rhcl-observability

name	rhcl-observability
metadata	{"author":"rhoai3-coding-demo","version":"1.0.0","platform-family":"rhcl","platform-baseline":"repo","ocp-baseline":"repo","skill-group":"OpenShift Platform"}
description	Use when configuring observability, monitoring gateways, APIs, and applications on OpenShift with Connectivity Link: metrics and Prometheus monitoring, Grafana dashboards (platform engineer, application developer, business user), ServiceMonitor and PodMonitor configuration, control-plane and data-plane tracing with OpenTelemetry, Envoy access logs and request correlation, MCP gateway observability with OpenTelemetry spans and audit trails, and AuthPolicy identity injection for audit logging. Do NOT use for Connectivity Link installation, API management, or gateway deployment; use the relevant rhcl-* skill. Do NOT use for OCP-level monitoring, logging, or Cluster Observability Operator; use the relevant ocp-* skill. Do NOT invent CR fields, API versions, or telemetry configurations not documented in the official source.

RHCL Observability

Use this skill to ground Connectivity Link observability guidance in the official Red Hat Connectivity Link 1.4 documentation for the active baseline in docs/PLATFORM_BASELINE.md.

Source Grounding

Read references/source-capture.md before using product behavior. Official Red Hat documentation is product authority. This skill captures the Connectivity Link observability model including metrics, tracing, access logs, dashboards, MCP gateway observability, and audit trails.

Key Concepts

Observability Features

• Metrics: Prometheus metrics for gateway and policy performance via Connectivity Link components, Gateway API state metrics, and Envoy metrics.
• Tracing: control-plane traces (operator reconciliation) and data-plane traces (request flows through Istio, Authorino, Limitador, wasm-shim).
• Access Logs: Envoy access logs with request correlation via x-request-id headers.
• Dashboards: pre-built Grafana dashboards for platform engineers, application developers, and business users.

Enabling Observability

Set spec.observability.enable: true in the Kuadrant CR (kuadrant.io/v1beta1) to create ServiceMonitor and PodMonitor CRDs automatically.

MCP Gateway Observability

Uses Red Hat build of OpenTelemetry with OpenTelemetryCollector CR for traces and logs. Router spans follow OpenTelemetry MCP Semantic Conventions. Audit trails capture caller identity, tool names, and session context through AuthPolicy identity injection and Istio Telemetry access logs.

Workflow

1. Confirm the active RHCL baseline in docs/PLATFORM_BASELINE.md.
2. Read references/official-doc-extraction.md.
3. Identify whether the task concerns:
   • enabling Connectivity Link observability monitoring
   • configuring Grafana dashboards and alerts
   • setting up control-plane or data-plane tracing
   • configuring Envoy access logs and filtering
   • request correlation across components
   • MCP gateway OpenTelemetry setup
   • MCP gateway audit trail configuration
   • AuthPolicy identity injection for audit logs
4. For manifests, verify all API versions, CRDs, fields, namespaces, and credentials before committing.
5. Validate the output with the official source and cluster schema.

Related Skills

• Use rhcl-develop for API management, console plugin, RBAC, and API product workflows.
• Use rhcl-troubleshoot for diagnosing gateway, routing, policy, and MCP gateway issues.
• Use ocp-observability for OCP-level monitoring, logging, and COO.
• Use ocp-opentelemetry for Red Hat build of OpenTelemetry Operator and Collector configuration.
• Use ocp-distributed-tracing for Tempo Operator, TempoStack, and tracing platform configuration.
• Use ocp-grafana-operator for GitOps-managed Grafana Operator instances and dashboards.

References

• references/source-capture.md
• references/official-doc-extraction.md