Novell Cluster Services supports NCP client access to cluster-enabled NSS volumes by using a unique volume ID to mount the volume in the cluster. The volume ID is used by an NCP client only for automatic reconnects to the volume after failover or migration of a cluster resource.
Valid volume ID values are 0 to 254 (up to 255 mounted volumes per server). When you create a new volume on a cluster-enabled pool, Cluster Services automatically assigns it a volume ID that is unique in the entire cluster and writes the value to the cluster resource load script for the pool. Values start at 254 for the first volume in the cluster and decrease for each new volume. You can view the volume IDs assigned on a node by using the ncpcon volumes command.
In older operating systems, there was a mounted volume limit of 64 volumes (values 0 to 63). Some older applications might have hardcoded the old maximum limit of 64 mounted volumes, and might not be able to handle volume IDs greater than 63. You can use the Clusters plug-in to iManager to modify the volume ID in the scripts for a given cluster resource in order to specify a value that works for the application.
Changing the volume ID does not affect the ability to log in to, back up, or access the data. However, there is a brief disruption of service as the cluster resource is offlined and onlined to apply the script changes. If you modify the volume ID for a volume in the cluster resource scripts, ensure that you do the following:
Volume IDs that you manually assign must be unique across every volume on all servers in cluster.
After the value is changed, you must offline and online the cluster resource for the volume in order to mount the volume with its new volume ID.
After the volume is mounted with its new ID, the clients must log out and log in to the volume in order to reconnect to the volume with its new volume ID. Automatic reconnection after cluster resource failovers or migrations occurs properly after this one-time reset.
Some clients might cache the volume IDs. To reset the cached value, the client must be rebooted and reconnected to the volume.
After the volume is mounted with its new ID, if the backup software is running on a client connection, you might need to restart the backup to reconnect to the volume with its new volume ID. Automatic reconnection after cluster resource failovers or migrations occurs properly after this one-time reset.