Take a look at exim. It has mail filtering built into the MTA, no need to wait till delivery to check for spam. It filter on header contents too. Much better if you remember that your method generates at least one outgoing for every spam in, internet-wide this means a spam of 50,000 emails generates a potential 50,000 more. Better to quietly can the mail before it ever gets to the user if it is from a known spam domain than to send it on to a user's spam filter that jams the mail server even more.
On 25 Jun 1997, Manoj Srivastava wrote: > Hi, > > I have mailagent then configured to look at the adresses in > that list and send an automated message to the sender asking to be > removed from the list (I could just delete the mail, I guess), with a > copy to the postmaster. George Bonser [EMAIL PROTECTED], [EMAIL PROTECTED] -- TO UNSUBSCRIBE FROM THIS MAILING LIST: e-mail the word "unsubscribe" to [EMAIL PROTECTED] . Trouble? e-mail to [EMAIL PROTECTED] .
