As everybody knows that recent viruses spread via sending spoofed 'sender address'.
fex.
I am a person '[EMAIL PROTECTED]' and got so called 'return mail' from '[EMAIL PROTECTED]' telling that mail sent by me (which I never sent in a first place) cannot be delivered. Obviously containg somekind malware as an attachment.
This kind of 'spread method' could easily stopped if the mail servers include some kind of fingerprint to the passing E-mail.
If the return mail (mail receiver server checks this against an private key or something) does not contain an fingerprint then the returned mail should not be delivered 'back to the sender'.
Rather clever way to counterfeit the sender address, it might double the infection if the bounce to the 'sender' leads to infection.
Now, what this kind of 'hardening' might need is...
- E-mail receiving server could check that 'very first original' From: line and if it is same than the receiver address ie. '[EMAIL PROTECTED]'
Perform an check to see if the 'sender identification' ie. salted public key, GUID or something (X-Authenticated-Guid: #0a845d299ca340087140) exists in mail header.
Delivery should be done only if an 'sender identification' exist and the key matches.
Otherwise mail should be trashed to dev/null :)
Waiting for comments and succestions...
_________________________________________________________________
MSN 8 helps eliminate e-mail viruses. Get 2 months FREE*. http://join.msn.com/?page=features/virus
_______________________________________________ Full-Disclosure - We believe in it. Charter: http://lists.netsys.com/full-disclosure-charter.html
