ietf-smtp
[Top] [All Lists]

Re: not delivering, and History of fallback to A

2008-03-30 23:36:03

ned+ietf-smtp(_at_)mrochek(_dot_)com writes:

No, of course not.  It's going to change the way that MTAs are set up
which will make it harder for spammers to abuse.

I completely fail to see how the availability or nonavailability of DNS
fallback behavior will lead to increased spam abuse. As I pointed out
previously, if spammers percieve an advantage to sending directly to
hosts named in A/AAA records they are going to do so no matter what. By
the same token, if they don't perceive an advantage they won't do it.

Trying to explain this one point just for the sake of clarity, since I
think I see what Jon's getting at:

A huge amount of the spam that's out there involves spammers scraping the
Internet for anything that looks like an address and dumping it into an
MTA (generally by way of a compromised system).  That MTA then does its
normal thing with that mail.  The widespread use of other anti-spam
tactics is pushing more spammers into abusing the MTAs accessible
compromised systems instead of sending spam directly themselves.

If that address points to some host that doesn't actually receive mail
(message IDs cause this to happen all the time), that spam creates a bunch
of attempted deliveries, bounces back to the (forged) envelope sender,
etc.  Without the A record fallback, it could be immediately discarded.

Now, in my opinion, like every other antispam measure, such as the
greylisting that makes this MTA behavior now more common, this wouldn't
result in a net decrease of spam in the long run.  Most anti-spam measures
just make the mail system more complex and fragile for short-lived gains
that quickly disappear.  However, this one has the advantage of being,
from a certain perspective, also more formally correct in that it
represents the reality of hosts that don't accept e-mail more clearly and
allows quick discarding of lots of spam to addresses that will never
exist.  As anti-spam measures go, it's a fairly straightforward and
uncomplicated one.

Practical questions of deployment are another matter which I won't get
into.  I just wanted to try to take another shot at explaining the above
sentence, since it makes sense to me too.

-- 
Russ Allbery (rra(_at_)stanford(_dot_)edu)             
<http://www.eyrie.org/~eagle/>