Wayne Driscoll's responses were all on target.

What am I confused about is why you thought you needed wait/post, 
particularly since the interface provides no way for you to identify the 
ECB.

>I read somewhere that storage reference by an SRB had to be in common 
storage.
I surely hope this was not in IBM documentation. It is certainly untrue.

>(I thought the task recovery routines would create a dump but they 
didn’t)
If your recovery routine got control and asked for a dump, then a dump 
would be created.
So one quickly concludes that your recovery routine did not get control.
SRB to task percolation of abends does not apply to a SYNCH=YES SRB since 
the completion status of the SRB can be ascertained from the available 
IEAMSCHD keywords.

Your example and FRR show you are coding for a non-reentrant program. 
While not intrinsically a problem, certainly atypical.

>RETREGS=YES,FRESDWA=YES, 
The first is unnecessary (and has no effect) for an FRR. The second is 
simply not applicable to an FRR.

>DUMP=YES
This has no effect for an SRB (it applies only to ESTAE-type recovery)

SDATA=(NOSQA,RGN,CSA) 
A dump without SQA is unlikely to be helpful (and possibly not even 
readable). Much of the data needed to analyze a dump is in SQA (such as 
the ASCB/ASSB).
Similarly a dump without LSQA is unlikely to be helpful (you'll probably 
find that your site has set things up so that dumps automatically get SQA 
and possibly LSQA.
I don't know if your "NOSQA" would override that.

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