Hi,

I have a cgi script installed that reads the e-mails from a catchall POP3 account in order to use a virtual mailing system.

My problem is that when my script recieves a mail to a user that doesn't exist. My script sends a bounced message through my SMTP via sendamail to the sender of the original message. The sender of the original message naturally has an invalid address and then the error message that my script sent bounces of their server and sends an undeliverable message back to the "root" user of my SMTP server which I then recieve in my normal POP3 inbox through the admin mail of the RAQ server.

I have removed the adress in the reply-to and from field in my error message. So that should remove some of the bounced messages, but I am still getting a huge amount of bounced messages that use the SMTP "root" address!

My question is, what is the difference between using my script system and using a real POP3 account? When a message normally reaches a POP3 account to a non existant user the server sends back a bounced message and we start the same loop, but this time I don't recieve any bounced messages, so what is the difference??

Is there anything I can do so that the other mail servers can't find the address of the sender? I don't want to receive these double bounced messages...

Thanks