A new generation can be passed when created. That generation can then be cataloged in a succeeding job step, or deleted at the end of the job as in normal disposition processing when DISP=(,PASS) is specified on the DD statement.

However, after a generation has been created with DISP=(NEW,PASS) specified on the DD statement, another new generation for that data group must not be cataloged until the passed version has been deleted or cataloged. To catalog another generation causes the wrong generation to be used when referencing the passed generation data set. If that data set is later cataloged, a bad generation is cataloged and a good one lost.

For example, if A.B.C(+1) is created with DISP=(NEW,PASS) specified on the DD statement, then A.B.C.(+2) must not be created with DISP=(NEW,CATLG) until A.B.C(+1) has been cataloged or deleted.