Re: [Exim] Local users spam filtering

Top Page
Delete this message
Reply to this message
Author: Pat Lashley
Date:  
To: Ron McKeating, Exim-Users (E-mail)
Subject: Re: [Exim] Local users spam filtering
--On Thursday, December 18, 2003 10:56:15 +0000 Ron McKeating <R.J.Mckeating@???> wrote:

> Hi all, hope you are all getting ready for a nice festive season.
>
> Just thinking, today we upgraded all our servers to exim 4.3 and
> spamassassin 2.61. We still get some of our local users email caught in
> our spamtrap because they use OE with html and do silly things like odd
> fonts etc.
>
> I was thinking of either not spam filtering email from our own ip range,
> or adding a spamassassin rule to give a big neg score for email from our
> own ip range. Is anybody doing similar and if so how are you doing it?


Not spam filtering for local->local usually makes sense.

Not filtering outgoing messages can save resources; but your
users should be aware that their messages are quite likely to
pass through a spam filter on the receiving end; and those
filters are likely to just drop any message with a high spam
score.

When filtering a locally generated message it should usually be
safe to bounce it instead of just dropping it; which lets your
user know that their message had a high spam score, and if allowed
through might run afoul of the recipients' filters. If the bounce
also contains the Spam Report from SpamAssassin, they will be
able to see what they did that caused the high score; and attempt
to correct it.

You could probably craft your ACLs so that locally generated
messages with high spam scores will cause a warning to be returned
to the sender; but still attempt delivery of the original message.



-Pat