> The overt issue I am having is a S0C1 after doing a POINT macro when the > data is on the second volume of a multi-volume (BSAM) dataset. When I > first started looking at the issue, I read that an EXTENDED striped dataset > would act as though it were on a single volume to the program. I would > have to add BLOCKSIZE=LARGE to my DCBE and handle the NOTE/POINT data > correctly, but that was about it as far as changes went. > > However, when I started looking at the data returned from NOTE, I am > getting values of 1, 2, 3, etc. which doesn't quite jive with the > documentation (I can't believe everything is on track 0). However, when it > (apparently) goes to the second volume, the return value from NOTE starts > over at 1. > > I'm at a loss. Everything I read says this is supposed to work and I am > sure I have something missing, I just can't figure out what.
With BLOCKTOKENSIZE=LARGE on the DCBE, the tokens are not TTRs. They are simple a relative block number. A striped extended format dataset (i.e. stripe count > 1) appears to a BSAM user as if it was a single volume. The note values are sequential regardless of which physical volume contains the block. For a non-striped multi-volume extended format data set (i.e. stripe count = 1), the NOTE values start over for each volume. The program is responsible for positioning to the correct volume. Jim Mulder z/OS System Test IBM Corp. Poughkeepsie, NY ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
