My, was  that a  tricky bug...  LISTSERV had a  file in  HOLD status  in its
reader before  the file  it distributed to  itself. The call  to Diag  14 then
ended up picking info  for that file, which had the SFBXFER  bit turned on. It
then saw a file from itself that had been transferred, and rejected it.
 
  Many thanks to Kerry for finding this out, it wasn't an easy bug... :-) I'll
be sure  to fix that  in 1.5j.  I'll probably be  using subcode X'FFF'  with a
nonzero spoolid, but as a temporary fix you can add a
 
       Call Diag 8,'ORDER R' spoolid
 
before  the  Diag(14,'RSFD',blablabla).  Don't  bother to  do  it  unless  you
encounter the problem though :-)
 
  Eric