MOUNT

The MOUNT statement defines the file systems to be mounted at initialization and where in the file hierarchy they are to be mounted. All HFS data sets specified on MOUNT statements in BPXPRMxx must be available at IPL time. If an HFS data set is migrated by hierarchical storage management (HSM), then the initialization of z/OS UNIX and HSM will cause an indefinite wait for kernel services. DFHSM will initialize, but any msgARC0055A issued for the migrated data set will need a reply.

If you specify the NOSETUID option, the setuid and setgid mode bits are not respected when a program in this file system is run. The program runs as though the setuid and setgid mode bits were not set. Also, the APF extended attribute (+a) and the program control extended attribute (+p) are not honored

Rule: The MOUNTPOINT keyword must specify an absolute path name.

Systems using shared file systems will have I/O to a z/OS UNIX couple data set (CDS). Because of these I/O operations to the CDS, each mount request requires additional system overhead. You will need to consider the effect that this change will have on your recovery time if a large number of mounts are required on any system that has shared file systems. For more information about shared file systems, see Sharing file systems in a sysplex.

You can use multiple MKDIR keywords on the MOUNT statement to define mount points in BPXPRMxx so that one or more directories is created in the mounted file system during z/OS UNIX initialization.

Restriction: MKDIR is intended to run during synchronous mounts on the system that is initializing. The directory might not be created if any of these situations exist:

The file system is mounted asynchronously, such as with NFS.
The SYSNAME value identifies a remote system.
The file system is already mounted on a remote system.

Tip: To ensure that the mounts will succeed, use SETOMVS SYNTAXCHECK to check the MVS™ catalog for the existence of the HFS or zFS file system name that is listed on each MOUNT statement.