In this guide I will show you how to connect Puppet Enterprise with your Control Repository using  GitHub. 

Why is it good to connect Puppet Enterprise with a Control Repository?

Ideally you want to connect Puppet Enterprise with a code repository as this will allow to keep all your files and folders that you are working on together in one place.

That way, you can continue to collaborate whilst applying best practices such as managing  change management, versioning, pull requests, etc. In this code control repository you can store your code and data to deploy your environments, just as you would with your usual repositories, such as GitHub, BitBucket, and/or GitLab.

• Private key: /etc/puppetlabs/puppetserver/ssh/id-control_repo.rsa
• Public key: /etc/puppetlabs/puppetserver/ssh/id-control_repo.rsa.pub
  
  

Step 2 - Install Git

We want to install Git on our Puppet Master (if it is not already installed on my Puppet Master). We will need this later on.

For Centos, you can install Git using the following command. 

$ yum install git -y

For Ubuntu, you can install Git using the following command. 

$ sudo apt-get git -y

We can then set our initial GitHub settings:

$ git init
$ git config --global user.name "Your Name"
$ git config --global user.email your.email@address.com

Step 3 - Create a repo by cloning the Puppet Control Repo template

We want to create a repository (repo) and will be leveraging the template repo made available by Puppet, the Puppet Control Repo. 

We now want to create a repository called control-repo in your Github user account or organization. So we will clone the Puppet Control Repo template for this. 

On the Puppet Control Repo on Github, you will see something like the following:

Click on Use this template. 

Give it an appropriate name such as "control-repo", etc.

Then click on Create repository from template. 

Then make a note of your new repository URL (you may see a HTTPS or SSH repository link. You can refer to Github's remote URL guide if you are unsure). 

In my example, it is: 

https://github.com/vulongtran/control-repo 

Step 4 - Clone this GitHub repo to our Puppet Master machine

Let's clone this new GitHub repo to our Puppet Master machine. 

To be organised, we can now create a folder called "code" or similar, so we can place our Github files into it.

$ cd ~
$ mkdir code
$ cd code

Then we can clone the GitHub repo to our Puppet Master Machine.

$ git clone https://github.com/vulongtran/control-repo
$ cd control-repo

Then we want to ensure that the repository as your default source is our GitHub repo (Else it may refer to the Puppet Control Repo template). 

$ git remote -v

$ git remote remove origin

$ git remote add origin https://github.com/vulongtran/control-repo

Then we want to check that this is set up correctly. As git remote -v will list the URL of the remote repo.

$ git remote -v
origin https://github.com/vulongtran/control-repo (fetch)
origin https://github.com/vulongtran/control-repo (push)

Step 5 - Connect to our Github repository from our Puppet Master machine

When you are ready, we can try connect to our Github repo by making a push request to GitHub. We are doing this to test connectivity only since we will have no active changes or updates to push.

$ git push origin master

It will ask you for a password after this. So depending on your set up, you may need to create a Github Personal Access Token to be able to log in via command line.

This is a screenshot of the Github personal access token that I later used as the Github password for this step.  

Once successful, you should see something like this with the message "Everything up-to-date". 

Perfect, you have connectivity with our Puppet Master machine and Github!

Step 6 - Set up Code Manager to engage with your GitHub repository

We want to set up Code Manager in the Puppet Enterprise Console to engage with our GitHub repository.

The reason we want to use the Puppet Code Manager is that it automates the management and deployment of our Puppet files and folders and keep it synchronised. 

For example, when a member of your team pushes code updates to your GitHub repo, then Puppet Code Manager will syncs the code to your Puppet Master, which in turn will ensure that all your servers will start running new code at the same time. 

So in your Puppet Enterprise console, go to the puppet_enterprise::profile::master class in your Puppet Enterprise (PE) Master node group. 

This is located under Inventory> Node groups> PE Master> Classes.

Inventory> Node groups

PE Master

Classes

Class: puppet_enterprise::profile::master

So now we need to find the puppet_enterprise::profile::master class in the list of classes or add it into the PE Master node group if it is not there. 

Once we have the puppet_enterprise::profile::master class, we can start adding parameters  to it. 

We want to set these fields:

• r10k_remote - The location for your GitHub repository.
• r10k_private_key - The path to the file containing the private key used to access the applicable GitHub repositories. Note this is required when using the SSH protocol; optional in all other cases.
• code_manager_auto_configure - Specifies whether to autoconfigure Code Manager.

In my example:

• r10k_remote = "'git@github.com:vulongtran/control-repo.git"
• r10k_private_key = '/etc/puppetlabs/puppetserver/ssh/id-control_repo.rsa'
• code_manager_auto_configure = true

Step 7 - Set up authentication for Code Manager

Step 8 - Test our Code Manager is working properly

Now we have the right authentication, we can test our Code Manager is working properly and is connected to our control repo. 

On your Puppet Master machine, we can run the following command to check this:

$ puppet-code deploy --dry-run