Selecting a Java dump data set

Specification of a Java™ dump data set to be analyzed is done by first selecting File > Analyze MVS Dump Data Set from the Fault Entry List display.

This option brings up the Analyze MVS™ Dump Data Set display on which the MVS dump data set name can be entered. An example of this display is shown in Selecting a CICS dump data set.

If Java activity is detected in the dump being analyzed, then a history file fault entry is required to be created early while asynchronous DTFJ processing is performed. For this reason, the Create Java Fault Entry display is presented, as shown in Figure 1.

Figure 1. Sample Create Java Fault Entry display
┌─────────────────────────────────────────────────────────────────────────────┐
│ Create Java Fault Entry                                    Line 1 Col 1 76  │
│ Command ===>                                              Scroll ===> CSR   │
│                                                                             │
│ To create a fault entry for this Java dump, specify a history file data     │
│ set name and press Enter, or press PF3/PF12 to cancel the fault entry       │
│ creation and continue the analysis with incomplete Java information.        │
│                                                                             │
│ History file DSN. . . . . : 'NWILKES.HIST'                                  │
│                                                                             │
│ *** Bottom of data.                                                         │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
│                                                                             │
└─────────────────────────────────────────────────────────────────────────────┘

Specify a history file that is managed by the HFZS subsystem and to which you have at least UPDATE access. Press Enter to create a fault entry and return to the Fault Entry List display. A MATCH is automatically performed to show only the fault entry that was just created.

Note: Z Abend Investigator Java DTFJ processing requires that the MVS post-dump exit HFZXTSEL is installed (for details, see Installing the MVS post-dump exit HFZXTSEL) and the HFZS subsystem is started (for details, see Using the Z Abend Investigator HFZS subsystem).

If the DTFJ_Status column has been selected for display in the Fault Entry List, then you can keep pressing Enter to refresh the value and wait for DTFJ_Status to show Finished. By then, the asynchronous DTFJ processing has added information to the fault entry so that a subsequent reanalysis using the I line command will present a complete analysis report that includes all available Java information.

If PF3 or PF12 is pressed on the Create Java Fault Entry display, the early fault entry creation is skipped but the analysis still continues. In this case, the dump analysis is performed as if no Java activity was detected, and the user is prompted to create a fault entry when exiting the analysis instead.