Re: DKIM and DMARC, Email explained from first principles
"John Levine" <[email protected]>
| Newsgroups | gmane.ietf.smtp |
|---|---|
| Organization | Taughannock Networks |
| Message-ID | <[email protected]> |
It appears that Sam Varshavchik <[email protected]> said: >I'm struggling to identify some tangible value-added that DKIM/DMARC brings >to the table. > >Ostensibly, these signatures prove that the mail really comes from the >domain it purported to come from. > >Ok, that's cool, but what is the point? For DKIM, the point is to have a reliable identifier for the message that is better than an IP address. Small mailers that share an IP can use separate signing domaisn to separate their mail streams, large mailers can aggregate the repuation of all of their outbound IPs. As you say, merely having a DKIM signature tells you nothing, but after you watch a mail stream for a while, you see that some DKIM signers send clean mail and some send lousy mail and adjust your filters appropriately. Large mail systems all do this. We hoped that there would be shared DKIM reputation lists like there are shared IP lists but so far that hasn't happened. The original point of DMARC was for B2C or B2B mail from heavily phished domains like Paypal, that could say please discard anything from us that fails DMARC and we understand that might be some real mail. (All of Paypal's mail just says "something happened, look at our web site".) It still works pretty well for that. Unfortunately, AOL and Yahoo had separate giant security failures and allowed crooks to steal people's address books, so spammers could take pairs of addresses and send spam that appeared to be from a friend, leading to huge numbers of support calls at AOL and Yahoo. They decided to outsource the cost of their security failures to the rest of the Internet by abusing DMARC p=reject to make that spam disappear, along with a lot of real person-to-person and list mail. They knew this would break every mailing list and they didn't care, according to someone who was in the room at the time. The people who designed DKIM and DMARC knew then and know now what it can and can't do, but we have done a poor job of explaining them to people who want them to be a magic FUSSP. R's, John