[2620] in SIPB_Linux_Development
Re: MIT Mailhubs Linux-Athena mail from outside net-18
daemon@ATHENA.MIT.EDU (Ron Hoffmann)
Sun Feb 14 11:50:15 1999
Date: Sun, 14 Feb 1999 11:49:56 -0500
From: Ron Hoffmann <hoffmann@MIT.EDU>
To: warlord@MIT.EDU
Cc: postmaster@MIT.EDU, linux-dev@MIT.EDU
In-Reply-To: <sjmiud5546u.fsf@datkins.ihtfp.org> (message from Derek Atkins on
14 Feb 1999 11:40:41 -0500)
From: Derek Atkins <warlord@MIT.EDU>
Date: 14 Feb 1999 11:40:41 -0500
FYI, I tracked down what's going on... This is what sendmail.cf says:
DDMIT.EDU
[snip]
R$* $#ether$@ATHENA.$D$:$1 All else to ATHENA
Of course, 'athena.mit.edu' mx's to pca and ssa. We could change this
to outgoing, but then shouldn't _all_ clients point to outgoing?
Better yet, why don't all clients send all non-MIT mail directly,
anyways?
OK, lets back up. Now I'm not certain what the right thing
is here.
The "athena clients" (if we mean cluster machines) are on campus
and that's different.
Also, we deliberately requested that those machines continue to
use "athena" as the relay so we continue to have a handle on them
and could remap that name as necessary to optimize service.
The reason we don't have "athena clients" deliver their own mail
is to avoid queueing on the workstations such that private email
(yeah, electronic postcards) isn't left in the queue for the next
user to examine.
Lastly, it's clear that ssa and pca should exhibit the same
behavior and they apparently do not. We'll get that fixed. I
would say for now to leave the redhat-based machines as they
are, using the "athena" sendmail.cf .
