Skip to main content

Observability

Logs​

  • All services emit structured JSON logs.
  • Ship to your preferred aggregator (Loki, ELK, Datadog).
  • Include the commission UUID in every log line related to a service run so you can reconstruct the full audit trail.

Metrics​

  • Prometheus scrape endpoints exposed on :9090/metrics.
  • Recommended dashboards:
    • Platform health — API latency, error rate, queue depth
    • Service runs — heartbeat lag, success rate, mean processing time
    • Sensor ingest — observations/sec, dropped messages, late observations

Health checks​

  • /health/live — process is alive
  • /health/ready — dependencies (DB, MQ, OIDC) are reachable

Audit log​

  • The platform writes a structured audit log for every IAM and clearing-house event (grant, revocation, contract issuance, data delivery).
  • See Concepts · Data Sovereignty.