Ask Your Question

Revision history [back]

click to hide/show revision 1
initial version

My replies are getting pretty long winded, so I'll turn this into a dedicated answer.

You said you already have a configuration set up with a master and two agents. That's a great start.

Previously linked trainings page

For your next step, I would suggest some of the following options.

1) Learn about module design. The puppetlabs docs are really great on this topic. One gotcha is containment and anchoring, That bit me a few times until I really started to understand how it worked.

2) Get involved with a module that you use/plan on using from the puppetforge. Read the codebase. Read the tests. Read the issues/pull requests.

3) Start learning about mcollective. mcollective seems to be the industry standard for cluster orchestration and management for systems using puppet. I would look at the mcollective-vagrant project. Be sure to look though the puppet modules used to craft the Virtualbox VM. They are a great example of a common pattern for module layout and design. This same style recommendation is reflected in the official puppet docs linked above.

4) Lots of great examples on rnelson0's blog. I would recommend learning about the r10k based workflow, integrating rspec-puppet tests into your modules, and integrating hiera for seperating your data from your code, as well as the roles and profiles pattern.

5) For more of the same from #3 but with a fun perspective to read, check out Shit Gary Says The talks on the workflow, and using hiera contain lots of great information.

My replies are getting pretty long winded, so I'll turn this into a dedicated answer.

You said you already have a configuration set up with a master and two agents. That's a great start.

Previously linked trainings page

https://puppetlabs.com/blog/quest-learn-puppet-new-learning-vm?_ga=1.165506300.2122898636.1436269010 and http://learn.puppetlabs.com/ are great places to start.

For your next step, I would suggest some of the following options.

1) Learn about module design. The puppetlabs docs are really great on this topic. One gotcha is containment and anchoring, That bit me a few times until I really started to understand how it worked.

2) Get involved with a module that you use/plan on using from the puppetforge. Read the codebase. Read the tests. Read the issues/pull requests.

3) Start learning about mcollective. mcollective seems to be the industry standard for cluster orchestration and management for systems using puppet. I would look at the mcollective-vagrant project. Be sure to look though the puppet modules used to craft the Virtualbox VM. They are a great example of a common pattern for module layout and design. This same style recommendation is reflected in the official puppet docs linked above.

4) Lots of great examples on rnelson0's blog. I would recommend learning about the r10k based workflow, integrating rspec-puppet tests into your modules, and integrating hiera for seperating your data from your code, as well as the roles and profiles pattern.

5) For more of the same from #3 but with a fun perspective to read, check out Shit Gary Says The talks on the workflow, and using hiera contain lots of great information.