Stealth Blocking

Eric A. Hall ehall at
Thu May 24 18:33:13 UTC 2001

> Last week, I got one spam ("get a free motorola pager") which came
> through 168 different open relays, bound for 4428 different recipients

I just peeked in my trash folder, and 6 out of the last 10 spams that I
received were sent directly from dial-up spam blowers.

Certainly we can agree that there are many paths spammers will take. If
rate-limiting eliminates/curbs the throwaway dial-up abusers, then surely
it is an effective tool in the fight. I'm not calling a cure-all.

> That's why I think that port 25 blocking is the only way.  That, and
> closing open relays, of course.

I would say that default blocking of port 25 is a good position to take,
but you can't deny that has its own problems. For one thing, the
exceptions become the rule. I've noticed a trend in spam from small
businesses, cable users, etc., many of whom are behind non-throwaway
lines. Going to a model where "legitimate" users are unfiltered doesn't
stop all spam, it only delays it at best.

In this regard, rate-limiting and port-blocks are just tools in the belt,
neither of them is perfect.

Eric A. Hall                              
Internet Core Protocols

More information about the NANOG mailing list