@Jim, not much faith in LE dumps, eh? <g>
I guess I will work on duplicating the problem with the LE dump disabled in
favor of SYSUDUMP.
Meanwhile, here is LE's take on the issue: (zero editing other than where I
say "snip")
Condition Information for (DSA address 000236E0)
CIB Address: 00024A30
Current Condition:
CEE3250C The system or user abend S202 R=00000000 was issued.
Location:
Program Unit: Entry: DIRECTOR Statement: Offset: +000000B6
Machine State:
ILC..... 0004 Interruption Code..... 0011
PSW..... 070D1400 B1012472
<snip>
ABEND code: 00202000 Reason code: 00000000
Storage dump near condition, beginning at location: 31012462
+000000 31012462 10000A6B 41000001 41106058 13110A01 07004510
C53C0000 00205810 10000A6B |...,......-.........E..........,|
You can pretty well see that the offending instruction is an SVC 1.
I know that LE makes some funny decisions about what is and what is not
worth dumping, but I have trouble believing they "relocated" the ABEND from
the POST to the WAIT.
By the way, I don't know which POST and ECB was the problem. There are three
ECBs in the ECBLIST, and the POST might have been for either of two of the
three.
Charles
-----Original Message-----
From: IBM Mainframe Discussion List [mailto:[email protected]] On
Behalf Of Jim Mulder
Sent: Wednesday, September 28, 2016 8:49 PM
To: [email protected]
Subject: Re: Interpreting an S202-0000
> It is definitely, unquestionably being reported on the WAIT, not the
POST.
>
> I don't know what is in the ECB -- or even which of three ECBs is in
error
> -- because LE in its wisdom decided the storage is not worth dumping.
>
> An interesting question would be "if the ECB has an invalid RB
> address,
how
> did POST know what task to ABEND?"
Since I know of no way for WAIT to issue a 202 abend, I wouldn't believe
that without seeing it in a dump myself (a dump taken by z/OS with a system
trace - not some LE concoction of a dump).
POST would abend the work unit which is doing the POST.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
