Date : Mon, 20 Dec 1982 17:12:39-EST (Mon)
From : Michael Muuss <mike@BRL>
Subject: On UUCP <-> ARPA Mail Traffic
"The ARPANET is an operational DoD network and is not intended to compete with
comparable commercial service. Accordingly, before ARPANET service is
provided to any non-U.S. Government activity, it must be determined that
adequate comercial service is not available."
"The ARPANET is intended to be used solely for the conduct of or in support
of official U.S. Government business. ...use of the ARPANET must not violate
any applicable privacy laws." [pg. 12, ARPANET Directory]
"If it is possible to gain access to the ARPANET from another network
(gateway) ..., it is the respnosibility of that host to provide software
protection which will permit only authorized ARPANET users to access
the network." [pg 18, ARPANET Directory]
The only reason we are able to provide any kind of relaying service AT ALL
is because of a ambiuity of wording in the regulations governing the
use of the ArpaNet. Basically, they say that outside users can not
ORIGINATE traffic on the net, but say nothing about outsiders getting
COPIES of things sent on the net. So, our mail gateway software implements
this policy, with one special case. It works like this:
*) Mail sent from ARPANET to any of our other networks is passed.
*) Mail sent from any other network to ARPANET is rejected, with explanation.
*) Mail sent from any other network to a MAIL FORWARDER on our machine is
DELIVERED, regardless as to where it forwards to. Hence, mailing to
...!brl-bmd!info-cpm@mit-ai
is rejected, as it asks for explicit routing to ARPANET, where as
...!brl-bmd!info-cpm
is DELIVERED, because we automatically forward mail for that list.
Believe me, I wish this was not so, but we have to play by the rules.
Otherwise... "If corrective action is not taken in a reasonable time,
DCA reserves the right to disconnect the host/terminal from the network."
[pg 14, ARPANET Directory]
If you wish to further discuss this issue, the proper forum is in the
TCP-IP Digest ("The InterNet Digest"). It is fed into NETNEWS as net.tcp-ip;
ArpaNet viewers may request direct mailing by sending a request to
TCP-IP-REQUEST @ BRL
Letters for publication in the digest should be sent to
TCP-IP @ BRL
Network Liaison and Host Administrator of BRL-BMD,
-Mike Muuss
PS: Attached is an automatic rejection notice, and a letter on this subject.
----- Forwarded message # 1:
Date: 18 Dec 1982 22:13-EST (Sat)
From: Memo Service (MMDF) <mmdf@BRL-BMD>
To: anton@hcr
Cc: Gurus@Brl-Bmd
Subject: Illegal Address (pourne@mit-mc)
Via: (decvax); 18 Dec 82 22:14-EST
Your letter has been intercepted trying to access
a restricted access host (e.g. an ARPANET host). A copy
of your letter has been sent to the system administrators.
The text of your letter follows.
--------------- Returned Mail Follows --------------
>From hcr!anton Fri Dec 17 11:47:45 1982 remote from decvax
Date: Fri Dec 17 11:01:15 1982
To: decvax!brl-bmd!pourne@mit-mc, info-cpm@BRL
Subject: somebody gotta DO something
Cc: rrg, utcsrgv!utcsstat!geoff, utzoo!henry
I have just got a nasty letter from an automatic mailer telling me
I have been reported for attempting to breach the security system
of ARPAnet. All I did was reply to dave Flystra's note about
sending all new members of the info-cpm list a idots guide to
things like being taken off the list.
I very much resent being accused of trying to break a secure system
when all I did was use the unix mailer's 'r' command.
Yes, Jerry, this junk about routing is getting pernicious.
IT MUST BE STOPPED. Abouyt 30% of my network replies are getting sent back now.
The only simple way out seems to be have ARPA and usenet seperate and
have all of usenet that relays news FORCED to relay mail. System
administrators won't like that, so we will probably need intelligent mailers,
which means everyone has to keep their tables up to date.
Sigh !
/anton aylward Human Computing Resources. Toronto.
(p.s. I hope this DOES get through....)
--------------- End of Returned Mail ---------------
----- End of forwarded messages