JCL error to create partitioned data sets

0
votes

Here is my JCL command

enter image description here

enter image description here

someone please help me why i got this error message, i got so frustated

2
mainframezosjcl
The notify (2nd line) is to far to the right. Shift it left 4 chars - Bruce Martin
Please copy the Text into the Question !! - Bruce Martin
Also, try typing in JEM in the ISPF editor. If you access to it, it will syntax check the JCL for you - SaggingRufus

2 Answers

2
votes

You need a blank after the JOB keyword.

//TUTOR001 JOB (123),.....

Because the submit program did not find a JOB card it generates a default JOB statement for you.

1
votes

There might be a few issues here, depending on what you're trying to do...

First is the issue mentioned by Fritz - you need a space after "JOB". The TSO submit command parses the JCL you submit, and if it thinks there's no JOB statement, it automatically generates one for you based on the information associated with your TSO session. You can see in the JCL that this is the case.

A little tidbit of information here is that if you're happy with the JOB statement generated by SUBMIT, then you don't need to include one in your JCL...no reason your JCL couldn't just start with the // EXEC PGM=IEFBR14 line. Sometimes this is done so that different users can submit the same JCL without needing to alter the JOB statement information.

Second, your question says that you're trying to create a partitioned dataset, but what you have coded is a sequential file. If you really mean to create a PDS, then you'll need to make two simple changes:

  1. Change the DSORG from PS (sequential) to PO (partitioned)
  2. Add a directory block count to the SPACE...you have (1,1) - that's one track primary allocation, and one track secondary allocation. This would need to have a third number for the number of directory blocks to allocate, such as (1,1,1). If you don't know how many to specify, a good rule of thumb is that you can have about four PDS members for every directory block.

One final comment is the RLSE...since IEFBR14 doesn't actually open the dataset you just created, RLSE doesn't really do what you expect. The typical use of RLSE is for programs that create files of varying sizes...you tend to set the allocation to the largest you expect, and RLSE trims back to the nearest extent should you write less.