SpamC score adaptation
Added 11 days ago
MailCleaner provides a lot of SpamC rules and the best adjustment we can to meet the global need.
Since each mail flow is unique, you may want to adjust your SpamC rules scores.
This can be done in
Configuration->Domain-> [domain name] -> SpamC rules adjustment
For example, we provide rules to detect big amounts of money. In fact, this rule is used in several meta rules (for example MC_LOTS_OF_MONEY is the rule about bigs amounts, as such, it already get some points, MC_LOTTERY is a rule detecting if a mail is about lottery and also gets some score but if both rules are met by an email then extra score is added).
There are positions (sales for example) which are susceptible to get quite some false positives with such rules. In that case, you can low the score of a rule.
In fact, the original score will still apply so you have to keep this in mind when doing a rule for this. Let's say I want the score for MC_LOTS_OF_MONEY to be 0.5 points. As it currently is 1.5 points, I d need to do an adjustment rule :
So that a mail would get
* 1,5 points for MC_LOTS_OF_MONEY
* -1.0 points for adjustment
So the mail finally gets 0.5 for being detected as containting MC_LOTS_OF_MONEY
This feature can be used to apply scores to mails
- coming from a specific sender/domain
- being sent to a specific user/domain
(In fact, the sender/recipient are strings searched for in the headers.)
MailCleaner proposes so many features, it is hard to keep a clear web interface. This adjustment feature can be pushed further but you then need to directly configured it in MailCleaner databases.
To do so :
- connect via SSH to your master server
- run mc_mysql -m mc_config to connect to MailCleaner databases
The rules for SpamC adjustment are stored in the table wwlists.
For example, to low the rule MC_LOTS_OF_MONEY by 1.0 point for the mails coming from @domain1.tld being sent to firstname.lastname@example.org run the SQL statement:
insert into wwlists (sender, recipient, type, comments) values ('@domain1.tld', 'email@example.com', 'SpamC', 'MC_LOTS_OF_MONEY -1.0');