[SURBL-Discuss] rDNS entries

Robert Brooks robb at hyperlink-interactive.co.uk
Tue Jun 29 10:50:02 CEST 2004


Mariano Absatz wrote:
> But, for me, SURBL is much better than plain smtp rbls because its collateral 
> damage is very, very, very low... regretfully, I can't play the BOFH and I 
> have to let pass legit e-mail from 'bad' or 'regular' ISPs (most ISPs in 
> Argentina, including the largest ones would be in these categories, and I 
> can't reject e-mail from the largest ISPs in the country).

this is why I'd suggest a separate data source and usage/scoring to suit your needs.

> I'm using relatively high scores for SURBL (between 3.0 and 4.0) and couldn't 
> be doing this with a higher rate of FPs... 

yes, personaly I'd probably only want to add 1.0 for such data but if it keeps a 
dirty email out of bayes it makes my life easier.

> See the thread 'Rule for email of "$43321" ?' in SpamAssassin-Users list... 
> SURBL caught most or all of them because it was highly scored... 

yes, unfortunately uri blacklists are looking to be the "best last defence" 
against spam.

>>It might also encourage the webhosting companies to clamp down on spammers 
>>hosting content with them.
> 
> I'll happily let SPEWS and the other IP RBLs deal with this.

I'd don't care who provides the data if it's valid, however the uri plugin would 
have to support ip resolution :)

ok, I think I've had my $0.02's worth :-)

-- 
Robert Brooks,           Network Manager,          Cable & Wireless UK
<robb at hyperlink-interactive.co.uk> http://hyperlink-interactive.co.uk/
Tel: +44 (0)20 7339 8600                      Fax: +44 (0)20 7339 8601
-  Help Microsoft stamp out piracy.  Give Linux to a friend today!   -


More information about the Discuss mailing list