Re: antiphishing idea
Martijn Grooten <[email protected]>
| Newsgroups | gmane.ietf.asrg |
|---|---|
| Message-ID | <18B53BA2A483AD45962AAD1397BE1325382EF38BFE@UK-EXCHMBX1.green.sophos> |
> SPF checks against envelope MAIL FROM: ...users dont know even that > this thing exists ! they blindly trust on mail header From: > How would be a relay attack? evils dont know every random message-id > of a mail server legit mail. Are they going to query "every" random > string? ....now you can query what you want on a DNS server...what is > the difference? Sorry, I meant rePlay attack. You get hold of one 'valid' messagge-id and use that for all your phishing messages. > sorry, I meant both of them ! you are obviously going to get the > validation from the evil domail but you also need one from phished > domain. The last one is only true if the mail was sent by paypal.com So if I were to use Gmail to send mail on behalf of a example.com, a domain which I hypothetically have an email address at (something which has worked well for years), I'd have to validate the message ID of every email I want to send at example.com? Martijn. Virus Bulletin Ltd, The Pentagon, Abingdon, OX14 3YP, England. Company Reg No: 2388295. VAT Reg No: GB 532 5598 33.