On Mon, 18 Aug 2003, Daniel Nilsson wrote:
I played around with this for a while, and I was able to make cyrus change the header of the reject message to be [EMAIL PROTECTED] But from looking at the source of cyrus, it seems like the from address in the envelope of the reject message is set to <> (lmtpd.c lines 502-503 in cyrus-imapd 2.0.16). I guess what happens then in my case is that the reject messages have an empty envelope from: address and when the spammers domain MTA doesn't want to receive the message from our mail server (since the user doesn't exist for example) sendmail decides to alert the local postmaster.
From RFC 2821:
This notification message must be from the SMTP server at the relay host or the host that first determines that delivery cannot be accomplished. Of course, SMTP servers MUST NOT send notification messages about problems transporting notification messages. One way to prevent loops in error reporting is to specify a null reverse-path in the MAIL command of a notification message. When such a message is transmitted the reverse-path MUST be set to null (see section 4.5.5 for additional discussion). A MAIL command with a null reverse-path appears as follows:
Rob,
Thanks, now I understand why cyrus works the way it does. So I guess my problem is more of a general sendmail configuration questions them, for which there are more appropriate lists. But I'll ask it here anyway, in order to complete the thread.
When the cyrus user on our mailserver tries to send a reject message it sets the null reverse-path and the empty FROM address per RFC 2821. The problem is when delivery of this reject message fails (due to a non-existing user to deliver it to for example), there is no sender to send the failure notice to and sendmail then decides to deliver it to postmaster on the local machine. Is there a way I can prevent sendmail from doing so ? That is: when a locally generated mail with an empty reverse-path and FROM address fails to deliver to the recipient, how can I make the failure notice go into the bitbucket /dev/null ?
Thanks -- Daniel Nilsson Signal Integrity Software Inc.
