Improve Kubernetes Platform

Operation

Implement the requested coherent change at the owning source and update directly affected consumers. An earlier audit is optional when the user and repository already establish the target. Validate the changed behavior, then stop when the requested scope is complete.

Goal

Implement one selected, evidence-backed Kubernetes Platform improvement at the authoritative owner. Complete the directly affected contract without broad cleanup or redesign.

Domain rules

• Modify the environment's actual source of truth—chart, kustomization, operator CR, or manifest—not a generated copy unless generation is authoritative.
• Separate startup, readiness, and liveness intent; preserve graceful drain and allow enough termination time for the measured shutdown path.
• Keep selectors immutable-compatible and coordinate service, policy, autoscaling, disruption, and workload changes as one rollout slice.
• Use least-privilege RBAC and pod security settings; do not solve connectivity by removing policies or granting cluster-wide permissions.
• For risky schema/state changes, use a staged rollout and explicit rollback/forward-fix condition rather than assuming Deployment rollback restores data compatibility.

Validation

• Run repository-native render/schema/policy checks and inspect the complete rendered delta for the target environment.
• Where a cluster is available, use dry-run/apply diff plus rollout, endpoint, event, and metrics evidence without destructive production experiments.
• Test startup delay, dependency unavailability, SIGTERM/drain, unavailable replica, node disruption, and rollback compatibility as applicable.