end-to-end observability baseline (trace IDs, metrics, dashboards)

[ashupednekar]

Summary

Establish an observability baseline across portal -> ingestor -> operator -> runtime.

Motivation

Cross-service failures are hard to debug without shared trace context and consistent metrics.

Proposal

• Propagate correlation/trace IDs through request and NATS message flows.
• Standardize structured logging fields across services.
• Add baseline metrics: request rate, latency, errors, queue lag, cold starts.
• Provide starter dashboards/alerts.

Acceptance criteria

• Single request can be traced end-to-end by trace/correlation ID.
• Metrics are emitted with per-project/per-function dimensions.
• Basic dashboard and alert rules included in docs or manifests.