On Wed, 10 Oct 2012 09:05:39 -0400, Gerhard Postpischil wrote:
>
>My recommendation for a documentation change is based on the
>fact that users of BSAM have the ability to write meaningful
>code using short records; just the ability to write a variable
>number of null intermediate segments offers itself for nefarious
>uses not readily detectable without source or data inspection.
> 
Somewhat reminiscent of past DoS attacks using ping with
oversized ICMP packets.

I'd go with your earlier suggestion that QSAM should report as
a data error any SDW of 0004xxxx other than 00040000, and
return a no-data record for the latter.  The reporting mechanism
must exist.  What does QSAM do on encountering an RDW/SDW
with count exceeding BLKSIZE?

-- gil

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

Reply via email to