Negative #s

[Doug Wolfgram <do...@gfx.com>]

This has probably been covered here before, but I missed it. I have been 
swamped in the past few weeks with spam that scores around -4.9.  What is 
this and how to I get rid of it??? Is there some parameter that says never 
go negative? These spammers are smart enough now to create negative scores 
to be sure they'll get through.

D

_________________________________________

"If you're not in e-business ... you're not in business.."
_________________________________________

Doug Wolfgram
GRAFX Group, Inc.
Cell: 949.433.3641
http://www.gfx.com

Re: Negative #s

[William Stearns <ws...@pobox.com>]

Good evening, Doug,

On Sun, 19 Sep 2004, Doug Wolfgram wrote:

> This has probably been covered here before, but I missed it. I have been 
> swamped in the past few weeks with spam that scores around -4.9.  What is 
> this and how to I get rid of it??? Is there some parameter that says never 

	For a given message, you'll have to look at the X-Spam-Status or 
X-Spam-Report header.  If you're lucky enough to have the latter, it will 
show the negative score that brought the score down.  If you only have the 
former, you'll need to look at the individual scores.

> go negative? These spammers are smart enough now to create negative scores 
> to be sure they'll get through.

	These are the clues that usually help ham be marked as such.  
	Cheers,
	- Bill

---------------------------------------------------------------------------
        "5) what are people like spaf/chris rouland/lance then?
a) THEY ARE THE ENEMY. WHITEHATS = ENEMY."
        -- http://www.blackhatbloc.org/phrack/texts/faq1.txt
--------------------------------------------------------------------------
William Stearns (wstearns@pobox.com).  Mason, Buildkernel, freedups, p0f,
rsync-backup, ssh-keyinstall, dns-check, more at:   http://www.stearns.org
--------------------------------------------------------------------------

Re: Negative #s

[Loren Wilton <lw...@earthlink.net>]

-4.9 is probably something like BAYES_00.  It sounds like you have bayes
trained to recognize spam as ham.

        Loren