similar to: Managing manifests

This is in reference to David Schmitt''s writeup [1] I like that writeup a lot; I just had a few comments/clarifications: * Namespaces: For now, we just use a flat namespace for modules, i.e. you can''t have modules within modules, and you always know that in a fully qualified name ''foo::bar::baz'' that ''foo'' is the module name and

I''m running into a very odd occurrence. I started with a new puppet server and was able to add a few classes, however as I try and add new ones I get odd parsing error messages, the most common being Could not match ''}'' at whatever file. Here is the exact error I get err: Could not retrieve catalog: Could not parse for environment production: Could not match

Hi, in an attempt to simplify my configuration, and from a suggestion on IRC, I was thinking of using multiple values for file { source => [] }. Was just currious if anyone had any comments on this. Basically, I setup my file-server to use: <snip> [private] path /var/lib/puppet/files/private/%H allow * [shared] path /var/lib/puppet/files/shared allow *

I constantly get told I should just create a DSL in Ruby, instead of having a separate, custom language. This has, unsurprisingly, come up multiple times here at OSCON. So, I decided I''d see if I could hack something up and see how it looked. This is the first thing I could come up with in terms of the basic functionality: #!/usr/bin/ruby class Base file "/etc/passwd",

I just upgraded to 0.22.0 from 0.18.4 and I''m running into what I think are problems due to changes in scope rules. In my site.pp manifest, I have the following: import "classes/*" node nfsserver { include nfsserver } node webserver { include webserver } class cluster1_sites { website { "example.com": ...; "example2.com": ...; } } node

Hello folks, I''ve blended/updated the Glossary of Terms (<http://reductivelabs.com/trac/puppet/wiki/GlossaryOfTerms>). It seems the definition of "manifests" is a bit different amongst members. Some people consider a "manifest" to be a configuration file written in puppet, e.g. any .pp file. But at my organization, we''ve considered the totality

I posted this earlier on IRC, but it didn''t seem anyone was active: <mozillaman>Hey all! <mozillaman>Quick question for you :) <mozillaman>I installed puppet according to the directions on the site and got this: sudo /usr/bin/puppetmasterd --nonodes undefined method `retrieve'' for nil:NilClass <mozillaman>Anyone have any ideas? Any help would be

Hey folks, I''ve just uploaded an apache2 module[1] to our Google Code repository[2]. It''s based largely on the recipe available on the Puppet trac[3], but with a fair amount of modification and generalization... Basically, it provides abstractions for config file snippets, site definitions, and apxs modules -- and then "basic" and "debian"

I am new to Puppet and very eager to apply it to a project. But I am somewhat stymied by the learning curve. So far I''ve found many very simple examples of how to modify a file or add a user and a very complex example http://reductivelabs.com/trac/puppet/wiki/CompleteConfiguration . I have not been able to find any other examples of a total configuration tree (ie /etc/puppet/*)

Hello everyone, A long time ago, I posted the Stanford Best Practices and I''ve gone through and updated it today. I''d like to have people go through it and see if we can strip out some Stanford specific stuff and tag this as an official best practice. I think an official best practice will be important as more and more people consider making shareable modules, etc (mostly

Hello >From my understanding, modules are not autosearched. So, if I have: include ssh ... then puppet will look for a module names ssh and class named ssh defined in /modules/ssh/init.pp Moreover, if I have: include ssh::global ... then Puppet will for class global defined in /modules/ssh/init.pp or /modules/ssh/global.ss. Is that correct? Or am I leaving out some imports

Hi, I want to validate my puppet manifests before I check them into source control. I just want to make sure that the manifest syntaxes are legal, not to do any more complex testing - the checkin may be on my laptop and not on the system that the manifests may be intended to run on. Ideally I''d be able to do this from ruby code - and I did try to instantiate some classes and override

In my puppet client I have puppet.conf defined puppet server as mypuppet server = mypuppet.example.net Not sure why the puppet client puppet-test is still sending these noises to the syslog Jun 10 13:36:23 puppet-test puppetd[10863]: [ID 702911 daemon.error] Could not find server : getaddrinfo: node name or service name not known Jun 10 13:36:23 puppet-test puppetd[10863]: [ID 702911

Hi there, I''m running puppetmasterd manually while testing, and I was trying to split my manifest in several parts using import. If I do so whenever I run the puppetmasterd the manifest gets loaded and served correctly: cognos ~ # puppetmasterd -v info: Starting server for Puppet version 0.20.0 info: Parsed manifest in 0.02 seconds info: /var/lib/puppet/files: allowing *.example.org

Dear puppet wizards, http://docs.puppetlabs.com/puppet/3/reference/lang_defined_types.html does not suggest to me that nested parametrized defines would be forbidden, hence I expected the following code to work: add_to_ssh_authorized_keys.pp: -------- define a::ssh_keys::add_to_ssh_authorized_keys($homebasedir="/home", $targetuser="root", $keyowner, $keyownername,

Hello folks, At Stanford, we use modules to define every possible service, package, feature in the traditional sense (apache, ntp, iptables, etc). Then, for clients (such as other departments) and "services" (such as our mailman infrastructure, our ldap infrastructure, etc), we use manifests in /master/manifests and distributable files in /dist/ and templates in /master/templates.

Hello, I want a simple operation in a puppet node like restarting the ssh service if it was stopped. My site.pp is simple as this: import "services/*" node default { include ssh } The services directory as a ssh.pp : class ssh { service { ssh: ensure => running, subscribe => File["/etc/ssh/sshd_config"] } } I''ve stopped the ssh service in the

Hello all. Probably this question has already been posed in the context of another software, but I am unable to find it. I would like to implement a module which should manage munin-clients as well as the munin-server. Here is what I am thinking about implementing: - There are two classes: munin::client and munin::server - munin::clients'' configuration will be managed automatically.

Hi all, I''ve been turning my head upside down to find a solution on the following problem: I have a samba fileserver and need to make puppet do: 1. Create groups 2. Create directories with group rights 3. Create an smb-shares.conf from a template that can be included in the main smb.conf. To do this I need to have a combination of sharename and group _and_ create the

Hi all, I have three parameterized classes, named mq, mq_gsk and apache. The mq class always calls the mq_gsk class, but the apache class should only call the mq_gsk class when the mq class is not used in a hosts'' node definition Here is my configuration class mq ( $gsk_version = undef ) { require mq::params < ..... > class { ''mq_gsk'':