Observability

Observability uses logs, metrics, and traces to explain the behavior of a running system. It helps you find failures, follow requests, and understand capacity and latency.

Three views of a running service

Logs record messages from service instances. Metrics summarize activity and resource use over time. Traces follow individual requests through components and external calls. Together they help explain where a request spent time and which part failed.

The local console supports development; the Akka Console and configured exporters support deployed services. Exporting data lets you use your own monitoring and retention policies.

Operational telemetry and evidence

Telemetry and the ledger answer different questions. Telemetry helps you operate the service and can be sampled or bounded. The ledger records interactions and control verdicts for governance. A sampled trace is not a complete audit history.

Use telemetry to investigate performance, failures, and capacity. Use the ledger when a decision needs evidence of which action ran, under whose identity, and under which controls. See The ledger and evidence.

Next steps

Observability and monitoring explains signal coverage, the consoles, platform telemetry, and the ledger comparison.
Exporting metrics, logs, and traces configures exporters and sampling.
Alerting baselines suggests alerts.
Metrics reference lists metrics.