Unterschiede

Hier werden die Unterschiede zwischen zwei Versionen angezeigt.

Link zu dieser Vergleichsansicht

Beide Seiten der vorigen Revision Vorhergehende Überarbeitung
centos:dansg:virus [16.11.2011 14:06. ] – Seite fertiggestellt und freigegeben djangocentos:dansg:virus [20.04.2018 10:50. ] (aktuell) – Externe Bearbeitung 127.0.0.1
Zeile 1: Zeile 1:
 +====== Virenfilterung bei Dansguardian ======
 +{{:centos:dansguardian-logo.png?nolink&80|Dansguardian Logo}}\\
 +
 +Damit wir neben der inhaltlichen Bewertung des Webtraffics auch eine Überpfüfung des Verkehrs auf Schadcode vornehmen können, müssen wir erst einmal [[centos:clamav_c6|ClamAV unter CentOS 6.x]] installieren und konfigurien.
 +
 +Die nötigen Konfigurationsschritte für unseren WEB-Content-Scanner [[http://dansguardian.org|Dansgurdian]] werden nun in diesem Kapitel beschrieben.
 +
 +===== dansguardian.conf =====
 +Zur Aktivierung des Virenscanner beim Contentfiltern aktivieren wir nun den **clamd** in der //**/etc/dansguardian/dansguardian.conf**//. Ebenso geben wir an, wo die empfangenen Daten zwischengespeichert werden können/sollen.
 +
 +Die relevanten Konfigurationszeilen lautet hierzu:
 +  * **contentscanner = '/etc/dansguardian/contentscanners/clamdscan.conf'**
 +  * **filecachedir = '/tmp'**
 +
 +   # vim /etc/dansguardian/dansguardian.conf
 +
 +<file bash /etc/dansguardian/dansguardian.conf>
 +# DansGuardian config file for version 2.10.1.1
 +
 +# **NOTE** as of version 2.7.5 most of the list files are now in dansguardianf1.conf
 +
 +
 +# Web Access Denied Reporting (does not affect logging)
 +#
 +# -1 = log, but do not block - Stealth mode
 +#  0 = just say 'Access Denied'
 +#  1 = report why but not what denied phrase
 +#  2 = report fully
 +#  3 = use HTML template file (accessdeniedaddress ignored) - recommended
 +#
 +reportinglevel = 3
 +
 +# Language dir where languages are stored for internationalisation.
 +# The HTML template within this dir is only used when reportinglevel
 +# is set to 3. When used, DansGuardian will display the HTML file instead of
 +# using the perl cgi script.  This option is faster, cleaner
 +# and easier to customise the access denied page.
 +# The language file is used no matter what setting however.
 +#
 +languagedir = '/usr/share/dansguardian/languages'
 +
 +# language to use from languagedir.
 +# Django 2011-11-15
 +# Default: language = 'ukenglish'
 +language = 'german'
 +
 +# Logging Settings
 +#
 +# 0 = none  1 = just denied  2 = all text based  3 = all requests
 +# Django 2011-11-15 Anpassen des Log-Levels
 +# default : loglevel = 2
 +loglevel = 1
 +
 +# Log Exception Hits
 +# Log if an exception (user, ip, URL, phrase) is matched and so
 +# the page gets let through.  Can be useful for diagnosing
 +# why a site gets through the filter.
 +# 0 = never log exceptions
 +# 1 = log exceptions, but do not explicitly mark them as such
 +# 2 = always log & mark exceptions (default)
 +logexceptionhits = 2
 +
 +# Log File Format
 +# 1 = DansGuardian format (space delimited)
 +# 2 = CSV-style format
 +# 3 = Squid Log File Format
 +# 4 = Tab delimited
 +logfileformat = 1
 +
 +# truncate large items in log lines
 +#maxlogitemlength = 400
 +
 +# anonymize logs (blank out usernames & IPs)
 +#anonymizelogs = on
 +
 +
 +# Syslog logging
 +#
 +# Use syslog for access logging instead of logging to the file
 +# at the defined or built-in "loglocation"
 +#syslog = on
 +
 +# Log file location
 +
 +# Defines the log directory and filename.
 +#loglocation = '/var/log/dansguardian/access.log'
 +
 +
 +# Statistics log file location
 +#
 +# Defines the stat file directory and filename.
 +# Only used in conjunction with maxips > 0
 +# Once every 3 minutes, the current number of IPs in the cache, and the most
 +# that have been in the cache since the daemon was started, are written to this
 +# file. IPs persist in the cache for 7 days.
 +#statlocation = '/var/log/dansguardian/stats'
 +
 +
 +# Network Settings
 +
 +# the IP that DansGuardian listens on.  If left blank DansGuardian will
 +# listen on all IPs.  That would include all NICs, loopback, modem, etc.
 +# Normally you would have your firewall protecting this, but if you want
 +# you can limit it to a certain IP. To bind to multiple interfaces,
 +# specify each IP on an individual filterip line.
 +filterip =
 +
 +# the port that DansGuardian listens to.
 +filterport = 8080
 +
 +# the ip of the proxy (default is the loopback - i.e. this server)
 +proxyip = 127.0.0.1
 +
 +# the port DansGuardian connects to proxy on
 +proxyport = 3128
 +
 +# Whether to retrieve the original destination IP in transparent proxy
 +# setups and check it against the domain pulled from the HTTP headers.
 +#
 +# Be aware that when visiting sites which use a certain type of round-robin
 +# DNS for load balancing, DG may mark requests as invalid unless DG gets
 +# exactly the same answers to its DNS requests as clients.  The chances of
 +# this happening can be increased if all clients and servers on the same LAN
 +# make use of a local, caching DNS server instead of using upstream DNS
 +# directly.
 +#
 +# See http://www.kb.cert.org/vuls/id/435052
 +# on (default) | off
 +#!! Not compiled !! originalip = on
 +
 +# accessdeniedaddress is the address of your web server to which the cgi
 +# dansguardian reporting script was copied. Only used in reporting levels 1 and 2.
 +#
 +# This webserver must be either:
 +#  1. Non-proxied. Either a machine on the local network, or listed as an exception
 +#     in your browser's proxy configuration.
 +#  2. Added to the exceptionsitelist. Option 1 is preferable; this option is
 +#     only for users using both transparent proxying and a non-local server
 +#     to host this script.
 +#
 +# Individual filter groups can override this setting in their own configuration.
 +#
 +# Django 2011-11-15
 +# default : accessdeniedaddress = 'http://YOURSERVER.YOURDOMAIN/cgi-bin/dansguardian.pl'
 +# accessdeniedaddress = 'http://nausch.org/cgi-bin/dansguardian.pl'
 +
 +
 +# Non standard delimiter (only used with accessdeniedaddress)
 +# To help preserve the full banned URL, including parameters, the variables
 +# passed into the access denied CGI are separated using non-standard
 +# delimiters. This can be useful to ensure correct operation of the filter
 +# bypass modes. Parameters are split using "::" in place of "&", and "==" in
 +# place of "=".
 +# Default is enabled, but to go back to the standard mode, disable it.
 +nonstandarddelimiter = on
 +
 +
 +
 +# Banned image replacement
 +# Images that are banned due to domain/url/etc reasons including those
 +# in the adverts blacklists can be replaced by an image.  This will,
 +# for example, hide images from advert sites and remove broken image
 +# icons from banned domains.
 +# on (default) | off
 +usecustombannedimage = on
 +custombannedimagefile = '/usr/share/dansguardian/transparent1x1.gif'
 +
 +
 +
 +# Filter groups options
 +# filtergroups sets the number of filter groups. A filter group is a set of content
 +# filtering options you can apply to a group of users.  The value must be 1 or more.
 +# DansGuardian will automatically look for dansguardianfN.conf where N is the filter
 +# group.  To assign users to groups use the filtergroupslist option.  All users default
 +# to filter group 1.  You must have some sort of authentication to be able to map users
 +# to a group.  The more filter groups the more copies of the lists will be in RAM so
 +# use as few as possible.
 +# Django 2011-11-16
 +# Default: filtergroups = 1
 +# Definition der Filtergruppen
 +#
 +# -----------------------------
 +# Gruppe 1 = Default
 +# -----------------------------
 +# Gruppe 2 = Default mit ByPass
 +# -----------------------------
 +# Gruppe 3 = logging only
 +# -----------------------------
 +# Gruppe 4 = banned useres
 +# -----------------------------
 +# Gruppe 5 = Spezialisten 
 +# -----------------------------
 +# Gruppe 6 = Jugendschutz
 +# -----------------------------
 +#
 +filtergroups = 6
 +filtergroupslist = '/etc/dansguardian/lists/filtergroupslist'
 +
 +
 +
 +# Authentication files location
 +bannediplist = '/etc/dansguardian/lists/bannediplist'
 +exceptioniplist = '/etc/dansguardian/lists/exceptioniplist'
 +
 +
 +
 +# Show weighted phrases found
 +# If enabled then the phrases found that made up the total which excedes
 +# the naughtyness limit will be logged and, if the reporting level is
 +# high enough, reported. on | off
 +showweightedfound = on
 +
 +# Weighted phrase mode
 +# There are 3 possible modes of operation:
 +# 0 = off = do not use the weighted phrase feature.
 +# 1 = on, normal = normal weighted phrase operation.
 +# 2 = on, singular = each weighted phrase found only counts once on a page.
 +#
 +weightedphrasemode = 2
 +
 +
 +
 +# Positive (clean) result caching for URLs
 +# Caches good pages so they don't need to be scanned again.
 +# It also works with AV plugins.
 +# 0 = off (recommended for ISPs with users with disimilar browsing)
 +# 1000 = recommended for most users
 +# 5000 = suggested max upper limit
 +# If you're using an AV plugin then use at least 5000.
 +urlcachenumber = 1000
 +#
 +# Age before they are stale and should be ignored in seconds
 +# 0 = never
 +# 900 = recommended = 15 mins
 +urlcacheage = 900
 +
 +
 +
 +# Clean cache for content (AV) scan results
 +# By default, to save CPU, files scanned and found to be
 +# clean are inserted into the clean cache and NOT scanned
 +# again for a while.  If you don't like this then choose
 +# to disable it.
 +# (on|off) default = on.
 +scancleancache = on
 +
 +
 +
 +# Smart, Raw and Meta/Title phrase content filtering options
 +# Smart is where the multiple spaces and HTML are removed before phrase filtering
 +# Raw is where the raw HTML including meta tags are phrase filtered
 +# Meta/Title is where only meta and title tags are phrase filtered (v. quick)
 +# CPU usage can be effectively halved by using setting 0 or 1 compared to 2
 +# 0 = raw only
 +# 1 = smart only
 +# 2 = both of the above (default)
 +# 3 = meta/title
 +phrasefiltermode = 2
 +
 +# Lower casing options
 +# When a document is scanned the uppercase letters are converted to lower case
 +# in order to compare them with the phrases.  However this can break Big5 and
 +# other 16-bit texts.  If needed preserve the case.  As of version 2.7.0 accented
 +# characters are supported.
 +# 0 = force lower case (default)
 +# 1 = do not change case
 +# 2 = scan first in lower case, then in original case
 +preservecase = 0
 +
 +# Note:
 +# If phrasefiltermode and preserve case are both 2, this equates to 4 phrase
 +# filtering passes. If you have a large enough userbase for this to be a
 +# worry, and need to filter pages in exotic character encodings, it may be
 +# better to run two instances on separate servers: one with preservecase 1
 +# (and possibly forcequicksearch 1) and non ASCII/UTF-8 phrase lists, and one
 +# with preservecase 0 and ASCII/UTF-8 lists.
 +
 +
 +
 +# Hex decoding options
 +# When a document is scanned it can optionally convert %XX to chars.
 +# If you find documents are getting past the phrase filtering due to encoding
 +# then enable.  However this can break Big5 and other 16-bit texts.
 +# off = disabled (default)
 +# on = enabled
 +hexdecodecontent = off
 +
 +
 +
 +# Force Quick Search rather than DFA search algorithm
 +# The current DFA implementation is not totally 16-bit character compatible
 +# but is used by default as it handles large phrase lists much faster.
 +# If you wish to use a large number of 16-bit character phrases then
 +# enable this option.
 +# off (default) | on (Big5 compatible)
 +forcequicksearch = off
 +
 +
 +
 +# Reverse lookups for banned site and URLs.
 +# If set to on, DansGuardian will look up the forward DNS for an IP URL
 +# address and search for both in the banned site and URL lists.  This would
 +# prevent a user from simply entering the IP for a banned address.
 +# It will reduce searching speed somewhat so unless you have a local caching
 +# DNS server, leave it off and use the Blanket IP Block option in the
 +# bannedsitelist file instead.
 +reverseaddresslookups = off
 +
 +
 +
 +# Reverse lookups for banned and exception IP lists.
 +# If set to on, DansGuardian will look up the forward DNS for the IP
 +# of the connecting computer.  This means you can put in hostnames in
 +# the exceptioniplist and bannediplist.
 +# If a client computer is matched against an IP given in the lists, then the
 +# IP will be recorded in any log entries; if forward DNS is successful and a
 +# match occurs against a hostname, the hostname will be logged instead.
 +# It will reduce searching speed somewhat so unless you have a local DNS server, 
 +# leave it off.
 +reverseclientiplookups = off
 +
 +
 +# Perform reverse lookups on client IPs for successful requests.
 +# If set to on, DansGuardian will look up the forward DNS for the IP
 +# of the connecting computer, and log host names (where available) rather than
 +# IPs against requests.
 +# This is not dependent on reverseclientiplookups being enabled; however, if it
 +# is, enabling this option does not incur any additional forward DNS requests.
 +logclienthostnames = off
 +
 +
 +# Build bannedsitelist and bannedurllist cache files.
 +# This will compare the date stamp of the list file with the date stamp of
 +# the cache file and will recreate as needed.
 +# If a bsl or bul .processed file exists, then that will be used instead.
 +# It will increase process start speed by 300%.  On slow computers this will
 +# be significant.  Fast computers do not need this option. on | off
 +createlistcachefiles = on
 +
 +
 +
 +# POST protection (web upload and forms)
 +# does not block forms without any file upload, i.e. this is just for
 +# blocking or limiting uploads
 +# measured in kibibytes after MIME encoding and header bumph
 +# use 0 for a complete block
 +# use higher (e.g. 512 = 512Kbytes) for limiting
 +# use -1 for no blocking
 +#maxuploadsize = 512
 +#maxuploadsize = 0
 +maxuploadsize = -1
 +
 +
 +
 +# Max content filter size
 +# Sometimes web servers label binary files as text which can be very
 +# large which causes a huge drain on memory and cpu resources.
 +# To counter this, you can limit the size of the document to be
 +# filtered and get it to just pass it straight through.
 +# This setting also applies to content regular expression modification.
 +# The value must not be higher than maxcontentramcachescansize
 +# The size is in Kibibytes - eg 2048 = 2Mb
 +# use 0 to set it to maxcontentramcachescansize
 +maxcontentfiltersize = 256
 +
 +
 +
 +# Max content ram cache scan size
 +# This is only used if you use a content scanner plugin such as AV
 +# This is the max size of file that DG will download and cache
 +# in RAM.  After this limit is reached it will cache to disk
 +# This value must be less than or equal to maxcontentfilecachescansize.
 +# The size is in Kibibytes - eg 10240 = 10Mb
 +# use 0 to set it to maxcontentfilecachescansize
 +# This option may be ignored by the configured download manager.
 +maxcontentramcachescansize = 2000
 +
 +
 +
 +# Max content file cache scan size
 +# This is only used if you use a content scanner plugin such as AV
 +# This is the max size file that DG will download
 +# so that it can be scanned or virus checked.
 +# This value must be greater or equal to maxcontentramcachescansize.
 +# The size is in Kibibytes - eg 10240 = 10Mb
 +maxcontentfilecachescansize = 20000
 +
 +
 +
 +# File cache dir
 +# Where DG will download files to be scanned if too large for the
 +# RAM cache.
 +filecachedir = '/tmp'
 +
 +
 +
 +# Delete file cache after user completes download
 +# When a file gets save to temp it stays there until it is deleted.
 +# You can choose to have the file deleted when the user makes a sucessful
 +# download.  This will mean if they click on the link to download from
 +# the temp store a second time it will give a 404 error.
 +# You should configure something to delete old files in temp to stop it filling up.
 +# on|off (defaults to on)
 +deletedownloadedtempfiles = on
 +
 +
 +
 +# Initial Trickle delay
 +# This is the number of seconds a browser connection is left waiting
 +# before first being sent *something* to keep it alive.  The
 +# *something* depends on the download manager chosen.
 +# Do not choose a value too low or normal web pages will be affected.
 +# A value between 20 and 110 would be sensible
 +# This may be ignored by the configured download manager.
 +initialtrickledelay = 20
 +
 +
 +
 +# Trickle delay
 +# This is the number of seconds a browser connection is left waiting
 +# before being sent more *something* to keep it alive.  The
 +# *something* depends on the download manager chosen.
 +# This may be ignored by the configured download manager.
 +trickledelay = 10
 +
 +
 +
 +# Download Managers
 +# These handle downloads of files to be filtered and scanned.
 +# They differ in the method they deal with large downloads.
 +# Files usually need to be downloaded 100% before they can be
 +# filtered and scanned before being sent on to the browser.
 +# Normally the browser can just wait, but with content scanning,
 +# for example to AV, the browser may timeout or the user may get
 +# confused so the download manager has to do some sort of
 +# 'keep alive'.
 +#
 +# There are various methods possible but not all are included.
 +# The author does not have the time to write them all so I have
 +# included a plugin systam.  Also, not all methods work with all
 +# browsers and clients.  Specifically some fancy methods don't
 +# work with software that downloads updates.  To solve this,
 +# each plugin can support a regular expression for matching
 +# the client's user-agent string, and lists of the mime types
 +# and extensions it should manage.
 +#
 +# Note that these are the matching methods provided by the base plugin
 +# code, and individual plugins may override or add to them.
 +# See the individual plugin conf files for supported options.
 +#
 +# The plugins are matched in the order you specify and the last
 +# one is forced to match as the default, regardless of user agent
 +# and other matching mechanisms.
 +#
 +downloadmanager = '/etc/dansguardian/downloadmanagers/fancy.conf'
 +#downloadmanager = '/etc/dansguardian/downloadmanagers/trickle.conf'
 +downloadmanager = '/etc/dansguardian/downloadmanagers/default.conf'
 +
 +
 +
 +# Content Scanners (Also known as AV scanners)
 +# These are plugins that scan the content of all files your browser fetches
 +# for example to AV scan.  The options are limitless.  Eventually all of
 +# DansGuardian will be plugin based.  You can have more than one content
 +# scanner. The plugins are run in the order you specify.
 +# This is one of the few places you can have multiple options of the same name.
 +#
 +# Some of the scanner(s) require 3rd party software and libraries eg clamav.
 +# See the individual plugin conf file for more options (if any).
 +#
 +#!! Not compiled !! contentscanner = '/etc/dansguardian/contentscanners/clamav.conf'
 +# Django 2011-11-16 ClamAV-Contentscanner aktiviert
 +contentscanner = '/etc/dansguardian/contentscanners/clamdscan.conf'
 +#!! Unimplemented !! contentscanner = '/etc/dansguardian/contentscanners/kavav.conf'
 +#contentscanner = '/etc/dansguardian/contentscanners/kavdscan.conf'
 +#contentscanner = '/etc/dansguardian/contentscanners/icapscan.conf'
 +#contentscanner = '/etc/dansguardian/contentscanners/commandlinescan.conf'
 +
 +
 +
 +# Content scanner timeout
 +# Some of the content scanners support using a timeout value to stop
 +# processing (eg AV scanning) the file if it takes too long.
 +# If supported this will be used.
 +# The default of 60 seconds is probably reasonable.
 +contentscannertimeout = 60
 +
 +
 +
 +# Content scan exceptions
 +# If 'on' exception sites, urls, users etc will be scanned
 +# This is probably not desirable behavour as exceptions are
 +# supposed to be trusted and will increase load.
 +# Correct use of grey lists are a better idea.
 +# (on|off) default = off
 +contentscanexceptions = off
 +
 +
 +
 +# Auth plugins
 +# These replace the usernameidmethod* options in previous versions. They
 +# handle the extraction of client usernames from various sources, such as
 +# Proxy-Authorisation headers and ident servers, enabling requests to be
 +# handled according to the settings of the user's filter group.
 +# Multiple plugins can be specified, and will be queried in order until one
 +# of them either finds a username or throws an error. For example, if Squid
 +# is configured with both NTLM and Basic auth enabled, and both the 'proxy-basic'
 +# and 'proxy-ntlm' auth plugins are enabled here, then clients which do not support
 +# NTLM can fall back to Basic without sacrificing access rights.
 +#
 +# If you do not use multiple filter groups, you need not specify this option.
 +#
 +#authplugin = '/etc/dansguardian/authplugins/proxy-basic.conf'
 +#authplugin = '/etc/dansguardian/authplugins/proxy-digest.conf'
 +#authplugin = '/etc/dansguardian/authplugins/proxy-ntlm.conf'
 +#authplugin = '/etc/dansguardian/authplugins/ident.conf'
 +#authplugin = '/etc/dansguardian/authplugins/ip.conf'
 +# Django 2011-11-15 Username im Logfile mit ausgeben
 +# default : NULL
 +authplugin = '/etc/dansguardian/authplugins/proxy-basic.conf'
 +
 +# Re-check replaced URLs
 +# As a matter of course, URLs undergo regular expression search/replace (urlregexplist)
 +# *after* checking the exception site/URL/regexpURL lists, but *before* checking against
 +# the banned site/URL lists, allowing certain requests that would be matched against the
 +# latter in their original state to effectively be converted into grey requests.
 +# With this option enabled, the exception site/URL/regexpURL lists are also re-checked
 +# after replacement, making it possible for URL replacement to trigger exceptions based
 +# on them.
 +# Defaults to off.
 +recheckreplacedurls = off
 +
 +
 +
 +# Misc settings
 +
 +# if on it adds an X-Forwarded-For: <clientip> to the HTTP request
 +# header.  This may help solve some problem sites that need to know the
 +# source ip. on | off
 +forwardedfor = off
 +
 +
 +# if on it uses the X-Forwarded-For: <clientip> to determine the client
 +# IP. This is for when you have squid between the clients and DansGuardian.
 +# Warning - headers are easily spoofed. on | off
 +usexforwardedfor = off
 +
 +
 +# if on it logs some debug info regarding fork()ing and accept()ing which
 +# can usually be ignored.  These are logged by syslog.  It is safe to leave
 +# it on or off
 +logconnectionhandlingerrors = on
 +
 +
 +
 +# Fork pool options
 +
 +# If on, this causes DG to write to the log file whenever child processes are
 +# created or destroyed (other than by crashes). This information can help in
 +# understanding and tuning the following parameters, but is not generally
 +# useful in production.
 +logchildprocesshandling = off
 +
 +# sets the maximum number of processes to spawn to handle the incoming
 +# connections.  Max value usually 250 depending on OS.
 +# On large sites you might want to try 180.
 +# Django 2011-11-15
 +# default maxchildren = 120
 +maxchildren = 80
 +
 +
 +# sets the minimum number of processes to spawn to handle the incoming connections.
 +# On large sites you might want to try 32.
 +# Django 2011-11-15
 +# default minchildren = 8
 +minchildren = 4
 +
 +
 +# sets the minimum number of processes to be kept ready to handle connections.
 +# On large sites you might want to try 8.
 +# Django 2011-11-15
 +# default minsparechildren = 4
 +minsparechildren = 2
 +
 +
 +# sets the minimum number of processes to spawn when it runs out
 +# On large sites you might want to try 10.
 +# Django 2011-11-15
 +# default preforkchildren = 6
 +preforkchildren = 4
 +
 +
 +# sets the maximum number of processes to have doing nothing.
 +# When this many are spare it will cull some of them.
 +# On large sites you might want to try 64.
 +# Django 2011-11-15
 +# default maxsparechildren = 32
 +maxsparechildren = 8
 +
 +
 +# sets the maximum age of a child process before it croaks it.
 +# This is the number of connections they handle before exiting.
 +# On large sites you might want to try 10000.
 +maxagechildren = 500
 +
 +
 +# Sets the maximum number client IP addresses allowed to connect at once.
 +# Use this to set a hard limit on the number of users allowed to concurrently
 +# browse the web. Set to 0 for no limit, and to disable the IP cache process.
 +maxips = 0
 +
 +
 +
 +# Process options
 +# (Change these only if you really know what you are doing).
 +# These options allow you to run multiple instances of DansGuardian on a single machine.
 +# Remember to edit the log file path above also if that is your intention.
 +
 +# IPC filename
 +
 +# Defines IPC server directory and filename used to communicate with the log process.
 +ipcfilename = '/tmp/.dguardianipc'
 +
 +# URL list IPC filename
 +
 +# Defines URL list IPC server directory and filename used to communicate with the URL
 +# cache process.
 +urlipcfilename = '/tmp/.dguardianurlipc'
 +
 +# IP list IPC filename
 +#
 +# Defines IP list IPC server directory and filename, for communicating with the client
 +# IP cache process.
 +ipipcfilename = '/tmp/.dguardianipipc'
 +
 +# PID filename
 +
 +# Defines process id directory and filename.
 +#pidfilename = '/var/run/dansguardian.pid'
 +
 +# Disable daemoning
 +# If enabled the process will not fork into the background.
 +# It is not usually advantageous to do this.
 +# on|off (defaults to off)
 +nodaemon = off
 +
 +# Disable logging process
 +# on|off (defaults to off)
 +nologger = off
 +
 +# Enable logging of "ADs" category blocks
 +# on|off (defaults to off)
 +logadblocks = off
 +
 +# Enable logging of client User-Agent
 +# Some browsers will cause a *lot* of extra information on each line!
 +# on|off (defaults to off)
 +loguseragent = off
 +
 +# Daemon runas user and group
 +# This is the user that DansGuardian runs as.  Normally the user/group nobody.
 +# Uncomment to use.  Defaults to the user set at compile time.
 +# Temp files created during virus scanning are given owner and group read
 +# permissions; to use content scanners based on external processes, such as
 +# clamdscan, the two processes must run with either the same group or user ID.
 +#daemonuser = 'dansguardian'
 +#daemongroup = 'dansguardian'
 +
 +# Soft restart
 +# When on this disables the forced killing off all processes in the process group.
 +# This is not to be confused with the -g run time option - they are not related.
 +# on|off (defaults to off)
 +softrestart = off
 +
 +# Mail program
 +# Path (sendmail-compatible) email program, with options.
 +# Not used if usesmtp is disabled (filtergroup specific).
 +mailer = '/usr/sbin/sendmail -t'
 +</file>
 +
 +===== clamdscan.conf =====
 +Die weitere Konfiguration findet in der oben genannten Datei //**/etc/dansguardian/contentscanners/clamdscan.conf**// statt. Mit dem Editor unserer ersten Wahl **vim** öffnen wir die Datei und tragen dort den Socket __**/tmp/clamd.socket**__  ein, den wir bei der clamd.conf angelegt hatten. 
 +   # vim /etc/dansguardian/contentscanners/clamdscan.conf
 +
 +<file bash /etc/dansguardian/contentscanners/clamdscan.conf>
 +plugname = 'clamdscan'
 +
 +# edit this to match the location of your ClamD UNIX domain socket
 +# Django 2011-11-16
 +# default : #clamdudsfile = '/var/run/clamav/clamd.sock'
 +clamdudsfile = '/var/run/clamav/clamd.sock'
 +
 +# If this string is set, the text it contains shall be removed from the
 +# beginning of filenames when passing them to ClamD.
 +# Use it to - for example - support a ClamD running inside a chroot jail:
 +# if DG's filecachedir is set to "/var/clamdchroot/downloads/" and pathprefix
 +# is set to "/var/clamdchroot", then file names given to ClamD will be of the
 +# form "/downloads/tf*" instead of "/var/clamdchroot/downloads/tf*".
 +#pathprefix = '/var/clamdchroot'
 +
 +exceptionvirusmimetypelist = '/etc/dansguardian/lists/contentscanners/exceptionvirusmimetypelist'
 +exceptionvirusextensionlist = '/etc/dansguardian/lists/contentscanners/exceptionvirusextensionlist'
 +exceptionvirussitelist = '/etc/dansguardian/lists/contentscanners/exceptionvirussitelist'
 +exceptionvirusurllist = '/etc/dansguardian/lists/contentscanners/exceptionvirusurllist'
 +</file>
 +
 +===== Client-Test =====
 +Nun ist es an der Zeit, dass wir unsere Dansguardian Installation einigen Test unterziehen.
 +==== Virustest ====
 +Als erstes versuchen wir uns einen Virus von einer entsprechenden Seite herunterzuladen. Wenn alles richtig gelaufen ist, dann wird uns am Client eine entsprechende Fehlerseite präsentiert.
 +
 +{{ :centos:dansg:dansguardian_eicar.png?500 |Bildschrimhardcopy: Dansguardian Fehlerseite mit Hinweis auf Virenbefall}}
 +
 +Im Accesslogfile unseres Contenscanners wird dies entsprechend vermerkt.
 +   2011.11.16 14:24:53 django 10.0.0.20 http://www.omni128.de/var/test/eicar.com *INFECTED* *DENIED* Virus or bad content detected. Eicar-Test-Signature GET 68 0 Content scanning 2 403 text/plain  VIPs -
 +
 +==== Test auf gesperrte Datei ====
 +Als nächsten versuchen wir uns eine auf den Rechner zu laden, die im System als unerwünscht deklariert wurde. Bei diesem Beispiel handelt es sich um eine **.PPS**-Datei.
 +
 +{{ :centos:dansg:dansguardian_banned_extension.png?500 |Bildschrimhardcopy: Dansguardian Fehlerseite mit Hinweis auf unerwünschte Datei}}
 +
 +Auch hier erfolgt ein entsprechender Eintrag im Accesslog unseres DAnsguardian-Servers.
 +   2011.11.16 14:36:38 django 10.0.0.20 http://www.boredatuni.com/files/illusion.pps *DENIED* Verbotene Datei-Erweiterung: .pps GET 0 0 Banned extension 2 403 application/vnd.ms-powerpoint  VIPs -
 +
 +==== Download-Test ====
 +Nach diesen beiden Negativbeispielen, wollen wir uns auch noch kurz ansehen, wie ein Download mit Virenscan des heruntergeladenen Inhalts abläuft. 
 +
 +Hierzu laden wir uns ein beliebiges ZIP-Archiv auf unseren Rechner, so z.B. die Aaktuelle Version des quelloffenen WEB-Analysetools [[http://piwik.org|Piwik]].
 +
 +Klicken wir auf den entsprechenden Downloadlink so wird uns am Bildschirm der fortlaufende Stand des Downloads angezeigt.
 +
 +{{ :centos:dansg:dansguardian_download.png?433 |Bidschirmhardcopy: Anzeige des Downloadvorganges}}
 +
 +Nach Abschluß des Downloadvorganges von der ursprünglichen Seite wird uns angezeigt, dass der Download einem Virenscan unterzogen wird.
 +
 +{{ :centos:dansg:dansguardian_scan.png?248 |Bidschirmhardcopy: Anzeige des Virusscanvorganges}}
 +
 +Ist dieser erfolgreich wird unserem Client ein Link angeboten, mit Hilfe man die Datei auf den Clientrechner laden und dort abspeichern kann.
 +
 +{{ :centos:dansg:dansguardian_finished.png?220 |Bidschirmhardcopy: Anzeige des Downloadlinks zur gerade heruntergeladenen Datei}}
 +====== Links ======
 +  * **[[centos:squid_dansg_6|Zurück zum Kapitel >>Proxyserver Squid mit Web-Contentfilter Dansguardian und Virenfilter ClamAV unter CentOS 6.x<<]]**
 +  * **[[wiki:start|Zurück zu >>Projekte und Themenkapitel<<]]**
 +  * **[[http://dokuwiki.nausch.org/doku.php/|Zurück zur Startseite]]**
 +