Skip to content

Expanding

Expanding a Persistent Volume (PV) in Kubernetes allows for increasing the size of a volume without downtime, ensuring applications continue running with sufficient storage. Simplyblock supports online expansion of Logical Volumes (LVs) through its CSI driver, making it possible to resize volumes dynamically as storage requirements grow.

Info

To enable volume expansion, Kubernetes 1.16 or later is required.

Enable Volume Expansion

To enable volume expansion, the StorageClass has to be configured accordingly. To enable volume expansion, the property allowVolumeExpansion has to be set to true.

The StorageClass the operator creates for a cluster's default pool (simplyblock-<namespace>-<cluster>) already allows volume expansion. An authored StorageClass sets the property itself:

Allowing volume expansion in StorageClass
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: tenant-a-fast
  labels:
    storage.simplyblock.io/namespace: simplyblock
    storage.simplyblock.io/cluster: production
    storage.simplyblock.io/pool: tenant-a
provisioner: csi.simplyblock.io
parameters:
  cluster_id: <CLUSTER_UUID>
  pool_name: tenant-a
  csi.storage.k8s.io/fstype: xfs
reclaimPolicy: Delete
volumeBindingMode: WaitForFirstConsumer
allowVolumeExpansion: true # <- Enable volume expansion

A volume cannot grow beyond the max_size parameter of its StorageClass, when it is set.

Expand a PersistentVolume

To expand an existing volume, update the field spec.resources.requests.storage in the existing resource descriptor.

Updating the volume size
apiVersion: v1
kind: PersistentVolumeClaim
metadata:
  name: my-example-pvc
spec:
  resources:
    requests:
      storage: 500Gi # <- Was 100Gi before

Then apply the change.

Apply resource update
kubectl apply -f pvc.yaml

Note

Simplyblock allocates logical volumes on GiB boundaries. A requested size is rounded up to the next full GiB, so an expansion that stays inside the current GiB does not change the size of the logical volume.

Resize the Filesystem

For a volume mounted as a filesystem, the resize is performed by the CSI driver after the logical volume has been expanded. The filesystem is grown in place, and no filesystem-specific command has to be run inside the pod.

For a raw block volume, no filesystem resize is performed. The block device is expanded at the storage layer, and the consuming application picks up the new size on its own.

Shrinking a Volume

A volume cannot be shrunk. Kubernetes rejects a decrease of spec.resources.requests.storage on an existing persistent volume claim. When a smaller volume is needed, a snapshot is created and restored onto a new volume of the smaller size.