automating spam handling

[Mathias Homann <ad...@eregion.de>]

Hi,


i want to automate my spam handling a bit further.

here's what i got by now:

- fetchmail gets mails from various pop3 accounts, injects into local mail system (postfix)
- postfix pushes mail through clamsmtp and then through spamassassin
- mail gets delivered into local cyrus imap which pushes it through sieve filters
  |_ mails with a score > 12 are discarded; mails with a 5 < score < 12 go in public.spam.CHECK
  |_ mails with score < 5 go through various other filters, or end up in INBOX

so right now, I have to check public.spam.check on a regular basis, then move all "true" SPAM
into public.spam.SPAM, and all false positives into public.spam.HAM.

then, a cronjob that runs every 5 minutes learns everything in public.spam.SPAM as spam, then
deletes it, and everything in public.spam.HAM as ham and keeps it so that the users can get
their mails from there.

What I want is to extend the cronjob ike this:
- mails in public.spam.SPAM are (after being learned as spam) bundled up in a mime digest, and
attached to a mail to spamcop and the ftc spam complaint office, that mail is sent, and the
spams are deleted.
- mails in public.spam.HAM are "re-injected" into the mail system, so that they get delivered
normally.   <---- this is the point where i'm stuck at right now; when i pipe such a mail
through /usr/bin/sendmail, it gets delivered, and then cyrus discards it because it has a
message-id it already knows.


Anyone got hints for me?



bye,
   MH




-- 
Die unaufgeforderte Zusendung einer Werbemail an Privatleute verstößt gegen §1 UWG und §823 I
BGB (Beschluß des LG Berlin vom 2.8.1998 Az: 16 O 201/98). Jede kommerzielle Nutzung der
übermittelten persönlichen Daten sowie deren Weitergabe an Dritte ist ausdrücklich untersagt!