Steve, the point you are missing is that my spamassassin "gives up" 
before it even scores it because of the Content Length Mismatch:

Dec  2 14:02:03 ns spamd[10870]: connection from localhost [127.0.0.1] at port9497
Dec  2 14:02:03 ns spamd[10930]: info: setuid to xxx succeeded
Dec  2 14:02:03 ns spamd[10930]: processing message 
<[EMAIL PROTECTED]> for xxx:520, expecting  17103 bytes.
Dec  2 14:02:03 ns spamd[10930]: bad protocol: header error: (Content-length mismatch: 
17103 vs. 17092)
Dec  2 14:02:03 ns spamc[10929]: failed sanity check, 256008 bytes claimed, 0 btes seen

So, because it fails the sanity check, it bails out and delivers 
message to procmail WITHOUT aadding any scores.

THAT, is what I need help with.

mark

On Mon, 2 Dec 2002, Steve Thomas wrote:

> | http://mark.cervarich.com/gb2312.spam.txt
> | contains 3 sample spams that get by spamassassin every time.
> 
> I got scores of 10.7, 9.7 and 9.8 respectively for those three messages
> using 2.43.
> 
> 
> | Although I'd love it if someone told me how to stop this kind of spam
> | using spamassassin.
> 
> All three of them triggered the SUBJ_FULL_OF_8BITS rule. I route any mail
> that:
> 
>  1) scores higher than my default threshhold, AND
>  2) triggers the SUBJ_FULL_OF_8BITS rule
> 
> straight to /dev/null. I never even see that I received them, unless I'm
> looking over my procmail logs or spam archive for some reason.
> 
> St-
> 

-- 
Mark R. Cervarich
Shelfspace.com



-------------------------------------------------------
This SF.net email is sponsored by: Get the new Palm Tungsten T 
handheld. Power & Color in a compact size! 
http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0002en
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to