I run a list (CLIMLIST) for people working in climatology and related
disciplines.  It is moderated and only the list owner (me) can send to the
list.  CLIMLIST is run from LISTSERV@PSUVM (aka [log in to unmask]).  I
distribute to it by email from Ohio State (magnus.acs.ohio-state.edu).
 
I was very surprised on Monday when an item was distributed to the
list which I had not sent.  It was a copy of last week's Digest and the
'Original sender' warning identified the sender as a CLIMLIST
subscriber in Manitoba who had elected to receive the list in Digest
form.
 
I checked with the subscriber and it turns out that he had just moved
his email base-of-operations from one machine to another and had been
mailing files from his old home to his new one.  He was very
embarrassed to admit that in the middle of the operation he became
confused, typed the word 'mail' and noted that his 'active file' was
mailed out to some location other than the intended one.  He claims he
did not specify an address however.  I am not sure what his mail
system is but the machine at which the mail originated is an IBM 3090
running MVS/XA - reference to an 'active file' might suggest Wylbur
mail or something similar, but that is all I know.
 
My questions (after this long-winded preamble) are as follows.
 
(1)  Why might the mail have been sent back to CLIMLIST@PSUVM?
(2)  Even if this did happen, why did LISTSERV then distribute it to
the list given that CLIMLIST is moderated?
 
With respect to (2), I am frequently contacted by people wanting to send to the
list who have had their postings rejected as 'unauthorized', so I know that, in
general, the restriction on sending works.  I guess the mail must have looked
as though it came from me to LISTSERV.
 
Here are the headers on the Digest as they were in the subscriber's space on
his old host machine.  If his mailer made assumptions about where the mail was
to go, presumably the basis for these assumptions is in these headers.  In
addition, if the mail arrived at LISTSERV looking as if it came from me, the
'From:' line must have been constructed from the information in these headers.
 
One thing I notice is that the 'Reply-To:' points back to the list.  This
is not the way the mailings leave PSUVM - they have a 'Reply-To:'
pointing back to my email address, which is not even at PSUVM (or
even Penn State).  That header is apparently getting rewritten
somewhere on the trip!
 
 
Received: (from VM1.MCGILL.CA for <[log in to unmask]> via BSMTP)
Received: (from MAILER@MCGILL1 for MAILER@UOFMCC via NJE)
         (PSUVM-5539;    2306 LINES); Sat, 13 Nov 93 23:05:11 CST
Received: from VM1.MCGILL.CA by VM1.MCGILL.CA (Mailer R2.10 ptf000) with BSMTP
id 5435; Sun, 14 Nov 93 00:00:33 EST
Date:     Sun, 14 Nov 1993 00:00:00 -0500
Sender:   CLIMLIST Climatology Distribution List <CLIMLIST@PSUVM>
Reply-To: CLIMLIST Climatology Distribution List <[log in to unmask]>
From:     Automatic digest processor <[log in to unmask]>
Subject:  CLIMLIST Digest - 6 Nov 1993 to 13 Nov 1993
To:       Recipients of CLIMLIST digests <[log in to unmask]>
 
I welcome any suggestions about this.  This is the first time something
like this has happened in many years of running the list.
 
Thanks in advance!
 
 
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
 +-+   John Arnfield                 +-+   Department of Geography     +-+
 +-+   BITNET/EARN: AJA+@OHSTMAIL    +-+   The Ohio State University   +-+
 +-+   Internet: [log in to unmask]        +-+   Columbus, OH 43210-1361     +-+
 +-+   Telephone: +1 614 292 7954    +-+   USA                         +-+
 +-+   Fax: +1 614 292 6213          +-+                               +-+
 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+