The directory you want JSpamFilter to create log files in. The files will be
named "jspamfilteryyyymmdd.log", based on the current date.
This will override LogFile, if specified. New files will be created as midnight
NOTE: you must specify a valid directory name for the logging directory or an error will occur!
The name of the file you want JSpamFilter to log to. The file will be appended
to (not overwritten) when JSpamFilter restarts, and may quickly grow very large.
Instead, the LogDir option is recommended. If neither LogDir nor LogFile are
specified, JSpamFilter will log to the console.
The amount of data you want to log; the default is "Verbose". "Actions"
will only log actions taken by JSpamFilter; "Verbose" will log all
messages, and "Debug" logs all SMTP commands.
If you want the log file to show the filter entries hit, add the following line to your jspamfilter.conf file:
Depending on the log-level used and the amount of email traffic, JSpamFilter logs can grow rapidly. For many businesses, daily log files
will likely be under a megabyte per day but, however, we recommend that you watch the log files after first implementing JSpamFilter to get an idea of how quickly they will grow in your system. Establish
a regular maintenance schedule to keep track of logs, in particular if disk space is a consideration.
JSpamFilter now gives you the option of a "dead letter mailbox" that lets you direct messages blocked after filtering to a specific mailbox. To use this feature, add the following line to your jspamfilter.conf file:
To prevent abuse of Inter-Exchange connections based on the proprietary Microsoft XEXCH50 protocol from bypassing JSpamFilter, you can add the following line to your jspamfilter.conf:
Optional, default 10000. This controls how many bytes of each message are
scanned for filter keywords. Increasing this value will increase the CPU utilization
of JSpamFilter, but will also defeat spammer tricks such as putting 7k of whitespace
at the top of HTML e-mails.
The first 10k of the message is scanned for these keyphrases, and the sum of scores (for each different keyphrase found) is totaled.
The full path to the filter.txt file, which determines the SPAM scoring.
A sample filter.txt is included in the distribution zip file.