C-Command Software Forum

They find a way to round SpamSieve


I look for information about this problem at forum, but i cant find any about this way spammers have to round the spam detection.

Thats an example of they do it:

De: Viagra.com <info@faq-mac.com>
Asunto: Online Doc Tammie
Para: info@faq-mac.com
Return-Path: <info@fapdec.org>
Return-Path: info@faq-mac.com
Delivered-To: faqmac-faq-mac:com-ximac@faq-mac.com
Delivered-To: faqmac-faq-mac:com-info@faq-mac.com
X-Envelope-To: ximac@faq-mac.com

faq-mac.com is a valid domain for me, but i cant find a way to round this problem. I receive … mmmmm … over 50 of this kind (from different email directions from my domain) by day.

I’ll try to avoid this behaviour … but no luck. If the email adress go to the white list, catch all the spam as legitimate: if i train it as spam, i lost all these valid mails.

I receive over 500 smaps by day … sometimes even more.

thanks in advance!!

If you think that a spam message is getting through SpamSieve, the first thing you should do is check the log to make sure that SpamSieve has been making predictions for the message. (If not, there is a setup problem.) If the log shows that SpamSieve keeps predicting that a certain type of spam message is good, please send me a report.

If the address alone is not enough to determine whether a message is spam, then SpamSieve must classify it by examining the message contents rather than just the address. This will happen automatically. After you train SpamSieve with both good and spam messages from that address, the address will be on both the whitelist and blocklist, but SpamSieve will disable it in order to avoid the two problems that you mention.