can someone explain whats going on behind the curtains of ispconfig regarding filtering of emails?
I mean explain a little bit the order of filters, what happens?
who jands over the message to whom? whats trashcan? why are there now clamav x-headers so I can see the message was scanned?
The messages are handled by procmail (postfix or sendmail => procmail as lda). To see whats going on, go to the home directory of one of your users and execute the command "ls -la" to see the scripts involved. In the file .procmailrc you can see the orders of the filters.
Trashscan is the script that calls clamAV. Trashscan is part of the clamAV package.
what about sa - autolearn? bayesdb for each user?
Inside the home directory of the user is the .user_prefs directory. There is the bayesdb stored.
the last tools I used was very well documented and working this way:
a mail came in and was handed over to clamav, if virus found, header set, subject changed and moved into a .virus folder of the users mailbox.
if no virus found, wrote x-header, handed over to sa, wrote its x-header, tested with pyzor, razor,dcc and if spam delivered to users .spam folder taged and subject changed.
if ham landed in spam moving it to inbox was enough (cronjob running in the background to do sa-learn inbox=ham) if spam was in inbox moving to spamhad the same effect.
additionally a bayes db was used for each user (if I , the admin decided so)
maildrop was being used for mail delivery.
I could set global settings like: use bayes, number of required hits, use razor, use pyzor, use dcc, what to do with spammail and the "reseller" could change his default for his sites, and the end user finally could change it as well
any chance that works here too?
please shed some light into this matters as I really like ispconfig but I am quite confused by the documentation ;-))
To see the available options in ISPConfig, open a mailuser in the controlpanel and have a look the spamfilter tab or read the documentation, the available options are listed there.