If the target data set exists, DFSMShsm deletes the existing data set, uses the Access Method Services IMPORT command or DFSMSdss as the data mover, and passes the storage class name associated with the data set. Access Method Services or DFSMSdss determines the volume to which the data set is recovered and creates space for the data set.
DFSMShsm recovers VSAM data sets to a level 0 volume by using the Access Method Services IMPORT command or DFSMSdss as the data mover. However, because the backup version of a VSAM data set contains DFSMShsm control information, a user-issued IMPORT command or DFSMSdss command cannot be used to recover a backup version.
If DFSMShsm recovers a VSAM data set to a volume on a type of device that is different from the device from which DFSMShsm backed up the VSAM data set, the allocation for the buffer that the Access Method Services IMPORT command uses could be too small. If this occurs, you can define the target VSAM data set (base cluster and all alternate indexes) and then reissue the command.
The recovery of an integrated catalog facility catalog is restricted to those users who are recognized as DFSMShsm-authorized users. If TOVOLUME is specified, the volume specified must have the same volume serial number and device type as the volume from which the catalog was backed up. For more information about recovering the integrated catalog facility catalog, see Backing up and recovering a catalog.
The NEWNAME parameter limits the recovery of VSAM data sets to spheres that do not have two or more paths defined on the base cluster, and recovers only the base cluster even when alternate indexes exist. After DFSMShsm recovers the base cluster, you can rebuild any necessary alternate indexes using the Access Method Services BLDINDEX command.
For VSAM data sets not cataloged, if the data set with the new name does not exist, DFSMShsm generates the new names for the data, index, and path components based on the character length of the new user-supplied cluster name:
| New cluster name length | Component naming scheme |
|---|---|
| 1–38 characters | cluster.DATA, cluster.INDEX, cluster.PATH |
| 39–42 characters | cluster.D, cluster.I, cluster.P |
| 43–44 characters | fourqual.Dhhmmss, fourqual.Ihhmmss, fourqual.Phhmmss |
Where cluster is the new user-supplied cluster name, fourqual is the first four qualifiers of the new user-supplied cluster name, and hhmmss is a time stamp.
If the data set with the new name already exists and the type (KSDS, ESDS, or RRDS) is the same, DFSMShsm reuses the entry names of the data component and index component. For information about how DFSMSdss generates new names for the data and index components, refer to the z/OS DFSMSdss Storage Administration.
A message is printed containing the generated or existing name.
If you specified the NEWNAME parameter and the REPLACE parameter and if a data set with the name you specified already exists, DFSMShsm deletes the existing data set before it recovers the data set you specified with the RECOVER command.
The new data set name and the old data set name can be cataloged in the same or different catalogs when you specify the NEWNAME parameter.
For VSAM data sets not cataloged, if you specify NEWNAME the data and index objects of the data set being recovered must be the same as the object names exported. The object name of the data set being recovered is passed to Access Method Services IMPORT to rename the NEWNAME data set. Access Method Services verifies if the name is the same as the one exported.