Caution

Grafana Agent has reached End-of-Life (EOL) on November 1, 2025. Agent is no longer receiving vendor support and will no longer receive security or bug fixes. Current users of Agent Static mode, Agent Flow mode, and Agent Operator should proceed with migrating to Grafana Alloy. If you have already migrated to Alloy, no further action is required. Read more about why we recommend migrating to Grafana Alloy.

Important: This documentation is about an older version. It's relevant only to the release noted, many of the features and functions have been updated or replaced. Please view the current version.

Documentationbreadcrumb arrow Grafana Agentbreadcrumb arrow Flow modebreadcrumb arrow Monitoring Grafana Agent Flowbreadcrumb arrow Controller metrics
Open source

Controller metrics

The Grafana Agent Flow component controller exposes Prometheus metrics which can be used to investigate the controller state.

Metrics for the controller are exposed at the /metrics HTTP endpoint of the Grafana Agent HTTP server, which defaults to listening on http://localhost:12345.

The documentation for the grafana-agent run command describes how to modify the address Grafana Agent listens on for HTTP traffic.

The controller exposes the following metrics:

  • agent_component_controller_evaluating (Gauge): Set to 1 whenever the component controller is currently evaluating components. Note that this value may be misrepresented depending on how fast evaluations complete or how often evaluations occur.
  • agent_component_controller_running_components (Gauge): The current number of running components by health. The health is represented in the health_type label.
  • agent_component_evaluation_seconds (Histogram): The number of completed graph evaluations performed by the component controller with how long they took.