IDI0500E I/O ERROR ON DFSURWF1 FILE.

Document ID : KB000019290
Last Modified Date : 14/02/2018
Show Technical Document Details

Description:

Executing Secondary Index Builder to rebuild primary and secondary indexes after recovering a main database.

FUNCTION=INDEXCREATE,DBDNAME=dbdname,INDD=DFSURWF1

The job is failing with error message IDI0500E I/O ERROR ON DFSURWF1 FILE
and also the SORT E35 exit routine terminates with messages
WER045C END SORT PHASE
WER135I TASK CALL/E35 TERMINATED PREMATURELY

How do we rebuild the indexes in this situation?

Solution:

The DFSURWF1 file defines the work file data set where records are written to build secondary indexes and resolve pointers in logical relationships.

FUNCTION=INDEXCREATE,DBDNAME=dbdname

This example does the following:

  • Scans the indexed database that has the DBD name specified.

  • Extracts information for every index. If the indexed database is HIDAM, also extract primary index information.

  • Sorts the extracted information. The input to the sort is passed using an E15 exit.

  • Builds each index. The indexes are built using an E35 exit.

  • For every index defined for the named DBD that had no source segments, creates them as empty indexes.

The E35 exit can not build the index because the default input file for Secondary Index Builder is a file in SIB file format.
If the input files are DFSURWF1 you must tell SIB that by specifying FILEFORMAT=DFSURWF1.

For the INDEXCREATE function in CA Secondary Index Builder, this control statement applies only when INDD is specified.
This control statement has the following format:

FILEFORMAT= DFSURWF1

Uses a file format that is compatible with the IMS work file format. Specify this option for the following functions:

  • An extract function where the output file is used by other utilities such as prefix resolution.
  • A build function where the input file was created during the initial load or reorganization reload of the indexes database. For a build operation, this option results in a default of SORT=YES, which sorts the file first.