Docs/Platform/Monitoring

Monitoring

Health checks, Prometheus metrics, Grafana, and operational monitoring guidance.

FireSonic exposes several layers of operational visibility: the health endpoint for current runtime state, Prometheus metrics for time-series collection, and the official Grafana dashboard for visual monitoring and capacity analysis.

Health endpoint

The health endpoint exposes the current application, database, listener, processing, script, rule, and operational state. It is suitable for liveness checks, operational diagnostics, and integration with infrastructure monitoring.

http://localhost:8181/health

Prometheus metrics Commercial

The Prometheus /metrics endpoint provides time-series metrics for the message pipeline and runtime. Typical measurements include intake and processing rates, ACK/NAK activity, FHIR delivery behavior, failures, latency, storage growth, database state, backup state, JVM/runtime metrics, and operational health.

http://localhost:8181/metrics

Prometheus can scrape these metrics and drive alerting or the official FireSonic Grafana dashboard.

Official Grafana dashboard Commercial

The official dashboard bundle provides a consolidated operational view of throughput, delivery, failures, storage, backup, MLLP connectivity, ACK/NAK behavior, success rates, and runtime health.

FireSonic Grafana dashboard platform overview and message pipeline — Platform overview with received, accepted, processed and delivered totals, pipeline throughput, ACK/NAK rates, and operational alerts.

FireSonic Grafana dashboard storage and backup panels — Storage and backup monitoring including filesystem capacity, SQLite database state, reusable space, backup status, age, duration, and size.

FireSonic Grafana dashboard message totals and MLLP connection metrics — Message totals, ACK/NAK and delivery success rates, plus MLLP connection activity.

Operational monitoring guidance

Area	What to watch
Health checks	Application, database, listener, scripts/rules, processing leases, and downstream availability.
Failures	Processing failures, delivery failures, rejected messages, NAKs, and changes in retryable-message counts.
Capacity	Received/processed/delivered rates, queue pressure, processing latency, and sustained throughput.
Storage	Filesystem free space, SQLite database size, reusable space, retention behavior, and growth trends.
Backup	Backup success, age, duration, size, and external-copy state where configured.
Runtime	CPU, memory, JVM behavior, and resource recovery after traffic spikes.

Alert thresholds should reflect the expected traffic profile and infrastructure capacity of the deployment rather than generic fixed values.