[2051] in Moira
Re: access service for outgoing mailers
daemon@ATHENA.MIT.EDU (Jonathon Weiss)
Wed Oct 29 03:31:15 2003
Date: Wed, 29 Oct 2003 03:31:09 -0500 (EST)
Message-Id: <200310290831.h9T8V9HQ021493@bearing-an-hourglass.mit.edu>
To: Tom Coppeto <tom@MIT.EDU>
CC: Jonathon Weiss <jweiss@MIT.EDU>, moiradev@MIT.EDU,
Mark Silis
<mark@MIT.EDU>
In-reply-to: "[2047] in Moira"
From: Jonathon Weiss <jweiss@MIT.EDU>
Hi Tom / Mark,
I've got a couple of questions followups:
1) Do I infer correctly from the existance of the "efl-access" file in
the script that valid-user@alum.mit.edu will still be able to send
mail through outgoing?
2) I realized that owner-listname@mit.edu (and several other things
along the same lines for mailman lists) are valid addresses in that
they appear in the aliases file on the mailhub, but are not in the
list of addresses allowed to send mail. I'm not sure if anyone
currently sends mail from one of these addresses, or if you care,
but I figured I'd mention it on the off chance it hadn't already
been considered.
3) Have you had an chance to look through the logs for mail from
{root,mailer-daemon}@machine.mit.edu and come up with a plan, or at
least determine the extent of the problem? (Unfortunately, the
patches I developed for athena for this case blew up in our face.
Once we get our smt-auth code deployed (and I don't know when that
will happen, but I don't think we can make it by the 3rd though it
is in testing on beta machines now) the problem will be somewhat
mitigated (at least for machines that take the relevant patch
release) because they will do direct delivery if they are unable to
do authenticated delivery (read if sendmail doesn't have access to
the user's kerberos tickets)).
Thanks,
Jonathon
