On Thu, 30 Oct 2025 15:28:11 +0200, Binyamin Dissen 
<[email protected]> wrote:

>It is called debugging.

I agree. The OP needs to do basic debugging. The OP was told how to take a dump 
and that it doesn't occur from TSO TEST. Instead, he must call the program from 
TSO, batch or console dump to obtain the dump. A SLIP S0C4 dump would be best 
but explaining how to obtain it may be a little tricky.

>CALLRTM with 0C4? Doesn't sound likely.

The OP is giving us a WSAG instead of looking at the trace table to verify it's 
a CALLRTM S0C4. The trace table should be available using IPCS DSN ACTIVE but 
you have to worry about wrapping.

I agree this is unlikely but don't forget this involves TSO, TSO TEST and 
IEWBIND. Each having unusual recovery logic. Unlike typical address spaces, TSO 
might be passing the abend to another TCB. Unlikely, but the OP could easily 
verify before leading us on a wild goose chase.

>Do W to find the PSW.

TSO TEST might not be best for debugging IEWBIND abend. IEWBIND is outside the 
control of TSO TEST and it's recovery may be masking something from TSO TEST 
(e.g. freeing important storage before TSO TEST receives control). TSO TEST 
seems to be using IEWBIND internally which could affect how it functions in the 
OP's program. The OP could easily eliminate TSO TEST as part of the problem by 
using TSO LOADGO. Additionally, the trace table is not easily available from 
TSO TEST and it runs the risk of wrapping depending upon the workload.

If TSO TEST isn't providing a clear picture, then a dump will be needed. 
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to