In following JCL, datasets on FMIN01X DD have same DCBs and the datasets are defined on PST with same DCB, (RECFM=VB,LRECL=13604,BLKSIZE=28200).
//DATA01 DD DISP=SHR,DCB=(RECFM=VB,LRECL=13604,BLKSIZE=28200),
// SUBSYS=(ZSAM,SHRK,SUPEREXP,DATA01X)
//DATA01X DD DSN=ABC0010.EFD03,DISP=SHR
// DD DSN=ABC0010.EFD04,DISP=SHR
// DD DSN=ABC0010.EFD05,DISP=SHR
// DD DSN=ABC0010.EFD06,DISP=SHR
In this case, is SUBSYS specification required ?
Is it true that SUBSYS specification required because I/O error occurs on the first dataset of concatenated datasets without SUBSYS ?
Release : 5.5
Component : CA Compress Data Compression
Under some circumstances, concatenated SUBSYS data sets encounter I/O errors after the first data set.
To avoid this problem, code the concatenation on an associated ddname and refer to the concatenation with a single SUBSYS ddname. The JCL in the issue is one of the examples.
This method is the safest and most efficient because it avoids multiple OPENs by SAM-SI.
If the files are defined in the Compress Table then no SUBSYS DD is required because the dynamic functions of ZSAM will intercept the open to continue the process of compression/decompression.