similar to: puppet thinks yum fails when it doesn't

Quick and dirty: --- lib/facter.rb (revision 203) +++ lib/facter.rb (working copy) @@ -989,6 +989,18 @@ %x{/usr/sbin/scutil --get LocalHostName} end end + Facter.add(:interfaces) do + confine :kernel => :linux + setcode do + %x{/sbin/ifconfig -a -s}.split($/)[1..-1].collect {|line|

For anyone that has experienced the odd End of file or Cannot describe errors, we''ve found that switching to Mongrel has fixed this problem for us. We''re currently running 5 instance of puppetmaster under mongrel (with the apache proxy in front) and things are going great. For more information on setting up Mongrel, visit:

--On Monday, May 14, 2007 9:16 PM -0700 Jos Backus <jos@catnook.com> wrote: > ObPuppet: we ramping up our deployment this week to around 200 hosts. So > far everything has been going smoothly. We don''t have nearly as many and yet, we see occasional errors ("End of file reached") which seems to indicate the network cutting out. I can''t imagine what would

Just curious as to the functionality of puppet. Does Puppet ensure that a service is up and running as long as puppet is running? Ie, I want to make sure ssh is always running, if for some reason ssh get''s shut down, does puppet start it back up when it does it''s config sync run? Thanks! --------------------------------- Pinpoint customers who are looking for

This is particularly nasty, since it''s on the puppetmaster itself: [root@puppet tmp]# rpm -q yum puppet puppet-server yum-3.0.5-1.el5.centos.5 puppet-0.24.0-2.el5 puppet-server-0.24.0-2.el5 [root@puppet tmp]# cat /etc/issue CentOS release 5 (Final) Here''s the output of puppetd --debug: notice: Starting catalog run debug: Loaded state in 0.01 seconds debug: Prefetching rpm

Is there a way to specify on a per-object basis that Puppet should merely report that an object needs to be updated without actually performing the update? This would make it possible to detect changes to critical objects (e.g. config files) that Puppet shouldn''t try to fix automaticaly. -- Jos Backus jos at catnook.com

I''m sure everyone has read and committed the Puppet Best Practice to memory by now (I joke). One of the things I''ve written in there deals with trailing commas, which I adopted from the way I used to do multiline in Perl, but I''ve noticed that most people don''t tend to do that in Puppet. Should I revise this or should we all start using the trailing commas?

FYI: We have a custom `checkout'' type which stopped working when migrating from 0.22.4 to 0.23.2-ish. This turned out to have been caused by the location of property.rb changing between 0.22.4 (puppet/type/property.rb) and 0.23.2-ish (puppet/property.rb). (I say 0.23.2-is because I''m using 40491ebe7ca9692b57fb533412ece8fb694b7d4c since it only has a few extra bugfixes over

I''d like to keep the desired and current states of a machine regarding it''s configuration. Sometimes a chain of administrative commands is needed to get a special thing (re)configured on a machine, e.g. reconfig of a suncluster. This cannot be done with puppet in an easy way. In the majority of cases, these commands change appropriate files, which reflect the current

Hi, I was wondering if there is any of getting the gem package provider to run non-interactively? When upgrading mongrel it fails because it gets prompted for which version to install: Attempting remote update of mongrel Select which gem to install for your platform (x86_64-linux) 1. mongrel 1.1.1 (ruby) 2. mongrel 1.1.1 (jruby) 3. mongrel 1.1.1 (mswin32) 4. mongrel 1.1 (mswin32) 5.

Morning all I''m working on converting some of my NetApp providers to prefetch/flush style to try and optimize performance. I''ve hit an issue on my Netapp_user provider, around handling resource creation versus resource modification? What''s the easiest way to differentiate? Current code is here:

Hi all, I''ve just rewritten the Puppet introduction from scratch: https://reductivelabs.com/trac/puppet/wiki/PuppetIntroduction I think it''s better, but I''m not convinced it''s a whole heckuva lot better. Comments are very much appreciated, especially if you can recommend a better approach to the document. I tried to organize it how I have been giving

Hi, We''re trying to follow the instructions at http://reductivelabs.com/trac/puppet/wiki/UsingMongrel (Just using a single puppetmaster/mongrel instance and plain mod_proxy for now, instead of balancer one) We got past the "Server is not a class" error by modifying mongrel.rb We also kind-of got past the: /opt/bin/puppetmasterd:293: undefined method `daemonize'' for

This patch updates the files under popt/ to the latest vendor drop. The only change is the inclusion of a FreeBSD-specific patch to popt.c. This is needed in case somebody decides to build rsync on that platform without using the port. I'm not happy about the wording in popt/README.rsync so I may change it. The patch is available at http://www.catnook.com/rsync-popt-1.5.1.patch Comments

Hi all; I am sure this question has been brought up before, but I could not find any information in the FAQ or docs etc. Quick question; Is there a way of syncing multiple servers simultaneously from 1 source server? Thanks, -- Vaibhav Goel extension 6347

Hi, I have an application which is dying horrible deaths (i.e. segmentation faults) in mid-flight, in production... And of course, I should fix it. But while I find and fix the bugs, I found something I think should be different - I can work on submitting a patch, as it is quite simple, but I might be losing something on my rationale. When Mongrel segfaults, it does not -obviously- get to clean

Perhaps now is a good time to commit this patch which updates the included popt to version 1.7. This has been tested on FreeBSD and Solaris. http://www.catnook.com/patches/rsync-popt-1.7.diff Comments? -- Jos Backus _/ _/_/_/ Sunnyvale, CA _/ _/ _/ _/ _/_/_/ _/ _/ _/

Hello, we have one puppet server managing approx servers, all FreeBSD. The problem is that the puppet clients all hangs, one by one, eventually. They will run for anything between an hour up to a week. But eventually the all hang. We really like puppet and want to keep using it. But the way it is now is completely unusable. We''ve tried many versions, from 24.x, and now 2.6.1rc1. But

One particular use of Puppet would be to serve as a machine fact/configuration retrieval tool. So in order to produce a list of packages with version-release info installed on each machine something along the lines of the following code could be run: require ''puppet'' Puppet::Type.type(:package).defaultprovider.list.each do |package| puts

Hi Luke and other people working on 0.24. I hope this is constructive. So, I have our configuration working against a checkout from git as of last night. Subject to (sorry, here''s the big but :) * external_nodes script is not being called (#951) * Custom types aren''t working unless I drop them into $rubysitedir/puppet/{type,provider}. Previously they worked in