Squid (squid.conf)

Categoria: Miscelânea

Software: Squid

[ Hits: 4.080 ]

Por: Paulo Cesar


Este arquivo mostra como configurar o Squid sem logs e com cache em outro diretorio, e ainda com uma acl para evitar a passagem por ele do msn com o hotmail.


#   WELCOME TO SQUID 2
#   ------------------
#
#   This is the default Squid configuration file. You may wish
#   to look at the Squid home page (http://www.squid-cache.org/)
#   for the FAQ and other documentation.
#
#   The default Squid config file shows what the defaults for
#   various options happen to be.  If you don't need to change the
#   default, you shouldn't uncomment the line.  Doing so may cause
#   run-time problems.  In some cases "none" refers to no default
#   setting at all, while in other cases it refers to a valid
#   option - the comments for that keyword indicate if this is the
#   case.
#


# NETWORK OPTIONS
# -----------------------------------------------------------------------------

#  TAG: http_port
#   Usage:   port
#      hostname:port
#      1.2.3.4:port
#
#   The socket addresses where Squid will listen for HTTP client
#   requests.  You may specify multiple socket addresses.
#   There are three forms: port alone, hostname with port, and
#   IP address with port.  If you specify a hostname or IP
#   address, then Squid binds the socket to that specific
#   address.  This replaces the old 'tcp_incoming_address'
#   option.  Most likely, you do not need to bind to a specific
#   address, so you can use the port number alone.
#
#   The default port number is 3128.
#
#   If you are running Squid in accelerator mode, then you
#   probably want to listen on port 80 also, or instead.
#
#   The -a command line option will override the *first* port
#   number listed here.   That option will NOT override an IP
#   address, however.
#
#   You may specify multiple socket addresses on multiple lines.
#
#   If you run Squid on a dual-homed machine with an internal
#   and an external interface then we recommend you to specify the
#   internal address:port in http_port. This way Squid will only be
#   visible on the internal address.
#
#Default:
# http_port 3128

#  TAG: https_port
# Note: This option is only available if Squid is rebuilt with the
#       --enable-ssl option
#
#        Usage:  [ip:]port cert=certificate.pem [key=key.pem] [options...]
#
#        The socket address where Squid will listen for HTTPS client
#        requests.
#
#        This is really only useful for situations where you are running
#        squid in accelerator mode and you want to do the SSL work at the
#        accelerator level.
#
#   You may specify multiple socket addresses on multiple lines,
#   each with their own SSL certificate and/or options.
#               
#   Options:
#
#      cert=   Path to SSL certificate (PEM format)
#      
#      key=      Path to SSL private key file (PEM format)
#         if not specified, the certificate file is
#         assumed to be a combined certificate and
#         key file
#
#      version=   The version of SSL/TLS supported
#             1   automatic (default)
#             2   SSLv2 only
#             3   SSLv3 only
#             4   TLSv1 only
#
#      cipher=   Colon separated list of supported ciphers
#
#      options=   Varions SSL engine options. The most important
#         being:
#             NO_SSLv2  Disallow the use of SSLv2
#             NO_SSLv3  Disallow the use of SSLv3
#             NO_TLSv1  Disallow the use of TLSv1
#         See src/ssl_support.c or OpenSSL documentation
#         for a more complete list.
#
#Default:
# none

#  TAG: ssl_unclean_shutdown
# Note: This option is only available if Squid is rebuilt with the
#       --enable-ssl option
#
#   Some browsers (especially MSIE) bugs out on SSL shutdown
#   messages.
#
#Default:
# ssl_unclean_shutdown off

#  TAG: icp_port
#   The port number where Squid sends and receives ICP queries to
#   and from neighbor caches.  Default is 3130.  To disable use
#   "0".  May be overridden with -u on the command line.
#
#Default:
# icp_port 3130

#  TAG: htcp_port
# Note: This option is only available if Squid is rebuilt with the
#       --enable-htcp option
#
#   The port number where Squid sends and receives HTCP queries to
#   and from neighbor caches.  Default is 4827.  To disable use
#   "0".
#
#Default:
# htcp_port 4827

#  TAG: mcast_groups
#   This tag specifies a list of multicast groups which your server
#   should join to receive multicasted ICP queries.
#
#   NOTE!  Be very careful what you put here!  Be sure you
#   understand the difference between an ICP _query_ and an ICP
#   _reply_.  This option is to be set only if you want to RECEIVE
#   multicast queries.  Do NOT set this option to SEND multicast
#   ICP (use cache_peer for that).  ICP replies are always sent via
#   unicast, so this option does not affect whether or not you will
#   receive replies from multicast group members.
#
#   You must be very careful to NOT use a multicast address which
#   is already in use by another group of caches.
#
#   If you are unsure about multicast, please read the Multicast
#   chapter in the Squid FAQ (http://www.squid-cache.org/FAQ/).
#
#   Usage: mcast_groups 239.128.16.128 224.0.1.20
#
#   By default, Squid doesn't listen on any multicast groups.
#
#Default:
# none

#  TAG: udp_incoming_address
#  TAG: udp_outgoing_address
#   udp_incoming_address   is used for the ICP socket receiving packets
#            from other caches.
#   udp_outgoing_address   is used for ICP packets sent out to other
#            caches.
#
#   The default behavior is to not bind to any specific address.
#
#   A udp_incoming_address value of 0.0.0.0 indicates that Squid should
#   listen for UDP messages on all available interfaces.
#
#   If udp_outgoing_address is set to 255.255.255.255 (the default)
#   then it will use the same socket as udp_incoming_address. Only
#   change this if you want to have ICP queries sent using another
#   address than where this Squid listens for ICP queries from other
#   caches.
#
#   NOTE, udp_incoming_address and udp_outgoing_address can not
#   have the same value since they both use port 3130.
#
#Default:
# udp_incoming_address 0.0.0.0
# udp_outgoing_address 255.255.255.255


# OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
# -----------------------------------------------------------------------------

#  TAG: cache_peer
#   To specify other caches in a hierarchy, use the format:
#
#      cache_peer hostname type http_port icp_port
#
#   For example,
#
#   #                                        proxy  icp
#   #          hostname             type     port   port  options
#   #          -------------------- -------- ----- -----  -----------
#   cache_peer parent.foo.net       parent    3128  3130  [proxy-only]
#   cache_peer sib1.foo.net         sibling   3128  3130  [proxy-only]
#   cache_peer sib2.foo.net         sibling   3128  3130  [proxy-only]
#
#         type:  either 'parent', 'sibling', or 'multicast'.
#
#   proxy_port:  The port number where the cache listens for proxy
#           requests.
#
#     icp_port:  Used for querying neighbor caches about
#           objects.  To have a non-ICP neighbor
#           specify '7' for the ICP port and make sure the
#           neighbor machine has the UDP echo port
#           enabled in its /etc/inetd.conf file.
#
#       options: proxy-only
#           weight=n
#           ttl=n
#           no-query
#           default
#           round-robin
#           multicast-responder
#           closest-only
#           no-digest
#           no-netdb-exchange
#           no-delay
#           login=user:password | PASS | *:password
#           connect-timeout=nn
#           digest-url=url
#           allow-miss
#           max-conn
#
#           use 'proxy-only' to specify that objects fetched
#           from this cache should not be saved locally.
#
#           use 'weight=n' to specify a weighted parent.
#           The weight must be an integer.  The default weight
#           is 1, larger weights are favored more.
#
#           use 'ttl=n' to specify a IP multicast TTL to use
#           when sending an ICP queries to this address.
#           Only useful when sending to a multicast group.
#           Because we don't accept ICP replies from random
#           hosts, you must configure other group members as
#           peers with the 'multicast-responder' option below.
#
#           use 'no-query' to NOT send ICP queries to this
#           neighbor.
#
#           use 'default' if this is a parent cache which can
#           be used as a "last-resort." You should probably
#           only use 'default' in situations where you cannot
#           use ICP with your parent cache(s).
#
#           use 'round-robin' to define a set of parents which
#           should be used in a round-robin fashion in the
#           absence of any ICP queries.
#
#           'multicast-responder' indicates that the named peer
#           is a member of a multicast group.  ICP queries will
#           not be sent directly to the peer, but ICP replies
#           will be accepted from it.
#
#           'closest-only' indicates that, for ICP_OP_MISS
#           replies, we'll only forward CLOSEST_PARENT_MISSes
#           and never FIRST_PARENT_MISSes.
#
#           use 'no-digest' to NOT request cache digests from
#           this neighbor.
#
#           'no-netdb-exchange' disables requesting ICMP
#           RTT database (NetDB) from the neighbor.
#
#           use 'no-delay' to prevent access to this neighbor
#           from influencing the delay pools.
#
#           use 'login=user:password' if this is a personal/workgroup
#           proxy and your parent requires proxy authentication.
#           Note: The string can include URL escapes (i.e. %20 for
#           spaces). This also means that % must be written as %%.
#
#           use 'login=PASS' if users must authenticate against
#           the upstream proxy. This will pass the users credentials
#           as they are to the peer proxy. This only works for the
#           Basic HTTP authentication sheme. Note: To combine this
#           with proxy_auth both proxies must share the same user
#           database as HTTP only allows for one proxy login.
#           Also be warned that this will expose your users proxy
#           password to the peer. USE WITH CAUTION
#
#           use 'login=*:password' to pass the username to the
#           upstream cache, but with a fixed password. This is meant
#           to be used when the peer is in another administrative
#           domain, but it is still needed to identify each user.
#           The star can optionally be followed by some extra
#           information which is added to the username. This can
#           be used to identify this proxy to the peer, similar to
#           the login=username:password option above.
#
#           use 'connect-timeout=nn' to specify a peer
#           specific connect timeout (also see the
#           peer_connect_timeout directive)
#
#           use 'digest-url=url' to tell Squid to fetch the cache
#           digest (if digests are enabled) for this host from
#           the specified URL rather than the Squid default
#           location.
#
#           use 'allow-miss' to disable Squid's use of only-if-cached
#           when forwarding requests to siblings. This is primarily
#           useful when icp_hit_stale is used by the sibling. To
#           extensive use of this option may result in forwarding
#           loops, and you should avoid having two-way peerings
#           with this option. (for example to deny peer usage on
#           requests from peer by denying cache_peer_access if the
#           source is a peer)
#
#           use 'max-conn' to limit the amount of connections Squid
#           may open to this peer.
#
#   NOTE: non-ICP neighbors must be specified as 'parent'.
#
#Default:
# none

#  TAG: cache_peer_domain
#   Use to limit the domains for which a neighbor cache will be
#   queried.  Usage:
#
#   cache_peer_domain cache-host domain [domain ...]
#   cache_peer_domain cache-host !domain
#
#   For example, specifying
#
#      cache_peer_domain parent.foo.net   .edu
#
#   has the effect such that UDP query packets are sent to
#   'bigserver' only when the requested object exists on a
#   server in the .edu domain.  Prefixing the domainname
#   with '!' means that the cache will be queried for objects
#   NOT in that domain.
#
#   NOTE:   * Any number of domains may be given for a cache-host,
#        either on the same or separate lines.
#      * When multiple domains are given for a particular
#        cache-host, the first matched domain is applied.
#      * Cache hosts with no domain restrictions are queried
#        for all requests.
#      * There are no defaults.
#      * There is also a 'cache_peer_access' tag in the ACL
#        section.
#
#Default:
# none

#  TAG: neighbor_type_domain
#   usage: neighbor_type_domain parent|sibling domain domain ...
#
#   Modifying the neighbor type for specific domains is now
#   possible.  You can treat some domains differently than the the
#   default neighbor type specified on the 'cache_peer' line.
#   Normally it should only be necessary to list domains which
#   should be treated differently because the default neighbor type
#   applies for hostnames which do not match domains listed here.
#
#EXAMPLE:
#   cache_peer  parent cache.foo.org 3128 3130
#   neighbor_type_domain cache.foo.org sibling .com .net
#   neighbor_type_domain cache.foo.org sibling .au .de
#
#Default:
# none

#  TAG: icp_query_timeout   (msec)
#   Normally Squid will automatically determine an optimal ICP
#   query timeout value based on the round-trip-time of recent ICP
#   queries.  If you want to override the value determined by
#   Squid, set this 'icp_query_timeout' to a non-zero value.  This
#   value is specified in MILLISECONDS, so, to use a 2-second
#   timeout (the old default), you would write:
#
#      icp_query_timeout 2000
#
#Default:
# icp_query_timeout 0

#  TAG: maximum_icp_query_timeout   (msec)
#   Normally the ICP query timeout is determined dynamically.  But
#   sometimes it can lead to very large values (say 5 seconds).
#   Use this option to put an upper limit on the dynamic timeout
#   value.  Do NOT use this option to always use a fixed (instead
#   of a dynamic) timeout value. To set a fixed timeout see the
#   'icp_query_timeout' directive.
#
#Default:
# maximum_icp_query_timeout 2000

#  TAG: mcast_icp_query_timeout   (msec)
#   For Multicast peers, Squid regularly sends out ICP "probes" to
#   count how many other peers are listening on the given multicast
#   address.  This value specifies how long Squid should wait to
#   count all the replies.  The default is 2000 msec, or 2
#   seconds.
#
#Default:
# mcast_icp_query_timeout 2000

#  TAG: dead_peer_timeout   (seconds)
#   This controls how long Squid waits to declare a peer cache
#   as "dead."  If there are no ICP replies received in this
#   amount of time, Squid will declare the peer dead and not
#   expect to receive any further ICP replies.  However, it
#   continues to send ICP queries, and will mark the peer as
#   alive upon receipt of the first subsequent ICP reply.
#
#   This timeout also affects when Squid expects to receive ICP
#   replies from peers.  If more than 'dead_peer' seconds have
#   passed since the last ICP reply was received, Squid will not
#   expect to receive an ICP reply on the next query.  Thus, if
#   your time between requests is greater than this timeout, you
#   will see a lot of requests sent DIRECT to origin servers
#   instead of to your parents.
#
#Default:
# dead_peer_timeout 10 seconds

#  TAG: hierarchy_stoplist
#   A list of words which, if found in a URL, cause the object to
#   be handled directly by this cache.  In other words, use this
#   to not query neighbor caches for certain objects.  You may
#   list this option multiple times.
#We recommend you to use at least the following line.
hierarchy_stoplist cgi-bin ?

#  TAG: no_cache
#   A list of ACL elements which, if matched, cause the request to
#   not be satisfied from the cache and the reply to not be cached.
#   In other words, use this to force certain objects to never be cached.
#
#   You must use the word 'DENY' to indicate the ACL names which should
#   NOT be cached.
#
#We recommend you to use the following two lines.
acl QUERY urlpath_regex cgi-bin \?
no_cache deny QUERY


# OPTIONS WHICH AFFECT THE CACHE SIZE
# -----------------------------------------------------------------------------

#  TAG: cache_mem   (bytes)
#   NOTE: THIS PARAMETER DOES NOT SPECIFY THE MAXIMUM PROCESS SIZE.
#   IT ONLY PLACES A LIMIT ON HOW MUCH ADDITIONAL MEMORY SQUID WILL
#   USE AS A MEMORY CACHE OF OBJECTS. SQUID USES MEMORY FOR OTHER
#   THINGS AS WELL. SEE THE SQUID FAQ SECTION 8 FOR DETAILS.
#
#   'cache_mem' specifies the ideal amount of memory to be used
#   for:
#      * In-Transit objects
#      * Hot Objects
#      * Negative-Cached objects
#
#   Data for these objects are stored in 4 KB blocks.  This
#   parameter specifies the ideal upper limit on the total size of
#   4 KB blocks allocated.  In-Transit objects take the highest
#   priority.
#
#   In-transit objects have priority over the others.  When
#   additional space is needed for incoming data, negative-cached
#   and hot objects will be released.  In other words, the
#   negative-cached and hot objects will fill up any unused space
#   not needed for in-transit objects.
#
#   If circumstances require, this limit will be exceeded.
#   Specifically, if your incoming request rate requires more than
#   'cache_mem' of memory to hold in-transit objects, Squid will
#   exceed this limit to satisfy the new requests.  When the load
#   decreases, blocks will be freed until the high-water mark is
#   reached.  Thereafter, blocks will be used to store hot
#   objects.
#
#Default:
# cache_mem 8 MB

#  TAG: cache_swap_low   (percent, 0-100)
#  TAG: cache_swap_high   (percent, 0-100)
#
#   The low- and high-water marks for cache object replacement.
#   Replacement begins when the swap (disk) usage is above the
#   low-water mark and attempts to maintain utilization near the
#   low-water mark.  As swap utilization gets close to high-water
#   mark object eviction becomes more aggressive.  If utilization is
#   close to the low-water mark less replacement is done each time.
#   
#   Defaults are 90% and 95%. If you have a large cache, 5% could be
#   hundreds of MB. If this is the case you may wish to set these
#   numbers closer together.
#
#Default:
# cache_swap_low 90
# cache_swap_high 95

#  TAG: maximum_object_size   (bytes)
#   Objects larger than this size will NOT be saved on disk.  The
#   value is specified in kilobytes, and the default is 4MB.  If
#   you wish to get a high BYTES hit ratio, you should probably
#   increase this (one 32 MB object hit counts for 3200 10KB
#   hits).  If you wish to increase speed more than your want to
#   save bandwidth you should leave this low.
#
#   NOTE: if using the LFUDA replacement policy you should increase
#   this value to maximize the byte hit rate improvement of LFUDA!
#   See replacement_policy below for a discussion of this policy.
#
#Default:
# maximum_object_size 4096 KB

#  TAG: minimum_object_size   (bytes)
#   Objects smaller than this size will NOT be saved on disk.  The
#   value is specified in kilobytes, and the default is 0 KB, which
#   means there is no minimum.
#
#Default:
# minimum_object_size 0 KB

#  TAG: maximum_object_size_in_memory   (bytes)
#        Objects greater than this size will not be attempted to kept in
#        the memory cache. This should be set high enough to keep objects
#        accessed frequently in memory to improve performance whilst low
#        enough to keep larger objects from hoarding cache_mem .
#
#Default:
# maximum_object_size_in_memory 8 KB

#  TAG: ipcache_size   (number of entries)
#  TAG: ipcache_low   (percent)
#  TAG: ipcache_high   (percent)
#   The size, low-, and high-water marks for the IP cache.
#
#Default:
# ipcache_size 1024
# ipcache_low 90
# ipcache_high 95

#  TAG: fqdncache_size   (number of entries)
#   Maximum number of FQDN cache entries.
#
#Default:
# fqdncache_size 1024

#  TAG: cache_replacement_policy
#   The cache replacement policy parameter determines which
#   objects are evicted (replaced) when disk space is needed.
#
#       lru       : Squid's original list based LRU policy
#       heap GDSF : Greedy-Dual Size Frequency
#       heap LFUDA: Least Frequently Used with Dynamic Aging
#       heap LRU  : LRU policy implemented using a heap
#
#   Applies to any cache_dir lines listed below this.
#
#   The LRU policies keeps recently referenced objects.
#
#   The heap GDSF policy optimizes object hit rate by keeping smaller
#   popular objects in cache so it has a better chance of getting a
#   hit.  It achieves a lower byte hit rate than LFUDA though since
#   it evicts larger (possibly popular) objects.
#
#   The heap LFUDA policy keeps popular objects in cache regardless of
#   their size and thus optimizes byte hit rate at the expense of
#   hit rate since one large, popular object will prevent many
#   smaller, slightly less popular objects from being cached.
#
#   Both policies utilize a dynamic aging mechanism that prevents
#   cache pollution that can otherwise occur with frequency-based
#   replacement policies.
#
#   NOTE: if using the LFUDA replacement policy you should increase
#   the value of maximum_object_size above its default of 4096 KB to
#   to maximize the potential byte hit rate improvement of LFUDA.
#
#   For more information about the GDSF and LFUDA cache replacement
#   policies see http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html
#   and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html.
#
#Default:
# cache_replacement_policy lru

#  TAG: memory_replacement_policy
#   The memory replacement policy parameter determines which
#   objects are purged from memory when memory space is needed.
#
#   See cache_replacement_policy for details.
#
#Default:
# memory_replacement_policy lru


# LOGFILE PATHNAMES AND CACHE DIRECTORIES
# -----------------------------------------------------------------------------

#  TAG: cache_dir
#   Usage:
#   
#   cache_dir Type Directory-Name Fs-specific-data [options]
#
#   cache_dir diskd Maxobjsize Directory-Name MB L1 L2 Q1 Q2
#
#   You can specify multiple cache_dir lines to spread the
#   cache among different disk partitions.
#
#   Type specifies the kind of storage system to use. Only "ufs"
#   is built by default. To eanble any of the other storage systems
#   see the --enable-storeio configure option.
#
#   'Directory' is a top-level directory where cache swap
#   files will be stored.  If you want to use an entire disk
#   for caching, then this can be the mount-point directory.
#   The directory must exist and be writable by the Squid
#   process.  Squid will NOT create this directory for you.
#
#   The ufs store type:
#
#   "ufs" is the old well-known Squid storage format that has always
#   been there.
#
#   cache_dir ufs Directory-Name Mbytes L1 L2 [options]
#
#   'Mbytes' is the amount of disk space (MB) to use under this
#   directory.  The default is 100 MB.  Change this to suit your
#   configuration.
#
#   'Level-1' is the number of first-level subdirectories which
#   will be created under the 'Directory'.  The default is 16.
#
#   'Level-2' is the number of second-level subdirectories which
#   will be created under each first-level directory.  The default
#   is 256.
#
#   The aufs store type:
#
#   "aufs" uses the same storage format as "ufs", utilizing
#   POSIX-threads to avoid blocking the main Squid process on
#   disk-I/O. This was formerly known in Squid as async-io.
#
#   cache_dir aufs Directory-Name Mbytes L1 L2 [options]
#
#   see argument descriptions under ufs above
#
#   The diskd store type:
#
#   "diskd" uses the same storage format as "ufs", utilizing a
#   separate process to avoid blocking the main Squid process on
#   disk-I/O.
#
#   cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]
#
#   see argument descriptions under ufs above
#
#   Q1 specifies the number of unacknowledged I/O requests when Squid
#   stops opening new files. If this many messages are in the queues,
#   Squid won't open new files. Default is 64
#
#   Q2 specifies the number of unacknowledged messages when Squid
#   starts blocking.  If this many messages are in the queues,
#   Squid blocks until it recevies some replies. Default is 72
#
#   Common options:
#
#   read-only, this cache_dir is read only.
#
#   max-size=n, refers to the max object size this storedir supports.
#   It is used to initially choose the storedir to dump the object.
#   Note: To make optimal use of the max-size limits you should order
#   the cache_dir lines with the smallest max-size value first and the
#   ones with no max-size specification last.
#
#Default:
 cache_dir ufs /squid/var/cache 2700 16 256

#  TAG: cache_access_log
#   Logs the client request activity.  Contains an entry for
#   every HTTP and ICP queries received. To disable, enter "none".
#
#Default:
# cache_access_log /squid/var/logs/access.log
cache_access_log /dev/null
#  TAG: cache_log
#   Cache logging file. This is where general information about
#   your cache's behavior goes. You can increase the amount of data
#   logged to this file with the "debug_options" tag below.
#
#Default:
# cache_log /squid/var/logs/cache.log
cache_log /dev/null
#  TAG: cache_store_log
#   Logs the activities of the storage manager.  Shows which
#   objects are ejected from the cache, and which objects are
#   saved and for how long.  To disable, enter "none". There are
#   not really utilities to analyze this data, so you can safely
#   disable it.
#
#Default:
 cache_store_log /squid/var/logs/store.log
cache_store_log /dev/null
#  TAG: cache_swap_log
#   Location for the cache "swap.log."  This log file holds the
#   metadata of objects saved on disk.  It is used to rebuild the
#   cache during startup.  Normally this file resides in each
#   'cache_dir' directory, but you may specify an alternate
#   pathname here.  Note you must give a full filename, not just
#   a directory. Since this is the index for the whole object
#   list you CANNOT periodically rotate it!
#
#   If %s can be used in the file name then it will be replaced with a
#   a representation of the cache_dir name where each / is replaced
#   with '.'. This is needed to allow adding/removing cache_dir
#   lines when cache_swap_log is being used.
#   
#   If have more than one 'cache_dir', and %s is not used in the name
#   then these swap logs will have names such as:
#
#      cache_swap_log.00
#      cache_swap_log.01
#      cache_swap_log.02
#
#   The numbered extension (which is added automatically)
#   corresponds to the order of the 'cache_dir' lines in this
#   configuration file.  If you change the order of the 'cache_dir'
#   lines in this file, then these log files will NOT correspond to
#   the correct 'cache_dir' entry (unless you manually rename
#   them).  We recommend that you do NOT use this option.  It is
#   better to keep these log files in each 'cache_dir' directory.
#
#Default:
# none

#  TAG: emulate_httpd_log   on|off
#   The Cache can emulate the log file format which many 'httpd'
#   programs use.  To disable/enable this emulation, set
#   emulate_httpd_log to 'off' or 'on'.  The default
#   is to use the native log format since it includes useful
#   information that Squid-specific log analyzers use.
#
#Default:
# emulate_httpd_log off

#  TAG: log_ip_on_direct   on|off
#   Log the destination IP address in the hierarchy log tag when going
#   direct. Earlier Squid versions logged the hostname here. If you
#   prefer the old way set this to off.
#
#Default:
# log_ip_on_direct on

#  TAG: mime_table
#   Pathname to Squid's MIME table. You shouldn't need to change
#   this, but the default file contains examples and formatting
#   information if you do.
#
#Default:
# mime_table /usr/local/squid/etc/mime.conf

#  TAG: log_mime_hdrs   on|off
#   The Cache can record both the request and the response MIME
#   headers for each HTTP transaction.  The headers are encoded
#   safely and will appear as two bracketed fields at the end of
#   the access log (for either the native or httpd-emulated log
#   formats).  To enable this logging set log_mime_hdrs to 'on'.
#
#Default:
# log_mime_hdrs off

#  TAG: useragent_log
# Note: This option is only available if Squid is rebuilt with the
#       --enable-useragent-log option
#
#   Squid will write the User-Agent field from HTTP requests
#   to the filename specified here.  By default useragent_log
#   is disabled.
#
#Default:
# none

#  TAG: referer_log
# Note: This option is only available if Squid is rebuilt with the
#       --enable-referer-log option
#
#   Squid will write the Referer field from HTTP requests to the
#   filename specified here.  By default referer_log is disabled.
#
#Default:
# none

#  TAG: pid_filename
#   A filename to write the process-id to.  To disable, enter "none".
#
#Default:
# pid_filename /usr/local/squid/var/logs/squid.pid

#  TAG: debug_options
#   Logging options are set as section,level where each source file
#   is assigned a unique section.  Lower levels result in less
#   output,  Full debugging (level 9) can result in a very large
#   log file, so be careful.  The magic word "ALL" sets debugging
#   levels for all sections.  We recommend normally running with
#   "ALL,1".
#
#Default:
# debug_options ALL,1

#  TAG: log_fqdn   on|off
#   Turn this on if you wish to log fully qualified domain names
#   in the access.log. To do this Squid does a DNS lookup of all
#   IP's connecting to it. This can (in some situations) increase
#   latency, which makes your cache seem slower for interactive
#   browsing.
#
#Default:
# log_fqdn off

#  TAG: client_netmask
#   A netmask for client addresses in logfiles and cachemgr output.
#   Change this to protect the privacy of your cache clients.
#   A netmask of 255.255.255.0 will log all IP's in that range with
#   the last digit set to '0'.
#
#Default:
# client_netmask 255.255.255.255


# OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
# -----------------------------------------------------------------------------

#  TAG: ftp_user
#   If you want the anonymous login password to be more informative
#   (and enable the use of picky ftp servers), set this to something
#   reasonable for your domain, like wwwuser@somewhere.net
#
#   The reason why this is domainless by default is that the
#   request can be made on the behalf of a user in any domain,
#   depending on how the cache is used.
#   Some ftp server also validate that the email address is valid
#   (for example perl.com).
#
#Default:
# ftp_user Squid@

#  TAG: ftp_list_width
#   Sets the width of ftp listings. This should be set to fit in
#   the width of a standard browser. Setting this too small
#   can cut off long filenames when browsing ftp sites.
#
#Default:
# ftp_list_width 32

#  TAG: ftp_passive
#   If your firewall does not allow Squid to use passive
#   connections, then turn off this option.
#
#Default:
# ftp_passive on

#  TAG: ftp_sanitycheck
#   For security and data integrity reasons Squid by default performs
#   sanity checks of the addresses of FTP data connections ensure the
#   data connection is to the requested server. If you need to allow
#   FTP connections to servers using another IP address for the data
#   connection then turn this off.
#
#Default:
# ftp_sanitycheck on

#  TAG: cache_dns_program
# Note: This option is only available if Squid is rebuilt with the
#       --disable-internal-dns option
#
#   Specify the location of the executable for dnslookup process.
#
#Default:
# cache_dns_program /usr/local/squid/libexec/

#  TAG: dns_children
# Note: This option is only available if Squid is rebuilt with the
#       --disable-internal-dns option
#
#   The number of processes spawn to service DNS name lookups.
#   For heavily loaded caches on large servers, you should
#   probably increase this value to at least 10.  The maximum
#   is 32.  The default is 5.
#
#   You must have at least one dnsserver process.
#
#Default:
# dns_children 5

#  TAG: dns_retransmit_interval
#   Initial retransmit interval for DNS queries. The interval is
#   doubled each time all configured DNS servers have been tried.
#
#
#Default:
# dns_retransmit_interval 5 seconds

#  TAG: dns_timeout
#   DNS Query timeout. If no response is received to a DNS query
#   within this time then all DNS servers for the queried domain
#   is assumed to be unavailable.
#
#Default:
# dns_timeout 5 minutes

#  TAG: dns_defnames   on|off
# Note: This option is only available if Squid is rebuilt with the
#       --disable-internal-dns option
#
#   Normally the 'dnsserver' disables the RES_DEFNAMES resolver
#   option (see res_init(3)).  This prevents caches in a hierarchy
#   from interpreting single-component hostnames locally.  To allow
#   dnsserver to handle single-component names, enable this
#   option.
#
#Default:
# dns_defnames off

#  TAG: dns_nameservers
#   Use this if you want to specify a list of DNS name servers
#   (IP addresses) to use instead of those given in your
#   /etc/resolv.conf file.
#   On Windows platforms, if no value is specified here or in
#   the /etc/resolv.conf file, the list of DNS name servers are
#   taken from the Windows registry, both static and dynamic DHCP
#   configurations are supported.
#
#   Example: dns_nameservers 10.0.0.1 192.172.0.4
#
#Default:
# none

#  TAG: hosts_file
#   Location of the host-local IP name-address associations
#   database.  Most Operating Systems have such a file: under
#   Un*X it's by default in /etc/hosts MS-Windows NT/2000 places
#   that in %SystemRoot%(by default
#   c:\winnt)\system32\drivers\etc\hosts, while Windows 9x/ME
#   places that in %windir%(usually c:\windows)\hosts
#
#   The file contains newline-separated definitions, in the
#   form ip_address_in_dotted_form name [name ...] names are
#   whitespace-separated.  lines beginnng with an hash (#)
#   character are comments.
#
#   The file is checked at startup and upon configuration.  If
#   set to 'none', it won't be checked.  If append_domain is
#   used, that domain will be added to domain-local (i.e. not
#   containing any dot character) host definitions.
#
#Default:
# hosts_file /etc/hosts

#  TAG: diskd_program
#   Specify the location of the diskd executable.
#   Note that this is only useful if you have compiled in
#   diskd as one of the store io modules.
#
#Default:
# diskd_program /usr/local/squid/libexec/

#  TAG: unlinkd_program
#   Specify the location of the executable for file deletion process.
#
#Default:
# unlinkd_program /usr/local/squid/libexec/unlinkd

#  TAG: pinger_program
# Note: This option is only available if Squid is rebuilt with the
#       --enable-icmp option
#
#   Specify the location of the executable for the pinger process.
#
#Default:
# pinger_program /usr/local/squid/libexec/

#  TAG: redirect_program
#   Specify the location of the executable for the URL redirector.
#   Since they can perform almost any function there isn't one included.
#   See the FAQ (section 15) for information on how to write one.
#   By default, a redirector is not used.
#
#Default:
# none

#  TAG: redirect_children
#   The number of redirector processes to spawn. If you start
#   too few Squid will have to wait for them to process a backlog of
#   URLs, slowing it down. If you start too many they will use RAM
#   and other system resources.
#
#Default:
# redirect_children 5

#  TAG: redirect_rewrites_host_header
#   By default Squid rewrites any Host: header in redirected
#   requests.  If you are running an accelerator then this may
#   not be a wanted effect of a redirector.
#
#Default:
# redirect_rewrites_host_header on

#  TAG: redirector_access
#   If defined, this access list specifies which requests are
#   sent to the redirector processes.  By default all requests
#   are sent.
#
#Default:
# none

#  TAG: auth_param
#   This is used to pass parameters to the various authentication
#   schemes.
#   format: auth_param scheme parameter [setting]
#   
#   auth_param basic program /usr/local/squid/bin/ncsa_auth /usr/local/squid/etc/passwd   
#   would tell the basic authentication scheme it's program parameter.
#
#   The order that authentication prompts are presented to the client_agent
#   is dependant on the order the scheme first appears in config file.
#   IE has a bug (it's not rfc 2617 compliant) in that it will use the basic
#   scheme if basic is the first entry presented, even if more secure schemes
#   are presented. For now use the order in the file below. If other browsers
#   have difficulties (don't recognise the schemes offered even if you are using
#   basic) then either put basic first, or disable the other schemes (by commenting
#   out their program entry).
#
#   Once an authentication scheme is fully configured, it can only be shutdown
#   by shutting squid down and restarting. Changes can be made on the fly and
#   activated with a reconfigure. I.E. You can change to a different helper,
#   but not unconfigure the helper completely.
#
#   === Parameters for the basic scheme follow. ===
#   
#   "program" cmdline
#   Specify the command for the external authenticator.  Such a
#   program reads a line containing "username password" and replies
#   "OK" or "ERR" in an endless loop.  If you use an authenticator,
#   make sure you have 1 acl of type proxy_auth.  By default, the
#   authenticate_program is not used.
#
#   If you want to use the traditional proxy authentication,
#   jump over to the ../auth_modules/NCSA directory and
#   type:
#      % make
#      % make install
#
#   Then, set this line to something like
#
#   auth_param basic program /usr/local/squid/bin/ncsa_auth /usr/local/squid/etc/passwd
#
#   "children" numberofchildren
#   The number of authenticator processes to spawn (no default).
#   If you start too few Squid will have to wait for them to
#   process a backlog of usercode/password verifications, slowing
#   it down. When password verifications are done via a (slow)
#   network you are likely to need lots of authenticator
#   processes.
#   auth_param basic children 5
#
#   "realm" realmstring
#   Specifies the realm name which is to be reported to the
#   client for the basic proxy authentication scheme (part of
#   the text the user will see when prompted their username and
#   password). There is no default.
#   auth_param basic realm Squid proxy-caching web server
#
#   "credentialsttl" timetolive
#   Specifies how long squid assumes an externally validated
#   username:password pair is valid for - in other words how
#   often the helper program is called for that user. Set this
#   low to force revalidation with short lived passwords.  Note
#   that setting this high does not impact your susceptability
#   to replay attacks unless you are using an one-time password
#   system (such as SecureID).  If you are using such a system,
#   you will be vulnerable to replay attacks unless you also
#   enable the IP ttl is strict option.
#
#   === Parameters for the digest scheme follow ===
#
#   "program" cmdline
#   Specify the command for the external authenticator.  Such
#   a program reads a line containing "username":"realm" and
#   replies with the appropriate H(A1) value base64 encoded.
#   See rfc 2616 for the definition of H(A1).  If you use an
#   authenticator, make sure you have 1 acl of type proxy_auth.
#   By default, authentication is not used.
#
#   If you want to use build an authenticator,
#   jump over to the ../digest_auth_modules directory and choose the
#   authenticator to use. It it's directory type
#           % make
#           % make install
#
#   Then, set this line to something like
#
#   auth_param digest program /usr/local/squid/bin/digest_auth_pw /usr/local/squid/etc/digpass
#
#
#   "children" numberofchildren
#   The number of authenticator processes to spawn (no default).
#   If you start too few Squid will have to wait for them to
#   process a backlog of H(A1) calculations, slowing it down.
#   When the H(A1) calculations are done via a (slow) network
#   you are likely to need lots of authenticator processes.
#   auth_param digest children 5
#
#   "realm" realmstring
#   Specifies the realm name which is to be reported to the
#   client for the digest proxy authentication scheme (part of
#   the text the user will see when prompted their username and
#   password). There is no default.
#   auth_param digest realm Squid proxy-caching web server
#
#   "nonce_garbage_interval" timeinterval
#   Specifies the interval that nonces that have been issued
#   to client_agent's are checked for validity.
#
#   "nonce_max_duration" timeinterval
#   Specifies the maximum length of time a given nonce will be
#   valid for.
#
#   "nonce_max_count" number
#   Specifies the maximum number of times a given nonce can be
#   used.
#
#   "nonce_strictness" on|off
#   Determines if squid requires increment-by-1 behaviour for
#   nonce counts (on - the default), or strictly incrementing
#   (off - for use when useragents generate nonce counts that
#   occasionally miss 1 (ie, 1,2,4,6)).
#
#   === NTLM scheme options follow ===
#
#   "program" cmdline
#   Specify the command for the external ntlm authenticator.
#   Such a program reads a line containing the uuencoded NEGOTIATE
#   and replies with the ntlm CHALLENGE, then waits for the
#   response and answers with "OK" or "ERR" in an endless loop.
#   If you use an ntlm authenticator, make sure you have 1 acl
#   of type proxy_auth.  By default, the ntlm authenticator_program
#   is not used.
#
#   auth_param ntlm program /usr/local/squid/bin/ntlm_auth
#
#   "children" numberofchildren
#   The number of authenticator processes to spawn (no default).
#   If you start too few Squid will have to wait for them to
#   process a backlog of credential verifications, slowing it
#   down. When crendential verifications are done via a (slow)
#   network you are likely to need lots of authenticator
#   processes.
#   auth_param ntlm children 5
#
#   "max_challenge_reuses" number
#   The maximum number of times a challenge given by a ntlm
#   authentication helper can be reused. Increasing this number
#   increases your exposure to replay attacks on your network.
#   0 means use the challenge only once.  (disable challenge
#   caching) See max_ntlm_challenge_lifetime for more information.
#   auth_param ntlm max_challenge_reuses 0
#
#   "max_challenge_lifetime" timespan
#   The maximum time period that a ntlm challenge is reused
#   over.  The actual period will be the minimum of this time
#   AND the number of reused challenges.
#   auth_param ntlm max_challenge_lifetime 2 minutes
#
#Recommended minimum configuration:
#auth_param digest program <uncomment and complete this line>
#auth_param digest children 5
#auth_param digest realm Squid proxy-caching web server
#auth_param digest nonce_garbage_interval 5 minutes
#auth_param digest nonce_max_duration 30 minutes
#auth_param digest nonce_max_count 50
#auth_param ntlm program <uncomment and complete this line to activate>
#auth_param ntlm children 5
#auth_param ntlm max_challenge_reuses 0
#auth_param ntlm max_challenge_lifetime 2 minutes
#auth_param basic program <uncomment and complete this line>
auth_param basic children 5
auth_param basic realm Squid proxy-caching web server
auth_param basic credentialsttl 2 hours

#  TAG: authenticate_cache_garbage_interval
#   The time period between garbage collection across the
#   username cache.  This is a tradeoff between memory utilisation
#   (long intervals - say 2 days) and CPU (short intervals -
#   say 1 minute). Only change if you have good reason to.
#
#Default:
# authenticate_cache_garbage_interval 1 hour

#  TAG: authenticate_ttl
#   The time a user & their credentials stay in the logged in
#   user cache since their last request. When the garbage
#   interval passes, all user credentials that have passed their
#   TTL are removed from memory.
#
#Default:
# authenticate_ttl 1 hour

#  TAG: authenticate_ip_ttl
#   With this option you control how long a proxy authentication
#   will be bound to a specific IP address. If a request using
#   the same user name is received during this time then access
#   will be denied and both users are required to reauthenticate
#   them selves.  The idea behind this is to make it annoying
#   for people to share their password to their friends, but
#   yet allow a dialup user to reconnect on a different dialup
#   port.
#
#   The default is 0 to disable the check. Recommended value
#   if you have dialup users are no more than 60 seconds to allow
#   the user to redial without hassle. If all your users are
#   stationary then higher values may be used.
#
#   See also the acl max_user_ip. The max_user_ip acl replaces
#   the authenticate_ip_ttl_is_strict option found in earlier
#   Squid versions.
#
#Default:
# authenticate_ip_ttl 0 seconds

#  TAG: external_acl_type
#   This option defines external acl classes using a helper program
#   to look up the status
#   
#     external_acl_type name [options] FORMAT.. /path/to/helper [helper arguments..]
#   
#   Options:
#
#     ttl=n      TTL in seconds for cached results (defaults to 3600
#           for 1 hour)
#     negative_ttl=n
#           TTL for cached negative lookups (default same
#           as ttl)
#     concurrency=n   Concurrency level / number of processes spawn
#         to service external acl lookups of this type.
#     cache=n   result cache size, 0 is unbounded (default)
#   
#   FORMAT specifications
#
#     %LOGIN   Authenticated user login name
#     %IDENT   Ident user name
#     %SRC      Client IP
#     %DST      Requested host
#     %PROTO   Requested protocol
#     %PORT      Requested port
#     %METHOD   Request method
#     %{Header}   HTTP request header
#     %{Hdr:member}   HTTP request header list member
#     %{Hdr:;member}
#           HTTP request header list member using ; as
#           list separator. ; can be any non-alphanumeric
#         character.
#
#   In addition, any string specified in the referencing acl will
#   also be included in the helper request line, after the specified
#   formats (see the "acl external" directive)
#
#   The helper receives lines per the above format specification,
#   and returns lines starting with OK or ERR indicating the validity
#   of the request and optionally followed by additional keywords with
#   more details.
#
#   General result syntax:
#   
#     OK/ERR keyword=value ...
#
#   Defined keywords:
#
#     user=      The users name (login)
#     error=   Error description (only defined for ERR results)
#
#   Keyword values need to be enclosed in quotes if they may contain
#   whitespace, or the whitespace escaped using \. Any quotes or \
#   characters within the keyword value must be \ escaped.
#
#Default:
# none


# OPTIONS FOR TUNING THE CACHE
# -----------------------------------------------------------------------------

#  TAG: wais_relay_host
#  TAG: wais_relay_port
#   Relay WAIS request to host (1st arg) at port (2 arg).
#
#Default:
# wais_relay_port 0

#  TAG: request_header_max_size   (KB)
#   This specifies the maximum size for HTTP headers in a request.
#   Request headers are usually relatively small (about 512 bytes).
#   Placing a limit on the request header size will catch certain
#   bugs (for example with persistent connections) and possibly
#   buffer-overflow or denial-of-service attacks.
#
#Default:
# request_header_max_size 10 KB

#  TAG: request_body_max_size   (KB)
#   This specifies the maximum size for an HTTP request body.
#   In other words, the maximum size of a PUT/POST request.
#   A user who attempts to send a request with a body larger
#   than this limit receives an "Invalid Request" error message.
#   If you set this parameter to a zero (the default), there will
#   be no limit imposed.
#
#Default:
# request_body_max_size 0 KB

#  TAG: refresh_pattern
#   usage: refresh_pattern [-i] regex min percent max [options]
#
#   By default, regular expressions are CASE-SENSITIVE.  To make
#   them case-insensitive, use the -i option.
#
#   'Min' is the time (in minutes) an object without an explicit
#   expiry time should be considered fresh. The recommended
#   value is 0, any higher values may cause dynamic applications
#   to be erroneously cached unless the application designer
#   has taken the appropriate actions.
#
#   'Percent' is a percentage of the objects age (time since last
#   modification age) an object without explicit expiry time
#   will be considered fresh.
#
#   'Max' is an upper limit on how long objects without an explicit
#   expiry time will be considered fresh.
#
#   options: override-expire
#       override-lastmod
#       reload-into-ims
#       ignore-reload
#
#      override-expire enforces min age even if the server
#      sent a Expires: header. Doing this VIOLATES the HTTP
#      standard.  Enabling this feature could make you liable
#      for problems which it causes.
#
#      override-lastmod enforces min age even on objects
#      that was modified recently.
#
#      reload-into-ims changes client no-cache or ``reload''
#      to If-Modified-Since requests. Doing this VIOLATES the
#      HTTP standard. Enabling this feature could make you
#      liable for problems which it causes.
#
#      ignore-reload ignores a client no-cache or ``reload''
#      header. Doing this VIOLATES the HTTP standard. Enabling
#      this feature could make you liable for problems which
#      it causes.
#      
#   Please see the file doc/Release-Notes-1.1.txt for a full
#   description of Squid's refresh algorithm.  Basically a
#   cached object is: (the order is changed from 1.1.X)
#
#      FRESH if expires < now, else STALE
#      STALE if age > max
#      FRESH if lm-factor < percent, else STALE
#      FRESH if age < min
#      else STALE
#
#   The refresh_pattern lines are checked in the order listed here.
#   The first entry which matches is used.  If none of the entries
#   match, then the default will be used.
#
#   Note, you must uncomment all the default lines if you want
#   to change one. The default setting is only active if none is
#   used.
#
#Suggested default:
refresh_pattern ^ftp:      1440   20%   10080
refresh_pattern ^gopher:   1440   0%   1440
refresh_pattern .      0   20%   4320

#  TAG: quick_abort_min   (KB)
#  TAG: quick_abort_max   (KB)
#  TAG: quick_abort_pct   (percent)
#   The cache by default continues downloading aborted requests
#   which are almost completed (less than 16 KB remaining). This
#   may be undesirable on slow (e.g. SLIP) links and/or very busy
#   caches.  Impatient users may tie up file descriptors and
#   bandwidth by repeatedly requesting and immediately aborting
#   downloads.
#
#   When the user aborts a request, Squid will check the
#   quick_abort values to the amount of data transfered until
#   then.
#
#   If the transfer has less than 'quick_abort_min' KB remaining,
#   it will finish the retrieval.
#
#   If the transfer has more than 'quick_abort_max' KB remaining,
#   it will abort the retrieval.
#
#   If more than 'quick_abort_pct' of the transfer has completed,
#   it will finish the retrieval.
#
#   If you do not want any retrieval to continue after the client
#   has aborted, set both 'quick_abort_min' and 'quick_abort_max'
#   to '0 KB'.
#
#   If you want retrievals to always continue if they are being
#   cached then set 'quick_abort_min' to '-1 KB'.
#
#Default:
# quick_abort_min 16 KB
# quick_abort_max 16 KB
# quick_abort_pct 95

#  TAG: negative_ttl   time-units
#   Time-to-Live (TTL) for failed requests.  Certain types of
#   failures (such as "connection refused" and "404 Not Found") are
#   negatively-cached for a configurable amount of time.  The
#   default is 5 minutes.  Note that this is different from
#   negative caching of DNS lookups.
#
#Default:
# negative_ttl 5 minutes

#  TAG: positive_dns_ttl   time-units
#   Time-to-Live (TTL) for positive caching of successful DNS lookups.
#   Default is 6 hours (360 minutes).  If you want to minimize the
#   use of Squid's ipcache, set this to 1, not 0.
#
#Default:
# positive_dns_ttl 6 hours

#  TAG: negative_dns_ttl   time-units
#   Time-to-Live (TTL) for negative caching of failed DNS lookups.
#
#Default:
# negative_dns_ttl 5 minutes

#  TAG: range_offset_limit   (bytes)
#   Sets a upper limit on how far into the the file a Range request
#   may be to cause Squid to prefetch the whole file. If beyond this
#   limit then Squid forwards the Range request as it is and the result
#   is NOT cached.
#
#   This is to stop a far ahead range request (lets say start at 17MB)
#   from making Squid fetch the whole object up to that point before
#   sending anything to the client.
#
#   A value of -1 causes Squid to always fetch the object from the
#   beginning so that it may cache the result. (2.0 style)
#
#   A value of 0 causes Squid to never fetch more than the
#   client requested. (default)
#
#Default:
# range_offset_limit 0 KB


# TIMEOUTS
# -----------------------------------------------------------------------------

#  TAG: connect_timeout   time-units
#   Some systems (notably Linux) can not be relied upon to properly
#   time out connect(2) requests.  Therefore the Squid process
#   enforces its own timeout on server connections.  This parameter
#   specifies how long to wait for the connect to complete.  The
#   default is two minutes (120 seconds).
#
#Default:
# connect_timeout 2 minutes

#  TAG: peer_connect_timeout   time-units
#   This parameter specifies how long to wait for a pending TCP
#   connection to a peer cache.  The default is 30 seconds.   You
#   may also set different timeout values for individual neighbors
#   with the 'connect-timeout' option on a 'cache_peer' line.
#
#Default:
# peer_connect_timeout 30 seconds

#  TAG: read_timeout   time-units
#   The read_timeout is applied on server-side connections.  After
#   each successful read(), the timeout will be extended by this
#   amount.  If no data is read again after this amount of time,
#   the request is aborted and logged with ERR_READ_TIMEOUT.  The
#   default is 15 minutes.
#
#Default:
# read_timeout 15 minutes

#  TAG: request_timeout
#   How long to wait for an HTTP request after initial
#   connection establishment.
#
#Default:
# request_timeout 5 minutes

#  TAG: persistent_request_timeout
#   How long to wait for the next HTTP request on a persistent
#   connection after the previous request completes.
#
#Default:
# persistent_request_timeout 1 minute

#  TAG: client_lifetime   time-units
#   The maximum amount of time that a client (browser) is allowed to
#   remain connected to the cache process.  This protects the Cache
#   from having a lot of sockets (and hence file descriptors) tied up
#   in a CLOSE_WAIT state from remote clients that go away without
#   properly shutting down (either because of a network failure or
#   because of a poor client implementation).  The default is one
#   day, 1440 minutes.
#
#   NOTE:  The default value is intended to be much larger than any
#   client would ever need to be connected to your cache.  You
#   should probably change client_lifetime only as a last resort.
#   If you seem to have many client connections tying up
#   filedescriptors, we recommend first tuning the read_timeout,
#   request_timeout, persistent_request_timeout and quick_abort values.
#
#Default:
# client_lifetime 1 day

#  TAG: half_closed_clients
#   Some clients may shutdown the sending side of their TCP
#   connections, while leaving their receiving sides open.   Sometimes,
#   Squid can not tell the difference between a half-closed and a
#   fully-closed TCP connection.  By default, half-closed client
#   connections are kept open until a read(2) or write(2) on the
#   socket returns an error.  Change this option to 'off' and Squid
#   will immediately close client connections when read(2) returns
#   "no more data to read."
#
#Default:
# half_closed_clients on

#  TAG: pconn_timeout
#   Timeout for idle persistent connections to servers and other
#   proxies.
#
#Default:
# pconn_timeout 120 seconds

#  TAG: ident_timeout
#   Maximum time to wait for IDENT lookups to complete.
#   
#   If this is too high, and you enabled IDENT lookups from untrusted
#   users, then you might be susceptible to denial-of-service by having
#   many ident requests going at once.
#
#Default:
# ident_timeout 10 seconds

#  TAG: shutdown_lifetime   time-units
#   When SIGTERM or SIGHUP is received, the cache is put into
#   "shutdown pending" mode until all active sockets are closed.
#   This value is the lifetime to set for all open descriptors
#   during shutdown mode.  Any active clients after this many
#   seconds will receive a 'timeout' message.
#
#Default:
# shutdown_lifetime 30 seconds


# ACCESS CONTROLS
# -----------------------------------------------------------------------------

#  TAG: acl
#   Defining an Access List
#
#   acl aclname acltype string1 ...
#   acl aclname acltype "file" ...
#
#   when using "file", the file should contain one item per line
#
#   acltype is one of the types described below
#
#   By default, regular expressions are CASE-SENSITIVE.  To make
#   them case-insensitive, use the -i option.
#
#   acl aclname src      ip-address/netmask ... (clients IP address)
#   acl aclname src      addr1-addr2/netmask ... (range of addresses)
#   acl aclname dst      ip-address/netmask ... (URL host's IP address)
#   acl aclname myip     ip-address/netmask ... (local socket IP address)
#
#   acl aclname srcdomain   .foo.com ...    # reverse lookup, client IP
#   acl aclname dstdomain   .foo.com ...    # Destination server from URL
#   acl aclname srcdom_regex [-i] xxx ...   # regex matching client name
#   acl aclname dstdom_regex [-i] xxx ...   # regex matching server
#     # For dstdomain and dstdom_regex  a reverse lookup is tried if a IP
#     # based URL is used. The name "none" is used if the reverse lookup
#     # fails.
#
#   acl aclname time     [day-abbrevs]  [h1:m1-h2:m2]
#       day-abbrevs:
#      S - Sunday
#      M - Monday
#      T - Tuesday
#      W - Wednesday
#      H - Thursday
#      F - Friday
#      A - Saturday
#       h1:m1 must be less than h2:m2
#   acl aclname url_regex [-i] ^http:// ...   # regex matching on whole URL
#   acl aclname urlpath_regex [-i] \.gif$ ...   # regex matching on URL path
#   acl aclname port     80 70 21 ...
#   acl aclname port     0-1024 ...      # ranges allowed
#   acl aclname myport   3128 ...      # (local socket TCP port)
#   acl aclname proto    HTTP FTP ...
#   acl aclname method   GET POST ...
#   acl aclname browser  [-i] regexp ...
#     # pattern match on User-Agent header
#        acl aclname referer_regex  [-i] regexp ...
#          # pattern match on Referer header
#          # Referer is highly unreliable, so use with care
#   acl aclname ident    username ...
#   acl aclname ident_regex [-i] pattern ...
#     # string match on ident output.
#     # use REQUIRED to accept any non-null ident.
#   acl aclname src_as   number ...
#   acl aclname dst_as   number ...
#     # Except for access control, AS numbers can be used for
#     # routing of requests to specific caches. Here's an
#     # example for routing all requests for AS#1241 and only
#     # those to mycache.mydomain.net:
#     # acl asexample dst_as 1241
#     # cache_peer_access mycache.mydomain.net allow asexample
#     # cache_peer_access mycache_mydomain.net deny all
#
#   acl aclname proxy_auth username ...
#   acl aclname proxy_auth_regex [-i] pattern ...
#     # list of valid usernames
#     # use REQUIRED to accept any valid username.
#     #
#     # NOTE: when a Proxy-Authentication header is sent but it is not
#     # needed during ACL checking the username is NOT logged
#     # in access.log.
#     #
#     # NOTE: proxy_auth requires a EXTERNAL authentication program
#     # to check username/password combinations (see
#     # authenticate_program).
#     #
#     # WARNING: proxy_auth can't be used in a transparent proxy. It
#     # collides with any authentication done by origin servers. It may
#     # seem like it works at first, but it doesn't.
#
#   acl aclname snmp_community string ...
#     # A community string to limit access to your SNMP Agent
#     # Example:
#     #
#     #   acl snmppublic snmp_community public
#
#   acl aclname maxconn number
#     # This will be matched when the client's IP address has
#     # more than <number> HTTP connections established.
#
#   acl aclname max_user_ip [-s] number
#     # This will be matched when the user attempts to log in from more
#     # than <number> different ip address's. The authenticate_ip_ttl
#     # parameter controls the timeout on the ip entries.
#     # If -s is specified then the limit is strict, denying browsing
#     # from any further IP addresses until the ttl has expired. Without
#     # -s Squid will just annoy the user by "randomly" deny requests.
#     # (the counter is then reset each time the limit is reached and a
#     # request is denied)
#     # NOTE: in acceleration mode or where there is mesh of child proxies,
#     # clients may appear to come from multiple address's if they are
#     # going through proxy farms, so a limit of 1 may cause user problems.
#
#   acl aclname req_mime_type mime-type1 ...
#     # regex match agains the mime type of the request generated
#     # by the client. Can be used to detect file upload or some
#     # types HTTP tunelling requests.
#     # NOTE: This does NOT match the reply. You cannot use this
#     # to match the returned file type.
#
#   acl aclname rep_mime_type mime-type1 ...
#     # regex match against the mime type of the reply recieved by
#     # squid. Can be used to detect file download or some
#     # types HTTP tunelling requests.
#     # NOTE: This has no effect in http_access rules. It only has
#     # effect in rules that affect the reply data stream such as
#     # http_reply_access.
#
#   acl acl_name external class_name [arguments...]
#     # external ACL lookup via a helper class defined by the
#     # external_acl_type directive.
#
#Examples:
#acl myexample dst_as 1241
#acl password proxy_auth REQUIRED
#acl fileupload req_mime_type -i ^multipart/form-data$
#acl javascript rep_mime_type -i ^application/x-javascript$
#
#Recommended minimum configuration:
acl all src 0.0.0.0/0.0.0.0
acl manager proto cache_object
acl localhost src 127.0.0.1/255.255.255.255
acl to_localhost dst 127.0.0.0/8
acl SSL_ports port 443 563
acl Safe_ports port 80      # http
acl Safe_ports port 21      # ftp
acl Safe_ports port 443 563   # https, snews
acl Safe_ports port 70      # gopher
acl Safe_ports port 210      # wais
acl Safe_ports port 1025-65535   # unregistered ports
acl Safe_ports port 280      # http-mgmt
acl Safe_ports port 488      # gss-http
acl Safe_ports port 591      # filemaker
acl Safe_ports port 777      # multiling http
acl hotmail_domains dstdomain .hotmail.msn.com
header_access Accept-Encoding deny hotmail_domains
acl CONNECT method CONNECT

#  TAG: http_access
#   Allowing or Denying access based on defined access lists
#
#   Access to the HTTP port:
#   http_access allow|deny [!]aclname ...
#
#   NOTE on default values:
#
#   If there are no "access" lines present, the default is to deny
#   the request.
#
#   If none of the "access" lines cause a match, the default is the
#   opposite of the last line in the list.  If the last line was
#   deny, then the default is allow.  Conversely, if the last line
#   is allow, the default will be deny.  For these reasons, it is a
#   good idea to have an "deny all" or "allow all" entry at the end
#   of your access lists to avoid potential confusion.
#
#Default:
  http_access allow all
# http_access deny all
#
#Recommended minimum configuration:
#
# Only allow cachemgr access from localhost
http_access allow manager localhost
http_access deny manager
# Deny requests to unknown ports
http_access deny !Safe_ports
# Deny CONNECT to other than SSL ports
http_access deny CONNECT !SSL_ports
#
# We strongly recommend to uncomment the following to protect innocent
# web applications running on the proxy server who think that the only
# one who can access services on "localhost" is a local user
#http_access deny to_localhost
#
# INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS

# Exampe rule allowing access from your local networks. Adapt
# to list your (internal) IP networks from where browsing should
# be allowed
#acl our_networks src 192.168.1.0/24 192.168.2.0/24
#http_access allow our_networks

# And finally deny all other access to this proxy
http_access deny all

#  TAG: http_reply_access
#        Allow replies to client requests. This is complementary to http_access.
#
#        http_reply_access allow|deny [!] aclname ...
#
#        NOTE: if there are no access lines present, the default is to allow
#   all replies
#
#        If none of the access lines cause a match, then the opposite of the
#        last line will apply. Thus it is good practice to end the rules
#        with an "allow all" or "deny all" entry.
#
#Default:
# http_reply_access allow all
#
#Recommended minimum configuration:
#
# Insert your own rules here.
#
#
# and finally allow by default
http_reply_access allow all

#  TAG: icp_access
#   Allowing or Denying access to the ICP port based on defined
#   access lists
#
#   icp_access  allow|deny [!]aclname ...
#
#   See http_access for details
#
#Default:
# icp_access deny all
#
#Allow ICP queries from everyone
icp_access allow all

#  TAG: miss_access
#   Use to force your neighbors to use you as a sibling instead of
#   a parent.  For example:
#
#      acl localclients src 172.16.0.0/16
#      miss_access allow localclients
#      miss_access deny  !localclients
#
#   This means that only your local clients are allowed to fetch
#   MISSES and all other clients can only fetch HITS.
#
#   By default, allow all clients who passed the http_access rules
#   to fetch MISSES from us.
#
#Default setting:
# miss_a
  


Comentários

Nenhum comentário foi encontrado.


Contribuir com comentário

  



Patrocínio

Site hospedado pelo provedor RedeHost.
Linux banner

Destaques

Artigos

Dicas

Tópicos

Top 10 do mês

Scripts