Storage class for creating consistency group volumes

Create PVCs in a consistency group on a file system that is owned by the primary cluster or a file system that is owned by a different cluster other than the primary cluster.

To create the consistency group volumes, use the storageClass details that are provided in the following example.

In this example, it is assumed that you want to create PVCs under the file system, gpfs0, as in the sample deployment explained in Remote cluster support. The same storageClass format is applicable for the sample deployment that is explained in Operator.

Consistency group storage class:
apiVersion: storage.k8s.io/v1
kind: StorageClass
metadata:
  name: ibm-spectrum-scale-csi-consistency-group
provisioner: spectrumscale.csi.ibm.com
parameters:
    version: "2"        
    volBackendFs: gpfs0
reclaimPolicy: Delete
allowVolumeExpansion: true
The following fields come under the parameters section of storageClass. Parameters section is mandatory for IBM Storage Scale CSI driver storageClass:
Field name Description
version Represents the version of the storage class.
  • Value 1 indicates that the storage class is not for a consistency group.
  • Value 2 indicates that the storage class is for a consistency group.
Note: The default value is 1.
volBackendFs (mandatory) The name of the file system under which the fileset must be created. File system name is the name of the remotely mounted file system on the primary cluster.
clusterId (optional) The Cluster ID of the owning cluster for remote files system.

The Cluster ID of the primary cluster for local file system.
uid (optional) The username of the fileset. The uid/gid must exist on the IBM Storage Scale GUI node of accessing and owning clusters. The default value is "root".
gid (optional) The group name of the fileset. The gid/group name must exist on the IBM Storage Scale GUI node of the accessing and owning clusters. The default value is "root".
inodeLimit (optional) Inode limit for consistency group. If not specified, inodeLimit is set to 1M.
Note: Start of changeWith IBM Storage Scale 5.1.4 and later, auto inode expansion can be used instead of inodeLimit. For more information, see the mmchfs command.End of change
Start of changeshared (optional)End of change Start of changeUse shared=true, if you have pods with non-root users that are using PVCs with ReadWriteMany(RWX) access mode. Default value is false.End of change
compression (optional) Specifies whether the compression is enabled.
Allowed values are:
  • true
  • false
The default value is false.
tier (optional) Name of the tier or pool where the volume data is to be placed.
Note: Tiering feature requires IBM Storage Scale file system 27.00 or later.
nodeClass (optional) The user can specify a pre-existing nodeClass in storageClass, so that the pre-existing nodeClass is used when a volume is created from an existing volume or snapshot. If a nodeClass is specified, only the node belonging to it participates in data copy.
Note: For more information about nodeClass, see IBM Storage Scale nodeClass command.
Note: For using the consistency group feature, you must use IBM Storage Scale 5.1.3.0 or later.