CosmicPerl.com Support wrote:
making individual probably-spam mailboxs for each virtual server.
:0:
* ^X-Spam-Status: Yes
$HOME/probably-spam
Or maybe somewhat like this:
:0
* ^X-Spam-Status: Yes
| procmail -d probably-spam
(but also add X-Loop checking)
Although this is giving each mail user a probably-spam box rather than each
virtual server. Although this might work just as well. The question is how
would the user actually download the probably spam if they so choosed?
Is webmail/IMAP involved? Then deliver to the IMAP maildir-folders.
Or I've updated my Webmin/Virtualmin configuration so that usernames are now
put out as the email address. I.e. mailbox for
probably-spam(_at_)domain(_dot_)com
would be /var/spool/mail/probably-spam(_at_)domain(_dot_)com
Given this would it be possible to pickup the domain.com part from each mail
user and if the mail is marked as spam direct it to
/var/spool/mail/probably-spam(_at_)domain(_dot_)com rather than
/var/spool/mail/whatever(_at_)domain(_dot_)com? If so how is it done?
There might not be a failsafe way to do that. What if a message is addressed
to two users? What if the address is only on the envelope and not in the
message? Your MTA could give the Envelope-To adress to procmail via a
parameter, or duplicate the message and add an Envelope-To header.
--
Affijn, Ruud
____________________________________________________________
procmail mailing list Procmail homepage: http://www.procmail.org/
procmail(_at_)lists(_dot_)RWTH-Aachen(_dot_)DE
http://MailMan.RWTH-Aachen.DE/mailman/listinfo/procmail