I think the problem has been found....
 
Darren
-------------------------------------------------------------------------
From VMXA-L:
 
On Friday (January 26) we had a severe problem with our RSCS 1.3 running
under VM/XA-SP 2, apparently due to a bad CP fix on PUT 8904.  This
problem caused files sent through our RSCS 1.3 machine (UKCCB) to be
transmitted and marked as purged, but they were not actually purged.
The next time the file queue was reordered, such as when a line was
restarted, RSCS would "discover" them and queue them to be transmitted
again.
 
UKCCB is our main BITNET node and files were sent multiple times to
UCCCVM1 and points north and to UTKVM1 and points south, as well has to
a number of instate nodes.  This probably caused a lot of duplicate mail
to be sent all over the network.  We cut off the links in all directions
for several hours while we worked on the problem; we finally reloaded
without the 8904 fixes.  We're sorry for the inconvenience.
 
The fix to watch out for is APAR VM38979 and PTF UM11628.  We're still
working with IBM to actually fix the problem.  Our RSCS 2.3 nodes did
not seem to be affected.  We are (still) planning to convert our BITNET
links to the 2.3 system Real Soon Now.
 
                                    Herman Collins
 
128 McVey Hall                      Bitnet: SYSHERM@UKCC
University of Kentucky              Internet: [log in to unmask]
Lexington, Ky.  40506-0045          Phone: 606-257-2256