[Mimedefang] implementing a small-scale spam filtering and recovery system

[Tony Nugent]

I have been thinking of ways to implement a crude but workable way
of safely filtering spam-flagged email out of locally delivered
email.

When defang does the spam filtering with SA, instead of tagging and
then delivering such messages, I would like to implement a policy
where a spam score over the threshold simply diverts a message to
some place where it can later be easily examined (and recovered if
necessary).

I have mimedefang up and running on a my mail server, it is an MX
and relay host living in an office environemt where all the clients
are windows-based (mostly using outlook).  The clients, up to 50 or
so, collect their email from the server with imap or pop (actually,
imaps and pops).

My idea (untested) goes like this... go ahead and divert spam into a
"spammer" account so that the original person does not see it.

On one or more (or every?) client, set up a second imap account
which gives them access to all the spammer email where they may
delete or recover messages.  (Trust and confidentality is assumed
within the office).  A ~spammer/.procmailrc could even be used to
automatically sort the spammer mail into local imap "mail folder"
files, according to the intended recipient (or sorted in other
ways).

It seems so simple that I'm suspicious that it isn't :)

Is anyone doing it like this?

How are others managing spam?  (CanIt seems to be very cool, using a
white/blacklist databases and so on...)

Thanks for any feedback.

Cheers
Tony