SURBLs indeed catch them fine. I have done RBL(at MTA level)+MailScanner+SpamAssassin+DCC+SURBLs+Razor2 along with SARE rulesets and some custom cfs. In last one month thats after implementing SURBL and Razor2 i have managed to gain 100% control on these kind of Spam messages and for the few of which (1/100) were left out i had taken care of it by customized cfs containing those URLs in the message body.

Well as bottom line i wld give full marks to SURL lookup for really doing a good job.

Rakesh


Shawn R. Beairsto wrote:

We are getting hit pretty hard with these spams as well, enough that my
users are complaining for the first time in a long time. It seems they
do not hit any of the rulesets from SARE and only a couple from the main
rules:

Content analysis details:   (0.2 points, 5.0 required)
pts rule name              description

---- ----------------------
------------------------------------------------

0.0 BAYES_56               BODY: Bayesian spam probability is 56 to 60%
                           [score: 0.5833]
0.1 HTML_50_60             BODY: Message is 50% to 60% HTML
0.1 HTML_MESSAGE           BODY: HTML included in message


Once they've been reported, SURBL's catch them fine. I'd appreciate any
helpful info.


Thanks,

--
Shawn Beairsto
Network Administrator
Data Kinetics Ltd.


-----Original Message-----
From: Allen Wayne [mailto:[EMAIL PROTECTED] Sent: Monday, August 09, 2004 8:52 AM
To: [EMAIL PROTECTED]
Subject: RE: How to stop porn spam






Hello!
We're receiving more and more spam which consists of a large
porn picture and several paragraphs of text grabbed from some books or sites like this:


"Party leaders saw economic reform as a way to regain their
and their Our goal was finished we safely landed a man on the moon. We have proven that our heroes were successful in leading us on a great journey through the first stage of space exploration, but were will we land next will a new president lead us on a journey to mars? Will there be new heroes to look up to? Only time will tell. "


I'm sure bayes would help, but it is impossible to teach our
customers to correctly train bayes. :( Eugene





Tell them to stop visiting porn sites and giving out their email


address!


:) Sorry for the obvious...



Has anyone got anything slighty more helpful? We are experiencing similar, with an image and a chuck of text, seemingly news based. I don't actually feel that a chunk of our users have all visited the same site and left their email addresses, all incidentally since last Thursay.







Reply via email to