BLOCK CONTAINS ZERO RECORDS

ssowjanya · New User Joined: 29 Apr 2005 Posts: 13

Hi,
What does it mean....
block contains zero records.
thanx,
sowjanya

maverick05 · New User Joined: 14 Apr 2005 Posts: 54 Location: Earth

Hi,
When we specify BLOCK CONTAINS 0 RECORDS ,the system will decide the optimum size for the file based on the device used for storing the file.
It can be understood as initialisation of a block.

mmwife · Super Moderator Joined: 30 May 2003 Posts: 1592

Hi Maverick,

Although one can do what you stated when BCZR is used in a COBOL pgm, it is not the only thing that can be done.

What BCZR does is take the COBOL pgm out of the DCB merge chain for the BKLSIZE param. Otherwise, the blksize in the pgm (or the default of the rec len) would be used to override what might have been provided in the dd stmt or the DSCB/label/catalog.

So, for new datasets, you could use use your own blksize, do what you suggested or let SMS (if your site has it) do it. If you don't use BCZR the the blksize in the pgm will be used.

Rajen · New User Joined: 05 May 2005 Posts: 61

When we write BLOCK CONTAINS 0 RECORDS
Whichever is appropriate will be taken by the system,but ususal practice is like we take care of Block Size in JCL code,which is very important.

Thanks,
Raj

mmwife · Super Moderator Joined: 30 May 2003 Posts: 1592

I should have mentioned this in my previous post. The approach mentioned by Maverick, while not the only approach, is the preferred approach because the system calculates the optimum blksize to use for the device assigned to the dataset.

This is very important since new device types are introduced with differing track capacities. BLKSIZE=0 in the DD stmt allows reallocation of the ds to the new device type without JCL changes, i.e. it is transparent to the application.

You must code both "BLOCK CONTAINS 0 RECORDS" in the FD of dataset and BLKSIZE=0 in the DD stmt to get the system to do the blksize calculation.