Production Observability

Instrument user-visible operations and operational boundaries, not every method.

Baseline

• Add Actuator and one metrics registry selected by the deployment platform.
• Use Micrometer Observation for application metrics and traces.
• Use Micrometer Tracing with the chosen bridge; do not mix tracing APIs throughout business code.
• Export through OTLP when the platform standardizes on OpenTelemetry collectors.
• Expose only required actuator endpoints and secure every non-public endpoint.

Observation conventions

• Name observations by stable operation, such as orders.create.
• Keep metric tags low-cardinality: method, outcome, region, or bounded status.
• Put request IDs, user IDs, order IDs, and exception messages only in traces or logs.
• Propagate context across @Async, executor, and Reactor boundaries.
• Record latency, throughput, failures, and saturation for every external dependency.