Peter, I went to the section that you are reading and see that IBM uses the term model DSCB most of the time and uses the term model dataset once. What you are actually doing is described in the section Referring to a Cataloged Dataset. Assume you have a GDG called MY.GDG.DSN that is cataloged in SYS1.USERCAT1 on volume CATVOL1. Your model DSCB would be a dataset called MY.GDG.DSN on volume CATVOL1. Notice the model DSCB name exactly matches the GDG base name and only can be used for one gdg. When creating a new generation, you would use the following DD: //SYSUT2 DD DSN=MY.GDG.DSN(+1),DISP=(NEW,CATLG), // UNIT=DISK,SPACE=(TRK,(1,1)), // DCB=(LRECL=80,RECFM=FB) Note that no dataset name was specified in the DCB= parameter. The system will look on the volume where the dataset will be cataloged to see if the model can be found.
I have never seen anyone using model DSCBs. They have always used the referring to a cataloged dataset method and you only need one dataset to use as a model instead of one for each GDG. The cataloged dataset can also reside on any volume. Since System Determined Blocksize and SMS has been around, I have tagged a default dataclass with almost no attributes on any GDG allocation that did not specify a dataclass in their jcl and encourage people to omit the dataset from the DCB= parameter. This works for SMS-managed and non SMS-managed GDG datasets going to dasd or tape. I apologize for using the term model dataset but I thought it was the correct term. I recommend that you cross out this section with a comment to see referring to a cataloged dataset. Regards, John ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
