On this page
Principle – Observability by Default
Summary
Every platform capability must ship with measurable telemetry, health, and evidence of compliance.
Intent
Ensure operational visibility, faster incident response, and verifiable compliance across the platform.
Scope and Applicability
Applies to services, infrastructure, automation pipelines, and security controls.
Principles
- Emit logs, metrics, and traces that support troubleshooting and audit needs.
- Define SLOs and evidence expectations alongside the service design.
- Make observability part of the delivery pipeline, not an afterthought.
Per-Domain Implementation
OpenShift
- Enforce cluster-level logging, metrics, and audit policy baselines.
Networking
- Capture pre-change and post-change evidence for fabric and routing changes.
Storage
- Track capacity, performance, and backup outcomes with standardized telemetry.
Security
- Ensure access and policy decisions are auditable and centrally visible.
Platform
- Provide shared observability tooling and standardized dashboards.
AI
- Emit telemetry for AI pipelines and inference endpoints (latency, error rates, and drift indicators where applicable) and retain evidence for audits.
Exceptions
Temporary exceptions require documented risk acceptance and a remediation plan.