FW: User Preferences

[Jeroen Koekkoek]

Hi

Wouldn’t it just be easier to use LDAP with MailScanner? You could just write a PHP script to update the LDAP tree, I believe MailScanner itself uses a pretty clever way of checking for modifications. Of course if using MySQL is a requirement you should update MailScanner ☺ to support it.

If  you continue to use the method you described you should implement a priority column, MailScanner goes through the rules and uses the first one it finds. Because of this it’s very important to use the following order:

local_part at domain.tld
@domain.tld
default

Please correct me if I’m wrong. Also it would be nice if some released a MailScanner PHP class that can be included and takes care of automatically determining in what order to place the rules.

Kind Regards,

Jeroen Koekkoek

From: mailscanner-bounces at lists.mailscanner.info [mailto:mailscanner-bounces at lists.mailscanner.info] On Behalf Of Paul Houselander
Sent: Thursday, April 26, 2007 1:44 PM
To: MailScanner discussion
Subject: User Preferences

Hi
 
Im just trying to spec out a web interface im going to put together to allow editing/amending of various MailScanner preferences.
 
At the moment ive got a basic PHP setup that writes info to a Mysql DB. At 5 minute intervals a perl script connects to the DB checks for any changes and rewrites the ruleset files.
 
So for example in MailScanner.conf I have
 
Virus Scanning = %rules-dir%/virus.scanning.rules
 
I have a Mysql table with columns
 
Direction
Domain
Value
 
With values like -  FromOrTo --- example.com --- yes
 
My perl script just loops round all the values and writes the flat file "virus.scanning.rules" - changes take affect based on the MailScanner.conf "Restart Every" directive.
 
This kind of worked ok with 1 MailScanner server but ive now introduced another for load balancing/resilience and wanted to put together a more sophisticated system.
 
I was intending writing custom functions for all my rulesets by following the examples in /usr/lib/MailScanner/MailScanner/CustomFunctions/ and also the SQLSpamSettings.pm and SQLBlackWhiteList.pm done for the mailwatch project.
 
I would consider myself very much an amatuer programmer and was looking for opinions and advice on the following:-
 
- Would there be much performance impact on using Custom Functions that read from a (potentially remote) database, rather than a flat file on the system? I can see myself writing custom functions for quite a number of the MailScanner.conf directives
- Any problems using LDAP instead of Mysql?
 
The reason im thinking of LDAP was im looking to move all my sendmail routing info to LDAP and would be nice to just maintain 1 system?
 
Any advice appreciated.
 
Kind Regards
 
Paul