worker-health-monitoring

Detect stuck or offline background workers using heartbeat-based health monitoring with configurable timeouts, calculating failure rates and analyzing rolling duration windows to flag degraded jobs.

Heartbeat-based monitoring tracks worker liveness by expecting periodic signals, detecting offline, degraded, or unhealthy states based on configurable thresholds and latency multipliers.

Yes, worker health monitoring supports both TypeScript and Python implementations, allowing integration into diverse backend systems for tracking queue depth and execution durations.

Configure failure detection thresholds by setting acceptable failure rates and latency multipliers, which the system uses to identify unhealthy workers and flag jobs running indefinitely.

Monitor background job success rates and execution time by implementing performance metrics tracking that analyzes rolling duration windows and queue depth to ensure optimal reliability.

Background workers may appear healthy but fail to complete due to stuck jobs, which heartbeat monitoring detects by flagging jobs that appear to run indefinitely beyond expected durations.