Boundary
Monitor Boundary
Visibility into various components of Boundary plays an important role in ensuring the health of its controllers and workers.
You can export metrics for controllers and workers and then configure them to be collected by Prometheus. Metrics provide insight into the current state of a Boundary deployment by enabling the diagnosis of deployed controllers and workers. Enabling metrics assists in the detection and mitigation of anomalies in a deployment. These docs describe the available metrics and the naming and labeling standards used.
Additionally, the health of controllers can be monitored using the
/health endpoint. The health endpoint is enabled by defining a listener in
Boundary's configuration file, and a health response may be configured to add a
shutdown grace period process.
You can also configure Boundary to monitor events. You can choose to monitor audit, observation, system, and telemetry events. Boundary supports both file and stderr sinks for events and you can send events to multiple sinks.
Listeners let you configure the addresses and ports on which Boundary responds to requests. You can configure TCP and Unix listeners. Boundary also supports custom response headers for any requests made on controllers. You can define headers based on the status code that is returned.
Tutorials
To learn how to configure event filtering and sinks, complete the Event filtering and sink configuration tutorial.
To learn how to configure metrics collection, complete the Visualize metrics with Prometheus tutorial.
More information
Refer to the following topics for the parameters you use to configure the addresses and ports on which Boundary responds to requests:
Refer to the following topics for more information on the various ways to monitor Boundary: