Before you use the vfiler migrate -m nocopy command to migrate a vFiler unit between the nodes of an HA pair, you must ensure that the nodes in the HA pair meet certain requirements, and also ensure that the required licenses are enabled.
The prerequisites are as follows:
- MultiStore must be enabled on each node of the configuration.
- Any license that is enabled on the source node must also be enabled on the destination node.
For example, if CIFS is licensed on the source node of the HA pair, CIFS must also be licensed on the destination node of the HA pair. Otherwise, moving the vFiler unit causes CIFS to be unavailable for that vFiler unit.
- Both the nodes of the HA pair must be connected correctly to the storage shelves in the HA pair.
The disks must be visible to the nodes of the source and destination HA pair.
- The destination node must have an Ethernet connection to the same subnet that the source node uses to ensure that software-based disk ownership changes are transparent to NFS users.
- The volumes assigned to the vFiler unit can either be traditional volumes or FlexVol volumes. If the volumes are FlexVol volumes, the containing aggregate must contain only volumes belonging to the migrating vFiler unit. For information about traditional and FlexVol volumes, see the Data ONTAP Storage Management Guide for 7-Mode.
- The vFiler unit’s storage units must all be composed of complete volumes; that is, the vFiler unit’s paths must use the form /vol/volname. Migration of storage units that name specific volume subdirectories—for example,/vol/volname/qtree—is not supported.
- The volume containing the configuration information of the vFiler unit (/vol/volname) must be writable.
- The destination storage system must not contain aggregates with the same names as the aggregates in the source storage system.
For more information about HA pair, see the Data ONTAP High Availability and MetroCluster Configuration Guide for 7-Mode.