Skip to content

Volume Expansion Failure

Symptom

The PVC size is patched upward, but the application still sees the old capacity or the resize stalls partway through the workflow.

Possible Causes

  • The StorageClass does not allow volume expansion.
  • The PVC was patched, but the filesystem resize has not completed yet.
  • The workload needs a restart or controlled remount to observe the new size.
  • The operator attempted an unsupported shrink operation instead of an expansion.

Diagnosis Steps

flowchart TD
    A[Resize appears stuck] --> B[Check PVC capacity and events]
    B --> C[Inspect StorageClass allowVolumeExpansion]
    C --> D[Check pod mount view of filesystem size]
    D --> E[Decide whether remount or restart is required]

  1. Inspect the PVC and its events.

    kubectl describe pvc "$PVC_NAME" \
    --namespace "$NAMESPACE"

  2. Inspect the StorageClass.

    kubectl get storageclass "$STORAGE_CLASS_NAME" \
    --output yaml

  3. Check the application view of the mounted filesystem.

    kubectl exec "$POD_NAME" \
    --namespace "$NAMESPACE" \
    -- df -h "$MOUNT_PATH"

Resolution

  • Enable or switch to a StorageClass that supports expansion.
  • Wait for filesystem resize to finish before measuring inside the container.
  • Restart or roll pods in a controlled way if the workload observes storage size only at startup.
  • If the request attempted a shrink, revert the manifest and use a migration plan instead.

Prevention

  • Treat PVC growth as a planned maintenance action with verification steps.
  • Standardize expandable storage classes for StatefulSets that are expected to grow.
  • Separate storage expansion from unrelated image or configuration rollouts.

See Also

Sources