>> How grown application written in ASM and C. Works fine since years, except 
>> for some "out of socket descriptor" problems every now and them. I was 
>> asked to help finding the cause of S23E or S53E (both DETCH) abends since 
>> the developer tries to fix the above "ourt of socket descriptor" problem. 
>> 
 >
​>What is the code in R15? Just as a guess, for the S23E, I'd bet it's x'08' 
>which says "tasking being DETACH'd is not a subtask of the task doing the 
>DETACH". 


Sorry for having forgotten to write this. 


The S23E had reason 00, meaning "The protection key of the address does not 
match the key of the issuer of the DETACH." Interesting, isn't it?


The S53E doesn't have different reasons. One of the possible causes listed in 
the manual is: "The protection key of the address does not match the key of the 
issuer of the DETACH." Seems likely since I see an S202 in the system trace 
before. I'm not in the office now, and I can't remember the S202's reason code 
for sure. I think it was 00, meaning "The system found an incorrect address for 
a request block (RB) in the 3 low-order bytes of the ECB specified by the 
problem program."


I know the program does subtasking via ATTACH and DETACH. Unfortunately, I only 
have an vague idea what the program does but don't have a clue (yet) how it 
exactly works. In fact there seems to be no-one left in the company who does. 


I would open a PMR with IBM asking for help, but thanks to great management, we 
are not allowed to send dumps. Therefore I have to get at least a basic 
understanding to be able to describe the symptoms good enough before I can even 
think of opening a PMR.


--
Peter Hunkeler


 

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to