Use this checklist as a review aid rather than a compliance certificate. Each item should have an owner, evidence and an operational procedure.
Availability
- Multiple control-plane nodes where high availability is required
- Worker capacity survives a planned node outage
- Pod disruption budgets are defined for critical workloads
- External dependencies are documented
Security
- RBAC follows least privilege
- Administrative access is auditable
- Network policies are defined for sensitive workloads
- Secret-management and rotation procedures exist
- Image provenance and vulnerability handling are defined
Operations
- Cluster and workload dashboards exist
- Alerts have owners and runbooks
- Backup and restore have been tested
- Upgrade procedure is rehearsed
- Capacity thresholds are monitored
Validation commands
kubectl get nodes
kubectl get pdb -A
kubectl get networkpolicy -A
kubectl get events -A --sort-by=.lastTimestamp | tail -50