<snip>
When the sdwaname is not present and sdwarbad is,
Would chaining the rb get you to where your program called the service that 
abended
</snip>

Surely you understand that that question is not answerable with a "yes" or 
"no". The RB chain is what the RB chain is. RB's have CDE's in pretty well 
known circumstances. Services can get called in many ways. Programs can get 
invoked in many ways. The answer is "maybe" or "sometimes".

<snip>
I have a number of CSECT in my main module
I always do a link eploc= to every CSECT
First thing I do at every CSECT is establish
An Estae passing parms such CSECT name
And retry address
</snip>

This is an extremely inefficient approach. It is also very inflexible. Is it 
viable? Sure.

<snip>
Is your code a good candidate for an ARR?
</snip>

Likely no, but could be a candidate for IEAARR.

As to whether using the RB chain to do much of anything at runtime (including 
recovery) is a good idea, I think that it's usually not (aside from SVC 
routines locating caller data, for example). Doing that is less clear than 
putting the data you want in a place you fully control (and would stop working 
if you changed your protocol).  SDWARBAD contains what it is documented to 
contain. RB chains can get complicated, especially when IRBs and/or recovery is 
involved.

Peter Relson
z/OS Core Technology Design


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

Reply via email to