Spam Protection
SpamAssassin
SpamAssassin is a powerful mail filter, which uses text analysis to determine the probability of a mail message being SPAM. Some of the factors for this decision are the presence of HTML parts in the mail, phrases common for SPAM ("drastically reduced", "save big money", etc.), excessive HTML comment tags (often used to 'hide' common phrases from mail filters), invisible text (using the same foreground color as the background color) and many others. Each occurrence of something suspicious raises the probability of the message being SPAM (the SPAM level). You can configure the SPAM level, above which the mail messages will be rejected (bounced).
The levels manage how strictly SpamAssassin reviews each message to consider whether it is SPAM or not.
Setting the SPAM level to a low value (1/10) will probably cause legitimate mail to be bounced. Setting it to a high value (10/10) may let some SPAM into your mailbox. The optimal level depends on the contents of the mail you receive - you will have to adjust it until you find the lowest level possible, at which no legitimate mail is rejected (you will need a feedback from your correspondents in order to define that optimal level).
Statistics show that 7/10 is a relatively safe threshold value, and setting it usually does not result in bouncing legitimate email.
Enabling SpamAssassin globally will make it process any message, sent to your domain, while enabling it per user basis will make it scan only the messages, destined for the respective mailbox.
[Back to Spam Protection Questions] - [Back to FAQ Topic Index]
Mail filters
Mail filtering allows you to selectively filter out mail messages based on the contents of their headers. Such emails can either be bounced back or they can be delivered to a Junk Mail folder, if you have this option activated for your mailbox. In the case when messages are bounced back to their sender, if a legitimate mail is rejected, the sender will be notified that you have a reject filter that matches the message, and will be able to try to contact you by other means (through another e-mail address or by telephone, for example).
Basically, all the options correspond to different email headers. Email headers are added to a message by the sending email software, and most of them are visible only when you view the email message source.
The Subject header contains the subject of the message. If you want to ban all messages that contain “viagra" in the subject, just select this option and type "viagra" (without the quotation marks) in the text box.
The From header contains the sender name and email address.
The To header contains the recipient email address. This would be normally your email address.
The CC header contains the other email addresses your message has been sent to. CC stands for "Carbon Copy", which is used in most email clients for this purpose.
The Received header is actually added by the mail server, and contains the IP address of the sender. If you want to block all messages from a given IP address, use this option.
The X-Spam-Level header is added by SpamAssassin – an email scanning software that can add this header to messages it has scanned. We are running SpamAssassin on our servers, but in case you are using some third party filtering services (SpamCop.net for example) you can take advantage of this setting. If you decide to use our SpamAssassin, this header will not be added to your messages.
The X-Mailer header identifies the email client used to send the email message. For example, if you select this option and type "Outlook" (without the quotation marks), you will ban all incoming messages that have been sent through MS Outlook and Outlook Express.
The X-Mailscanner header is added by an anti-virus software such as Mailscanner. We have added it upon user request because a particular virus was adding this header to all incoming messages, and banning messages with this header would ban all such viruses. However, it might also ban messages that have been scanned by Mailscanner, so we do not recommend using this option.
The Content-type header says what is the structure of the incoming mail message - it could be plain text, HTML, MIME encoded, etc.
Please note that using these filters requires some general knowledge of how the email system works.
[Back to Spam Protection Questions] - [Back to FAQ Topic Index]
Database of known spam sources
Our mail system utilizes a database, which contains IP addresses and networks that are known sources of unsolicited mail messages. Mail from these servers is rejected from our machines. The database that we use is extremely precise and is updated real-time.
[Back to Spam Protection Questions] - [Back to FAQ Topic Index]