I prefer spam blocking at smtp level, based on
adresses/domains/host names/ ip addresses/network addresses.
I have difficulties with SA's approach of allowing spammers to
steal my resources.
How much spam do you get, and what is an acceptable failure rate for
you? Just curious. I cannot see an MTA approach being more than 80-85%
effective, which is not good enough IMO.
I agree, and I also think that my approach is not workable for larger
enterprises or institutions. The old YMMV :)
I am perfectly willing to
trade some silent resources for 100% control of my mail :/
For the whole domain, in a four week period, we reject about 50% of all
mail on average; this rate goes up to 80%/day on weekends. I have reason
to believe that the number of false positives is better than 1% (*). In
absulute numbers, we are blocking about 25k spam emails (**) per week - an
average of 100 per staff, albeit very unevenly distributed.
Personally, I am targeted with about 2500 spam emails per month. About
5-10 make it through the mail gateway per day, and maybe 5 per week are not
caught by bogofilter, which I'm running through procmail (to get back on
topic :). Whenever there's some new Windoze worm outbreak, it is not unusual
for me to receive several thousand virus emails within one or two weeks,
plus virus "notifications" from email address forging.
(*) False positives usually come in clusters - a new user communicating with
a blocked country/ISP, or something like that. That's easily dealt with.
(**) Including spam bounces due to forged email addresses.
_______________________________________________
procmail mailing list
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail