On Tue, 25 Oct 2022 08:23:53 -0500, Paul Gorlinsky wrote: >Like so many of the things IBM does, it is a left over from when we had >several different DASD types going all the way back to OS/MVT/MFT days... 3120 >was optimum for any... > How about an RFE or even an SR to allow SDB to operate? For some products IBM has made such a change. What compatibility implications? A programmer might allow BLKSIZE=0 with OUTDSN then subsequently hard code 3120 reading the data set.
I deem this an acceptable risk, outweighed by the benefit. >But today, for the most part, BLKSIZE=0 is OK and the OS will calculate the >proper size for the device you are writing on. > Alas, coding BLKSIZE=0 in JCLor DYNALLOC has no effect. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
