Tiled gif spam

[Gareth]

No Bayes?

On Thu, 2007-08-23 at 14:21, Andrew MacLachlan wrote:
> Sorry - this is the analysis from Mailwatch:
> 
> Spam Report:	
> Score	Matching Rule 
> 	score=3.919	 
> 4	required	 
> 0.00	DKIM_SIGNED	 
> -0.00	DKIM_VERIFIED	 
> 2.50	HTML_IMAGE_ONLY_16	 
> 0.00	HTML_MESSAGE	 
> 1.42	SARE_GIF_ATTACH
> 
> As you can see, it was almost trapped. Of course I could always up the
> scores for SARE_GIF_ATTACH and HTML_IMAGE_ONLY_16, but I think an
> additional rule for side-by-side gifs might be a better approach? I'm
> not sure what the best score might be for such a rule - but something
> around the 1.0 mark would probably be appropriate.
> 
> -Andy
> 
> > -----Original Message-----
> > From: mailscanner-bounces at lists.mailscanner.info [mailto:mailscanner-
> > bounces at lists.mailscanner.info] On Behalf Of Andrew MacLachlan
> > Sent: 23 August 2007 14:03
> > To: MailScanner discussion
> > Subject: Tiled gif spam
> > 
> > I just forwarded this from my Yahoo acct through my MailScanner with
> > FuzzyOcr and it sailed straight through... I'm not sure what my MTA
> > level checks would have made of the original message though.
> > 
> > Does anyone have any special tricks or rules to combat these messages?
> > 
> > I'm running the latest everything on Centos 4.5
> > 
> > -Andy
> > 
> > 
> > --
> > This message was scanned by ESVA and is believed to be clean.
> > 
> 
> 
> 
> --
> This message was scanned by ESVA and is believed to be clean.