golang-observability

End-to-end observability in a Go service is set up by wiring structured JSON logs via slog, adding OpenTelemetry tracing with otelhttp, and exposing Prometheus histograms for request latency. This blueprint connects logs, metrics, and traces across replicas.

Yes, you can migrate Go logging from zap, logrus, or zerolog to slog without breaking existing code. The blueprint provides step-by-step migration guides and integrates otelslog bridges to maintain context-aware structured logging.

The best way to propagate OpenTelemetry traces across multiple Go microservices is by applying otelhttp transport instrumentation for outgoing calls. This ensures trace context propagation and correlation across replicas.

On-demand profiling in Go production services is enabled via environment variable toggles. This allows you to activate profiling controls dynamically to diagnose latency issues without redeploying.

Yes, the Go observability blueprint includes privacy-conscious Real User Monitoring (RUM) integration and provides production-ready templates for alerting and dashboards to ensure readiness.

You should use a histogram for Prometheus metrics in Go applications to track request latency and timing distributions. The blueprint provides reusable patterns for declaring metrics with histogram-based latency monitoring.