PTRERROR
The PTRERROR keyword is valid with the DBSCAN and UNLOAD commands.
You can specify this option in a global options module or in PLUSIN. The following table describes the keyword.
Table 1. Pointer error keyword
To do this | Specify this in PLUSIN |
|---|---|
Continue after pointer errors (no limit). | PTRERROR(ACCEPT) |
Continue after pointer errors (up to nnnnn errors). | PTRERROR(ACCEPT,nnnnn) |
Abend after a pointer error. | PTRERROR(ABEND) |
Internal default: ABEND | |
The Unload function uses the PCF and PTF (or HF) pointers to move in hierarchical sequence through a database record. If you specify PTRERROR(ABEND) and any of these pointers are bad, the unload terminates. You can use the PTRERROR keyword to bypass these errors. The Unload function can skip the bad segment or a portion of the bad record and continue.
When the Unload function encounters an error, it snaps the block containing the bad database pointer and generates a BMC message. When the unload is finished, you can use this information to fix the problem. For more information, see Bad-Block-Pointer-report.
If you specify PTRERROR(ACCEPT,nnnnn), the Unload function continues until it encounters nnnnn pointer errors. The value for nnnnn can be any number from 1 through 32,767. If you use the default PTRERROR(ABEND), the Unload function terminates when it detects the first pointer error.
You can set the default value for the PTRERROR option by using the Unload function ISPF panels or the equivalent batch job. Sample JCL is in member #DBUGLBL of the sample library.