Read syntax diagramSkip visual syntax diagram


This suboption is used to specify the number of hours elapse time, inside of which invocations of Z Abend Investigator, using the same history file, can be deemed duplicates of a previously recorded fault. This mechanism is the standard duplicate fault detection mechanism used, for example, for abending batch jobs.

The valid range of hours is 0 - 168 (168 is equivalent to one week). Specification of 0 hours means that "NORMAL" duplicate faults are not detected.

The default elapse time is 24 hours.

A fault is considered a duplicate of another already recorded in the same history file if the faults occurred within the specified elapse time (hours), the point-of-failure has been determined, and the following fault details are identical:
  • Initial abend codes
  • CICS® transaction IDs (applicable to CICS transaction faults only)
  • Point-of-failure module names
  • Point-of-failure CSECT (or program) names
  • Point-of-failure offsets from start of the CSECT (or program)
  • Point-of-failure module link-edit date and time (if available for the current fault as well as in the history file entry for the duplicate fault)
  • User titles.
  • The jobname (only if the Jobname suboption is specified; otherwise, the jobname is not included in the duplicate fault determination)

NoDup(NORMAL) signatures are based on the existing fault entries in a history file.

When a fault is deemed a duplicate of another based on the fault characteristics and the NoDup(NORMAL(hours)) option in effect, the default is to suppress both the writing of the history file entry and any dumps for the duplicate fault. However, the final decision on suppression can be overridden by an End Processing user exit.

When it has been determined that the fault is a duplicate of another fault, the duplicate count that is associated with the existing fault is incremented by one and message HFZ0044I is issued.