Defining the debug data set (SYSDEBUG)
When you compile from JCL or from TSO and specify the TEST(...,SEP,...) compiler
option, the symbolic debug information tables are written to the data
set that you specify in the SYSDEBUG DD statement.
About this task
//SYSDEBUG DD DSNAME=dsname,UNIT=SYSDA
The SYSDEBUG data set can be a sequential
data set, a PDS or PDSE member, or an HFS file. For details about
how to specify the record format, record length, and block size of
the SYSDEBUG data set, see the related reference
below about logical record length and block size.
Language Environment® uses SYSDEBUG for
its dump services. If the TEST|NOTEST(...,SEPARATE(DSNAME),...) compiler
option is in effect, the SYSDEBUG dataset name is
stored in the object program and is used as the default at run time. You
can change the name of that data set at run time by using the SYSDEBUG COBOL
debug file user exit, IGZIUXB. You can direct IBM® z/OS® Debugger to a renamed data set
using the SET DEFAULT LISTINGS command, user exit EQAUEDAT,
or the EQADEBUG DD statement.
The dataset name
that you specify in DDNAME SYSDEBUG might be used
by several IBM products, including Language Environment, IBM z/OS Debugger , Fault Analyzer, and Application Performance Analyzer. For details, see the documentation
of those individual products.
Language Environment Customization (Modifying the COBOL debug file name)
z/OS Debugger User's Guide (How does Debug Tool locate COBOL and PL/I separate debug files)