Data set characteristics

DFSORT accepts empty and null non-VSAM data sets, and DUMMY data sets, for sorting and copying (be sure to supply RECFM, LRECL and BLKSIZE). DFSORT also accepts empty VSAM data sets for sorting and copying provided VSAMEMT is in effect. For non-VSAM data sets, DFSORT examines the DS1LSTAR field in the format-1 DSCB to determine whether the data set is empty or null. If DS1LSTAR is zero, DFSORT treats the data set as empty or null. If the data set is a null multivolume data set and the DS1IND80 flag is off in the format-1 DSCB of the first volume of the multivolume data set, DFSORT opens the data set for output to force an end of file (EOF) mark before using the data set for input.

Note that a null data set is one that has been newly created, but never successfully closed. Null data sets cannot be processed successfully for a tape work data set sort. The "System Code" field in the data set label in the disk Volume Table of Contents (DSCB in the VTOC) indicates a data set created by the VSE operating system if it contains the letters DOS or VSE within it. Such data sets are never treated as null; however, they may be empty. DFSORT cannot process VSE disk data sets that do not have DOS or VSE within the System Code field.

DFSORT may set what it considers to be appropriate values for missing attributes (RECFM, LRECL, BLKSIZE) of input data sets based on other attributes, or may terminate due to a missing attribute. If a missing attribute results in termination, or you don't want to use a missing attribute set by DFSORT, specify that attribute explicitly (for example, specify RECFM=VB).

See Data set considerations for additional considerations.

The following rules apply to concatenated data sets:
  • RECFM must be either all fixed-length or all variable-length for the data sets in the concatenation.
  • BLKSIZE can vary. However, if a tape data set has the largest block size and is not first in the concatenation, you must specify BLKSIZE explicitly on its DD statement in the following two situations:
    • Blockset is selected and the tape data set has a block size greater than 32760 bytes, but the block size is not available from DFSMSrmm or ICETPEX.
    • Blockset is not selected.
  • With fixed-length records, LRECL must be the same for all data sets. With variable-length records, LRECL can vary. However:
    • If Blockset is selected: If a tape data set has the largest LRECL and is not first in the concatenation, you must specify LRECL explicitly on its DD statement if the LRECL is not available from DFSMSrmm or ICETPEX.
    • If Blockset is not selected, the first data set in the concatenation must have the largest LRECL (LRECL can be specified explicitly on its DD statement).
  • If the data sets are on unlike devices, you cannot use the EXLST parameter at user exit E18.
  • If Blockset is not selected and BSAM is used, all null data sets must precede all non-null data sets; otherwise, the results are unpredictable.
  • DFSORT forces an EOF mark on all null data sets whose format-1 DSCB DS1IND80 flag is off before using BSAM to process the null data sets.
  • If you define a data set using the DUMMY parameter, do not concatenate other data sets to it; the system ignores data sets concatenated to a DUMMY data set.
  • VSAM data sets must not be concatenated (system restriction).
  • Input cannot consist of both VSAM and non-VSAM data sets.
Parent topic: SORTIN DD statement