(I was on vacation; I cannot speak for why the consoles team has not 
responded).

OA36338 was mentioned.

But beyond that, any thought about "during IPL" vs "after IPL" will not 
apply.

Exit processing in this case is
-- anything defined via the PROGxx processed during IPL is "in limbo" for 
a while.
The exit is considered implicitly defined. 
-- once the exit is defined (by the consoles code in this case), the exit 
routines are loaded.
If there is a problem doing so, I believe that a message is written.

If the display indicates that the exit routine is active (and in your case 
has no limiting jobname or stoken), then if the 
CSVDYNEX REQUEST=CALL is made the exit routine will get control.

I therefore conclude either that the CSVDYNEX REQUEST=CALL is not made 
(which may relate to OA36338) or that the exit routine is not getting far 
enough to issue its WTO. I do not know the default for this exit with 
respect to "ABENDNUM" and "ABENDCONSEC" (for example, "disable the exit 
routine after N consecutive abends"). The exit defines these but the exit 
routine add can specify values for that particular exit). A typical 
approach is "disable upon two consecutive times that the exit routine did 
not return normally to the caller". But some implement "never disable" 
(leaving it up to the exit routine to handle its own errors). The default 
behavior for a specific exit should be documented.

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