Whitelist / Blacklist Amavis SpamAssassin Zimbra 8.6

Sometimes SpamAssassin scores email as False Positive spam, to avoid incoming emails to get junked, we can define globally in the config file /opt/zimbra/conf/amavisd.conf.in domain with a initial score.

#  read_hash("/var/amavis/sender_scores_sitewide"),

   { # a hash-type lookup table (associative array)
     'nobody@cert.org'                        => -3.0,
     'cert-advisory@us-cert.gov'              => -3.0,
     'owner-alert@iss.net'                    => -3.0,
     'slashdot@slashdot.org'                  => -3.0,
     ...

To whitelist a domain we add domain with a negative score:

#  read_hash("/var/amavis/sender_scores_sitewide"),

   { # a hash-type lookup table (associative array)
     'mydomain.com'                           => -3.0,
     'nobody@cert.org'                        => -3.0,
     'cert-advisory@us-cert.gov'              => -3.0,
     'owner-alert@iss.net'                    => -3.0,
     'slashdot@slashdot.org'                  => -3.0,
     ...

To blacklist a domain we add domain with positive score:

#  read_hash("/var/amavis/sender_scores_sitewide"),

   { # a hash-type lookup table (associative array)
     'spammer.com'                            =>  5.0,
     'nobody@cert.org'                        => -3.0,
     'cert-advisory@us-cert.gov'              => -3.0,
     'owner-alert@iss.net'                    => -3.0,
     'slashdot@slashdot.org'                  => -3.0,
     ...

To apply the modification restart Amavis:

su - zimbra
zmamavisdctl restart

Now from the source of incoming emails we can see when it’s coming from mydomain.com the initial score is -3 and spammer.com is 5.

X-Virus-Scanned: amavisd-new at mail.numberspeaks.com
X-Spam-Flag: NO
X-Spam-Score: -2
X-Spam-Level:
X-Spam-Status: No, score=-2 tagged_above=-10 required=6 tests=[AM.WBL=-3,
	BAYES_40=-0.001, HTML_MESSAGE=0.001, KAM_LAZY_DOMAIN_SECURITY=1]
	autolearn=no autolearn_force=no

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.