REPLACE


>>-+---------+-------------------------------------------------><
   '-REPlace-'

REPLACE specifies that DFSMSdss is to search the target volumes for usable preallocated data sets. If a usable preallocated data set is found, it will be replaced with the data set from the source volume. If no preallocated target is found, DFSMSdss attempts to allocate a data set.

DFSMSdss searches for preallocated data sets as follows:

For SMS-managed data sets, DFSMSdss first searches in the standard order of search for a catalog entry for the data set.
For VSAM or Non-VSAM data sets that are not SMS-managed, DFSMSdss searches for a preallocated data set, cataloged or uncataloged, in the following order:
1. In any output volumes specified.
2. If DFSMSdss finds no preallocated data set, it then searches the catalogs in the standard order of search.
3. If no catalog entry is found for the data set, DFSMSdss searches the volume that the data set was dumped from.
If no preallocated target is found, DFSMSdss attempts to allocate a data set. DFSMSdss invokes the ACS routines to determine if the data set should be SMS managed. If it should be SMS managed, then allocation is done according to the SMS constructs. If the data set should not be SMS managed, the output volumes specified are used. If no output volumes were specified, the volume the data set resided on at dump time is used. If allocation is successful, the data set is restored.
PURGE is accepted and is treated the same as REPLACE. REPLACE only applies if the data set is not being renamed. REPLACEUNCONDITIONAL should be used when renaming a data set and a preallocated target data set should be replaced.
The REPLACE and REPLACEUNCONDITIONAL keywords can not be specified together.

The volume that the data set was dumped from on system A is shared with system B, where the data set was being restored. If the data set was NONSMS when it was dumped, and the REPLACE keyword is specified on the RESTORE, the order of search, for a pre-allocated data set that is cataloged or uncataloged, will first search any output volumes specified. If no pre-allocated data set is found, then it will search the catalogs in the standard order of search. If no catalog entry is found for the data set, then it will search the volume the data set was dumped from.

Note:

REPLACE or REPLACEUnconditional must be specified to restore to preallocated data sets.
If the source data set is an extended-addressable VSAM data set, then the target must also be an extended-addressable VSAM data set.
If REPLACE is specified with the RESTORE command, the SMS constructs already associated with the preallocated data set remain the same.
CATALOG and RECATALOG are ignored for preallocated data sets.
The target data set name must match the source data set name. REPLACEUnconditional must be specified to replace a target data set that has a name matching the rename criteria.
If you delete and reallocate a striped VSAM data set that is being replaced, DFSMSdss uses the same rules that apply to a new allocation. In general, it is best to not specify output volumes, or the VOLCOUNT keyword, when you are replacing striped VSAM data sets.
If the target data set is smaller than the source data set, DFSMSdss scratches the target data set and reallocates it with the size of the source data set.
Specify the FORCE keyword with the REPLACE keyword if you want to restore the data from unmovable data sets whose extents do not match.
If the RENAME or RENAMEU keywords are specified in conjunction with the REPLACE keyword, only one of the keywords take effect for any particular data set. The RENAME or RENAMEU keyword takes precedence over the REPLACE keyword. If a source data set name matches the RENAME or RENAMEU criteria, then rename processing will be performed and replace processing will not be performed. If a preallocated target data set exists with the new name as chosen by the rename criteria, then the restore fails even if the REPLACE keyword was specified. If you want to replace a preallocated target with the new name, specify the REPLACEUNCONDITIONAL keyword. If a source data set name does not match the rename criteria and a preallocated target data set with the source name exists, the preallocated target data set is replaced.
If the data set being restored is a large format sequential data set, but the preallocated target is not a large format sequential data set, the preallocated target will be used and turned into a large format sequential data set as long as it has enough allocated space for the data set being restored. If the preallocated new name target does not have enough allocated space, it will be scratched and reallocated as a large format sequential data set with enough space for the data set being restored.
If REPLACE is specified with the RESTORE command, the CA Reclaim attribute already associated with the preallocated data set remains the same.
The preallocated data set is usable if all of the following conditions that apply to the data set being processed are met:
- The user is authorized to update the target data set.
- If VSAM, the cluster types match
- The DSORG matches.
- If the data set being processed is multivolume or single volume and the non-SMS preallocated data set matches the multivolume or single volume.

For more information about the REPLACE keyword, see Special considerations for RESTORE.