>Our LISTSERV machine has received many MANY copies of the following
>X-DEL jobs from LISTSERV at NCSUVM.
>
>  X-FOR JAG8B6E3@PANAM SIGNOFF * FWDED=2 (NETWIDE
 
It appears that what happened a couple of weeks ago down here  has happened
again. The original distribution (i.e., when LISTSERV distributes the job to
itself and everyone else) of a similar job had the entire DIST2 and paths
appeneded to it (basically, a nested DIST2). So, every LISTSERV ended up
distributing it to every other LISTSERV.
 
There is likely little to be done for the net's benefit at this point since it
was the initial distribution which marked the start of it-- that job has likely
long since gotten everywhere. You can save your own cycles by catching them,
but I imagine most of the jobs being passed around now do not have further
distributions, unless you're a core site.
 
See the logs of LSTSRV-L, where I and Eric described the problem. I don't
know what the cause of the apparently corrupt job is.
 
These two jobs, as far as I can tell from my log, have two things in common:
 
  1. They are both from users at PANAM or PANAM1. This may easily be
     coincidental, as the manager there sends all his X-DEL's to TAMVM1 and
     there are certainly many of these jobs from PANAM which cause no trouble.
     After the previous problem, BURNETT@PANAM explained his procedure to me
     and there didn't seem to be any flaw.
  2. They both contained (in the original version) "LISTSERV@UGABUS" in the
     job such that LISTSERV tried to interpret "LISTSERV@UGABUS" as a command,
     right before it redistributed the job.
 
I don't know from where "LISTSERV@UGABUS" is coming or why it got in the job.
 
--Shane