Install PE using PIM
Puppet Installation Manager (PIM) supports the deployment of standard, large, and extra-large PE architectures. For an interactive experience, choose the guided installation and follow the step-by-step process in your terminal to configure and install the PE infrastructure you require. Alternatively, if you do not require guidance, you can create a JSON file containing your custom installation parameters, and run the installation from the command line.
Regardless of the installation process you choose, you can use PIM on a jump host to install PE infrastructure components on remote nodes that run a supported PE operating system. Alternatively, you can install PE locally by using PIM on a machine running a supported PE operating system. In this scenario, if you require additional infrastructure nodes to host PE components, your local machine can serve as a jump host.
PE infrastructure nodes are the hosts where PE components are installed. The following table lists the infrastructure nodes you can include in your installation when you use PIM to install PE:
| PE infrastructure node | Description |
|---|---|
| Primary server (required) |
Essential for a PE installation. Can host all components and services for smaller scale environments that include up to 2,000 nodes. |
| Primary server replica (optional) | To set up disaster recovery, install a replica of the primary server. If your primary server fails, the replica takes over to continue critical operations. |
| Database server | In an extra-large PE installation, a dedicated database server hosts a PostgreSQL instance containing the PuppetDB database. |
| Database server replica (optional) | Provides backup support during failovers. |
| Compilers | Compilers process Puppet code and convert it into catalogs that can be applied to agent nodes. The primary server can handle requests and compile catalogs for up to 2,000 agent nodes. In large and extra-large PE installations, dedicated compiler nodes help accelerate catalog compilation. |
