The input data set for IEBGENER, as specified in SYSUT1, can contain fixed, variable, undefined, or variable spanned records. You cannot use concatenated data sets with unlike attributes except for block size or device type as input to IEBGENER. The rules for concatenating data sets on SYSUT1 are the same as those for BSAM unless you are concatenating data sets on DASD and tapes. Then the concatenating rules are the same as for QSAM.

If the SYSUT1 data set is in the system input stream and contains JCL statements, code SYSUT1 DD DATA and not SYSUT1 DD *.

Block size for the input data set must be available in the data set label (DSCB), or tape label, or the DD statement. For a tape data set the block size can exceed 32␠760 bytes.

The default record format is undefined (U) for the input data set. Record format must be specified if the data set is new, undefined, or a dummy data set.

The input logical record length must be specified when the record format is fixed blocked, variable spanned, or variable blocked spanned, or when the data set is new, or a dummy data set. If the input and output logical record length are not available and the record format is V or VB, IEBGENER sets the maximum record length to the block size -4, with a maximum of 32␠760. In all other cases, a default logical record length of 80 is used.

A partitioned data set or PDSE cannot be produced if an input data set contains spanned records.

If both the SYSUT1 and the SYSUT2 DD statements specify standard user labels (SUL), IEBGENER copies user labels from SYSUT1 to SYSUT2.