Read-only mirror of Apache SpamAssassin. Submit patches to Do not send pull requests
Perl C Shell Groff DIGITAL Command Language Python Other
Latest commit e8eca0f Sep 26, 2016 spamassassin_role updated scores for revision 1762298 active rules added since last mas…

git-svn-id: 13f79535-47bb-0310-9956-ffa450edef68
Failed to load latest commit information.
backend Fixing issues with extract_to_rsync_dir Jun 16, 2014
build Clarifying Copyright - bug 7263 Nov 16, 2015
contrib git-svn-id: Aug 29, 2016
debian Debian build file changes per bug 6685 Oct 28, 2011
ldap remove references to, replace with issues.a… Jun 29, 2006
lib Bug 6608: Make thinks that Digest::SHA is optional AND required - Mak… Jun 23, 2016
lm bug 5032: update ASF license text Aug 16, 2006
masses git-svn-id: Feb 28, 2016
powered_by fixing MANIFEST and powered by Apr 14, 2015
rules-extras git-svn-id: Sep 1, 2016
rules promotions validated Sep 25, 2016
rulesrc updated scores for revision 1762298 active rules added since last mas… Sep 27, 2016
spamc Bug 7321: impossible to disable ipv6 in spamc - document options -4 a… Jun 15, 2016
spamd-apache2 Cleanup of spamd-apache2 for httpd 2.4 - bug 7197 May 22, 2015
spamd Bug 7199: Bad SSL/TLS Version Default - applied Proposed Patch v2: su… Jun 15, 2016
sql fix for txrep sql that is not valid postgres Jun 29, 2015
t.rules Removed tests for FM_FRM_RN_L_BRACK Feb 5, 2014
t Bug 7307: Base64 decoded headers have invisible characters sometimes … Jun 23, 2016
tools git-svn-id: Apr 15, 2016
xt replacing dkim test with dkim2 Jan 11, 2013
CREDITS git-svn-id: Jul 1, 2016
Changes Preparing to release 3.4.0 Apr 28, 2015
INSTALL Another pass at bookkeeping for TxRep to succeed AWL Jan 12, 2015
INSTALL.VMS remove references to, replace with issues.a… Jun 29, 2006
LICENSE * incubator/spamassassin/ Jun 27, 2004
MANIFEST Bug 7307: Base64 decoded headers have invisible characters sometimes … Jun 20, 2016
MANIFEST.SKIP preparing to release 3.4.0-pre2 Jan 12, 2013
META.yml Bug 6599: Add Geo::IP support to RelayCountries Apr 1, 2012
Makefile.PL Windows option to enable building spamd Nov 10, 2015
NOTICE bug 5833: update ECCN status for SpamAssassin to note 5D002 status Feb 20, 2008
PACKAGING remove 'Building with another ExtUtils::MakeMaker' -- it's fragile, u… Jul 1, 2009
README More preparations for 3.4.1 Apr 17, 2015
TRADEMARK bug 3630: tweak text to match Apache License 2.0 better Aug 12, 2004
UPGRADE added normalize_charset 1 commented out in, and a note in UP… Apr 13, 2015
USAGE Another pass at bookkeeping for TxRep to succeed AWL Jan 12, 2015
procmailrc.example bug 4976: increase default message size limit from 250KB to 500KB, to… Jul 18, 2006
sa-awl.raw Bug 6793: reduce sa-awl memory usage Apr 24, 2012
sa-check_spamd.raw Clarifying Copyright - bug 7263 Nov 16, 2015
sa-compile.raw Clarifying Copyright - bug 7263 Nov 16, 2015
sa-learn.raw Bug 7232: Getting rid of "use bytes" crouches throughout Aug 6, 2015
sa-update.raw Bug 7266: scheme name is case insensitive, digits 1-8 are allowed too Nov 24, 2015
sample-nonspam.txt added new files Apr 22, 2001
sample-spam.txt * incubator/spamassassin/ Jun 27, 2004
spamassassin.raw Bug 7250: Changing a default setting of "report_charset" to UTF-8 Oct 8, 2015


Welcome to Apache SpamAssassin!

What Apache SpamAssassin Is

Apache SpamAssassin is the #1 Open Source anti-spam platform giving
system administrators a filter to classify email and block "spam"
(unsolicited bulk email).  It uses a robust scoring framework and plug-ins
to integrate a wide range of advanced heuristic and statistical analysis
tests on email headers and body text including text analysis, Bayesian 
filtering, DNS blocklists, and collaborative filtering databases.

Apache SpamAssassin is a project of the Apache Software Foundation (ASF).

What Apache SpamAssassin Is Not

Apache SpamAssassin is not a program to delete spam, route spam and ham to
separate mailboxes or folders, or send bounces when you receive spam.
Those are mail routing functions, and Apache SpamAssassin is not a mail
router.  Apache SpamAssassin is a mail filter or classifier.  It will examine
each message presented to it, and assign a score indicating the
likelihood that the mail is spam.  An external program must then
examine this score and do any routing the user wants done.  There are
many programs that will easily perform these functions after examining
the score assigned by Apache SpamAssassin.

How Apache SpamAssassin Works

Apache SpamAssassin uses a wide range of heuristic tests on mail headers and
body text to identify "spam", also known as unsolicited commercial

Once identified, the mail can then be optionally tagged as spam for
later filtering using the user's own mail user-agent application.

Apache SpamAssassin typically differentiates successfully between spam and
non-spam in between 95% and 100% of cases, depending on what kind of mail
you get and your training of its Bayesian filter.  Specifically,
Apache SpamAssassin has been shown to produce around 1.5% false negatives (spam
that was missed) and around 0.06% false positives (ham incorrectly marked
as spam).  See the rules/STATISTICS*.txt files for more information.

Apache SpamAssassin also includes plugins to support reporting spam messages
automatically or manually to collaborative filtering databases such as
Pyzor, DCC, and Vipul's Razor.

The distribution provides "spamassassin", a command line tool to
perform filtering, along with the "Mail::SpamAssassin" module set
which allows Apache SpamAssassin to be used in spam-protection proxy SMTP or
POP/IMAP server, or a variety of different spam-blocking scenarios.

In addition, "spamd", a daemonized version of Apache SpamAssassin which
runs persistently, is available.  Using its counterpart, "spamc",
a lightweight client written in C, an MTA can process large volumes of
mail through Apache SpamAssassin without having to fork/exec a perl interpreter
for each message.

Questions? Need Help?

If you have questions about Apache SpamAssassin, please check the Wiki[1] to
see if someone has already posted an answer to your question. (The
Wiki doubles as a FAQ.) Failing that, post a message to the
spamassassin-users mailing list[2]. If you've found a bug (and you're
sure it's a bug after checking the Wiki), please file a report in our


Please also be sure to read the man pages.

Upgrading Apache SpamAssassin

IMPORTANT: If you are upgrading from a previous major version of
Apache SpamAssassin, please be sure to read the notes in UPGRADE to find out
what has changed in a non- backward compatible way.

Installing Apache SpamAssassin

See the INSTALL file.

Customizing Apache SpamAssassin

These are the configuration files installed by Apache SpamAssassin.  The commands
that can be used therein are listed in the POD documentation for the
Mail::SpamAssassin::Conf class (run the following command to read it:
"perldoc Mail::SpamAssassin::Conf").  Note: The following directories are
the standard defaults that people use.  There is an explanation of all the
default locations that Apache SpamAssassin will look at the end.

  - /usr/share/spamassassin/*.cf:

	Distributed configuration files, with all defaults.  Do not modify
	these, as they are overwritten when you upgrade.

  - /var/lib/spamassassin/*/*.cf:

        Local state directory; updated rulesets, overriding the
        distributed configuration files, downloaded using "sa-update". Do
        not modify these, as they are overwritten when you run

  - /etc/mail/spamassassin/*.cf:

  	Site config files, for system admins to create, modify, and
	add local rules and scores to.  Modifications here will be
	appended to the config loaded from the above directory.

  - /etc/mail/spamassassin/*.pre:

        Plugin control files, installed from the distribution. These are
        used to control what plugins are loaded.  Modifications here will
        be loaded before any configuration loaded from the above
        You want to modify these files if you want to load additional
        plugins, or inhibit loading a plugin that is enabled by default.
        If the files exist in /etc/mail/spamassassin, they will not
        be overwritten during future installs.

  - /usr/share/spamassassin/user_prefs.template:

	Distributed default user preferences. Do not modify this, as it is
	overwritten when you upgrade.

  - /etc/mail/spamassassin/user_prefs.template:

	Default user preferences, for system admins to create, modify, and
	set defaults for users' preferences files.  Takes precedence over
	the above prefs file, if it exists.

        Do not put system-wide settings in here; put them in a file in the
        "/etc/mail/spamassassin" directory ending in ".cf". This file is
        just a template, which will be copied to a user's home directory
        for them to change.

  - $USER_HOME/.spamassassin:

  	User state directory.  Used to hold spamassassin state, such
	as a per-user automatic whitelist, and the user's preferences

  - $USER_HOME/.spamassassin/user_prefs:

  	User preferences file.  If it does not exist, one of the
	default prefs file from above will be copied here for the
	user to edit later, if they wish.

	Unless you're using spamd, there is no difference in
	interpretation between the rules file and the preferences file, so
	users can add new rules for their own use in the
	"~/.spamassassin/user_prefs" file, if they like.  (spamd disables
	this for security and increased speed.)

  - $USER_HOME/.spamassassin/bayes*

	Statistics databases used for Bayesian filtering.  If they do
	not exist, they will be created by Apache SpamAssassin.

	Spamd users may wish to create a shared set of bayes databases;
	the "bayes_path" and "bayes_file_mode" configuration settings
	can be used to do this.

	See "perldoc sa-learn" for more documentation on how
	to train this.

File Locations:

Apache SpamAssassin will look in a number of areas to find the default
configuration files that are used.  The "__*__" text are variables
whose value you can see by looking at the first several lines of the
"spamassassin" or "spamd" scripts.

They are set on install time and can be overridden with the Makefile.PL
command line options DATADIR (for __def_rules_dir__) and CONFDIR (for
__local_rules_dir__).  If none of these options were given, FHS-compliant
locations based on the PREFIX (which becomes __prefix__) are chosen.
These are:

  __prefix__    __def_rules_dir__              __local_rules_dir__
  /usr          /usr/share/spamassassin        /etc/mail/spamassassin
  /usr/local    /usr/local/share/spamassassin  /etc/mail/spamassassin
  /opt/$DIR     /opt/$DIR/share/spamassassin   /etc/opt/mail/spamassassin
  $DIR          $DIR/share/spamassassin        $DIR/etc/mail/spamassassin

The files themselves are then looked for in these paths:

  - Distributed Configuration Files

  - Site Configuration Files

  - Default User Preferences File

In addition, the "Distributed Configuration Files" location is overridden
by a "Local State Directory", used to store an updated copy of the

  __prefix__    __local_state_dir__
  /usr          /var/lib/spamassassin/__version__
  /usr/local    /var/lib/spamassassin/__version__
  /opt/$DIR     /var/opt/spamassassin/__version__
  $DIR          $DIR/var/spamassassin/__version__

This is normally written to by the "sa-update" script.  "__version__" is
replaced by a representation of the version number, so that multiple
versions of Apache SpamAssassin will not interfere with each other's rulesets.

After installation, try "perldoc Mail::SpamAssassin::Conf" to see what
can be set. Common first-time tweaks include:

  - required_score

	Set this higher to make Apache SpamAssassin less sensitive.
        If you are installing Apache SpamAssassin system-wide, this is
        **strongly** recommended!

        Statistics on how many false positives to expect at various
        different thresholds are available in the "STATISTICS.txt" file in
        the "rules" directory.

  - rewrite_header, add_header

        These options affect the way messages are tagged as spam or
	non-spam. This makes it easy to identify incoming mail.

  - ok_locales

	If you expect to receive mail in non-ISO-8859 character sets (ie.
	Chinese, Cyrillic, Japanese, Korean, or Thai) then set this.


Apache SpamAssassin includes a Bayesian learning filter, so it is worthwhile
training Apache SpamAssassin with your collection of non-spam and spam,
if possible.  This will make it more accurate for your incoming mail.
Do this using the "sa-learn" tools, like so:

	sa-learn --spam ~/Mail/saved-spam-folder
	sa-learn --ham ~/Mail/inbox
	sa-learn --ham ~/Mail/other-nonspam-folder

If these are mail folders in mbox format, use the --mbox switch, for
Maildirs use a trailing slash, like Maildir/cur/.

Use as many mailboxes as you like.  Note that Apache SpamAssassin will remember
what mails it has learnt from, so you can re-run this as often as you like.


All text displayed to users is taken from the configuration files.  This
means that you can translate messages, test descriptions, and templates
into other languages.

If you do so, we would *really* appreciate it if you could contribute
these translations, so that they can be added to the
distribution. Please file a bug in our Bugzilla[4], and attach your
translations. You will, of course, be credited for this work!


Disabled code

There are some tests and code in Apache SpamAssassin that are turned off by
default: experimental code, slow code, or code that depends on
non-open-source software or services that are not always free.  These
disabled tests include:

  - DCC: depends on non-open-source software (disabled in init.pre)
  - MAPS: commercial service (disabled in
  - TextCat: slow (disabled in init.pre)
  - various optional plugins, disabled for speed (disabled in *.pre)

To turn on tests disabled in, simply assign them a non-zero
score, e.g. by adding score lines to your ~/.spamassassin/user_prefs file.

To turn on tests disabled by commenting out the required plugin in
init.pre, you need to uncomment the loadplugin line and make sure the
prerequisites for proper operation of the plugin are present.

Automatic Reputation System

Apache SpamAssassin includes an automatic reputation system. The way it works is
by tracking for each sender address a rolling average score of messages
so far seen from there.  Then, it combines this long-term average score
for the sender with the score for the particular message being evaluated,
after all other rules have been applied.

This functionality can be enabled or disabled with the
"use_txrep" option.

For more information, read sql/README.txrep

(end of README)

// vim:tw=74: