NSS volumes are logical storage media that acquire space from pools of storage. When you create a logical volume, you can either assign it a fixed quota as the maximum size, or allow it to expand to the pool size. To grow a volume, you might need to add new segments to grow the pool first, up to the maximum pool size of 8 TB.
If a pool contains multiple volumes, the cumulative administrative maximum sizes of all volumes can exceed the pool’s maximum size by overbooking, although real total size is bound by physical limitations. Because space is allocated to volumes as needed, a volume might not reach its quota. As the overbooked volumes consume the available physical space, you need to add more disk space to the pool to accommodate the growth, or consider moving or splitting volumes to move data to other pools.
For example, suppose you have an 800 MB storage pool with eight volumes set at 100 MB each. The administrative size equals the physical limits. To overbook the pool, you can add volumes, set one or more of the volumes to expand to the pool size, or increase the size of existing volumes, with the understanding that these are administrative maximum sizes, not physical sizes.
Because volume sizes can be overbooked in a pool, NSS automatically considers what space is remaining in a pool in order to report the maximum size that is currently possible for the volume. In addition to other volumes that can consume space in a pool, NSS snapshots and third-party snapshots can consume space that might not be reported in all of the management tools that report space. NSS reports the total space possible for the volume and the amount of space used by the volume so that tools can properly calculate the maximum free space available.
NSS recognizes DOS, Macintosh, UNIX, and Long name spaces. Volume names, directory names, and filenames in NSS are case insensitive. This differs from Linux POSIX file systems, which are case sensitive by default. For information, see Lookup Namespace.
You must create at least one shared volume in a cluster-enabled pool. Typically, all volumes are created when you initially set up the cluster resource and before you need to cluster migrate or fail over the resource to other servers in the cluster.
You can add volumes to the pool later by cluster migrating the pool cluster resource back to the original server node in the cluster where the pool was created. Otherwise, you get an eDirectory error because the tools only look for the Pool object under its current server node, and not under the original node where it it was created.
To create or modify home directories, Distributed File Services junctions, or any other elements that are managed using eDirectory objects, you must cluster migrate the pool resource back to the node where it was created before you perform those management tasks. This restriction also applies to management tasks like renaming a pool or volume that changes information in the eDirectory objects for the shared pool or volume.
In a clustered storage area network with Novell Cluster Services™, NSS volumes can fail over between kernels, allowing for full data and file system feature preservation when migrating data to Linux. However, you cannot SAN boot cross-platform.
For information about using NSS volumes cross-platform, see the following:
For information about clustering, see the following: