A process can remain running while its data stops changing. For a receiver, that distinction matters: checking only the service state can miss a stalled feed. dump1090-healthcheck evaluates several signals and returns a health state with a reason you can inspect.
Check the evidence
The checks can consider data freshness, message rate, aircraft count, processing blocks, disk space, and TCP connectivity. These signals cover different failure modes. No single number works as a universal threshold, so configuration should reflect the station, its traffic patterns, and the components you actually run.
Avoid flapping on every fluctuation
Healthy, degraded, and failed states provide more detail than a binary result. Hysteresis helps keep short-lived changes from constantly flipping the reported state. Reasons make the output useful for diagnosis instead of leaving an operator to guess which condition caused the result.
Fit it into an existing monitor
Exit codes and integrations support command-line checks, container health checks, Nagios-style monitoring, and Prometheus. Start by observing normal behavior, then choose thresholds that distinguish a real interruption from expected quiet periods. The repository documents the options and output formats.
Explore dump1090-healthcheck → · Source and documentation ↗