> -----Original Message-----
> From: IBM Mainframe Discussion List On Behalf Of Patrick O'Keefe
>
> On Wed, 1 Mar 2006 14:03:15 -0600, Chase, John wrote:
>
> >...
> >regular production FTP job gives a corrupted ASCII file about 40% of
> >the time.
> >
> >This is going to be one of those "fun" problems to document.... :-(
> >...
>
> Can you run a packet trace during every transfer until you've
> traced a failure? It sure sounds like you might have a bug
> in the the client or server or TCP/IP stack, and it's a good
> bet that the bug makes itself felt during exceptional
> conditions - retransmissions, duplicate ACKs, etc. The file
> would not be corrupted unless you ran into one of those
> exceptional conditions.
It's beginning to look like that. We've "played with" FTP-ing the file
about a dozen times today, all without corruption. Of course the
Production job runs just before midnight....
Oh, BTW, using the JCL suggested by Brian Peterson I found that the
subject file did not have any embedded short blocks.
> I don't know of any way to trap this except to have the
> packet trace of each transmission.
That's probably what we'll end up having to do, unless I can "get lucky"
during the day.
-jc-
----------------------------------------------------------------------
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
