Puppet Enterprise metrics and status monitoring

You can use Puppet Enterprise (PE) metrics and status monitoring for your own performance tuning or provide the information to Support for troubleshooting.

There puppet_metrics_collector and pe_status_check modules are bundled with PE. These modules help you track the status of your PE installation as a whole.
Tip: The information reported by these modules is different from information presented in Infrastructure reports and Node run statuses, which report on the outcome of Puppet runs.

You can also use APIs or our Splunk plugin to View and manage Puppet Server metrics.

About the puppet_metrics_collector module

The puppet_metrics_collector module collects metrics from the status endpoints of Puppet Enterprise (PE) services.

The puppet_metrics_collector module is installed with PE and is partially enabled by default.

Important: If you have a version of this module, from the Forge or other sources, specified in the code, you must remove this version before upgrading to allow the version bundled with PE to be asserted.

The following two parameters control metrics collection:

puppet_enterprise::enable_metrics_collection
A Boolean specifying whether the primary server collects metrics from PE services, such as Puppet Server and PuppetDB .
Default: true
puppet_enterprise::enable_system_metrics_collection
A Boolean specifying whether your infrastructure nodes collect metrics from the operating system your PE services run on. To allow the collection of system metrics, sysstat must be installed and enabled on your operating system.
Default: false

Visit the puppet_metrics_collector Forge page to learn about this module's other classes and features, such as retention time, collection frequency, and parameters for specific services.

You can use the puppet_operational_dashboards module to view PE metrics.

Tip: You can also use APIs or our Splunk plugin to View and manage Puppet Server metrics.

Enable or disable metrics collection

By default, the puppet_metrics_collector module is partially enabled so that your primary server collects metrics for PE services. You can configure the metrics collection parameters in the PE console.

The metrics collection parameters accept Boolean values. Setting both parameters to false disables the module entirely.
  1. In the PE console, click Node groups and select the PE Infrastructure node group.
  2. Configure the following parameters according to your requirements:
    • puppet_enterprise::enable_metrics_collection: Set this parameter on the Configuration data tab to determine whether your primary server collects metrics for PE services.
    • puppet_enterprise::enable_system_metrics_collection: Set this parameter on the Classes tab to determine whether your infrastructure nodes collect metrics from the operating system your PE services run on.
      Note: To allow the collection of system metrics, sysstat must be installed and enabled on your operating system.
  3. Commit your changes and run Puppet.

About the pe_status_check module

The pe_status_check module can alert you when your Puppet Enterprise (PE) installation is not in an ideal state, based on preset indicators, and describe how you can resolve or improve the detected issue.

Important: If you have previously specified a version of this module, from the Forge or other sources, in your code, we recommend removing this version to allow the version bundled with PE to be asserted.

By default, the pe_status_check module uses the pe_status_check fact to compare your installation to a predetermined ideal state. This fact collects information about your PE infrastructure components. You can optionally enable the agent_status_check fact to collect information about agent nodes that are not infrastructure nodes. To enable this fact, classify pe_status_check::agent_status_enable to your nodes.

The pe_status_check module produces reports based on the information collected by the pe_status_check fact (and the agent_status_check fact, if it is enabled). The module recommends remedial actions you can take to either resolve a deficiency or improve your installation's tuning.

To get reports from the module, you can:

To enable notifications in reports, you must declare the pe_status_check class.

These sections of the module's Forge page explain how to interpret the module's reports and the information that the facts collect: