Monitoring

JamBridge exposes Prometheus metrics via Spring Boot Actuator at /actuator/prometheus.

Key metrics

Metric	Type	Description
jambridge_messages_received_total	Counter	Total messages received per port
jambridge_pipeline_duration_seconds	Histogram	End-to-end pipeline duration
jambridge_stage_duration_seconds	Histogram	Per-stage duration (labels: stage)
jambridge_consent_decisions_total	Counter	Consent decisions (labels: result=permit/deny/error)
jambridge_mpi_match_grade_total	Counter	MPI match grades (labels: grade)
jambridge_circuit_breaker_state	Gauge	HAPI circuit breaker (0=closed, 1=open, 2=half-open)
jambridge_queue_depth	Gauge	Secondary retry queue depth
jambridge_dedup_hits_total	Counter	Duplicate messages detected

Grafana dashboard

Import the JamBridge dashboard from Grafana ID 21847 (or from /monitoring/grafana-dashboard.json in the repo).

Panels include:

• Message throughput by port (messages/min)
• Pipeline stage latency heatmap
• Consent decision rate (permit vs deny)
• MPI match grade distribution
• Circuit breaker state timeline
• Queue depth over time
• ATNA audit event rate

Alerting rules

prometheus-rules.yaml

groups:
  - name: jambridge
    rules:
      - alert: JamBridgeCircuitOpen
        expr: jambridge_circuit_breaker_state == 1
        for: 2m
        annotations:
          summary: "JamBridge HAPI circuit breaker is OPEN"

      - alert: JamBridgeQueueDepthHigh
        expr: jambridge_queue_depth > 100
        for: 5m
        annotations:
          summary: "JamBridge retry queue depth > 100"

      - alert: JamBridgeConsentErrorRate
        expr: rate(jambridge_consent_decisions_total{result="error"}[5m]) > 0.1
        for: 2m
        annotations:
          summary: "JamBridge consent error rate > 10%"

      - alert: JamBridgeNoMessages
        expr: rate(jambridge_messages_received_total[10m]) == 0
        for: 10m
        annotations:
          summary: "JamBridge: no messages received in 10 minutes"