashboard/spool (4 Gib). shouldn't the files be deleted after they have been processed ?
[2011/10/23 23:51:28] @ MattyM joined channel #puppet
[2011/10/23 23:52:37] @ Quit: Stenbryggen: Quit: Leaving.
[2011/10/23 23:52:43] @ cosman2001 left channel #puppet ()
[2011/10/23 23:54:53] @ Quit: dgillies: Quit: Eaten by a grue
[2011/10/23 23:55:49] <kesor> do you have your delayed_jobs running?
[2011/10/23 23:56:21] @ csanchez joined channel #puppet
[2011/10/23 23:56:27] <kesor> i had a problem where redhat+puppet2.6.6 was sending reports that caused mysql errors, and the delayed_jobs died from that every time.

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