Puppet Enterprise 2016.2


The following instructions are for installing Puppet agents on nodes running *nix operating systems. For Windows instructions, see the Windows agent installation quick start guide.

A computer running the PE agent is usually referred to as an “agent”. The Puppet agent regularly pulls configuration catalogs from a Puppet master and applies them to the local system.

For more information about Puppet agents, as well as the components that make up your PE deployment, see the installation overview.

Tip: If you don’t have internet connectivity, refer to Installing agents in a Puppet Enterprise infrastructure without internet access to choose a suitable method.

Prerequisites: This guide assumes you’ve installed a monolithic PE deployment and have the Puppet master, the PE console, and PuppetDB up and running on one node.

Installing the Puppet Enterprise agent

Choose one of the following scenarios:

Installing agents with the same OS and architecture as the Puppet master

If your Puppet agent node is the same OS and architecture as the Puppet master, log into your agent node and run the following command: curl -k https://<MASTER HOSTNAME>:8140/packages/current/install.bash | sudo bash.

This script detects the OS on which the agent is running, sets up an apt, yum, or zipper repo that refers back to the Puppet master, and then pulls down and installs the puppet-agent packages.

Important: After installation is complete, continue on to Approving the certificate request to approve the new agent’s certificate request in the console.

Installing agents with different OS and architecture than the Puppet master

If your Puppet master OS and architecture are different from the agent, follow this example, changing the options to match your agent’s OS and architecture. This example describes adding an agent running Debian 6 on AMD64 hardware.

To install an agent with a different OS than the Puppet master, you first add the appropriate class for the repo that contains the agent packages. You then classify the PE Master node group with that class. Last, you run a cURL command from the agent to the master to retrieve the necessary packages to install the agent.

Note: If your Puppet master uses a proxy server to access the internet, refer to this known issue for a workaround.

  1. In the console, click Nodes > Classification.

  2. On the Classification page, click the PE Master group.

  3. Click the Classes tab.

  4. In the Add new class field, in the Class name box, begin typing pe_repo, and select the class pe_repo::platform::debian_6_amd64, or whichever class matches your agent, from the list of classes.

    Note: The repo classes are listed as pe_repo::platform::<agent_os_version_architecture>

  5. Click Add class.

  6. Click Commit 1 change.

    Note: The class you selected now appears in the list of classes for the PE Master group, but it has not yet been configured on your nodes. For that to happen, you need to kick off a Puppet run.

  7. From the command line on the Puppet master, run puppet agent -t.

    This configures the Puppet master node using the newly-assigned class.

    The new repo is created in /opt/puppetlabs/server/data/packages/public.

  8. SSH into the node where you want to install the agent, and run curl -k https://<master.example.com>:8140/packages/current/install.bash | sudo bash.

    You can optionally replace current in the script with a specific PE version number, in the form of 2015.x.x.

Note: Depending on your platform, the method for downloading the script may vary.

The installer installs and configures the Puppet Enterprise agent.

Important: After installation is complete, continue on to Approving the certificate request to approve the new agent’s certificate request in the console.

Approving the certificate request

During installation, the agent node contacts the Puppet master and requests a certificate. To add the node to the console and to start managing its configuration, approve its request on the Puppet master. This is most easily done in the console:

  1. Load a list of currently pending node requests by clicking Nodes > Unsigned certificates.

  2. Click the Accept All button to approve all the requests and add the nodes.

The Puppet agents can now retrieve configurations from the master the next time Puppet runs.

Testing the agent nodes

During this walk-through, you’ll run the Puppet agent manually. By default, the agent runs in the background and fetches configurations from the Puppet master every 30 minutes. (This interval is configurable with the runinterval setting in puppet.conf.) However, you can also trigger a Puppet run manually from the command line.

  1. On the agent, log in as root and run puppet agent --test on the command line. This triggers a single Puppet run on the agent with verbose logging.

    Note: If you receive a -bash: puppet: command not found error, then the directory that PE installs its binaries in, /opt/puppetlabs/bin, isn’t included in your default $PATH. To include these binaries in your default $PATH, add them by running PATH=/opt/puppetlabs/bin:$PATH;export PATH.

  2. Note the long string of log messages, ending with Notice: Applied catalog in [...] seconds.

You are now fully managing the agent node. It has checked in with the Puppet master for the first time and received its configuration info. It will continue to check in and fetch new configurations every 30 minutes. The node also appears in the console, where you can make changes to its configuration by assigning classes and modifying the values of class parameters.

Next: Hello, world!

Back to top
The page rank or the 1 our of 5 rating a user has given the page.
The email address of the user submitting feedback.
The URL of the page being ranked/rated.