Atlas: PVC Mount/Attach Failures

What this usually means

Provisioning succeeded, but the node cannot attach or mount the volume. The blocker is often CSI controller/daemon health, topology mismatch, or stale attachments.

What to inspect first

Start with pod events; they name the failing stage. Then inspect the PVC and driver posture.

• Read the events at the bottom of `describe pod`.
• Distinguish attach errors from mount errors.

kubectl describe pod <pod> -n <ns>
kubectl describe pvc <pvc> -n <ns>

kubectl get pods -A | rg -n "csi|storage" || true

Resolution guidance

Contain risk. Storage fixes can cause data damage when rushed.

• Restore CSI driver health first; do not thrash pods while the driver is unhealthy.
• Check topology/zone alignment between nodes and volumes.
• If stale attachments exist, follow your platform’s safe detach procedure; audit before forcing.
• Avoid manual deletion of PV/PVC objects unless the data is intentionally disposable.