|Main Archive Page > Month Archives > spamassassin-users archives|
On Sun, 2011-10-16 at 12:26 -0400, Brian J. Murrell wrote:
> I don't call spamd from the deliver end. Each user has a .procmailrc
> which pipes the mail through "spamc" so spamc is already being called by
> the recipients effective user-ID.
Have you thought of running spamc remotely? This way you could avoid the
need to login the the server just to process mail.
spamc takes -d -p and -u options, which should do exactly what you want:
-d gives the host name (default is localhost)
-p is the port (default 783)
-u is the username
This way you can go on calling spamc from the procmail recipe so it
would remain invisible to the users.
You'd store user preferences on the server as individual files or in a
MySQL database as others have described. The worst case would be that
your users may have to log in to the server to change their preferences,
unless, that is, you go the MySQL way and provide, say, a simple PHP
script to maintain them via an in-house Apache web server.