As part of a Mailman configuration, entries are made in /etc/aliases to feed pseudo accounts into a script. For example, when an email message is sent to "[EMAIL PROTECTED]," an aliases entry such as…

listname-subscribe: "|/usr/lib/mailman/mail/mailman subscribe listname"


…feeds the message into the system and generates a confirm-to- complete email challenge to subscribe the sender to the list.

Two of the pseudo accounts that Mailman uses are "owner" and "admin." These two only forward email to email address(es) specified within the Mailman administration utility (web-based). I am not sure if Mailman is doing the forward or if it is handing that off to the systems MTA (which in this case would be Sendmail) however. I am curious about this because the messages that Mailman is forwarding are being treated differently than others.

For example, I have messages redirected to a Google-Apps-for-my- domain account from all three of the following situations on my own server:

1.      an entry in /etc/aliases that redirects mail addressed to "root" to
        "[EMAIL PROTECTED]"

2.      an entry in /etc/mail/virtualusertable that redirects mail addressed
        to "[EMAIL PROTECTED]" to "[EMAIL PROTECTED]"

3.      the email address "[EMAIL PROTECTED]" specified
        in Mailman's administration utility as the address where all mail
        sent to "[EMAIL PROTECTED]" and
        "[EMAIL PROTECTED]" should be sent


…and it is all delivered properly. The problem is that Google's spam filter is not processes/catching mail that arrive via method #3. I use the word "processes" specifically because it is 100% consistent, meaning that all spam that arrives via methods 1 and 2 is caught with complete accuracy and spam that arrives via method 3 is not touched at all. So, it appears like Google is treating the messages differently for some reason and thus the reason that I am asking the group (opposed to it just being a Google issue… as I think Google's spam filter rocks and don't think it has anything to do with that). I have tried comparing headers from messages in the three different scenarios but have been unable t come up with anything.

So, my question is if anyone has any ideas. Even if you aren't a Mailman guru, do you have any thoughts on why mail received via a unix system alias redirect or sendmail virtual user redirect would appear differently than a message forwarded by a script?


wade

Attachment: smime.p7s
Description: S/MIME cryptographic signature

/*
PLUG: http://plug.org, #utah on irc.freenode.net
Unsubscribe: http://plug.org/mailman/options/plug
Don't fear the penguin.
*/

Reply via email to