InterWorx uses SpamAssassin to help keep levels of email spam reasonable. The page in NodeWorx deals with settings that are global across the entire server.
SpamAssassin can be configured to block spam at the SMTP leve before the message is even processed by the mail delivery system and sent to a local mailbox. SpamAssassin will refer to a rules database when checking incoming messages and assign a score to each message based on how likely it thinks a message is spam. A message with a higher spam score then a user-configurable threshold will automatically be refused by the SMTP service.
You may not want to enable this setting. It has no bearing on the SpamAssassin setting in SiteWorx, so e-mail messages may be scanned twice if this setting is enabled.
This option will scan all e-mail using the SpamAssassin content filter as the e-mail enters the server. The SMTP connection will be dropped only if the Spam Score is higher than the SMTP Spam Score option below. Otherwise, the e-mail will still be delivered to the recipient.
SMTP SPAM Score Threshold
This is the SpamAssassin score threshold at which the SMTP server will reject a message before local delivery begins. If you don't want to block any spam at the SMTP level, use a very high number.
Rewrite Header Subject
E-mails marked as SPAM will have the email subject rewritten with what is entered here. Corresponds to a server-wide spam preference setting of "rewrite_header subject <VALUE>".
Reporting Method
This option sets the method of delivering the actual e-mail spam message. The email can be sent unchanged but with the subject rewritten, it can be attached to an email from SpamAssassin, or it can be attached as plain text to an email from SpamAssassin. Corresponds to a server-wide spam preference setting of "report_safe" with a value of 0, 1, or 2 respectively.
Auto Whitelisting
Enable or disable the SpamAssassin AutoWhitelist functionality. See related documentation for details.
Related Documentation: • SpamAssassin Bayes Documentation
Bayesian filtering is essentially user-driven spam filtering dictated by users flagging messages as spam or non-spam (ham). By keeping an even number of spam and non-spam messages flagged, the system is able to filter messages based on words present in the message.
Bayes Scanning
Enable or Disable the Bayes Database system server-wide.
Bayes Auto Learning
Enable or disable Bayes auto learning. If enabled, the system will automatically train the Bayes database by analyzing messages that have a score that strongly suggests that they are spam or non-spam.
Bayes IMAP Training
Enable or disable IMAP training. With Bayes IMAP training enabled, there will be two folders created for all e-mail accounts on the system: "Learn Spam" and "Learn Ham". You can then put SPAM emails in the "Learn Spam" folder and non-spam in the "Learn Ham" folder. This will train the Bayes system on which emails are Spam and which are Ham (not Spam). The training occurs once daily, after which the folders' contents are purged.
Bayes IMAP Learning Limit
The maximum number of e-mails that will be trained from each of the IMAP Training Folders, "Learn Spam" and "Learn Ham," each day.
Global Bayes Database
Enable or disable the Bayes global database. If enabled, all the Bayes data will be stored as a single global user. If disabled, each e-mail box user will have their own Bayes database to store bayesian data. "Disabled" is considered more effective for fighting Spam, but requires all e-mail users to train their messages individually, and uses more disk space, compared to the global database option.
Related Documentation: • SpamAssassin Configuration Docs
SpamAssassin allows the ability to globally designate "good" and "bad" email addresses to specifically target specific spammers or domains and also to allow certain users/domains which are turning up false-positive.
Global Whitelist
Enter e-mail addresses or domains, one per line, that you want to guarantee will NOT be tagged as Spam.
Global Blacklist
Enter e-mail addresses or domains, one per line, that you want to guarantee WILL be tagged as Spam.
Trusted Networks
Enter networks or hosts that are 'trusted' in your setup. Trusted in this case means that relay hosts on these networks are considered to not be potentially operated by spammers, open relays, or open proxies. A trusted host could conceivably relay spam, but will not originate it, and will not forge header data.
Click the System Services menu item if it is not already open.
Click the Mail Server menu item if it is not already open.
Click the Spam Filtering menu item.
Locate the Spam Access Lists section.
Enter desired values into the large text boxes. Global Whitelist and Blacklist values should be either e-mail addresses or domains, one per line. Trusted Networks values should be IP addresses or CIDR format, one per line
Click the button in the SPAM Access Lists.
You will see the following message at the top of the screen: Access lists updated successfully.
SPAM Filtering
InterWorx uses SpamAssassin to help keep levels of email spam reasonable. The page in NodeWorx deals with settings that are global across the entire server.
Related Documentation: • SpamAssassin Configuration Docs • SpamAssassin Autowhitelist Docs
Table Of Contents
SpamAssassin SMTP Scanning and Global Options
SpamAssassin can be configured to block spam at the SMTP leve before the message is even processed by the mail delivery system and sent to a local mailbox. SpamAssassin will refer to a rules database when checking incoming messages and assign a score to each message based on how likely it thinks a message is spam. A message with a higher spam score then a user-configurable threshold will automatically be refused by the SMTP service.
SMTP and Global Scanning Options
SMTP Level Scanning
This option will scan all e-mail using the SpamAssassin content filter as the e-mail enters the server. The SMTP connection will be dropped only if the Spam Score is higher than the SMTP Spam Score option below. Otherwise, the e-mail will still be delivered to the recipient.
SMTP SPAM Score Threshold
This is the SpamAssassin score threshold at which the SMTP server will reject a message before local delivery begins. If you don't want to block any spam at the SMTP level, use a very high number.
Rewrite Header Subject
E-mails marked as SPAM will have the email subject rewritten with what is entered here. Corresponds to a server-wide spam preference setting of "rewrite_header subject <VALUE>".
Reporting Method
This option sets the method of delivering the actual e-mail spam message. The email can be sent unchanged but with the subject rewritten, it can be attached to an email from SpamAssassin, or it can be attached as plain text to an email from SpamAssassin. Corresponds to a server-wide spam preference setting of "report_safe" with a value of 0, 1, or 2 respectively.
Auto Whitelisting
Enable or disable the SpamAssassin AutoWhitelist functionality. See related documentation for details.
How to Configure in NodeWorx
SpamAssassin Bayesian Filtering
Related Documentation: • SpamAssassin Bayes Documentation
Bayesian filtering is essentially user-driven spam filtering dictated by users flagging messages as spam or non-spam (ham). By keeping an even number of spam and non-spam messages flagged, the system is able to filter messages based on words present in the message.
Bayesian Scanning Options
Bayes Scanning
Enable or Disable the Bayes Database system server-wide.
Bayes Auto Learning
Enable or disable Bayes auto learning. If enabled, the system will automatically train the Bayes database by analyzing messages that have a score that strongly suggests that they are spam or non-spam.
Bayes IMAP Training
Enable or disable IMAP training. With Bayes IMAP training enabled, there will be two folders created for all e-mail accounts on the system: "Learn Spam" and "Learn Ham". You can then put SPAM emails in the "Learn Spam" folder and non-spam in the "Learn Ham" folder. This will train the Bayes system on which emails are Spam and which are Ham (not Spam). The training occurs once daily, after which the folders' contents are purged.
Bayes IMAP Learning Limit
The maximum number of e-mails that will be trained from each of the IMAP Training Folders, "Learn Spam" and "Learn Ham," each day.
Global Bayes Database
Enable or disable the Bayes global database. If enabled, all the Bayes data will be stored as a single global user. If disabled, each e-mail box user will have their own Bayes database to store bayesian data. "Disabled" is considered more effective for fighting Spam, but requires all e-mail users to train their messages individually, and uses more disk space, compared to the global database option.
How to Configure in NodeWorx
SpamAssassin Access List-Based Filtering
Related Documentation: • SpamAssassin Configuration Docs
SpamAssassin allows the ability to globally designate "good" and "bad" email addresses to specifically target specific spammers or domains and also to allow certain users/domains which are turning up false-positive.
Access List Descriptions
Global Whitelist
Enter e-mail addresses or domains, one per line, that you want to guarantee will NOT be tagged as Spam.
Global Blacklist
Enter e-mail addresses or domains, one per line, that you want to guarantee WILL be tagged as Spam.
Trusted Networks
Enter networks or hosts that are 'trusted' in your setup. Trusted in this case means that relay hosts on these networks are considered to not be potentially operated by spammers, open relays, or open proxies. A trusted host could conceivably relay spam, but will not originate it, and will not forge header data.
Configure the SPAM Access Lists