-------------------------<snip>---------------------------
we faced a strange problem with a system code S106, as response to a
LOAD macro call.
Although the LOAD was not successful (I/O error on the PDS), the LOAD in
error added an entry into the CDE with the requested module's name, and
an entry point address X'80000000'.
This seems to be normal behaviour; the entry is made before the actual
processing of the PDS member. Even the AMODE is already known to be 31
(from the dictionary), so the entry get's the address X'80000000'.
The S106 error is captured by our own error processing, a message is
produced (including alert mail), and processing continues (IMS region).
Later on, the same module is to be fetched again. We do this by lookup
of the CDE, and now the problem is: we find an entry in the CDE, get the
address from there (which is X'80000000') and try to call the module at
this address, which, of course, leads to another error, in this case S0C1.
My question is:
what do you think about the strange CDE entry? Shouldn't it be removed,
following the S106?
-------------------------<unsnip>----------------------
My first thought is that the invalid CDE should be removed by LOAD
processing if the LOAD fails. But there's also a good case for fixing
the problem that led to the S106 Abend as well.
Rick
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
