[13] in testers

home help back first fref pref prev next nref lref last post

6.3A: /usr/lib/sendmail

daemon@ATHENA.MIT.EDU (daemon@ATHENA.MIT.EDU)
Sun Jun 25 06:54:37 1989

From: <probe@ATHENA.MIT.EDU>
Date: Sun, 25 Jun 89 06:54:23 -0400
To: testers@ATHENA.MIT.EDU
Reply-To: Richard Basch   <probe@ATHENA.MIT.EDU>

... the following rules certainly did not do what I expected:

R$*<@$+>$*:$*		$#tcp$@$2$:$1<@$4>$3		It resolved, use tcp
R$*<@$+>$*		$#tcp$@$2$:$1<@$2>$3		It resolved, use tcp

The result of a verbose mail is:

    testers-mtg@menelaus.local... Connecting to menelaus.mit.edu (tcp)...
    220 MENELAUS.MIT.EDU Sendmail 5.45/4.7 ready at Sun, 25 Jun 89 05:33:44 EDT
    >>> HELO ODIE.MIT.EDU
    250 MENELAUS.MIT.EDU Hello ODIE.MIT.EDU, pleased to meet you
    >>> MAIL From:<root@ODIE.MIT.EDU>
    250 <root@ODIE.MIT.EDU>... Sender ok
    >>> RCPT To:<testers-mtg@menelaus.MIT.EDU.local>
    250 <testers-mtg@menelaus.MIT.EDU.local>... Recipient ok
    >>> DATA
    354 Enter mail, end with "." on a line by itself
    >>> .
    250 Ok
    >>> QUIT
    221 MENELAUS.MIT.EDU closing connection
    testers-mtg@menelaus.local... Sent

If I change the above rules to:

R$*<@$+>$*:$*		$#tcp$@$2$:$1@$4$3		It resolved, use tcp
R$*<@$+>$*		$#tcp$@$2$:$1@$2$3		It resolved, use tcp

a verbose mail is now:

    testers-mtg@menelaus.local... Connecting to menelaus.mit.edu (tcp)...
    220 MENELAUS.MIT.EDU Sendmail 5.45/4.7 ready at Sun, 25 Jun 89 05:43:31 EDT
    >>> HELO ODIE.MIT.EDU
    250 MENELAUS.MIT.EDU Hello ODIE.MIT.EDU, pleased to meet you
    >>> MAIL From:<probe@ATHENA.MIT.EDU>
    250 <probe@ATHENA.MIT.EDU>... Sender ok
    >>> RCPT To:<testers-mtg@menelaus.local>
    250 <testers-mtg@menelaus.local>... Recipient ok
    >>> DATA
    354 Enter mail, end with "." on a line by itself
    >>> .
    250 Ok
    >>> QUIT
    221 MENELAUS.MIT.EDU closing connection
    testers-mtg@menelaus.local... Sent

Why is it that the addressee in the mailer specification is being
canonicalized?  My change will get around the problem with the
machine.LOCAL specification used on Athena servers, but it certainly is
not the correct solution in the general case (ie. wanting to
differentiate between CNAMEs).

-Richard

home help back first fref pref prev next nref lref last post