The ThinkAutomation Mail Server includes a number of features designed to reduce the number of spam emails that you receive.
When an email message is received by the mail server it will be checked against Trusted Hosts and Whitelists. Any emails that are not from trusted hosts or are not whitelisted will be checked against the security options and spam filter.
Enable Bayesian Filter
If this option is enabled then email messages will be scanned through the 'bayesian filter'. ThinkAutomation will split the incoming email into a list of tokens (including headers). These tokens are then looked up from a database. The database contains a count of good messages and spam messages for each token. A subset of the tokens are used to calculate a bayes score. The higher the score, the more likely the message is spam.
The filter will improve in accuracy over time as more tokens are added to its database or good and bad token counts.
For more information about Bayesian spam filtering, see: http://en.wikipedia.org/wiki/Baysian_filter
Index As Good All Outgoing Emails And Incoming Emails From Trusted Sources
This option should normally be enabled. It ensures that the bayes database is updated with good messages. All outgoing emails will be scanned and tokens added to the database as good. All incoming emails from Trusted Hosts or that are Whitelisted will also be indexed as good emails.
Use SURBL (URL Block List)
If this option is enabled then any incoming emails that contain HTML links will be checked against the SURBL list. This is a public list of links that commonly occur in spam messages. For more information about URL Block Lists, see: http://en.wikipedia.org/wiki/SURBL
Add Score To Subject
ThinkAutomation will assign a spam score of between 1 and 10 to each message. The higher the score, the more likely the email is spam. ThinkAutomation can add this score to the message subject allowing you to easily see the score when viewing messages.
Score Required To Flag Messages As Spam
This defaults to 7. You can lower the value to mark more messages as spam. If a message has a score equal or higher than this value then it will be flagged as spam and optionally moved to the spam folder (see below).
Score Required To Refuse Messages
If any message spam score is equal to or higher than this number then it will be rejected by the mail server.
Messages refused due to their spam score can either be deleted, or rejected. If the message is deleted, then the sender thinks you have received the email - you just won't see it. If the message is refused then the sender will receive a NDR (non delivery report - or a 'bounce' message).
Move Spam To Folder
For spam that is not rejected, you can move the messages to a folder in the Message Store, instead of the Inbox of the recipient. The Spam folder is shared for all users. You should then select the folder in the message store to use. A 'spam' folder is created automatically. The email alias 'spam' is assigned to this folder. Any email sent to the mail server to 'spam@' will be indexed as spam.
Delete Spam From Folder After
ThinkAutomation can delete old messages from the Spam folder after a certain number of days. Specify zero days to never delete.