-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello Mark,

Saturday, June 28, 2003, 11:52:25 AM, you wrote:

M> I've been encountering some Spam which plays a particularly nasty
M> trick.  However I'm not running my own mail server.  My web-hosting
M> company is doing that for me.  They run Spam Assassin, and I've set up
M> my mail client to filter on the "X-Spam-Score: " header line.  So, I
M> don't really know Spam Assassin or have access to the mail host.  On
M> the other hand, I've been a software developer for several decades,
M> and I can write regular expressions in my sleep.

That's my situation, with two different domain sets (and a third, where
SA isn't being used and I don't have access to install it).

Three weeks ago I was in your position -- I could change the value of a
rule's score, and I could add whitelists and blacklists (such as a
blacklist for valodata.com which would solve your immediate problem with
this immediate spammer), but I couldn't create my own rules for anything.

What I've now done is created a shell script (I program better/faster in
shell than in perl, and the overhead appears to be minimal) which
a) examines each inbound email message waiting in a flagged inbox
b) looks to see if SA has already flagged the message as spam
c) if so, it moves that message to my spamtrap (a POP3 inbox I collect
when convenient)
d) If not (yet) spam, then execute SA --remove-markup to remove the
initial SA headers, and pipe that into SA --local --exit-code, which
gives me the ability to now test the rules I have created and stored in
my own user_prefs file.
e) If now flagged as spam, again the message goes into the spamtrap.
f) If the message is not flagged as spam, then forward the email to its
correct destination.

This type of system wouldn't give you the ability to create a rule as
powerful as the one you suggest, but it would let you create one which
tests
> <IMG SRC="{valid URI}{yourdomain}{maybemore}"{maybemore}>
and I would think that would be almost as useful for you, yes?

M> As I said, I'm not running the mail server myself, and I don't have
M> the time to delve into the guts of Spam Assassin myself right now. But
M> if someone could impmlement the rule I described, it would be a BIG
M> help.

I haven't had the time to do a proper development and testing cycle on
this system of mine. It works, but it could certainly be improved. Are
there others who might be interested in sharing notes on this type of
thing, with eventually a small add-on package and how-to for SA end-users
like us?

Bob Menschel

-----BEGIN PGP SIGNATURE-----
Version: PGP 8.0

iQA/AwUBPv5p1pebK8E4qh1HEQKHpwCg3ls97uippNDxdh+Q+w7LndAKQOYAoMg0
+dyOZbkcRGGzn7VRHW60/DcJ
=ESyo
-----END PGP SIGNATURE-----




-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100006ave/direct;at.asp_061203_01/01
_______________________________________________
Spamassassin-talk mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/spamassassin-talk

Reply via email to