ruano .NET Observability
Goal
Improve diagnosability, operational visibility, and production supportability in .NET systems.
Use this skill for
- OpenTelemetry
- structured logging
- tracing
- metrics
- correlation IDs
- health checks
- diagnostics for APIs and background workers
Core rules
- Prefer structured logs over string-concatenated logs.
- Preserve correlation IDs across boundaries.
- Instrument inbound and outbound calls when relevant.
- Add metrics for throughput, latency, failure rate, and retries when useful.
- Use health checks for readiness and liveness.
- Never log secrets or sensitive data.
Logging
Prefer logs for:
- important business transitions
- external dependency failures
- retry attempts
- circuit breaker state changes
- terminal processing failures
Avoid:
- noisy logs in hot paths
- duplicate logs for the same failure at multiple layers
Tracing
- Propagate trace context across HTTP and messaging when possible.
- Instrument API, database, and external dependency boundaries.
Metrics
Useful examples:
- request duration
- dependency duration
- retry count
- failure count
- consumer lag
- queue depth