observability-and-instrumentation

Structured logging uses stable event names and fields to answer what happened, where, and why, ensuring production events are consistently captured and queryable for incident diagnosis.

Apply OpenTelemetry distributed tracing to new services and endpoints to achieve end-to-end request visibility, identifying bottlenecks across services and diagnosing incidents effectively.

Metrics need low-cardinality labels to prevent unbounded dimensional growth, ensuring dashboards using RED/USE patterns to quantify latency, errors, and throughput remain performant and queryable.

Configure symptom-based alerts with actionable guidance and runbooks so that production telemetry triggers meaningful notifications with clear, guided incident response steps.

Yes, the instrumentation applies to new services, endpoints, background jobs, and external integrations, ensuring all production code paths are monitored and diagnosable for reliability.