PC World 2007 June/PCWorld_2007-06_cd.bin/akce/Panda/IS2007.exe/Files/RAntispam/spamcatcher.conf

Text File | 2006-11-28 | 24.3 KB | 859 lines

###########################################################################
# spamcatcher.conf
#
# Sample configuration file for spamcatcher. Format of a line is:
#
# key=value
#
# Blank lines are ignored, as is anything starting with a '#' character
#
#
############################################################################

# Name: regkey
# Arguments: xxxx-xxxx-xxxx-xxxx
# Default: none
#
# Description:
# A license key is needed to activate the SpamCatcher SDK
regkey=A3%2%-5$5$$-EJ$%8-Z5%RE
snhost=sn30.mailshell.net

# Name: ruleupdate
# Arguments: 0 | 600 - (2^31-1)
# Default: 3600
#
# Description:
# How often to retrieve new rules from the Mailshell SpamLabs. The
# value is specified in units of integral seconds.
# Note that a value of "0" disables this feature and rule files will not be updated.
ruleupdate=0



# Name: sntimeout
# Arguments: 0 - 2^31-1
# Default: 5
#
# Description:
# Limit how long single request to the Mailshell SpamLabs can take. The value
# is specified in units of integral seconds.
# Note that a value of "0" disables this feature and no limit will be placed.
# sntimeout=5



# Name: use_both_mimesections
# Arguments: yes | no
# Default: yes
#
# Description:
# The SDK will analyze both text/plain and text/html MIME sections in a
# message. If additional performance is desired, it is possible to only analyze
# one section. If this option is set to "no", then only one section will be analyzed.
# use_both_mimesections=yes



# Name: use_score_history
# Arguments: yes | no
# Default: no
#
# Description:
# Enables the tracking of historical scores for repeat senders.
use_score_history=yes



# Name: use_score_offsets
# Arguments: yes | no
# Default: no
#
# Description:
# Enables fingerprint score offset training.
use_score_offsets=yes



# Name: use_https
# Arguments: yes | no
# Default: no
#
# Description:
# Communication between the SDK and the Mailshell SpamLabs is always
# encrypted. This option is now obsoleted and HTTP (port 80) is always
# used.
# use_https=no



# Name: verbose
# Arguments: yes | no
# Default: no
#
# Description:
# Enables increased verbose logging.
# verbose=no



# Name: dbg_logfile
# Arguments: <filename>
# Default: none
#
# Description:
# Redirects log output to a file in the conf directory.
# dbg_verbose_level=6
# dbg_logfile=mailshell.log


# Name: extended_rules
# Arguments: yes | no
# Default: yes
#
# Description:
# Enables the extended rule set. This replaces the old use_cache option. The
# old use_cache option is now equivalent to this option.
# This rule set extension is stored as as file sc1.bin.<date> and sc12.bin.<date>
# extended_rules=yes


# Name: extended_rules2
# Arguments: yes | no
# Default: yes
#
# Description:
# Enables the second extension to rule set.
# This rule set extension is stored as as file sc6.bin.<date>
extended_rules2=yes


# Name: enable_fingerprint_cache
# Arguments: yes | no
# Default: yes
#
# Description:
# Enables usage of a fingerprint cache.
# The fingerprint cache is stored as file sc10.bin.<date>.
enable_fingerprint_cache=yes


# Name: enable_domain_cache
# Arguments: yes|no
# Default: yes
#
# Description:
# Enables usage of a domain reputation cache.
# If enabled, domains are extracted from messages and compared against
# a domain reputation caches.
#
# The domain reputation cache is stored in file sc8.bin.<date>
enable_domain_cache=yes


# Name: netcheck
# Arguments: yes | no
# Default: no
#
# Description:
# Whether to communicate with the Mailshell SpamLabs to determine scoring.
netcheck=yes



# Name: pcre_match_limit
# Arguments: 0 - 2^32-1
# Default: 100 for FreeBSD, 10000 for others
#
# Description:
# Sets the maximum thread stack size to use. If the thread stack size is set
# to 64KB (e.g. FreeBSD), then this variable should be set to 100 or less. If the
# thread stack size is set to greater than 1MB (e.g. Linux, Solaris, and MacOSX),
# then this variable should be set to 10000 or less. If this variable is set
# below 200, accuracy can be reduced by a couple of percentages.
# pcre_match_limit=10000



# Name: proxy_host
# Arguments: host:port
# Default: none
#
# Description:
# Specifies the host name and port number of a HTTPS proxy to connect to
# the Mailshell servers.
# proxy_host=



# Name: proxy_userpwd
# Arguments: username:password
# Default: none
#
# Description:
# Specifies the user name and password of the HTTPS proxy to connect to
# the Mailshell servers.
# proxy_userpwd=



# Name: rbl_list
# Arguments: servername:response:offset,
#       servername:response:offset,...
# Default: none
#
# Description:
# Specifies a list of Realtime Blackhole List (RBL) servers to query when
# analyzing messages. Please refer to the RBL section in this document for
# further information.
# rbl_list=


# Name: rbl_multihit
# Arguments: yes | no
# Default: no
#
# Description:
# If this option is no, then the rbl servers are checked in the order that
# they are listed. Once an IP address is found to be listed on an RBL
# server, no further RBL queries are made.
#
# If this option is yes, then all of the rbl servers are checked in parallel
# and the hits which return are added together.
#
# rbl_list must be specified or this option is ignored.
# rbl_multihit=



# Name: enable_word_training
# Arguments: yes | no
# Default: yes
#
# Description:
# This option controls whether Bayesian Word Token analysis is
# used. Accuracy can be greatly improved but more memory is used and
# it is slightly slower.
#
# "no" - disables option
# "yes" - enables option
enable_word_training=yes



# Name: max_word_entries
# Arguments: integer
# Default: 50000
#
# Description:
# This option specifies the number of word tokens to cache at any time.
# The higher the number, the more memory is used but also the higher the
# accuracy. enable_word_training must be yes or this option is ignored.
# max_word_entries = 50000



# Name: enable_rules
# Arguments: yes | no
# Default: yes
#
# Description:
# This option controls whether heuristic rules (like before v3.0) are
# used. Accuracy can be greatly improved but more memory is used and it is much slower.
# The rules are stored as file sc1.bin.<date>
#
# "no" - disables option
# "yes" - enables option
# enable_rules=yes



# Name: rule_weights
# Arguments: ruleidA:weightA,ruleidB:weightB,...
# Default: none
#
# Description:
# This option allows overriding weights associated with individual
# Mailshell Rules.  The format is a comma delimited list of "ruleid" and
# "weight" pairs.  The pairs are themselves delimited with a colon.
# rule_weights=



# Name: enable_training_updates
# Arguments: yes | no
# Default: yes
#
# Description:
# This option controls whether the word, rules, and fingerprint training database
# can be modified or is read-only after initial load. A read-only training database is faster.
#
# "no" - the training databases can be modified.
# "yes" - the training databases are read-only.
enable_training_updates=yes



# Name: auto_training_threshold
# Arguments: low:high
# Default: 1:99
#
# Description:
# Sets a threshold for auto-training.  If a message is scored at or above the
# high threshold, that message is considered a definite spam and is then used
# to train all the enabled bayesian modules (rules and/or word) but not
# sender or fingerprint.
#
# If a message is scored at or below the low threshold, that message is considered
# a definite ham and is then used to train all the enabled bayesian modules
# (rules and/or word) but not sender or fingerprint. enable_word_training and
# enable_training_updates must be yes or this option is ignored.
auto_training_threshold=1:99



# Name: min_training
# Arguments: integer
# Default: 100
#
# Description:
# Initially, only the rule weights are used to compute the spam score. Once a minimum
# set of training data is achieved, rule/word training data replaces the rule weights.
# The default minimum is 100 which means that must train on at least 100 equivalent known
# ham messages and 100 equivalent spam messages for a total of 200 messages before
# the training data replaces the rule weights. If the number is too low then the accuracy could
# be poor due to insufficient data. If the number is too high, then the training
# data will not be fully taken advantage of. A value of 0 will cause
# rule weights to always be ignored.
min_training=100



# Name: full_training_weight
# Arguments: yes | no
# Default: no
#
# Description:
# Controls whether to give full weight to training data.  If this option
# is set to "yes" then scoring will be based solely on training data.
# If option is "no" then both rules and training data will be used.
# full_training_weight=no



# Name: training_write_buffer
# Arguments: an integer
# Default: 1000
#
# Description:
# While training, the SDK will process a configurable amount of messages
# before writing the training database to disk.  This option determines
# how many messages to process before writing to disk.
#
# Writing to disk is expensive so this number you should made as large
# as possible for maximum performance.
#
# If program is unexpectedly terminated before buffer has been written to
# disk, then training performed since the last disk write will be lost.
# The buffer is written to disk on normal termination.
# training_write_buffer=1000



# Name: ignored_ip_list
# Arguments: 1.2.3.4,2.3.4.5-2.3.4.8,2.3.4.0/24,5.6.0.0/16
# Default: none
#
# Description:
# This option allows specifying IPs which should be ignored when doing
# RBL checks.  Format is a comma delimited list of single IPs or ranges of
# IPs. The following are always implicitly ignored:
# 10.0.0.0/8, 127.0.0.0/8, 192.168.0.0/16, 172.16.0.0
#
# Ranges can be specified in three ways:
#   a) startingIP-endingIP
#   b) IP/netmask
#   c) IP
# ignored_ip_list=



# Name: approved_ip_list
# Arguments: 1.2.3.4,2.3.4.5-2.3.4.8,2.3.4.0/24,5.6.0.0/16
# Default: none
#
# Description:
# This option allows specifying IPs which should be approved
# Format is a comma delimited list of single IPs or ranges of IPs.
# Ranges can be specified in two ways:
#   a) startingIP-endingIP
#   b) IP/netmask
#   c) IP
#
# If the first non-ignored IP in Received: headers match any in this list
# then message is scored a 1 and no other checks are made.
# approved_ip_list=



# Name: blocked_ip_list
# Arguments: 1.2.3.4,2.3.4.5-2.3.4.8,2.3.4.0/24,5.6.0.0/16
# Default: none
#
# Description:
# This option allows specifying IPs which should be blocked.
# Format is a comma delimited list of single IPs or ranges of IPs.
#
# Ranges can be specified in three ways:
#   a) startingIP-endingIP
#   b) IP/netmask
#   c) IP
#
# If any IP addresses in Received: headers match any in this list
# then message is scored a 99 and no other checks are made.
# blocked_ip_list=



# Name: rbl_threshold
# Arguments: low:high
# Default: 1:99
#
# Description:
# Since RBL checks can introduce latency and a decrease in performance,
# this option allows running RBLs check conditionally based on the score
# prior to RBL checks.
#
# If score is greater than the "high" value then only those RBL servers
# which can bring score below "high" value are queried.
#
# If score is less than the "low" value then only those RBL servers
# which can bring score above "low" value are queried.
#
# If score is between "low" and "high" then all RBL servers are queried.
#
# rbl_list must be specified or this option is ignored.
# rbl_threshold=1:99



# Name: rbl_timeout
# Arguments: integer
# Default: 5
#
# Description:
# This option allows setting a maximum timeout for finishing all RBL
# queries.
# RBL responses are only used from those RBL servers which responded in
# time.
#
# If value is "0" then no timeout is enforced.
#
# rbl_list must be specified or this option is ignored.
# rbl_timeout=5



# Name: rbl_max_ips
# Arguments: integer
# Default: 4
#
# Description:
# This option allows limiting how many IP addresses are queried against
# the RBL server.  Note that the total number of RBL queries will be
# the number of IP addresses in the Received: headers (up to a max of
# rbl_maxcheck_ips) multiplied by the number of RBL servers specified in
# "rbl_list".
# If the value is "0" then unlimited number of received headers are checked.
#
# Note that IPs which match against the "ignored_ip_list" option do not
# count towards the rbl_max_ips limit.
#
# rbl_list must be specified or this option is ignored.
# rbl_max_ips=4


# Name: home_country_list
# Arguments: us,ca,kr,...
# Default: none
#
# Description:
# This option allows specifying a list of countries which are
# considered "home" countries.  Messages routed through a country
# which is not on this list will be scored more aggressively.
# If this option is empty then no penalty will occur.
#
# Countries are specified by their two-letter code as defined in ISO 3166
#
# home_country_list=



# Name: blocked_country_list
# Arguments: countryCode1:offset1,countryCode2:offset2...
# Default: none
#
# Description:
# Allows blocking by country.  The format is a comma delimited list of
# "country" and "offset" pairs which are themselves delimited by a
# colon.  Country is specified as two letter code (ISO-3166). Offsets
# are optional and default to a value of 100.
#
# If an IP address in a received header matches a listed country, then
# that offset is used and we stop checking any other IPs. If one
# is in Russia and the user has set "ru:someOffset", then it'll apply.
#
# The country codes aren't applied to sender addresses.
# If you want to block From addresses ending in .ru, you can use the email
# address block list.
#
# Note that it is possible for a message to have travelled through
# various countries before reaching the final destination.
#
# Note that this option is only 98% accurate so blocking countries can
# result in false positives.
# blocked_country_list=



# Name: blocked_charset_list
# Arguments: charset1:offset1,charset2:offset2...
# Default: none
#
# Description:
# Allows blocking by character-set.  The format is a comma delimited list of
# "char-set" and "offset" pairs which are themselves delimited by a
# colon.  A char-set to foreign language map can be found at:
# http://www.w3.org/International/O-charset-list.html
# Offsets are optional and default to a value of 100.
#
# Note that language to char-set mapping is not 100% accurate
# so blocking charsets can result in false positives.
# blocked_charset_list=



# Name: message_readsize
# Arguments: integer
# Default: 100000000

# Description:
# This option instructs the SDK not to read more than a configurable
# number of bytes from the message buffer when computing scores.
#
# This value is ignored if it is greater than the "size" parameter in the
# SCMessage::computeScore() or MailshellMsg_computeScore() functions.
# message_readsize=0


# Name: message_scansize
# Arguments: integer
# Default: 20000
#
# Description:
# This option instructs the SDK not to read more than "message_scansize"
# bytes when computing the message fingerprint.  This is useful to
# for getting consistent fingerprints.
#
# This value is ignored if it is greater than either the message_scansize
# value or the "size" parameter in the SCMessage::computeScore() or
# MailshellMsg_computeScore() functions.
# message_scansize=20000


# Name: netcheck_threshold
# Arguments: low:high
# Default: 1:99
#
# Description:
# Since networks can introduce latency and a decrease in performance,
# this option allows running network checks conditionally based on the
# score. netcheck must be yes or this option is ignored.
#
# Network is only queried if score is at or between the "low" and "high" range
# specified via this option.
netcheck_threshold=85:95


# Name: spoofed_sender_list
# Arguments: address1:iprange1:offset1,address2:iprange2:offset2...
# Default: none
#
# Description:
# Allows blocking spammers who spoof your domain name and
# other domain names. For example, spammers often use the recipient's
# domain name as the From: domain name. This list allows you to specify which
# mail servers are allowed to use which domain names in the From: address
# spoofed_sender_list=


# Name: scan_attachments
# Arguments: yes | no
# Default: no
#
# Description:
# This option controls whether the SDK will scan and consider attachments
# when computing the spam score.
# "no" - disables option
# "yes" - enables option
scan_attachments=no


# Name: enable_spamcompiler_cache
# Arguments: yes|no
# Default: yes
# Description: If this option is set to yes, SpamCompiler will store the
# compiled data on disk instead of memory to reduce memory usage.
# enable_spamcompiler_cache=yes


# Name: enable_legitrepute_cache
# Arguments: yes|no
# Default: yes
# Description: Enables usage of a LegitRepute cache to reduce false positives
# especially for newsletters.
# enable_legitrepute_cache=yes


# Name: convert_unicode
# Arguments: yes|no
# Default: yes
# Description: Improves accuracy and throughput for email message bodies in Unicode
# especially double-byte languages by converting the message into single-bytes.
# convert_unicode=yes

# Name: retrieverules_list
# Arguments: index1,index2,...
# Default: none
# Description: This option specifies which rules files are downloaded when
# retrieveRules() is called.  If the option is blank then retrieveRules()
# will only download files which are necessary to support a given
# configuration (based on other options).
# retrieverules_list=1,2,3,4,5,6,7,8,9,10,11,12,14,15


# Name: enable_spf
# Arguments: yes|no
# Default: yes
# Description:
# This option controls whether or not to do Sender Policy Framework (SPF)
# checks. If set to yes, then attempt to validate that sender is allowed
# to send from a particular domain based on the domain's published policy.
# enable_spf=yes


# Name: enable_all_spf
# Arguments: yes|no
# Default: no
# Description:
# This option controls whether or not domains which are not listed in
# the spf_list option or in Mailshell's rule file will undergo SPF checks.
# This option requires enable_realtime_spf to be set to yes.
# enable_all_spf=no


# Name: enable_realtime_spf
# Arguments: yes|no
# Default: no
# Description:
# This option controls whether live DNS queries will be performed for SPF
# checks. This may result in greater latency.
# enable_realtime_spf=no


# Name: spf_list
# Arguments:  domain1:"weights1":"spf record1",domain2:"weights2":"spf record2": ...
# Default: none
# Description:
# This option allows you to override a domain's SPF record. Through the "weights",
# you can specify how a particular SPF result will affect the message score.
# spf_list=


# Name: spf_pass_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Pass for the from domain.
# spf_pass_weight=0


# Name: spf_fail_weight
# Arguments: integer
# Default: 10
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Fail for the from domain.
# spf_fail_weight=10


# Name: spf_neutral_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Neutral for the from domain.
# spf_neutral_weight=0


# Name: spf_none_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is None for the from domain.
# spf_none_weight=0


# Name: spf_softfail_weight
# Arguments: integer
# Default: 5
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is SoftFail for the from domain.
# spf_softfail_weight=5


# Name: spf_permerror_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Permanent Error for the from domain.
# spf_permerror_weight=0


# Name: spf_temperror_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Temporary Error for the from domain.
# spf_temperror_weight=0


# Name: spf_helo_pass_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Pass for the HELO domain.
# spf_helo_pass_weight=0


# Name: spf_helo_fail_weight
# Arguments: integer
# Default: 10
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Fail for the HELO domain.
# spf_helo_fail_weight=10


# Name: spf_helo_neutral_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Neutral for the HELO domain.
# spf_helo_neutral_weight=0


# Name: spf_helo_none_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is None for the HELO domain.
# spf_helo_none_weight=0


# Name: spf_helo_softfail_weight
# Arguments: integer
# Default: 5
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is SoftFail for the HELO domain.
# spf_helo_softfail_weight=5


# Name: spf_helo_permerror_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Permanent Error for the HELO domain.
# spf_helo_permerror_weight=0


# Name: spf_helo_temperror_weight
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the default offset to apply to scores when
# the SPF result is Temporary Error for the HELO domain.
# spf_helo_temperror_weight=0


# Name: spam_threshold
# Arguments: integer
# Default: 100
# Description:
# This option allows you to tell the SDK to stop analyzing the message once
# a score has been reached.  This can reduce the number of rules and other
# checks that are performed, thus improving throughput.
# spam_threshold=100


# Name: target_throughput
# Arguments: integer
# Default: 0
# Description:
# This option allows you to specify the desired throughput in messages per
# second.  The Mailshell SDK will attempt to reach that level by optimizing
# the rules that are run. It is possible that accuracy may be reduced.
# A value of 0 disables the option.
target_throughput=0


# Name: enable_filecleanup_on_retrieve
# Arguments: yes | no
# Default:  yes
# Description:
# The SDK, by default, will clean up older rule files from the
# configuration directory when a new file is retrieved from the Mailshell
# SpamCatcher network.  However, some users of the SDK will want to
# archive older rule files.  This can be done by disabling the cleanup feature.
# enable_filecleanup_on_retrieve=yes


# Name: enable_filemerge_on_reload
# Arguments:  yes | no
# Default: yes
# Description:
# The SDK, by default, will merge multiple incr files and a full file into a
# single updated full file.  This is done to reduce file clutter in the
# configuration directory.
# enable_filemerge_on_reload=yes


# Name: retrieve_incr_only
# Arguments: yes | no
# Default: no
# Description:
# The SDK, by default, will attempt to download the most size efficient
# combination of full and incr file.  The SDK can be forced to only download
# incr file by setting this option to yes.
# retrieve_incr_only=no


# Name: home_language_list
# Arguments: languageCode1,languageCode2,...
# Default: none
# Description:
# This option permits you to set languages which are preferred in your email messages.
# The language codes are two character ISO-639 language codes. Please refer to the
# SDK documentation for the list of supported codes.
# home_language_list=