· 7 years ago · Nov 09, 2018, 05:20 PM
1# WELCOME TO SQUID 3.5.27
2# ----------------------------
3#
4# This is the documentation for the Squid configuration file.
5# This documentation can also be found online at:
6# http://www.squid-cache.org/Doc/config/
7#
8# You may wish to look at the Squid home page and wiki for the
9# FAQ and other documentation:
10# http://www.squid-cache.org/
11# http://wiki.squid-cache.org/SquidFaq
12# http://wiki.squid-cache.org/ConfigExamples
13#
14# This documentation shows what the defaults for various directives
15# happen to be. If you don't need to change the default, you should
16# leave the line out of your squid.conf in most cases.
17#
18# In some cases "none" refers to no default setting at all,
19# while in other cases it refers to the value of the option
20# - the comments for that keyword indicate if this is the case.
21#
22
23# Configuration options can be included using the "include" directive.
24# Include takes a list of files to include. Quoting and wildcards are
25# supported.
26#
27# For example,
28#
29# include /path/to/included/file/squid.acl.config
30#
31# Includes can be nested up to a hard-coded depth of 16 levels.
32# This arbitrary restriction is to prevent recursive include references
33# from causing Squid entering an infinite loop whilst trying to load
34# configuration files.
35#
36# Values with byte units
37#
38# Squid accepts size units on some size related directives. All
39# such directives are documented with a default value displaying
40# a unit.
41#
42# Units accepted by Squid are:
43# bytes - byte
44# KB - Kilobyte (1024 bytes)
45# MB - Megabyte
46# GB - Gigabyte
47#
48# Values with spaces, quotes, and other special characters
49#
50# Squid supports directive parameters with spaces, quotes, and other
51# special characters. Surround such parameters with "double quotes". Use
52# the configuration_includes_quoted_values directive to enable or
53# disable that support.
54#
55# Squid supports reading configuration option parameters from external
56# files using the syntax:
57# parameters("/path/filename")
58# For example:
59# acl whitelist dstdomain parameters("/etc/squid/whitelist.txt")
60#
61# Conditional configuration
62#
63# If-statements can be used to make configuration directives
64# depend on conditions:
65#
66# if <CONDITION>
67# ... regular configuration directives ...
68# [else
69# ... regular configuration directives ...]
70# endif
71#
72# The else part is optional. The keywords "if", "else", and "endif"
73# must be typed on their own lines, as if they were regular
74# configuration directives.
75#
76# NOTE: An else-if condition is not supported.
77#
78# These individual conditions types are supported:
79#
80# true
81# Always evaluates to true.
82# false
83# Always evaluates to false.
84# <integer> = <integer>
85# Equality comparison of two integer numbers.
86#
87#
88# SMP-Related Macros
89#
90# The following SMP-related preprocessor macros can be used.
91#
92# ${process_name} expands to the current Squid process "name"
93# (e.g., squid1, squid2, or cache1).
94#
95# ${process_number} expands to the current Squid process
96# identifier, which is an integer number (e.g., 1, 2, 3) unique
97# across all Squid processes of the current service instance.
98#
99# ${service_name} expands into the current Squid service instance
100# name identifier which is provided by -n on the command line.
101#
102
103# TAG: broken_vary_encoding
104# This option is not yet supported by Squid-3.
105#Default:
106# none
107
108# TAG: cache_vary
109# This option is not yet supported by Squid-3.
110#Default:
111# none
112
113# TAG: error_map
114# This option is not yet supported by Squid-3.
115#Default:
116# none
117
118# TAG: external_refresh_check
119# This option is not yet supported by Squid-3.
120#Default:
121# none
122
123# TAG: location_rewrite_program
124# This option is not yet supported by Squid-3.
125#Default:
126# none
127
128# TAG: refresh_stale_hit
129# This option is not yet supported by Squid-3.
130#Default:
131# none
132
133# TAG: hierarchy_stoplist
134# Remove this line. Use always_direct or cache_peer_access ACLs instead if you need to prevent cache_peer use.
135#Default:
136# none
137
138# TAG: log_access
139# Remove this line. Use acls with access_log directives to control access logging
140#Default:
141# none
142
143# TAG: log_icap
144# Remove this line. Use acls with icap_log directives to control icap logging
145#Default:
146# none
147
148# TAG: ignore_ims_on_miss
149# Remove this line. The HTTP/1.1 feature is now configured by 'cache_miss_revalidate'.
150#Default:
151# none
152
153# TAG: chunked_request_body_max_size
154# Remove this line. Squid is now HTTP/1.1 compliant.
155#Default:
156# none
157
158# TAG: dns_v4_fallback
159# Remove this line. Squid performs a 'Happy Eyeballs' algorithm, the 'fallback' algorithm is no longer relevant.
160#Default:
161# none
162
163# TAG: emulate_httpd_log
164# Replace this with an access_log directive using the format 'common' or 'combined'.
165#Default:
166# none
167
168# TAG: forward_log
169# Use a regular access.log with ACL limiting it to MISS events.
170#Default:
171# none
172
173# TAG: ftp_list_width
174# Remove this line. Configure FTP page display using the CSS controls in errorpages.css instead.
175#Default:
176# none
177
178# TAG: ignore_expect_100
179# Remove this line. The HTTP/1.1 feature is now fully supported by default.
180#Default:
181# none
182
183# TAG: log_fqdn
184# Remove this option from your config. To log FQDN use %>A in the log format.
185#Default:
186# none
187
188# TAG: log_ip_on_direct
189# Remove this option from your config. To log server or peer names use %<A in the log format.
190#Default:
191# none
192
193# TAG: maximum_single_addr_tries
194# Replaced by connect_retries. The behaviour has changed, please read the documentation before altering.
195#Default:
196# none
197
198# TAG: referer_log
199# Replace this with an access_log directive using the format 'referrer'.
200#Default:
201# none
202
203# TAG: update_headers
204# Remove this line. The feature is supported by default in storage types where update is implemented.
205#Default:
206# none
207
208# TAG: url_rewrite_concurrency
209# Remove this line. Set the 'concurrency=' option of url_rewrite_children instead.
210#Default:
211# none
212
213# TAG: useragent_log
214# Replace this with an access_log directive using the format 'useragent'.
215#Default:
216# none
217
218# TAG: dns_testnames
219# Remove this line. DNS is no longer tested on startup.
220#Default:
221# none
222
223# TAG: extension_methods
224# Remove this line. All valid methods for HTTP are accepted by default.
225#Default:
226# none
227
228# TAG: zero_buffers
229#Default:
230# none
231
232# TAG: incoming_rate
233#Default:
234# none
235
236# TAG: server_http11
237# Remove this line. HTTP/1.1 is supported by default.
238#Default:
239# none
240
241# TAG: upgrade_http0.9
242# Remove this line. ICY/1.0 streaming protocol is supported by default.
243#Default:
244# none
245
246# TAG: zph_local
247# Alter these entries. Use the qos_flows directive instead.
248#Default:
249# none
250
251# TAG: header_access
252# Since squid-3.0 replace with request_header_access or reply_header_access
253# depending on whether you wish to match client requests or server replies.
254#Default:
255# none
256
257# TAG: httpd_accel_no_pmtu_disc
258# Since squid-3.0 use the 'disable-pmtu-discovery' flag on http_port instead.
259#Default:
260# none
261
262# TAG: wais_relay_host
263# Replace this line with 'cache_peer' configuration.
264#Default:
265# none
266
267# TAG: wais_relay_port
268# Replace this line with 'cache_peer' configuration.
269#Default:
270# none
271
272# OPTIONS FOR SMP
273# -----------------------------------------------------------------------------
274
275# TAG: workers
276# Number of main Squid processes or "workers" to fork and maintain.
277# 0: "no daemon" mode, like running "squid -N ..."
278# 1: "no SMP" mode, start one main Squid process daemon (default)
279# N: start N main Squid process daemons (i.e., SMP mode)
280#
281# In SMP mode, each worker does nearly all what a single Squid daemon
282# does (e.g., listen on http_port and forward HTTP requests).
283#Default:
284# SMP support disabled.
285
286# TAG: cpu_affinity_map
287# Usage: cpu_affinity_map process_numbers=P1,P2,... cores=C1,C2,...
288#
289# Sets 1:1 mapping between Squid processes and CPU cores. For example,
290#
291# cpu_affinity_map process_numbers=1,2,3,4 cores=1,3,5,7
292#
293# affects processes 1 through 4 only and places them on the first
294# four even cores, starting with core #1.
295#
296# CPU cores are numbered starting from 1. Requires support for
297# sched_getaffinity(2) and sched_setaffinity(2) system calls.
298#
299# Multiple cpu_affinity_map options are merged.
300#
301# See also: workers
302#Default:
303# Let operating system decide.
304
305# OPTIONS FOR AUTHENTICATION
306# -----------------------------------------------------------------------------
307
308# TAG: auth_param
309# This is used to define parameters for the various authentication
310# schemes supported by Squid.
311#
312# format: auth_param scheme parameter [setting]
313#
314# The order in which authentication schemes are presented to the client is
315# dependent on the order the scheme first appears in config file. IE
316# has a bug (it's not RFC 2617 compliant) in that it will use the basic
317# scheme if basic is the first entry presented, even if more secure
318# schemes are presented. For now use the order in the recommended
319# settings section below. If other browsers have difficulties (don't
320# recognize the schemes offered even if you are using basic) either
321# put basic first, or disable the other schemes (by commenting out their
322# program entry).
323#
324# Once an authentication scheme is fully configured, it can only be
325# shutdown by shutting squid down and restarting. Changes can be made on
326# the fly and activated with a reconfigure. I.E. You can change to a
327# different helper, but not unconfigure the helper completely.
328#
329# Please note that while this directive defines how Squid processes
330# authentication it does not automatically activate authentication.
331# To use authentication you must in addition make use of ACLs based
332# on login name in http_access (proxy_auth, proxy_auth_regex or
333# external with %LOGIN used in the format tag). The browser will be
334# challenged for authentication on the first such acl encountered
335# in http_access processing and will also be re-challenged for new
336# login credentials if the request is being denied by a proxy_auth
337# type acl.
338#
339# WARNING: authentication can't be used in a transparently intercepting
340# proxy as the client then thinks it is talking to an origin server and
341# not the proxy. This is a limitation of bending the TCP/IP protocol to
342# transparently intercepting port 80, not a limitation in Squid.
343# Ports flagged 'transparent', 'intercept', or 'tproxy' have
344# authentication disabled.
345#
346# === Parameters common to all schemes. ===
347#
348# "program" cmdline
349# Specifies the command for the external authenticator.
350#
351# By default, each authentication scheme is not used unless a
352# program is specified.
353#
354# See http://wiki.squid-cache.org/Features/AddonHelpers for
355# more details on helper operations and creating your own.
356#
357# "key_extras" format
358# Specifies a string to be append to request line format for
359# the authentication helper. "Quoted" format values may contain
360# spaces and logformat %macros. In theory, any logformat %macro
361# can be used. In practice, a %macro expands as a dash (-) if
362# the helper request is sent before the required macro
363# information is available to Squid.
364#
365# By default, Squid uses request formats provided in
366# scheme-specific examples below (search for %credentials).
367#
368# The expanded key_extras value is added to the Squid credentials
369# cache and, hence, will affect authentication. It can be used to
370# autenticate different users with identical user names (e.g.,
371# when user authentication depends on http_port).
372#
373# Avoid adding frequently changing information to key_extras. For
374# example, if you add user source IP, and it changes frequently
375# in your environment, then max_user_ip ACL is going to treat
376# every user+IP combination as a unique "user", breaking the ACL
377# and wasting a lot of memory on those user records. It will also
378# force users to authenticate from scratch whenever their IP
379# changes.
380#
381# "realm" string
382# Specifies the protection scope (aka realm name) which is to be
383# reported to the client for the authentication scheme. It is
384# commonly part of the text the user will see when prompted for
385# their username and password.
386#
387# For Basic the default is "Squid proxy-caching web server".
388# For Digest there is no default, this parameter is mandatory.
389# For NTLM and Negotiate this parameter is ignored.
390#
391# "children" numberofchildren [startup=N] [idle=N] [concurrency=N]
392#
393# The maximum number of authenticator processes to spawn. If
394# you start too few Squid will have to wait for them to process
395# a backlog of credential verifications, slowing it down. When
396# password verifications are done via a (slow) network you are
397# likely to need lots of authenticator processes.
398#
399# The startup= and idle= options permit some skew in the exact
400# amount run. A minimum of startup=N will begin during startup
401# and reconfigure. Squid will start more in groups of up to
402# idle=N in an attempt to meet traffic needs and to keep idle=N
403# free above those traffic needs up to the maximum.
404#
405# The concurrency= option sets the number of concurrent requests
406# the helper can process. The default of 0 is used for helpers
407# who only supports one request at a time. Setting this to a
408# number greater than 0 changes the protocol used to include a
409# channel ID field first on the request/response line, allowing
410# multiple requests to be sent to the same helper in parallel
411# without waiting for the response.
412#
413# Concurrency must not be set unless it's known the helper
414# supports the input format with channel-ID fields.
415#
416# NOTE: NTLM and Negotiate schemes do not support concurrency
417# in the Squid code module even though some helpers can.
418#
419#
420#
421# === Example Configuration ===
422#
423# This configuration displays the recommended authentication scheme
424# order from most to least secure with recommended minimum configuration
425# settings for each scheme:
426#
427##auth_param negotiate program <uncomment and complete this line to activate>
428##auth_param negotiate children 20 startup=0 idle=1
429##auth_param negotiate keep_alive on
430##
431##auth_param digest program <uncomment and complete this line to activate>
432##auth_param digest children 20 startup=0 idle=1
433##auth_param digest realm Squid proxy-caching web server
434##auth_param digest nonce_garbage_interval 5 minutes
435##auth_param digest nonce_max_duration 30 minutes
436##auth_param digest nonce_max_count 50
437##
438##auth_param ntlm program <uncomment and complete this line to activate>
439##auth_param ntlm children 20 startup=0 idle=1
440##auth_param ntlm keep_alive on
441#
442auth_param basic program /usr/lib/squid/basic_ncsa_auth /etc/squid/passwords
443auth_param basic realm proxy
444acl authenticated proxy_auth REQUIRED
445http_access allow authenticated
446
447##auth_param basic program <uncomment and complete this line>
448##auth_param basic children 5 startup=5 idle=1
449##auth_param basic realm Squid proxy-caching web server
450##auth_param basic credentialsttl 2 hours
451#Default:
452# none
453
454# TAG: authenticate_cache_garbage_interval
455# The time period between garbage collection across the username cache.
456# This is a trade-off between memory utilization (long intervals - say
457# 2 days) and CPU (short intervals - say 1 minute). Only change if you
458# have good reason to.
459#Default:
460# authenticate_cache_garbage_interval 1 hour
461
462# TAG: authenticate_ttl
463# The time a user & their credentials stay in the logged in
464# user cache since their last request. When the garbage
465# interval passes, all user credentials that have passed their
466# TTL are removed from memory.
467#Default:
468# authenticate_ttl 1 hour
469
470# TAG: authenticate_ip_ttl
471# If you use proxy authentication and the 'max_user_ip' ACL,
472# this directive controls how long Squid remembers the IP
473# addresses associated with each user. Use a small value
474# (e.g., 60 seconds) if your users might change addresses
475# quickly, as is the case with dialup. You might be safe
476# using a larger value (e.g., 2 hours) in a corporate LAN
477# environment with relatively static address assignments.
478#Default:
479# authenticate_ip_ttl 1 second
480
481# ACCESS CONTROLS
482# -----------------------------------------------------------------------------
483
484# TAG: external_acl_type
485# This option defines external acl classes using a helper program
486# to look up the status
487#
488# external_acl_type name [options] FORMAT.. /path/to/helper [helper arguments..]
489#
490# Options:
491#
492# ttl=n TTL in seconds for cached results (defaults to 3600
493# for 1 hour)
494#
495# negative_ttl=n
496# TTL for cached negative lookups (default same
497# as ttl)
498#
499# grace=n Percentage remaining of TTL where a refresh of a
500# cached entry should be initiated without needing to
501# wait for a new reply. (default is for no grace period)
502#
503# cache=n The maximum number of entries in the result cache. The
504# default limit is 262144 entries. Each cache entry usually
505# consumes at least 256 bytes. Squid currently does not remove
506# expired cache entries until the limit is reached, so a proxy
507# will sooner or later reach the limit. The expanded FORMAT
508# value is used as the cache key, so if the details in FORMAT
509# are highly variable, a larger cache may be needed to produce
510# reduction in helper load.
511#
512# children-max=n
513# Maximum number of acl helper processes spawned to service
514# external acl lookups of this type. (default 5)
515#
516# children-startup=n
517# Minimum number of acl helper processes to spawn during
518# startup and reconfigure to service external acl lookups
519# of this type. (default 0)
520#
521# children-idle=n
522# Number of acl helper processes to keep ahead of traffic
523# loads. Squid will spawn this many at once whenever load
524# rises above the capabilities of existing processes.
525# Up to the value of children-max. (default 1)
526#
527# concurrency=n concurrency level per process. Only used with helpers
528# capable of processing more than one query at a time.
529#
530# protocol=2.5 Compatibility mode for Squid-2.5 external acl helpers.
531#
532# ipv4 / ipv6 IP protocol used to communicate with this helper.
533# The default is to auto-detect IPv6 and use it when available.
534#
535#
536# FORMAT specifications
537#
538# %LOGIN Authenticated user login name
539# %un A user name. Expands to the first available name
540# from the following list of information sources:
541# - authenticated user name, like %ul or %LOGIN
542# - user name sent by an external ACL, like %EXT_USER
543# - SSL client name, like %us in logformat
544# - ident user name, like %ui in logformat
545# %EXT_USER Username from previous external acl
546# %EXT_LOG Log details from previous external acl
547# %EXT_TAG Tag from previous external acl
548# %IDENT Ident user name
549# %SRC Client IP
550# %SRCPORT Client source port
551# %URI Requested URI
552# %DST Requested host
553# %PROTO Requested URL scheme
554# %PORT Requested port
555# %PATH Requested URL path
556# %METHOD Request method
557# %MYADDR Squid interface address
558# %MYPORT Squid http_port number
559# %PATH Requested URL-path (including query-string if any)
560# %USER_CERT SSL User certificate in PEM format
561# %USER_CERTCHAIN SSL User certificate chain in PEM format
562# %USER_CERT_xx SSL User certificate subject attribute xx
563# %USER_CA_CERT_xx SSL User certificate issuer attribute xx
564# %ssl::>sni SSL client SNI sent to Squid
565# %ssl::<cert_subject SSL server certificate DN
566# %ssl::<cert_issuer SSL server certificate issuer DN
567#
568# %>{Header} HTTP request header "Header"
569# %>{Hdr:member}
570# HTTP request header "Hdr" list member "member"
571# %>{Hdr:;member}
572# HTTP request header list member using ; as
573# list separator. ; can be any non-alphanumeric
574# character.
575#
576# %<{Header} HTTP reply header "Header"
577# %<{Hdr:member}
578# HTTP reply header "Hdr" list member "member"
579# %<{Hdr:;member}
580# HTTP reply header list member using ; as
581# list separator. ; can be any non-alphanumeric
582# character.
583#
584# %ACL The name of the ACL being tested.
585# %DATA The ACL arguments. If not used then any arguments
586# is automatically added at the end of the line
587# sent to the helper.
588# NOTE: this will encode the arguments as one token,
589# whereas the default will pass each separately.
590#
591# %% The percent sign. Useful for helpers which need
592# an unchanging input format.
593#
594#
595# General request syntax:
596#
597# [channel-ID] FORMAT-values [acl-values ...]
598#
599#
600# FORMAT-values consists of transaction details expanded with
601# whitespace separation per the config file FORMAT specification
602# using the FORMAT macros listed above.
603#
604# acl-values consists of any string specified in the referencing
605# config 'acl ... external' line. see the "acl external" directive.
606#
607# Request values sent to the helper are URL escaped to protect
608# each value in requests against whitespaces.
609#
610# If using protocol=2.5 then the request sent to the helper is not
611# URL escaped to protect against whitespace.
612#
613# NOTE: protocol=3.0 is deprecated as no longer necessary.
614#
615# When using the concurrency= option the protocol is changed by
616# introducing a query channel tag in front of the request/response.
617# The query channel tag is a number between 0 and concurrency-1.
618# This value must be echoed back unchanged to Squid as the first part
619# of the response relating to its request.
620#
621#
622# The helper receives lines expanded per the above format specification
623# and for each input line returns 1 line starting with OK/ERR/BH result
624# code and optionally followed by additional keywords with more details.
625#
626#
627# General result syntax:
628#
629# [channel-ID] result keyword=value ...
630#
631# Result consists of one of the codes:
632#
633# OK
634# the ACL test produced a match.
635#
636# ERR
637# the ACL test does not produce a match.
638#
639# BH
640# An internal error occurred in the helper, preventing
641# a result being identified.
642#
643# The meaning of 'a match' is determined by your squid.conf
644# access control configuration. See the Squid wiki for details.
645#
646# Defined keywords:
647#
648# user= The users name (login)
649#
650# password= The users password (for login= cache_peer option)
651#
652# message= Message describing the reason for this response.
653# Available as %o in error pages.
654# Useful on (ERR and BH results).
655#
656# tag= Apply a tag to a request. Only sets a tag once,
657# does not alter existing tags.
658#
659# log= String to be logged in access.log. Available as
660# %ea in logformat specifications.
661#
662# clt_conn_tag= Associates a TAG with the client TCP connection.
663# Please see url_rewrite_program related documentation
664# for this kv-pair.
665#
666# Any keywords may be sent on any response whether OK, ERR or BH.
667#
668# All response keyword values need to be a single token with URL
669# escaping, or enclosed in double quotes (") and escaped using \ on
670# any double quotes or \ characters within the value. The wrapping
671# double quotes are removed before the value is interpreted by Squid.
672# \r and \n are also replace by CR and LF.
673#
674# Some example key values:
675#
676# user=John%20Smith
677# user="John Smith"
678# user="J. \"Bob\" Smith"
679#Default:
680# none
681
682# TAG: acl
683# Defining an Access List
684#
685# Every access list definition must begin with an aclname and acltype,
686# followed by either type-specific arguments or a quoted filename that
687# they are read from.
688#
689# acl aclname acltype argument ...
690# acl aclname acltype "file" ...
691#
692# When using "file", the file should contain one item per line.
693#
694# Some acl types supports options which changes their default behaviour.
695# The available options are:
696#
697# -i,+i By default, regular expressions are CASE-SENSITIVE. To make them
698# case-insensitive, use the -i option. To return case-sensitive
699# use the +i option between patterns, or make a new ACL line
700# without -i.
701#
702# -n Disable lookups and address type conversions. If lookup or
703# conversion is required because the parameter type (IP or
704# domain name) does not match the message address type (domain
705# name or IP), then the ACL would immediately declare a mismatch
706# without any warnings or lookups.
707#
708# -- Used to stop processing all options, in the case the first acl
709# value has '-' character as first character (for example the '-'
710# is a valid domain name)
711#
712# Some acl types require suspending the current request in order
713# to access some external data source.
714# Those which do are marked with the tag [slow], those which
715# don't are marked as [fast].
716# See http://wiki.squid-cache.org/SquidFaq/SquidAcl
717# for further information
718#
719# ***** ACL TYPES AVAILABLE *****
720#
721# acl aclname src ip-address/mask ... # clients IP address [fast]
722# acl aclname src addr1-addr2/mask ... # range of addresses [fast]
723# acl aclname dst [-n] ip-address/mask ... # URL host's IP address [slow]
724# acl aclname localip ip-address/mask ... # IP address the client connected to [fast]
725#
726# acl aclname arp mac-address ... (xx:xx:xx:xx:xx:xx notation)
727# # [fast]
728# # The 'arp' ACL code is not portable to all operating systems.
729# # It works on Linux, Solaris, Windows, FreeBSD, and some other
730# # BSD variants.
731# #
732# # NOTE: Squid can only determine the MAC/EUI address for IPv4
733# # clients that are on the same subnet. If the client is on a
734# # different subnet, then Squid cannot find out its address.
735# #
736# # NOTE 2: IPv6 protocol does not contain ARP. MAC/EUI is either
737# # encoded directly in the IPv6 address or not available.
738#
739# acl aclname srcdomain .foo.com ...
740# # reverse lookup, from client IP [slow]
741# acl aclname dstdomain [-n] .foo.com ...
742# # Destination server from URL [fast]
743# acl aclname srcdom_regex [-i] \.foo\.com ...
744# # regex matching client name [slow]
745# acl aclname dstdom_regex [-n] [-i] \.foo\.com ...
746# # regex matching server [fast]
747# #
748# # For dstdomain and dstdom_regex a reverse lookup is tried if a IP
749# # based URL is used and no match is found. The name "none" is used
750# # if the reverse lookup fails.
751#
752# acl aclname src_as number ...
753# acl aclname dst_as number ...
754# # [fast]
755# # Except for access control, AS numbers can be used for
756# # routing of requests to specific caches. Here's an
757# # example for routing all requests for AS#1241 and only
758# # those to mycache.mydomain.net:
759# # acl asexample dst_as 1241
760# # cache_peer_access mycache.mydomain.net allow asexample
761# # cache_peer_access mycache_mydomain.net deny all
762#
763# acl aclname peername myPeer ...
764# # [fast]
765# # match against a named cache_peer entry
766# # set unique name= on cache_peer lines for reliable use.
767#
768# acl aclname time [day-abbrevs] [h1:m1-h2:m2]
769# # [fast]
770# # day-abbrevs:
771# # S - Sunday
772# # M - Monday
773# # T - Tuesday
774# # W - Wednesday
775# # H - Thursday
776# # F - Friday
777# # A - Saturday
778# # h1:m1 must be less than h2:m2
779#
780# acl aclname url_regex [-i] ^http:// ...
781# # regex matching on whole URL [fast]
782# acl aclname urllogin [-i] [^a-zA-Z0-9] ...
783# # regex matching on URL login field
784# acl aclname urlpath_regex [-i] \.gif$ ...
785# # regex matching on URL path [fast]
786#
787# acl aclname port 80 70 21 0-1024... # destination TCP port [fast]
788# # ranges are alloed
789# acl aclname localport 3128 ... # TCP port the client connected to [fast]
790# # NP: for interception mode this is usually '80'
791#
792# acl aclname myportname 3128 ... # *_port name [fast]
793#
794# acl aclname proto HTTP FTP ... # request protocol [fast]
795#
796# acl aclname method GET POST ... # HTTP request method [fast]
797#
798# acl aclname http_status 200 301 500- 400-403 ...
799# # status code in reply [fast]
800#
801# acl aclname browser [-i] regexp ...
802# # pattern match on User-Agent header (see also req_header below) [fast]
803#
804# acl aclname referer_regex [-i] regexp ...
805# # pattern match on Referer header [fast]
806# # Referer is highly unreliable, so use with care
807#
808# acl aclname ident username ...
809# acl aclname ident_regex [-i] pattern ...
810# # string match on ident output [slow]
811# # use REQUIRED to accept any non-null ident.
812#
813# acl aclname proxy_auth [-i] username ...
814# acl aclname proxy_auth_regex [-i] pattern ...
815# # perform http authentication challenge to the client and match against
816# # supplied credentials [slow]
817# #
818# # takes a list of allowed usernames.
819# # use REQUIRED to accept any valid username.
820# #
821# # Will use proxy authentication in forward-proxy scenarios, and plain
822# # http authenticaiton in reverse-proxy scenarios
823# #
824# # NOTE: when a Proxy-Authentication header is sent but it is not
825# # needed during ACL checking the username is NOT logged
826# # in access.log.
827# #
828# # NOTE: proxy_auth requires a EXTERNAL authentication program
829# # to check username/password combinations (see
830# # auth_param directive).
831# #
832# # NOTE: proxy_auth can't be used in a transparent/intercepting proxy
833# # as the browser needs to be configured for using a proxy in order
834# # to respond to proxy authentication.
835#
836# acl aclname snmp_community string ...
837# # A community string to limit access to your SNMP Agent [fast]
838# # Example:
839# #
840# # acl snmppublic snmp_community public
841#
842# acl aclname maxconn number
843# # This will be matched when the client's IP address has
844# # more than <number> TCP connections established. [fast]
845# # NOTE: This only measures direct TCP links so X-Forwarded-For
846# # indirect clients are not counted.
847#
848# acl aclname max_user_ip [-s] number
849# # This will be matched when the user attempts to log in from more
850# # than <number> different ip addresses. The authenticate_ip_ttl
851# # parameter controls the timeout on the ip entries. [fast]
852# # If -s is specified the limit is strict, denying browsing
853# # from any further IP addresses until the ttl has expired. Without
854# # -s Squid will just annoy the user by "randomly" denying requests.
855# # (the counter is reset each time the limit is reached and a
856# # request is denied)
857# # NOTE: in acceleration mode or where there is mesh of child proxies,
858# # clients may appear to come from multiple addresses if they are
859# # going through proxy farms, so a limit of 1 may cause user problems.
860#
861# acl aclname random probability
862# # Pseudo-randomly match requests. Based on the probability given.
863# # Probability may be written as a decimal (0.333), fraction (1/3)
864# # or ratio of matches:non-matches (3:5).
865#
866# acl aclname req_mime_type [-i] mime-type ...
867# # regex match against the mime type of the request generated
868# # by the client. Can be used to detect file upload or some
869# # types HTTP tunneling requests [fast]
870# # NOTE: This does NOT match the reply. You cannot use this
871# # to match the returned file type.
872#
873# acl aclname req_header header-name [-i] any\.regex\.here
874# # regex match against any of the known request headers. May be
875# # thought of as a superset of "browser", "referer" and "mime-type"
876# # ACL [fast]
877#
878# acl aclname rep_mime_type [-i] mime-type ...
879# # regex match against the mime type of the reply received by
880# # squid. Can be used to detect file download or some
881# # types HTTP tunneling requests. [fast]
882# # NOTE: This has no effect in http_access rules. It only has
883# # effect in rules that affect the reply data stream such as
884# # http_reply_access.
885#
886# acl aclname rep_header header-name [-i] any\.regex\.here
887# # regex match against any of the known reply headers. May be
888# # thought of as a superset of "browser", "referer" and "mime-type"
889# # ACLs [fast]
890#
891# acl aclname external class_name [arguments...]
892# # external ACL lookup via a helper class defined by the
893# # external_acl_type directive [slow]
894#
895# acl aclname user_cert attribute values...
896# # match against attributes in a user SSL certificate
897# # attribute is one of DN/C/O/CN/L/ST or a numerical OID [fast]
898#
899# acl aclname ca_cert attribute values...
900# # match against attributes a users issuing CA SSL certificate
901# # attribute is one of DN/C/O/CN/L/ST or a numerical OID [fast]
902#
903# acl aclname ext_user username ...
904# acl aclname ext_user_regex [-i] pattern ...
905# # string match on username returned by external acl helper [slow]
906# # use REQUIRED to accept any non-null user name.
907#
908# acl aclname tag tagvalue ...
909# # string match on tag returned by external acl helper [fast]
910# # DEPRECATED. Only the first tag will match with this ACL.
911# # Use the 'note' ACL instead for handling multiple tag values.
912#
913# acl aclname hier_code codename ...
914# # string match against squid hierarchy code(s); [fast]
915# # e.g., DIRECT, PARENT_HIT, NONE, etc.
916# #
917# # NOTE: This has no effect in http_access rules. It only has
918# # effect in rules that affect the reply data stream such as
919# # http_reply_access.
920#
921# acl aclname note name [value ...]
922# # match transaction annotation [fast]
923# # Without values, matches any annotation with a given name.
924# # With value(s), matches any annotation with a given name that
925# # also has one of the given values.
926# # Names and values are compared using a string equality test.
927# # Annotation sources include note and adaptation_meta directives
928# # as well as helper and eCAP responses.
929#
930# acl aclname adaptation_service service ...
931# # Matches the name of any icap_service, ecap_service,
932# # adaptation_service_set, or adaptation_service_chain that Squid
933# # has used (or attempted to use) for the master transaction.
934# # This ACL must be defined after the corresponding adaptation
935# # service is named in squid.conf. This ACL is usable with
936# # adaptation_meta because it starts matching immediately after
937# # the service has been selected for adaptation.
938#
939# acl aclname any-of acl1 acl2 ...
940# # match any one of the acls [fast or slow]
941# # The first matching ACL stops further ACL evaluation.
942# #
943# # ACLs from multiple any-of lines with the same name are ORed.
944# # For example, A = (a1 or a2) or (a3 or a4) can be written as
945# # acl A any-of a1 a2
946# # acl A any-of a3 a4
947# #
948# # This group ACL is fast if all evaluated ACLs in the group are fast
949# # and slow otherwise.
950#
951# acl aclname all-of acl1 acl2 ...
952# # match all of the acls [fast or slow]
953# # The first mismatching ACL stops further ACL evaluation.
954# #
955# # ACLs from multiple all-of lines with the same name are ORed.
956# # For example, B = (b1 and b2) or (b3 and b4) can be written as
957# # acl B all-of b1 b2
958# # acl B all-of b3 b4
959# #
960# # This group ACL is fast if all evaluated ACLs in the group are fast
961# # and slow otherwise.
962#
963# Examples:
964# acl macaddress arp 09:00:2b:23:45:67
965# acl myexample dst_as 1241
966# acl password proxy_auth REQUIRED
967# acl fileupload req_mime_type -i ^multipart/form-data$
968# acl javascript rep_mime_type -i ^application/x-javascript$
969#
970#Default:
971# ACLs all, manager, localhost, and to_localhost are predefined.
972#
973#
974# Recommended minimum configuration:
975#
976
977# Example rule allowing access from your local networks.
978# Adapt to list your (internal) IP networks from where browsing
979# should be allowed
980acl localnet src 10.0.0.0/8 # RFC1918 possible internal network
981#acl localnet src 172.16.0.0/12 # RFC1918 possible internal network
982#acl localnet src 192.168.0.0/16 # RFC1918 possible internal network
983#acl localnet src fc00::/7 # RFC 4193 local private network range
984#acl localnet src fe80::/10 # RFC 4291 link-local (directly plugged) machines
985
986acl SSL_ports port 443
987acl Safe_ports port 80 # http
988acl Safe_ports port 21 # ftp
989acl Safe_ports port 443 # https
990acl Safe_ports port 70 # gopher
991acl Safe_ports port 210 # wais
992acl Safe_ports port 1025-65535 # unregistered ports
993acl Safe_ports port 280 # http-mgmt
994acl Safe_ports port 488 # gss-http
995acl Safe_ports port 591 # filemaker
996acl Safe_ports port 777 # multiling http
997acl CONNECT method CONNECT
998
999# TAG: proxy_protocol_access
1000# Determine which client proxies can be trusted to provide correct
1001# information regarding real client IP address using PROXY protocol.
1002#
1003# Requests may pass through a chain of several other proxies
1004# before reaching us. The original source details may by sent in:
1005# * HTTP message Forwarded header, or
1006# * HTTP message X-Forwarded-For header, or
1007# * PROXY protocol connection header.
1008#
1009# This directive is solely for validating new PROXY protocol
1010# connections received from a port flagged with require-proxy-header.
1011# It is checked only once after TCP connection setup.
1012#
1013# A deny match results in TCP connection closure.
1014#
1015# An allow match is required for Squid to permit the corresponding
1016# TCP connection, before Squid even looks for HTTP request headers.
1017# If there is an allow match, Squid starts using PROXY header information
1018# to determine the source address of the connection for all future ACL
1019# checks, logging, etc.
1020#
1021# SECURITY CONSIDERATIONS:
1022#
1023# Any host from which we accept client IP details can place
1024# incorrect information in the relevant header, and Squid
1025# will use the incorrect information as if it were the
1026# source address of the request. This may enable remote
1027# hosts to bypass any access control restrictions that are
1028# based on the client's source addresses.
1029#
1030# This clause only supports fast acl types.
1031# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1032#Default:
1033# all TCP connections to ports with require-proxy-header will be denied
1034
1035# TAG: follow_x_forwarded_for
1036# Determine which client proxies can be trusted to provide correct
1037# information regarding real client IP address.
1038#
1039# Requests may pass through a chain of several other proxies
1040# before reaching us. The original source details may by sent in:
1041# * HTTP message Forwarded header, or
1042# * HTTP message X-Forwarded-For header, or
1043# * PROXY protocol connection header.
1044#
1045# PROXY protocol connections are controlled by the proxy_protocol_access
1046# directive which is checked before this.
1047#
1048# If a request reaches us from a source that is allowed by this
1049# directive, then we trust the information it provides regarding
1050# the IP of the client it received from (if any).
1051#
1052# For the purpose of ACLs used in this directive the src ACL type always
1053# matches the address we are testing and srcdomain matches its rDNS.
1054#
1055# On each HTTP request Squid checks for X-Forwarded-For header fields.
1056# If found the header values are iterated in reverse order and an allow
1057# match is required for Squid to continue on to the next value.
1058# The verification ends when a value receives a deny match, cannot be
1059# tested, or there are no more values to test.
1060# NOTE: Squid does not yet follow the Forwarded HTTP header.
1061#
1062# The end result of this process is an IP address that we will
1063# refer to as the indirect client address. This address may
1064# be treated as the client address for access control, ICAP, delay
1065# pools and logging, depending on the acl_uses_indirect_client,
1066# icap_uses_indirect_client, delay_pool_uses_indirect_client,
1067# log_uses_indirect_client and tproxy_uses_indirect_client options.
1068#
1069# This clause only supports fast acl types.
1070# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1071#
1072# SECURITY CONSIDERATIONS:
1073#
1074# Any host from which we accept client IP details can place
1075# incorrect information in the relevant header, and Squid
1076# will use the incorrect information as if it were the
1077# source address of the request. This may enable remote
1078# hosts to bypass any access control restrictions that are
1079# based on the client's source addresses.
1080#
1081# For example:
1082#
1083# acl localhost src 127.0.0.1
1084# acl my_other_proxy srcdomain .proxy.example.com
1085# follow_x_forwarded_for allow localhost
1086# follow_x_forwarded_for allow my_other_proxy
1087#Default:
1088# X-Forwarded-For header will be ignored.
1089
1090# TAG: acl_uses_indirect_client on|off
1091# Controls whether the indirect client address
1092# (see follow_x_forwarded_for) is used instead of the
1093# direct client address in acl matching.
1094#
1095# NOTE: maxconn ACL considers direct TCP links and indirect
1096# clients will always have zero. So no match.
1097#Default:
1098# acl_uses_indirect_client on
1099
1100# TAG: delay_pool_uses_indirect_client on|off
1101# Controls whether the indirect client address
1102# (see follow_x_forwarded_for) is used instead of the
1103# direct client address in delay pools.
1104#Default:
1105# delay_pool_uses_indirect_client on
1106
1107# TAG: log_uses_indirect_client on|off
1108# Controls whether the indirect client address
1109# (see follow_x_forwarded_for) is used instead of the
1110# direct client address in the access log.
1111#Default:
1112# log_uses_indirect_client on
1113
1114# TAG: tproxy_uses_indirect_client on|off
1115# Controls whether the indirect client address
1116# (see follow_x_forwarded_for) is used instead of the
1117# direct client address when spoofing the outgoing client.
1118#
1119# This has no effect on requests arriving in non-tproxy
1120# mode ports.
1121#
1122# SECURITY WARNING: Usage of this option is dangerous
1123# and should not be used trivially. Correct configuration
1124# of follow_x_forwarded_for with a limited set of trusted
1125# sources is required to prevent abuse of your proxy.
1126#Default:
1127# tproxy_uses_indirect_client off
1128
1129# TAG: spoof_client_ip
1130# Control client IP address spoofing of TPROXY traffic based on
1131# defined access lists.
1132#
1133# spoof_client_ip allow|deny [!]aclname ...
1134#
1135# If there are no "spoof_client_ip" lines present, the default
1136# is to "allow" spoofing of any suitable request.
1137#
1138# Note that the cache_peer "no-tproxy" option overrides this ACL.
1139#
1140# This clause supports fast acl types.
1141# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1142#Default:
1143# Allow spoofing on all TPROXY traffic.
1144
1145# TAG: http_access
1146# Allowing or Denying access based on defined access lists
1147#
1148# To allow or deny a message received on an HTTP, HTTPS, or FTP port:
1149# http_access allow|deny [!]aclname ...
1150#
1151# NOTE on default values:
1152#
1153# If there are no "access" lines present, the default is to deny
1154# the request.
1155#
1156# If none of the "access" lines cause a match, the default is the
1157# opposite of the last line in the list. If the last line was
1158# deny, the default is allow. Conversely, if the last line
1159# is allow, the default will be deny. For these reasons, it is a
1160# good idea to have an "deny all" entry at the end of your access
1161# lists to avoid potential confusion.
1162#
1163# This clause supports both fast and slow acl types.
1164# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1165#
1166#Default:
1167# Deny, unless rules exist in squid.conf.
1168#
1169
1170#
1171# Recommended minimum Access Permission configuration:
1172#
1173# Deny requests to certain unsafe ports
1174http_access deny !Safe_ports
1175
1176# Deny CONNECT to other than secure SSL ports
1177http_access deny CONNECT !SSL_ports
1178
1179# Only allow cachemgr access from localhost
1180http_access allow localhost manager
1181http_access deny manager
1182
1183# We strongly recommend the following be uncommented to protect innocent
1184# web applications running on the proxy server who think the only
1185# one who can access services on "localhost" is a local user
1186#http_access deny to_localhost
1187
1188#
1189# INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS
1190#
1191
1192# Example rule allowing access from your local networks.
1193# Adapt localnet in the ACL section to list your (internal) IP networks
1194# from where browsing should be allowed
1195http_access allow localnet
1196http_access allow localhost
1197
1198# And finally deny all other access to this proxy
1199http_access deny all
1200
1201# TAG: adapted_http_access
1202# Allowing or Denying access based on defined access lists
1203#
1204# Essentially identical to http_access, but runs after redirectors
1205# and ICAP/eCAP adaptation. Allowing access control based on their
1206# output.
1207#
1208# If not set then only http_access is used.
1209#Default:
1210# Allow, unless rules exist in squid.conf.
1211
1212# TAG: http_reply_access
1213# Allow replies to client requests. This is complementary to http_access.
1214#
1215# http_reply_access allow|deny [!] aclname ...
1216#
1217# NOTE: if there are no access lines present, the default is to allow
1218# all replies.
1219#
1220# If none of the access lines cause a match the opposite of the
1221# last line will apply. Thus it is good practice to end the rules
1222# with an "allow all" or "deny all" entry.
1223#
1224# This clause supports both fast and slow acl types.
1225# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1226#Default:
1227# Allow, unless rules exist in squid.conf.
1228
1229# TAG: icp_access
1230# Allowing or Denying access to the ICP port based on defined
1231# access lists
1232#
1233# icp_access allow|deny [!]aclname ...
1234#
1235# NOTE: The default if no icp_access lines are present is to
1236# deny all traffic. This default may cause problems with peers
1237# using ICP.
1238#
1239# This clause only supports fast acl types.
1240# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1241#
1242## Allow ICP queries from local networks only
1243##icp_access allow localnet
1244##icp_access deny all
1245#Default:
1246# Deny, unless rules exist in squid.conf.
1247
1248# TAG: htcp_access
1249# Allowing or Denying access to the HTCP port based on defined
1250# access lists
1251#
1252# htcp_access allow|deny [!]aclname ...
1253#
1254# See also htcp_clr_access for details on access control for
1255# cache purge (CLR) HTCP messages.
1256#
1257# NOTE: The default if no htcp_access lines are present is to
1258# deny all traffic. This default may cause problems with peers
1259# using the htcp option.
1260#
1261# This clause only supports fast acl types.
1262# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1263#
1264## Allow HTCP queries from local networks only
1265##htcp_access allow localnet
1266##htcp_access deny all
1267#Default:
1268# Deny, unless rules exist in squid.conf.
1269
1270# TAG: htcp_clr_access
1271# Allowing or Denying access to purge content using HTCP based
1272# on defined access lists.
1273# See htcp_access for details on general HTCP access control.
1274#
1275# htcp_clr_access allow|deny [!]aclname ...
1276#
1277# This clause only supports fast acl types.
1278# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1279#
1280## Allow HTCP CLR requests from trusted peers
1281#acl htcp_clr_peer src 192.0.2.2 2001:DB8::2
1282#htcp_clr_access allow htcp_clr_peer
1283#htcp_clr_access deny all
1284#Default:
1285# Deny, unless rules exist in squid.conf.
1286
1287# TAG: miss_access
1288# Determines whether network access is permitted when satisfying a request.
1289#
1290# For example;
1291# to force your neighbors to use you as a sibling instead of
1292# a parent.
1293#
1294# acl localclients src 192.0.2.0/24 2001:DB8::a:0/64
1295# miss_access deny !localclients
1296# miss_access allow all
1297#
1298# This means only your local clients are allowed to fetch relayed/MISS
1299# replies from the network and all other clients can only fetch cached
1300# objects (HITs).
1301#
1302# The default for this setting allows all clients who passed the
1303# http_access rules to relay via this proxy.
1304#
1305# This clause only supports fast acl types.
1306# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1307#Default:
1308# Allow, unless rules exist in squid.conf.
1309
1310# TAG: ident_lookup_access
1311# A list of ACL elements which, if matched, cause an ident
1312# (RFC 931) lookup to be performed for this request. For
1313# example, you might choose to always perform ident lookups
1314# for your main multi-user Unix boxes, but not for your Macs
1315# and PCs. By default, ident lookups are not performed for
1316# any requests.
1317#
1318# To enable ident lookups for specific client addresses, you
1319# can follow this example:
1320#
1321# acl ident_aware_hosts src 198.168.1.0/24
1322# ident_lookup_access allow ident_aware_hosts
1323# ident_lookup_access deny all
1324#
1325# Only src type ACL checks are fully supported. A srcdomain
1326# ACL might work at times, but it will not always provide
1327# the correct result.
1328#
1329# This clause only supports fast acl types.
1330# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1331#Default:
1332# Unless rules exist in squid.conf, IDENT is not fetched.
1333
1334# TAG: reply_body_max_size size [acl acl...]
1335# This option specifies the maximum size of a reply body. It can be
1336# used to prevent users from downloading very large files, such as
1337# MP3's and movies. When the reply headers are received, the
1338# reply_body_max_size lines are processed, and the first line where
1339# all (if any) listed ACLs are true is used as the maximum body size
1340# for this reply.
1341#
1342# This size is checked twice. First when we get the reply headers,
1343# we check the content-length value. If the content length value exists
1344# and is larger than the allowed size, the request is denied and the
1345# user receives an error message that says "the request or reply
1346# is too large." If there is no content-length, and the reply
1347# size exceeds this limit, the client's connection is just closed
1348# and they will receive a partial reply.
1349#
1350# WARNING: downstream caches probably can not detect a partial reply
1351# if there is no content-length header, so they will cache
1352# partial responses and give them out as hits. You should NOT
1353# use this option if you have downstream caches.
1354#
1355# WARNING: A maximum size smaller than the size of squid's error messages
1356# will cause an infinite loop and crash squid. Ensure that the smallest
1357# non-zero value you use is greater that the maximum header size plus
1358# the size of your largest error page.
1359#
1360# If you set this parameter none (the default), there will be
1361# no limit imposed.
1362#
1363# Configuration Format is:
1364# reply_body_max_size SIZE UNITS [acl ...]
1365# ie.
1366# reply_body_max_size 10 MB
1367#
1368#Default:
1369# No limit is applied.
1370
1371# NETWORK OPTIONS
1372# -----------------------------------------------------------------------------
1373
1374# TAG: http_port
1375# Usage: port [mode] [options]
1376# hostname:port [mode] [options]
1377# 1.2.3.4:port [mode] [options]
1378#
1379# The socket addresses where Squid will listen for HTTP client
1380# requests. You may specify multiple socket addresses.
1381# There are three forms: port alone, hostname with port, and
1382# IP address with port. If you specify a hostname or IP
1383# address, Squid binds the socket to that specific
1384# address. Most likely, you do not need to bind to a specific
1385# address, so you can use the port number alone.
1386#
1387# If you are running Squid in accelerator mode, you
1388# probably want to listen on port 80 also, or instead.
1389#
1390# The -a command line option may be used to specify additional
1391# port(s) where Squid listens for proxy request. Such ports will
1392# be plain proxy ports with no options.
1393#
1394# You may specify multiple socket addresses on multiple lines.
1395#
1396# Modes:
1397#
1398# intercept Support for IP-Layer NAT interception delivering
1399# traffic to this Squid port.
1400# NP: disables authentication on the port.
1401#
1402# tproxy Support Linux TPROXY (or BSD divert-to) with spoofing
1403# of outgoing connections using the client IP address.
1404# NP: disables authentication on the port.
1405#
1406# accel Accelerator / reverse proxy mode
1407#
1408# ssl-bump For each CONNECT request allowed by ssl_bump ACLs,
1409# establish secure connection with the client and with
1410# the server, decrypt HTTPS messages as they pass through
1411# Squid, and treat them as unencrypted HTTP messages,
1412# becoming the man-in-the-middle.
1413#
1414# The ssl_bump option is required to fully enable
1415# bumping of CONNECT requests.
1416#
1417# Omitting the mode flag causes default forward proxy mode to be used.
1418#
1419#
1420# Accelerator Mode Options:
1421#
1422# defaultsite=domainname
1423# What to use for the Host: header if it is not present
1424# in a request. Determines what site (not origin server)
1425# accelerators should consider the default.
1426#
1427# no-vhost Disable using HTTP/1.1 Host header for virtual domain support.
1428#
1429# protocol= Protocol to reconstruct accelerated and intercepted
1430# requests with. Defaults to HTTP/1.1 for http_port and
1431# HTTPS/1.1 for https_port.
1432# When an unsupported value is configured Squid will
1433# produce a FATAL error.
1434# Values: HTTP or HTTP/1.1, HTTPS or HTTPS/1.1
1435#
1436# vport Virtual host port support. Using the http_port number
1437# instead of the port passed on Host: headers.
1438#
1439# vport=NN Virtual host port support. Using the specified port
1440# number instead of the port passed on Host: headers.
1441#
1442# act-as-origin
1443# Act as if this Squid is the origin server.
1444# This currently means generate new Date: and Expires:
1445# headers on HIT instead of adding Age:.
1446#
1447# ignore-cc Ignore request Cache-Control headers.
1448#
1449# WARNING: This option violates HTTP specifications if
1450# used in non-accelerator setups.
1451#
1452# allow-direct Allow direct forwarding in accelerator mode. Normally
1453# accelerated requests are denied direct forwarding as if
1454# never_direct was used.
1455#
1456# WARNING: this option opens accelerator mode to security
1457# vulnerabilities usually only affecting in interception
1458# mode. Make sure to protect forwarding with suitable
1459# http_access rules when using this.
1460#
1461#
1462# SSL Bump Mode Options:
1463# In addition to these options ssl-bump requires TLS/SSL options.
1464#
1465# generate-host-certificates[=<on|off>]
1466# Dynamically create SSL server certificates for the
1467# destination hosts of bumped CONNECT requests.When
1468# enabled, the cert and key options are used to sign
1469# generated certificates. Otherwise generated
1470# certificate will be selfsigned.
1471# If there is a CA certificate lifetime of the generated
1472# certificate equals lifetime of the CA certificate. If
1473# generated certificate is selfsigned lifetime is three
1474# years.
1475# This option is disabled by default. See the ssl-bump
1476# option above for more information.
1477#
1478# dynamic_cert_mem_cache_size=SIZE
1479# Approximate total RAM size spent on cached generated
1480# certificates. If set to zero, caching is disabled.
1481#
1482# TLS / SSL Options:
1483#
1484# cert= Path to SSL certificate (PEM format).
1485#
1486# key= Path to SSL private key file (PEM format)
1487# if not specified, the certificate file is
1488# assumed to be a combined certificate and
1489# key file.
1490#
1491# version= The version of SSL/TLS supported
1492# 1 automatic (default)
1493# 2 SSLv2 only
1494# 3 SSLv3 only
1495# 4 TLSv1.0 only
1496# 5 TLSv1.1 only
1497# 6 TLSv1.2 only
1498#
1499# cipher= Colon separated list of supported ciphers.
1500# NOTE: some ciphers such as EDH ciphers depend on
1501# additional settings. If those settings are
1502# omitted the ciphers may be silently ignored
1503# by the OpenSSL library.
1504#
1505# options= Various SSL implementation options. The most important
1506# being:
1507# NO_SSLv2 Disallow the use of SSLv2
1508# NO_SSLv3 Disallow the use of SSLv3
1509# NO_TLSv1 Disallow the use of TLSv1.0
1510# NO_TLSv1_1 Disallow the use of TLSv1.1
1511# NO_TLSv1_2 Disallow the use of TLSv1.2
1512# SINGLE_DH_USE Always create a new key when using
1513# temporary/ephemeral DH key exchanges
1514# NO_TICKET Disables TLS tickets extension
1515#
1516# SINGLE_ECDH_USE
1517# Enable ephemeral ECDH key exchange.
1518# The adopted curve should be specified
1519# using the tls-dh option.
1520#
1521# ALL Enable various bug workarounds
1522# suggested as "harmless" by OpenSSL
1523# Be warned that this reduces SSL/TLS
1524# strength to some attacks.
1525# See OpenSSL SSL_CTX_set_options documentation for a
1526# complete list of options.
1527#
1528# clientca= File containing the list of CAs to use when
1529# requesting a client certificate.
1530#
1531# cafile= File containing additional CA certificates to
1532# use when verifying client certificates. If unset
1533# clientca will be used.
1534#
1535# capath= Directory containing additional CA certificates
1536# and CRL lists to use when verifying client certificates.
1537#
1538# crlfile= File of additional CRL lists to use when verifying
1539# the client certificate, in addition to CRLs stored in
1540# the capath. Implies VERIFY_CRL flag below.
1541#
1542# tls-dh=[curve:]file
1543# File containing DH parameters for temporary/ephemeral DH key
1544# exchanges, optionally prefixed by a curve for ephemeral ECDH
1545# key exchanges.
1546# See OpenSSL documentation for details on how to create the
1547# DH parameter file. Supported curves for ECDH can be listed
1548# using the "openssl ecparam -list_curves" command.
1549# WARNING: EDH and EECDH ciphers will be silently disabled if
1550# this option is not set.
1551#
1552# sslflags= Various flags modifying the use of SSL:
1553# DELAYED_AUTH
1554# Don't request client certificates
1555# immediately, but wait until acl processing
1556# requires a certificate (not yet implemented).
1557# NO_DEFAULT_CA
1558# Don't use the default CA lists built in
1559# to OpenSSL.
1560# NO_SESSION_REUSE
1561# Don't allow for session reuse. Each connection
1562# will result in a new SSL session.
1563# VERIFY_CRL
1564# Verify CRL lists when accepting client
1565# certificates.
1566# VERIFY_CRL_ALL
1567# Verify CRL lists for all certificates in the
1568# client certificate chain.
1569#
1570# sslcontext= SSL session ID context identifier.
1571#
1572# Other Options:
1573#
1574# connection-auth[=on|off]
1575# use connection-auth=off to tell Squid to prevent
1576# forwarding Microsoft connection oriented authentication
1577# (NTLM, Negotiate and Kerberos)
1578#
1579# disable-pmtu-discovery=
1580# Control Path-MTU discovery usage:
1581# off lets OS decide on what to do (default).
1582# transparent disable PMTU discovery when transparent
1583# support is enabled.
1584# always disable always PMTU discovery.
1585#
1586# In many setups of transparently intercepting proxies
1587# Path-MTU discovery can not work on traffic towards the
1588# clients. This is the case when the intercepting device
1589# does not fully track connections and fails to forward
1590# ICMP must fragment messages to the cache server. If you
1591# have such setup and experience that certain clients
1592# sporadically hang or never complete requests set
1593# disable-pmtu-discovery option to 'transparent'.
1594#
1595# name= Specifies a internal name for the port. Defaults to
1596# the port specification (port or addr:port)
1597#
1598# tcpkeepalive[=idle,interval,timeout]
1599# Enable TCP keepalive probes of idle connections.
1600# In seconds; idle is the initial time before TCP starts
1601# probing the connection, interval how often to probe, and
1602# timeout the time before giving up.
1603#
1604# require-proxy-header
1605# Require PROXY protocol version 1 or 2 connections.
1606# The proxy_protocol_access is required to whitelist
1607# downstream proxies which can be trusted.
1608#
1609# If you run Squid on a dual-homed machine with an internal
1610# and an external interface we recommend you to specify the
1611# internal address:port in http_port. This way Squid will only be
1612# visible on the internal address.
1613#
1614#
1615
1616# Squid normally listens to port 3128
1617http_port 3128
1618
1619# TAG: https_port
1620# Note: This option is only available if Squid is rebuilt with the
1621# --with-openssl
1622#
1623# Usage: [ip:]port cert=certificate.pem [key=key.pem] [mode] [options...]
1624#
1625# The socket address where Squid will listen for client requests made
1626# over TLS or SSL connections. Commonly referred to as HTTPS.
1627#
1628# This is most useful for situations where you are running squid in
1629# accelerator mode and you want to do the SSL work at the accelerator level.
1630#
1631# You may specify multiple socket addresses on multiple lines,
1632# each with their own SSL certificate and/or options.
1633#
1634# Modes:
1635#
1636# accel Accelerator / reverse proxy mode
1637#
1638# intercept Support for IP-Layer interception of
1639# outgoing requests without browser settings.
1640# NP: disables authentication and IPv6 on the port.
1641#
1642# tproxy Support Linux TPROXY for spoofing outgoing
1643# connections using the client IP address.
1644# NP: disables authentication and maybe IPv6 on the port.
1645#
1646# ssl-bump For each intercepted connection allowed by ssl_bump
1647# ACLs, establish a secure connection with the client and with
1648# the server, decrypt HTTPS messages as they pass through
1649# Squid, and treat them as unencrypted HTTP messages,
1650# becoming the man-in-the-middle.
1651#
1652# An "ssl_bump server-first" match is required to
1653# fully enable bumping of intercepted SSL connections.
1654#
1655# Requires tproxy or intercept.
1656#
1657# Omitting the mode flag causes default forward proxy mode to be used.
1658#
1659#
1660# See http_port for a list of generic options
1661#
1662#
1663# SSL Options:
1664#
1665# cert= Path to SSL certificate (PEM format).
1666#
1667# key= Path to SSL private key file (PEM format)
1668# if not specified, the certificate file is
1669# assumed to be a combined certificate and
1670# key file.
1671#
1672# version= The version of SSL/TLS supported
1673# 1 automatic (default)
1674# 2 SSLv2 only
1675# 3 SSLv3 only
1676# 4 TLSv1 only
1677#
1678# cipher= Colon separated list of supported ciphers.
1679#
1680# options= Various SSL engine options. The most important
1681# being:
1682# NO_SSLv2 Disallow the use of SSLv2
1683# NO_SSLv3 Disallow the use of SSLv3
1684# NO_TLSv1 Disallow the use of TLSv1
1685#
1686# SINGLE_DH_USE Always create a new key when using
1687# temporary/ephemeral DH key exchanges
1688#
1689# SINGLE_ECDH_USE
1690# Enable ephemeral ECDH key exchange.
1691# The adopted curve should be specified
1692# using the tls-dh option.
1693#
1694# See src/ssl_support.c or OpenSSL SSL_CTX_set_options
1695# documentation for a complete list of options.
1696#
1697# clientca= File containing the list of CAs to use when
1698# requesting a client certificate.
1699#
1700# cafile= File containing additional CA certificates to
1701# use when verifying client certificates. If unset
1702# clientca will be used.
1703#
1704# capath= Directory containing additional CA certificates
1705# and CRL lists to use when verifying client certificates.
1706#
1707# crlfile= File of additional CRL lists to use when verifying
1708# the client certificate, in addition to CRLs stored in
1709# the capath. Implies VERIFY_CRL flag below.
1710#
1711# tls-dh=[curve:]file
1712# File containing DH parameters for temporary/ephemeral DH key
1713# exchanges, optionally prefixed by a curve for ephemeral ECDH
1714# key exchanges.
1715#
1716# sslflags= Various flags modifying the use of SSL:
1717# DELAYED_AUTH
1718# Don't request client certificates
1719# immediately, but wait until acl processing
1720# requires a certificate (not yet implemented).
1721# NO_DEFAULT_CA
1722# Don't use the default CA lists built in
1723# to OpenSSL.
1724# NO_SESSION_REUSE
1725# Don't allow for session reuse. Each connection
1726# will result in a new SSL session.
1727# VERIFY_CRL
1728# Verify CRL lists when accepting client
1729# certificates.
1730# VERIFY_CRL_ALL
1731# Verify CRL lists for all certificates in the
1732# client certificate chain.
1733#
1734# sslcontext= SSL session ID context identifier.
1735#
1736# generate-host-certificates[=<on|off>]
1737# Dynamically create SSL server certificates for the
1738# destination hosts of bumped SSL requests.When
1739# enabled, the cert and key options are used to sign
1740# generated certificates. Otherwise generated
1741# certificate will be selfsigned.
1742# If there is CA certificate life time of generated
1743# certificate equals lifetime of CA certificate. If
1744# generated certificate is selfsigned lifetime is three
1745# years.
1746# This option is disabled by default. See the ssl-bump
1747# option above for more information.
1748#
1749# dynamic_cert_mem_cache_size=SIZE
1750# Approximate total RAM size spent on cached generated
1751# certificates. If set to zero, caching is disabled.
1752#
1753# See http_port for a list of available options.
1754#Default:
1755# none
1756
1757# TAG: ftp_port
1758# Enables Native FTP proxy by specifying the socket address where Squid
1759# listens for FTP client requests. See http_port directive for various
1760# ways to specify the listening address and mode.
1761#
1762# Usage: ftp_port address [mode] [options]
1763#
1764# WARNING: This is a new, experimental, complex feature that has seen
1765# limited production exposure. Some Squid modules (e.g., caching) do not
1766# currently work with native FTP proxying, and many features have not
1767# even been tested for compatibility. Test well before deploying!
1768#
1769# Native FTP proxying differs substantially from proxying HTTP requests
1770# with ftp:// URIs because Squid works as an FTP server and receives
1771# actual FTP commands (rather than HTTP requests with FTP URLs).
1772#
1773# Native FTP commands accepted at ftp_port are internally converted or
1774# wrapped into HTTP-like messages. The same happens to Native FTP
1775# responses received from FTP origin servers. Those HTTP-like messages
1776# are shoveled through regular access control and adaptation layers
1777# between the FTP client and the FTP origin server. This allows Squid to
1778# examine, adapt, block, and log FTP exchanges. Squid reuses most HTTP
1779# mechanisms when shoveling wrapped FTP messages. For example,
1780# http_access and adaptation_access directives are used.
1781#
1782# Modes:
1783#
1784# intercept Same as http_port intercept. The FTP origin address is
1785# determined based on the intended destination of the
1786# intercepted connection.
1787#
1788# tproxy Support Linux TPROXY for spoofing outgoing
1789# connections using the client IP address.
1790# NP: disables authentication and maybe IPv6 on the port.
1791#
1792# By default (i.e., without an explicit mode option), Squid extracts the
1793# FTP origin address from the login@origin parameter of the FTP USER
1794# command. Many popular FTP clients support such native FTP proxying.
1795#
1796# Options:
1797#
1798# name=token Specifies an internal name for the port. Defaults to
1799# the port address. Usable with myportname ACL.
1800#
1801# ftp-track-dirs
1802# Enables tracking of FTP directories by injecting extra
1803# PWD commands and adjusting Request-URI (in wrapping
1804# HTTP requests) to reflect the current FTP server
1805# directory. Tracking is disabled by default.
1806#
1807# protocol=FTP Protocol to reconstruct accelerated and intercepted
1808# requests with. Defaults to FTP. No other accepted
1809# values have been tested with. An unsupported value
1810# results in a FATAL error. Accepted values are FTP,
1811# HTTP (or HTTP/1.1), and HTTPS (or HTTPS/1.1).
1812#
1813# Other http_port modes and options that are not specific to HTTP and
1814# HTTPS may also work.
1815#Default:
1816# none
1817
1818# TAG: tcp_outgoing_tos
1819# Allows you to select a TOS/Diffserv value for packets outgoing
1820# on the server side, based on an ACL.
1821#
1822# tcp_outgoing_tos ds-field [!]aclname ...
1823#
1824# Example where normal_service_net uses the TOS value 0x00
1825# and good_service_net uses 0x20
1826#
1827# acl normal_service_net src 10.0.0.0/24
1828# acl good_service_net src 10.0.1.0/24
1829# tcp_outgoing_tos 0x00 normal_service_net
1830# tcp_outgoing_tos 0x20 good_service_net
1831#
1832# TOS/DSCP values really only have local significance - so you should
1833# know what you're specifying. For more information, see RFC2474,
1834# RFC2475, and RFC3260.
1835#
1836# The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
1837# "default" to use whatever default your host has.
1838# Note that only multiples of 4 are usable as the two rightmost bits have
1839# been redefined for use by ECN (RFC 3168 section 23.1).
1840# The squid parser will enforce this by masking away the ECN bits.
1841#
1842# Processing proceeds in the order specified, and stops at first fully
1843# matching line.
1844#
1845# Only fast ACLs are supported.
1846#Default:
1847# none
1848
1849# TAG: clientside_tos
1850# Allows you to select a TOS/DSCP value for packets being transmitted
1851# on the client-side, based on an ACL.
1852#
1853# clientside_tos ds-field [!]aclname ...
1854#
1855# Example where normal_service_net uses the TOS value 0x00
1856# and good_service_net uses 0x20
1857#
1858# acl normal_service_net src 10.0.0.0/24
1859# acl good_service_net src 10.0.1.0/24
1860# clientside_tos 0x00 normal_service_net
1861# clientside_tos 0x20 good_service_net
1862#
1863# Note: This feature is incompatible with qos_flows. Any TOS values set here
1864# will be overwritten by TOS values in qos_flows.
1865#
1866# The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
1867# "default" to use whatever default your host has.
1868# Note that only multiples of 4 are usable as the two rightmost bits have
1869# been redefined for use by ECN (RFC 3168 section 23.1).
1870# The squid parser will enforce this by masking away the ECN bits.
1871#
1872#Default:
1873# none
1874
1875# TAG: tcp_outgoing_mark
1876# Note: This option is only available if Squid is rebuilt with the
1877# Packet MARK (Linux)
1878#
1879# Allows you to apply a Netfilter mark value to outgoing packets
1880# on the server side, based on an ACL.
1881#
1882# tcp_outgoing_mark mark-value [!]aclname ...
1883#
1884# Example where normal_service_net uses the mark value 0x00
1885# and good_service_net uses 0x20
1886#
1887# acl normal_service_net src 10.0.0.0/24
1888# acl good_service_net src 10.0.1.0/24
1889# tcp_outgoing_mark 0x00 normal_service_net
1890# tcp_outgoing_mark 0x20 good_service_net
1891#
1892# Only fast ACLs are supported.
1893#Default:
1894# none
1895
1896# TAG: clientside_mark
1897# Note: This option is only available if Squid is rebuilt with the
1898# Packet MARK (Linux)
1899#
1900# Allows you to apply a Netfilter mark value to packets being transmitted
1901# on the client-side, based on an ACL.
1902#
1903# clientside_mark mark-value [!]aclname ...
1904#
1905# Example where normal_service_net uses the mark value 0x00
1906# and good_service_net uses 0x20
1907#
1908# acl normal_service_net src 10.0.0.0/24
1909# acl good_service_net src 10.0.1.0/24
1910# clientside_mark 0x00 normal_service_net
1911# clientside_mark 0x20 good_service_net
1912#
1913# Note: This feature is incompatible with qos_flows. Any mark values set here
1914# will be overwritten by mark values in qos_flows.
1915#Default:
1916# none
1917
1918# TAG: qos_flows
1919# Allows you to select a TOS/DSCP value to mark outgoing
1920# connections to the client, based on where the reply was sourced.
1921# For platforms using netfilter, allows you to set a netfilter mark
1922# value instead of, or in addition to, a TOS value.
1923#
1924# By default this functionality is disabled. To enable it with the default
1925# settings simply use "qos_flows mark" or "qos_flows tos". Default
1926# settings will result in the netfilter mark or TOS value being copied
1927# from the upstream connection to the client. Note that it is the connection
1928# CONNMARK value not the packet MARK value that is copied.
1929#
1930# It is not currently possible to copy the mark or TOS value from the
1931# client to the upstream connection request.
1932#
1933# TOS values really only have local significance - so you should
1934# know what you're specifying. For more information, see RFC2474,
1935# RFC2475, and RFC3260.
1936#
1937# The TOS/DSCP byte must be exactly that - a octet value 0 - 255.
1938# Note that only multiples of 4 are usable as the two rightmost bits have
1939# been redefined for use by ECN (RFC 3168 section 23.1).
1940# The squid parser will enforce this by masking away the ECN bits.
1941#
1942# Mark values can be any unsigned 32-bit integer value.
1943#
1944# This setting is configured by setting the following values:
1945#
1946# tos|mark Whether to set TOS or netfilter mark values
1947#
1948# local-hit=0xFF Value to mark local cache hits.
1949#
1950# sibling-hit=0xFF Value to mark hits from sibling peers.
1951#
1952# parent-hit=0xFF Value to mark hits from parent peers.
1953#
1954# miss=0xFF[/mask] Value to mark cache misses. Takes precedence
1955# over the preserve-miss feature (see below), unless
1956# mask is specified, in which case only the bits
1957# specified in the mask are written.
1958#
1959# The TOS variant of the following features are only possible on Linux
1960# and require your kernel to be patched with the TOS preserving ZPH
1961# patch, available from http://zph.bratcheda.org
1962# No patch is needed to preserve the netfilter mark, which will work
1963# with all variants of netfilter.
1964#
1965# disable-preserve-miss
1966# This option disables the preservation of the TOS or netfilter
1967# mark. By default, the existing TOS or netfilter mark value of
1968# the response coming from the remote server will be retained
1969# and masked with miss-mark.
1970# NOTE: in the case of a netfilter mark, the mark must be set on
1971# the connection (using the CONNMARK target) not on the packet
1972# (MARK target).
1973#
1974# miss-mask=0xFF
1975# Allows you to mask certain bits in the TOS or mark value
1976# received from the remote server, before copying the value to
1977# the TOS sent towards clients.
1978# Default for tos: 0xFF (TOS from server is not changed).
1979# Default for mark: 0xFFFFFFFF (mark from server is not changed).
1980#
1981# All of these features require the --enable-zph-qos compilation flag
1982# (enabled by default). Netfilter marking also requires the
1983# libnetfilter_conntrack libraries (--with-netfilter-conntrack) and
1984# libcap 2.09+ (--with-libcap).
1985#
1986#Default:
1987# none
1988
1989# TAG: tcp_outgoing_address
1990# Allows you to map requests to different outgoing IP addresses
1991# based on the username or source address of the user making
1992# the request.
1993#
1994# tcp_outgoing_address ipaddr [[!]aclname] ...
1995#
1996# For example;
1997# Forwarding clients with dedicated IPs for certain subnets.
1998#
1999# acl normal_service_net src 10.0.0.0/24
2000# acl good_service_net src 10.0.2.0/24
2001#
2002# tcp_outgoing_address 2001:db8::c001 good_service_net
2003# tcp_outgoing_address 10.1.0.2 good_service_net
2004#
2005# tcp_outgoing_address 2001:db8::beef normal_service_net
2006# tcp_outgoing_address 10.1.0.1 normal_service_net
2007#
2008# tcp_outgoing_address 2001:db8::1
2009# tcp_outgoing_address 10.1.0.3
2010#
2011# Processing proceeds in the order specified, and stops at first fully
2012# matching line.
2013#
2014# Squid will add an implicit IP version test to each line.
2015# Requests going to IPv4 websites will use the outgoing 10.1.0.* addresses.
2016# Requests going to IPv6 websites will use the outgoing 2001:db8:* addresses.
2017#
2018#
2019# NOTE: The use of this directive using client dependent ACLs is
2020# incompatible with the use of server side persistent connections. To
2021# ensure correct results it is best to set server_persistent_connections
2022# to off when using this directive in such configurations.
2023#
2024# NOTE: The use of this directive to set a local IP on outgoing TCP links
2025# is incompatible with using TPROXY to set client IP out outbound TCP links.
2026# When needing to contact peers use the no-tproxy cache_peer option and the
2027# client_dst_passthru directive re-enable normal forwarding such as this.
2028#
2029#Default:
2030# Address selection is performed by the operating system.
2031
2032# TAG: host_verify_strict
2033# Regardless of this option setting, when dealing with intercepted
2034# traffic, Squid always verifies that the destination IP address matches
2035# the Host header domain or IP (called 'authority form URL').
2036#
2037# This enforcement is performed to satisfy a MUST-level requirement in
2038# RFC 2616 section 14.23: "The Host field value MUST represent the naming
2039# authority of the origin server or gateway given by the original URL".
2040#
2041# When set to ON:
2042# Squid always responds with an HTTP 409 (Conflict) error
2043# page and logs a security warning if there is no match.
2044#
2045# Squid verifies that the destination IP address matches
2046# the Host header for forward-proxy and reverse-proxy traffic
2047# as well. For those traffic types, Squid also enables the
2048# following checks, comparing the corresponding Host header
2049# and Request-URI components:
2050#
2051# * The host names (domain or IP) must be identical,
2052# but valueless or missing Host header disables all checks.
2053# For the two host names to match, both must be either IP
2054# or FQDN.
2055#
2056# * Port numbers must be identical, but if a port is missing
2057# the scheme-default port is assumed.
2058#
2059#
2060# When set to OFF (the default):
2061# Squid allows suspicious requests to continue but logs a
2062# security warning and blocks caching of the response.
2063#
2064# * Forward-proxy traffic is not checked at all.
2065#
2066# * Reverse-proxy traffic is not checked at all.
2067#
2068# * Intercepted traffic which passes verification is handled
2069# according to client_dst_passthru.
2070#
2071# * Intercepted requests which fail verification are sent
2072# to the client original destination instead of DIRECT.
2073# This overrides 'client_dst_passthru off'.
2074#
2075# For now suspicious intercepted CONNECT requests are always
2076# responded to with an HTTP 409 (Conflict) error page.
2077#
2078#
2079# SECURITY NOTE:
2080#
2081# As described in CVE-2009-0801 when the Host: header alone is used
2082# to determine the destination of a request it becomes trivial for
2083# malicious scripts on remote websites to bypass browser same-origin
2084# security policy and sandboxing protections.
2085#
2086# The cause of this is that such applets are allowed to perform their
2087# own HTTP stack, in which case the same-origin policy of the browser
2088# sandbox only verifies that the applet tries to contact the same IP
2089# as from where it was loaded at the IP level. The Host: header may
2090# be different from the connected IP and approved origin.
2091#
2092#Default:
2093# host_verify_strict off
2094
2095# TAG: client_dst_passthru
2096# With NAT or TPROXY intercepted traffic Squid may pass the request
2097# directly to the original client destination IP or seek a faster
2098# source using the HTTP Host header.
2099#
2100# Using Host to locate alternative servers can provide faster
2101# connectivity with a range of failure recovery options.
2102# But can also lead to connectivity trouble when the client and
2103# server are attempting stateful interactions unaware of the proxy.
2104#
2105# This option (on by default) prevents alternative DNS entries being
2106# located to send intercepted traffic DIRECT to an origin server.
2107# The clients original destination IP and port will be used instead.
2108#
2109# Regardless of this option setting, when dealing with intercepted
2110# traffic Squid will verify the Host: header and any traffic which
2111# fails Host verification will be treated as if this option were ON.
2112#
2113# see host_verify_strict for details on the verification process.
2114#Default:
2115# client_dst_passthru on
2116
2117# SSL OPTIONS
2118# -----------------------------------------------------------------------------
2119
2120# TAG: ssl_unclean_shutdown
2121# Note: This option is only available if Squid is rebuilt with the
2122# --with-openssl
2123#
2124# Some browsers (especially MSIE) bugs out on SSL shutdown
2125# messages.
2126#Default:
2127# ssl_unclean_shutdown off
2128
2129# TAG: ssl_engine
2130# Note: This option is only available if Squid is rebuilt with the
2131# --with-openssl
2132#
2133# The OpenSSL engine to use. You will need to set this if you
2134# would like to use hardware SSL acceleration for example.
2135#Default:
2136# none
2137
2138# TAG: sslproxy_client_certificate
2139# Note: This option is only available if Squid is rebuilt with the
2140# --with-openssl
2141#
2142# Client SSL Certificate to use when proxying https:// URLs
2143#Default:
2144# none
2145
2146# TAG: sslproxy_client_key
2147# Note: This option is only available if Squid is rebuilt with the
2148# --with-openssl
2149#
2150# Client SSL Key to use when proxying https:// URLs
2151#Default:
2152# none
2153
2154# TAG: sslproxy_version
2155# Note: This option is only available if Squid is rebuilt with the
2156# --with-openssl
2157#
2158# SSL version level to use when proxying https:// URLs
2159#
2160# The versions of SSL/TLS supported:
2161#
2162# 1 automatic (default)
2163# 2 SSLv2 only
2164# 3 SSLv3 only
2165# 4 TLSv1.0 only
2166# 5 TLSv1.1 only
2167# 6 TLSv1.2 only
2168#Default:
2169# automatic SSL/TLS version negotiation
2170
2171# TAG: sslproxy_options
2172# Note: This option is only available if Squid is rebuilt with the
2173# --with-openssl
2174#
2175# Colon (:) or comma (,) separated list of SSL implementation options
2176# to use when proxying https:// URLs
2177#
2178# The most important being:
2179#
2180# NO_SSLv2 Disallow the use of SSLv2
2181# NO_SSLv3 Disallow the use of SSLv3
2182# NO_TLSv1 Disallow the use of TLSv1.0
2183# NO_TLSv1_1 Disallow the use of TLSv1.1
2184# NO_TLSv1_2 Disallow the use of TLSv1.2
2185#
2186# SINGLE_DH_USE
2187# Always create a new key when using temporary/ephemeral
2188# DH key exchanges
2189#
2190# NO_TICKET
2191# Disable use of RFC5077 session tickets. Some servers
2192# may have problems understanding the TLS extension due
2193# to ambiguous specification in RFC4507.
2194#
2195# ALL Enable various bug workarounds suggested as "harmless"
2196# by OpenSSL. Be warned that this may reduce SSL/TLS
2197# strength to some attacks.
2198#
2199# See the OpenSSL SSL_CTX_set_options documentation for a
2200# complete list of possible options.
2201#
2202# WARNING: This directive takes a single token. If a space is used
2203# the value(s) after that space are SILENTLY IGNORED.
2204#Default:
2205# none
2206
2207# TAG: sslproxy_cipher
2208# Note: This option is only available if Squid is rebuilt with the
2209# --with-openssl
2210#
2211# SSL cipher list to use when proxying https:// URLs
2212#
2213# Colon separated list of supported ciphers.
2214#Default:
2215# none
2216
2217# TAG: sslproxy_cafile
2218# Note: This option is only available if Squid is rebuilt with the
2219# --with-openssl
2220#
2221# file containing CA certificates to use when verifying server
2222# certificates while proxying https:// URLs
2223#Default:
2224# none
2225
2226# TAG: sslproxy_capath
2227# Note: This option is only available if Squid is rebuilt with the
2228# --with-openssl
2229#
2230# directory containing CA certificates to use when verifying
2231# server certificates while proxying https:// URLs
2232#Default:
2233# none
2234
2235# TAG: sslproxy_session_ttl
2236# Note: This option is only available if Squid is rebuilt with the
2237# --with-openssl
2238#
2239# Sets the timeout value for SSL sessions
2240#Default:
2241# sslproxy_session_ttl 300
2242
2243# TAG: sslproxy_session_cache_size
2244# Note: This option is only available if Squid is rebuilt with the
2245# --with-openssl
2246#
2247# Sets the cache size to use for ssl session
2248#Default:
2249# sslproxy_session_cache_size 2 MB
2250
2251# TAG: sslproxy_foreign_intermediate_certs
2252# Note: This option is only available if Squid is rebuilt with the
2253# --with-openssl
2254#
2255# Many origin servers fail to send their full server certificate
2256# chain for verification, assuming the client already has or can
2257# easily locate any missing intermediate certificates.
2258#
2259# Squid uses the certificates from the specified file to fill in
2260# these missing chains when trying to validate origin server
2261# certificate chains.
2262#
2263# The file is expected to contain zero or more PEM-encoded
2264# intermediate certificates. These certificates are not treated
2265# as trusted root certificates, and any self-signed certificate in
2266# this file will be ignored.
2267#Default:
2268# none
2269
2270# TAG: sslproxy_cert_sign_hash
2271# Note: This option is only available if Squid is rebuilt with the
2272# --with-openssl
2273#
2274# Sets the hashing algorithm to use when signing generated certificates.
2275# Valid algorithm names depend on the OpenSSL library used. The following
2276# names are usually available: sha1, sha256, sha512, and md5. Please see
2277# your OpenSSL library manual for the available hashes. By default, Squids
2278# that support this option use sha256 hashes.
2279#
2280# Squid does not forcefully purge cached certificates that were generated
2281# with an algorithm other than the currently configured one. They remain
2282# in the cache, subject to the regular cache eviction policy, and become
2283# useful if the algorithm changes again.
2284#Default:
2285# none
2286
2287# TAG: ssl_bump
2288# Note: This option is only available if Squid is rebuilt with the
2289# --with-openssl
2290#
2291# This option is consulted when a CONNECT request is received on
2292# an http_port (or a new connection is intercepted at an
2293# https_port), provided that port was configured with an ssl-bump
2294# flag. The subsequent data on the connection is either treated as
2295# HTTPS and decrypted OR tunneled at TCP level without decryption,
2296# depending on the first matching bumping "action".
2297#
2298# ssl_bump <action> [!]acl ...
2299#
2300# The following bumping actions are currently supported:
2301#
2302# splice
2303# Become a TCP tunnel without decrypting proxied traffic.
2304# This is the default action.
2305#
2306# bump
2307# When used on step SslBump1, establishes a secure connection
2308# with the client first, then connect to the server.
2309# When used on step SslBump2 or SslBump3, establishes a secure
2310# connection with the server and, using a mimicked server
2311# certificate, with the client.
2312#
2313# peek
2314# Receive client (step SslBump1) or server (step SslBump2)
2315# certificate while preserving the possibility of splicing the
2316# connection. Peeking at the server certificate (during step 2)
2317# usually precludes bumping of the connection at step 3.
2318#
2319# stare
2320# Receive client (step SslBump1) or server (step SslBump2)
2321# certificate while preserving the possibility of bumping the
2322# connection. Staring at the server certificate (during step 2)
2323# usually precludes splicing of the connection at step 3.
2324#
2325# terminate
2326# Close client and server connections.
2327#
2328# Backward compatibility actions available at step SslBump1:
2329#
2330# client-first
2331# Bump the connection. Establish a secure connection with the
2332# client first, then connect to the server. This old mode does
2333# not allow Squid to mimic server SSL certificate and does not
2334# work with intercepted SSL connections.
2335#
2336# server-first
2337# Bump the connection. Establish a secure connection with the
2338# server first, then establish a secure connection with the
2339# client, using a mimicked server certificate. Works with both
2340# CONNECT requests and intercepted SSL connections, but does
2341# not allow to make decisions based on SSL handshake info.
2342#
2343# peek-and-splice
2344# Decide whether to bump or splice the connection based on
2345# client-to-squid and server-to-squid SSL hello messages.
2346# XXX: Remove.
2347#
2348# none
2349# Same as the "splice" action.
2350#
2351# All ssl_bump rules are evaluated at each of the supported bumping
2352# steps. Rules with actions that are impossible at the current step are
2353# ignored. The first matching ssl_bump action wins and is applied at the
2354# end of the current step. If no rules match, the splice action is used.
2355# See the at_step ACL for a list of the supported SslBump steps.
2356#
2357# This clause supports both fast and slow acl types.
2358# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2359#
2360# See also: http_port ssl-bump, https_port ssl-bump, and acl at_step.
2361#
2362#
2363# # Example: Bump all TLS connections except those originating from
2364# # localhost or those going to example.com.
2365#
2366# acl broken_sites ssl::server_name .example.com
2367# ssl_bump splice localhost
2368# ssl_bump splice broken_sites
2369# ssl_bump bump all
2370#Default:
2371# Become a TCP tunnel without decrypting proxied traffic.
2372
2373# TAG: sslproxy_flags
2374# Note: This option is only available if Squid is rebuilt with the
2375# --with-openssl
2376#
2377# Various flags modifying the use of SSL while proxying https:// URLs:
2378# DONT_VERIFY_PEER Accept certificates that fail verification.
2379# For refined control, see sslproxy_cert_error.
2380# NO_DEFAULT_CA Don't use the default CA list built in
2381# to OpenSSL.
2382#Default:
2383# none
2384
2385# TAG: sslproxy_cert_error
2386# Note: This option is only available if Squid is rebuilt with the
2387# --with-openssl
2388#
2389# Use this ACL to bypass server certificate validation errors.
2390#
2391# For example, the following lines will bypass all validation errors
2392# when talking to servers for example.com. All other
2393# validation errors will result in ERR_SECURE_CONNECT_FAIL error.
2394#
2395# acl BrokenButTrustedServers dstdomain example.com
2396# sslproxy_cert_error allow BrokenButTrustedServers
2397# sslproxy_cert_error deny all
2398#
2399# This clause only supports fast acl types.
2400# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2401# Using slow acl types may result in server crashes
2402#
2403# Without this option, all server certificate validation errors
2404# terminate the transaction to protect Squid and the client.
2405#
2406# SQUID_X509_V_ERR_INFINITE_VALIDATION error cannot be bypassed
2407# but should not happen unless your OpenSSL library is buggy.
2408#
2409# SECURITY WARNING:
2410# Bypassing validation errors is dangerous because an
2411# error usually implies that the server cannot be trusted
2412# and the connection may be insecure.
2413#
2414# See also: sslproxy_flags and DONT_VERIFY_PEER.
2415#Default:
2416# Server certificate errors terminate the transaction.
2417
2418# TAG: sslproxy_cert_sign
2419# Note: This option is only available if Squid is rebuilt with the
2420# --with-openssl
2421#
2422#
2423# sslproxy_cert_sign <signing algorithm> acl ...
2424#
2425# The following certificate signing algorithms are supported:
2426#
2427# signTrusted
2428# Sign using the configured CA certificate which is usually
2429# placed in and trusted by end-user browsers. This is the
2430# default for trusted origin server certificates.
2431#
2432# signUntrusted
2433# Sign to guarantee an X509_V_ERR_CERT_UNTRUSTED browser error.
2434# This is the default for untrusted origin server certificates
2435# that are not self-signed (see ssl::certUntrusted).
2436#
2437# signSelf
2438# Sign using a self-signed certificate with the right CN to
2439# generate a X509_V_ERR_DEPTH_ZERO_SELF_SIGNED_CERT error in the
2440# browser. This is the default for self-signed origin server
2441# certificates (see ssl::certSelfSigned).
2442#
2443# This clause only supports fast acl types.
2444#
2445# When sslproxy_cert_sign acl(s) match, Squid uses the corresponding
2446# signing algorithm to generate the certificate and ignores all
2447# subsequent sslproxy_cert_sign options (the first match wins). If no
2448# acl(s) match, the default signing algorithm is determined by errors
2449# detected when obtaining and validating the origin server certificate.
2450#
2451# WARNING: SQUID_X509_V_ERR_DOMAIN_MISMATCH and ssl:certDomainMismatch can
2452# be used with sslproxy_cert_adapt, but if and only if Squid is bumping a
2453# CONNECT request that carries a domain name. In all other cases (CONNECT
2454# to an IP address or an intercepted SSL connection), Squid cannot detect
2455# the domain mismatch at certificate generation time when
2456# bump-server-first is used.
2457#Default:
2458# none
2459
2460# TAG: sslproxy_cert_adapt
2461# Note: This option is only available if Squid is rebuilt with the
2462# --with-openssl
2463#
2464#
2465# sslproxy_cert_adapt <adaptation algorithm> acl ...
2466#
2467# The following certificate adaptation algorithms are supported:
2468#
2469# setValidAfter
2470# Sets the "Not After" property to the "Not After" property of
2471# the CA certificate used to sign generated certificates.
2472#
2473# setValidBefore
2474# Sets the "Not Before" property to the "Not Before" property of
2475# the CA certificate used to sign generated certificates.
2476#
2477# setCommonName or setCommonName{CN}
2478# Sets Subject.CN property to the host name specified as a
2479# CN parameter or, if no explicit CN parameter was specified,
2480# extracted from the CONNECT request. It is a misconfiguration
2481# to use setCommonName without an explicit parameter for
2482# intercepted or tproxied SSL connections.
2483#
2484# This clause only supports fast acl types.
2485#
2486# Squid first groups sslproxy_cert_adapt options by adaptation algorithm.
2487# Within a group, when sslproxy_cert_adapt acl(s) match, Squid uses the
2488# corresponding adaptation algorithm to generate the certificate and
2489# ignores all subsequent sslproxy_cert_adapt options in that algorithm's
2490# group (i.e., the first match wins within each algorithm group). If no
2491# acl(s) match, the default mimicking action takes place.
2492#
2493# WARNING: SQUID_X509_V_ERR_DOMAIN_MISMATCH and ssl:certDomainMismatch can
2494# be used with sslproxy_cert_adapt, but if and only if Squid is bumping a
2495# CONNECT request that carries a domain name. In all other cases (CONNECT
2496# to an IP address or an intercepted SSL connection), Squid cannot detect
2497# the domain mismatch at certificate generation time when
2498# bump-server-first is used.
2499#Default:
2500# none
2501
2502# TAG: sslpassword_program
2503# Note: This option is only available if Squid is rebuilt with the
2504# --with-openssl
2505#
2506# Specify a program used for entering SSL key passphrases
2507# when using encrypted SSL certificate keys. If not specified
2508# keys must either be unencrypted, or Squid started with the -N
2509# option to allow it to query interactively for the passphrase.
2510#
2511# The key file name is given as argument to the program allowing
2512# selection of the right password if you have multiple encrypted
2513# keys.
2514#Default:
2515# none
2516
2517# OPTIONS RELATING TO EXTERNAL SSL_CRTD
2518# -----------------------------------------------------------------------------
2519
2520# TAG: sslcrtd_program
2521# Note: This option is only available if Squid is rebuilt with the
2522# --enable-ssl-crtd
2523#
2524# Specify the location and options of the executable for ssl_crtd process.
2525# /usr/lib/squid/ssl_crtd program requires -s and -M parameters
2526# For more information use:
2527# /usr/lib/squid/ssl_crtd -h
2528#Default:
2529# sslcrtd_program /usr/lib/squid/ssl_crtd -s /var/lib/ssl_db -M 4MB
2530
2531# TAG: sslcrtd_children
2532# Note: This option is only available if Squid is rebuilt with the
2533# --enable-ssl-crtd
2534#
2535# The maximum number of processes spawn to service ssl server.
2536# The maximum this may be safely set to is 32.
2537#
2538# The startup= and idle= options allow some measure of skew in your
2539# tuning.
2540#
2541# startup=N
2542#
2543# Sets the minimum number of processes to spawn when Squid
2544# starts or reconfigures. When set to zero the first request will
2545# cause spawning of the first child process to handle it.
2546#
2547# Starting too few children temporary slows Squid under load while it
2548# tries to spawn enough additional processes to cope with traffic.
2549#
2550# idle=N
2551#
2552# Sets a minimum of how many processes Squid is to try and keep available
2553# at all times. When traffic begins to rise above what the existing
2554# processes can handle this many more will be spawned up to the maximum
2555# configured. A minimum setting of 1 is required.
2556#
2557# You must have at least one ssl_crtd process.
2558#Default:
2559# sslcrtd_children 32 startup=5 idle=1
2560
2561# TAG: sslcrtvalidator_program
2562# Note: This option is only available if Squid is rebuilt with the
2563# --with-openssl
2564#
2565# Specify the location and options of the executable for ssl_crt_validator
2566# process.
2567#
2568# Usage: sslcrtvalidator_program [ttl=n] [cache=n] path ...
2569#
2570# Options:
2571# ttl=n TTL in seconds for cached results. The default is 60 secs
2572# cache=n limit the result cache size. The default value is 2048
2573#Default:
2574# none
2575
2576# TAG: sslcrtvalidator_children
2577# Note: This option is only available if Squid is rebuilt with the
2578# --with-openssl
2579#
2580# The maximum number of processes spawn to service SSL server.
2581# The maximum this may be safely set to is 32.
2582#
2583# The startup= and idle= options allow some measure of skew in your
2584# tuning.
2585#
2586# startup=N
2587#
2588# Sets the minimum number of processes to spawn when Squid
2589# starts or reconfigures. When set to zero the first request will
2590# cause spawning of the first child process to handle it.
2591#
2592# Starting too few children temporary slows Squid under load while it
2593# tries to spawn enough additional processes to cope with traffic.
2594#
2595# idle=N
2596#
2597# Sets a minimum of how many processes Squid is to try and keep available
2598# at all times. When traffic begins to rise above what the existing
2599# processes can handle this many more will be spawned up to the maximum
2600# configured. A minimum setting of 1 is required.
2601#
2602# concurrency=
2603#
2604# The number of requests each certificate validator helper can handle in
2605# parallel. A value of 0 indicates the certficate validator does not
2606# support concurrency. Defaults to 1.
2607#
2608# When this directive is set to a value >= 1 then the protocol
2609# used to communicate with the helper is modified to include
2610# a request ID in front of the request/response. The request
2611# ID from the request must be echoed back with the response
2612# to that request.
2613#
2614# You must have at least one ssl_crt_validator process.
2615#Default:
2616# sslcrtvalidator_children 32 startup=5 idle=1 concurrency=1
2617
2618# OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
2619# -----------------------------------------------------------------------------
2620
2621# TAG: cache_peer
2622# To specify other caches in a hierarchy, use the format:
2623#
2624# cache_peer hostname type http-port icp-port [options]
2625#
2626# For example,
2627#
2628# # proxy icp
2629# # hostname type port port options
2630# # -------------------- -------- ----- ----- -----------
2631# cache_peer parent.foo.net parent 3128 3130 default
2632# cache_peer sib1.foo.net sibling 3128 3130 proxy-only
2633# cache_peer sib2.foo.net sibling 3128 3130 proxy-only
2634# cache_peer example.com parent 80 0 default
2635# cache_peer cdn.example.com sibling 3128 0
2636#
2637# type: either 'parent', 'sibling', or 'multicast'.
2638#
2639# proxy-port: The port number where the peer accept HTTP requests.
2640# For other Squid proxies this is usually 3128
2641# For web servers this is usually 80
2642#
2643# icp-port: Used for querying neighbor caches about objects.
2644# Set to 0 if the peer does not support ICP or HTCP.
2645# See ICP and HTCP options below for additional details.
2646#
2647#
2648# ==== ICP OPTIONS ====
2649#
2650# You MUST also set icp_port and icp_access explicitly when using these options.
2651# The defaults will prevent peer traffic using ICP.
2652#
2653#
2654# no-query Disable ICP queries to this neighbor.
2655#
2656# multicast-responder
2657# Indicates the named peer is a member of a multicast group.
2658# ICP queries will not be sent directly to the peer, but ICP
2659# replies will be accepted from it.
2660#
2661# closest-only Indicates that, for ICP_OP_MISS replies, we'll only forward
2662# CLOSEST_PARENT_MISSes and never FIRST_PARENT_MISSes.
2663#
2664# background-ping
2665# To only send ICP queries to this neighbor infrequently.
2666# This is used to keep the neighbor round trip time updated
2667# and is usually used in conjunction with weighted-round-robin.
2668#
2669#
2670# ==== HTCP OPTIONS ====
2671#
2672# You MUST also set htcp_port and htcp_access explicitly when using these options.
2673# The defaults will prevent peer traffic using HTCP.
2674#
2675#
2676# htcp Send HTCP, instead of ICP, queries to the neighbor.
2677# You probably also want to set the "icp-port" to 4827
2678# instead of 3130. This directive accepts a comma separated
2679# list of options described below.
2680#
2681# htcp=oldsquid Send HTCP to old Squid versions (2.5 or earlier).
2682#
2683# htcp=no-clr Send HTCP to the neighbor but without
2684# sending any CLR requests. This cannot be used with
2685# only-clr.
2686#
2687# htcp=only-clr Send HTCP to the neighbor but ONLY CLR requests.
2688# This cannot be used with no-clr.
2689#
2690# htcp=no-purge-clr
2691# Send HTCP to the neighbor including CLRs but only when
2692# they do not result from PURGE requests.
2693#
2694# htcp=forward-clr
2695# Forward any HTCP CLR requests this proxy receives to the peer.
2696#
2697#
2698# ==== PEER SELECTION METHODS ====
2699#
2700# The default peer selection method is ICP, with the first responding peer
2701# being used as source. These options can be used for better load balancing.
2702#
2703#
2704# default This is a parent cache which can be used as a "last-resort"
2705# if a peer cannot be located by any of the peer-selection methods.
2706# If specified more than once, only the first is used.
2707#
2708# round-robin Load-Balance parents which should be used in a round-robin
2709# fashion in the absence of any ICP queries.
2710# weight=N can be used to add bias.
2711#
2712# weighted-round-robin
2713# Load-Balance parents which should be used in a round-robin
2714# fashion with the frequency of each parent being based on the
2715# round trip time. Closer parents are used more often.
2716# Usually used for background-ping parents.
2717# weight=N can be used to add bias.
2718#
2719# carp Load-Balance parents which should be used as a CARP array.
2720# The requests will be distributed among the parents based on the
2721# CARP load balancing hash function based on their weight.
2722#
2723# userhash Load-balance parents based on the client proxy_auth or ident username.
2724#
2725# sourcehash Load-balance parents based on the client source IP.
2726#
2727# multicast-siblings
2728# To be used only for cache peers of type "multicast".
2729# ALL members of this multicast group have "sibling"
2730# relationship with it, not "parent". This is to a multicast
2731# group when the requested object would be fetched only from
2732# a "parent" cache, anyway. It's useful, e.g., when
2733# configuring a pool of redundant Squid proxies, being
2734# members of the same multicast group.
2735#
2736#
2737# ==== PEER SELECTION OPTIONS ====
2738#
2739# weight=N use to affect the selection of a peer during any weighted
2740# peer-selection mechanisms.
2741# The weight must be an integer; default is 1,
2742# larger weights are favored more.
2743# This option does not affect parent selection if a peering
2744# protocol is not in use.
2745#
2746# basetime=N Specify a base amount to be subtracted from round trip
2747# times of parents.
2748# It is subtracted before division by weight in calculating
2749# which parent to fectch from. If the rtt is less than the
2750# base time the rtt is set to a minimal value.
2751#
2752# ttl=N Specify a TTL to use when sending multicast ICP queries
2753# to this address.
2754# Only useful when sending to a multicast group.
2755# Because we don't accept ICP replies from random
2756# hosts, you must configure other group members as
2757# peers with the 'multicast-responder' option.
2758#
2759# no-delay To prevent access to this neighbor from influencing the
2760# delay pools.
2761#
2762# digest-url=URL Tell Squid to fetch the cache digest (if digests are
2763# enabled) for this host from the specified URL rather
2764# than the Squid default location.
2765#
2766#
2767# ==== CARP OPTIONS ====
2768#
2769# carp-key=key-specification
2770# use a different key than the full URL to hash against the peer.
2771# the key-specification is a comma-separated list of the keywords
2772# scheme, host, port, path, params
2773# Order is not important.
2774#
2775# ==== ACCELERATOR / REVERSE-PROXY OPTIONS ====
2776#
2777# originserver Causes this parent to be contacted as an origin server.
2778# Meant to be used in accelerator setups when the peer
2779# is a web server.
2780#
2781# forceddomain=name
2782# Set the Host header of requests forwarded to this peer.
2783# Useful in accelerator setups where the server (peer)
2784# expects a certain domain name but clients may request
2785# others. ie example.com or www.example.com
2786#
2787# no-digest Disable request of cache digests.
2788#
2789# no-netdb-exchange
2790# Disables requesting ICMP RTT database (NetDB).
2791#
2792#
2793# ==== AUTHENTICATION OPTIONS ====
2794#
2795# login=user:password
2796# If this is a personal/workgroup proxy and your parent
2797# requires proxy authentication.
2798#
2799# Note: The string can include URL escapes (i.e. %20 for
2800# spaces). This also means % must be written as %%.
2801#
2802# login=PASSTHRU
2803# Send login details received from client to this peer.
2804# Both Proxy- and WWW-Authorization headers are passed
2805# without alteration to the peer.
2806# Authentication is not required by Squid for this to work.
2807#
2808# Note: This will pass any form of authentication but
2809# only Basic auth will work through a proxy unless the
2810# connection-auth options are also used.
2811#
2812# login=PASS Send login details received from client to this peer.
2813# Authentication is not required by this option.
2814#
2815# If there are no client-provided authentication headers
2816# to pass on, but username and password are available
2817# from an external ACL user= and password= result tags
2818# they may be sent instead.
2819#
2820# Note: To combine this with proxy_auth both proxies must
2821# share the same user database as HTTP only allows for
2822# a single login (one for proxy, one for origin server).
2823# Also be warned this will expose your users proxy
2824# password to the peer. USE WITH CAUTION
2825#
2826# login=*:password
2827# Send the username to the upstream cache, but with a
2828# fixed password. This is meant to be used when the peer
2829# is in another administrative domain, but it is still
2830# needed to identify each user.
2831# The star can optionally be followed by some extra
2832# information which is added to the username. This can
2833# be used to identify this proxy to the peer, similar to
2834# the login=username:password option above.
2835#
2836# login=NEGOTIATE
2837# If this is a personal/workgroup proxy and your parent
2838# requires a secure proxy authentication.
2839# The first principal from the default keytab or defined by
2840# the environment variable KRB5_KTNAME will be used.
2841#
2842# WARNING: The connection may transmit requests from multiple
2843# clients. Negotiate often assumes end-to-end authentication
2844# and a single-client. Which is not strictly true here.
2845#
2846# login=NEGOTIATE:principal_name
2847# If this is a personal/workgroup proxy and your parent
2848# requires a secure proxy authentication.
2849# The principal principal_name from the default keytab or
2850# defined by the environment variable KRB5_KTNAME will be
2851# used.
2852#
2853# WARNING: The connection may transmit requests from multiple
2854# clients. Negotiate often assumes end-to-end authentication
2855# and a single-client. Which is not strictly true here.
2856#
2857# connection-auth=on|off
2858# Tell Squid that this peer does or not support Microsoft
2859# connection oriented authentication, and any such
2860# challenges received from there should be ignored.
2861# Default is auto to automatically determine the status
2862# of the peer.
2863#
2864#
2865# ==== SSL / HTTPS / TLS OPTIONS ====
2866#
2867# ssl Encrypt connections to this peer with SSL/TLS.
2868#
2869# sslcert=/path/to/ssl/certificate
2870# A client SSL certificate to use when connecting to
2871# this peer.
2872#
2873# sslkey=/path/to/ssl/key
2874# The private SSL key corresponding to sslcert above.
2875# If 'sslkey' is not specified 'sslcert' is assumed to
2876# reference a combined file containing both the
2877# certificate and the key.
2878#
2879# Notes:
2880#
2881# On Debian/Ubuntu systems a default snakeoil certificate is
2882# available in /etc/ssl and users can set:
2883#
2884# cert=/etc/ssl/certs/ssl-cert-snakeoil.pem
2885#
2886# and
2887#
2888# key=/etc/ssl/private/ssl-cert-snakeoil.key
2889#
2890# for testing.
2891#
2892# sslversion=1|2|3|4|5|6
2893# The SSL version to use when connecting to this peer
2894# 1 = automatic (default)
2895# 2 = SSL v2 only
2896# 3 = SSL v3 only
2897# 4 = TLS v1.0 only
2898# 5 = TLS v1.1 only
2899# 6 = TLS v1.2 only
2900#
2901# sslcipher=... The list of valid SSL ciphers to use when connecting
2902# to this peer.
2903#
2904# ssloptions=... Specify various SSL implementation options:
2905#
2906# NO_SSLv2 Disallow the use of SSLv2
2907# NO_SSLv3 Disallow the use of SSLv3
2908# NO_TLSv1 Disallow the use of TLSv1.0
2909# NO_TLSv1_1 Disallow the use of TLSv1.1
2910# NO_TLSv1_2 Disallow the use of TLSv1.2
2911#
2912# SINGLE_DH_USE
2913# Always create a new key when using
2914# temporary/ephemeral DH key exchanges
2915#
2916# NO_TICKET
2917# Disable use of RFC5077 session tickets. Some servers
2918# may have problems understanding the TLS extension due
2919# to ambiguous specification in RFC4507.
2920#
2921# ALL Enable various bug workarounds
2922# suggested as "harmless" by OpenSSL
2923# Be warned that this reduces SSL/TLS
2924# strength to some attacks.
2925#
2926# See the OpenSSL SSL_CTX_set_options documentation for a
2927# more complete list.
2928#
2929# sslcafile=... A file containing additional CA certificates to use
2930# when verifying the peer certificate.
2931#
2932# sslcapath=... A directory containing additional CA certificates to
2933# use when verifying the peer certificate.
2934#
2935# sslcrlfile=... A certificate revocation list file to use when
2936# verifying the peer certificate.
2937#
2938# sslflags=... Specify various flags modifying the SSL implementation:
2939#
2940# DONT_VERIFY_PEER
2941# Accept certificates even if they fail to
2942# verify.
2943# NO_DEFAULT_CA
2944# Don't use the default CA list built in
2945# to OpenSSL.
2946# DONT_VERIFY_DOMAIN
2947# Don't verify the peer certificate
2948# matches the server name
2949#
2950# ssldomain= The peer name as advertised in it's certificate.
2951# Used for verifying the correctness of the received peer
2952# certificate. If not specified the peer hostname will be
2953# used.
2954#
2955# front-end-https
2956# Enable the "Front-End-Https: On" header needed when
2957# using Squid as a SSL frontend in front of Microsoft OWA.
2958# See MS KB document Q307347 for details on this header.
2959# If set to auto the header will only be added if the
2960# request is forwarded as a https:// URL.
2961#
2962#
2963# ==== GENERAL OPTIONS ====
2964#
2965# connect-timeout=N
2966# A peer-specific connect timeout.
2967# Also see the peer_connect_timeout directive.
2968#
2969# connect-fail-limit=N
2970# How many times connecting to a peer must fail before
2971# it is marked as down. Standby connection failures
2972# count towards this limit. Default is 10.
2973#
2974# allow-miss Disable Squid's use of only-if-cached when forwarding
2975# requests to siblings. This is primarily useful when
2976# icp_hit_stale is used by the sibling. Excessive use
2977# of this option may result in forwarding loops. One way
2978# to prevent peering loops when using this option, is to
2979# deny cache peer usage on requests from a peer:
2980# acl fromPeer ...
2981# cache_peer_access peerName deny fromPeer
2982#
2983# max-conn=N Limit the number of concurrent connections the Squid
2984# may open to this peer, including already opened idle
2985# and standby connections. There is no peer-specific
2986# connection limit by default.
2987#
2988# A peer exceeding the limit is not used for new
2989# requests unless a standby connection is available.
2990#
2991# max-conn currently works poorly with idle persistent
2992# connections: When a peer reaches its max-conn limit,
2993# and there are idle persistent connections to the peer,
2994# the peer may not be selected because the limiting code
2995# does not know whether Squid can reuse those idle
2996# connections.
2997#
2998# standby=N Maintain a pool of N "hot standby" connections to an
2999# UP peer, available for requests when no idle
3000# persistent connection is available (or safe) to use.
3001# By default and with zero N, no such pool is maintained.
3002# N must not exceed the max-conn limit (if any).
3003#
3004# At start or after reconfiguration, Squid opens new TCP
3005# standby connections until there are N connections
3006# available and then replenishes the standby pool as
3007# opened connections are used up for requests. A used
3008# connection never goes back to the standby pool, but
3009# may go to the regular idle persistent connection pool
3010# shared by all peers and origin servers.
3011#
3012# Squid never opens multiple new standby connections
3013# concurrently. This one-at-a-time approach minimizes
3014# flooding-like effect on peers. Furthermore, just a few
3015# standby connections should be sufficient in most cases
3016# to supply most new requests with a ready-to-use
3017# connection.
3018#
3019# Standby connections obey server_idle_pconn_timeout.
3020# For the feature to work as intended, the peer must be
3021# configured to accept and keep them open longer than
3022# the idle timeout at the connecting Squid, to minimize
3023# race conditions typical to idle used persistent
3024# connections. Default request_timeout and
3025# server_idle_pconn_timeout values ensure such a
3026# configuration.
3027#
3028# name=xxx Unique name for the peer.
3029# Required if you have multiple peers on the same host
3030# but different ports.
3031# This name can be used in cache_peer_access and similar
3032# directives to identify the peer.
3033# Can be used by outgoing access controls through the
3034# peername ACL type.
3035#
3036# no-tproxy Do not use the client-spoof TPROXY support when forwarding
3037# requests to this peer. Use normal address selection instead.
3038# This overrides the spoof_client_ip ACL.
3039#
3040# proxy-only objects fetched from the peer will not be stored locally.
3041#
3042#Default:
3043# none
3044
3045# TAG: cache_peer_domain
3046# Use to limit the domains for which a neighbor cache will be
3047# queried.
3048#
3049# Usage:
3050# cache_peer_domain cache-host domain [domain ...]
3051# cache_peer_domain cache-host !domain
3052#
3053# For example, specifying
3054#
3055# cache_peer_domain parent.foo.net .edu
3056#
3057# has the effect such that UDP query packets are sent to
3058# 'bigserver' only when the requested object exists on a
3059# server in the .edu domain. Prefixing the domainname
3060# with '!' means the cache will be queried for objects
3061# NOT in that domain.
3062#
3063# NOTE: * Any number of domains may be given for a cache-host,
3064# either on the same or separate lines.
3065# * When multiple domains are given for a particular
3066# cache-host, the first matched domain is applied.
3067# * Cache hosts with no domain restrictions are queried
3068# for all requests.
3069# * There are no defaults.
3070# * There is also a 'cache_peer_access' tag in the ACL
3071# section.
3072#Default:
3073# none
3074
3075# TAG: cache_peer_access
3076# Restricts usage of cache_peer proxies.
3077#
3078# Usage:
3079# cache_peer_access peer-name allow|deny [!]aclname ...
3080#
3081# For the required peer-name parameter, use either the value of the
3082# cache_peer name=value parameter or, if name=value is missing, the
3083# cache_peer hostname parameter.
3084#
3085# This directive narrows down the selection of peering candidates, but
3086# does not determine the order in which the selected candidates are
3087# contacted. That order is determined by the peer selection algorithms
3088# (see PEER SELECTION sections in the cache_peer documentation).
3089#
3090# If a deny rule matches, the corresponding peer will not be contacted
3091# for the current transaction -- Squid will not send ICP queries and
3092# will not forward HTTP requests to that peer. An allow match leaves
3093# the corresponding peer in the selection. The first match for a given
3094# peer wins for that peer.
3095#
3096# The relative order of cache_peer_access directives for the same peer
3097# matters. The relative order of any two cache_peer_access directives
3098# for different peers does not matter. To ease interpretation, it is a
3099# good idea to group cache_peer_access directives for the same peer
3100# together.
3101#
3102# A single cache_peer_access directive may be evaluated multiple times
3103# for a given transaction because individual peer selection algorithms
3104# may check it independently from each other. These redundant checks
3105# may be optimized away in future Squid versions.
3106#
3107# This clause only supports fast acl types.
3108# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
3109#Default:
3110# No peer usage restrictions.
3111
3112# TAG: neighbor_type_domain
3113# Modify the cache_peer neighbor type when passing requests
3114# about specific domains to the peer.
3115#
3116# Usage:
3117# neighbor_type_domain neighbor parent|sibling domain domain ...
3118#
3119# For example:
3120# cache_peer foo.example.com parent 3128 3130
3121# neighbor_type_domain foo.example.com sibling .au .de
3122#
3123# The above configuration treats all requests to foo.example.com as a
3124# parent proxy unless the request is for a .au or .de ccTLD domain name.
3125#Default:
3126# The peer type from cache_peer directive is used for all requests to that peer.
3127
3128# TAG: dead_peer_timeout (seconds)
3129# This controls how long Squid waits to declare a peer cache
3130# as "dead." If there are no ICP replies received in this
3131# amount of time, Squid will declare the peer dead and not
3132# expect to receive any further ICP replies. However, it
3133# continues to send ICP queries, and will mark the peer as
3134# alive upon receipt of the first subsequent ICP reply.
3135#
3136# This timeout also affects when Squid expects to receive ICP
3137# replies from peers. If more than 'dead_peer' seconds have
3138# passed since the last ICP reply was received, Squid will not
3139# expect to receive an ICP reply on the next query. Thus, if
3140# your time between requests is greater than this timeout, you
3141# will see a lot of requests sent DIRECT to origin servers
3142# instead of to your parents.
3143#Default:
3144# dead_peer_timeout 10 seconds
3145
3146# TAG: forward_max_tries
3147# Controls how many different forward paths Squid will try
3148# before giving up. See also forward_timeout.
3149#
3150# NOTE: connect_retries (default: none) can make each of these
3151# possible forwarding paths be tried multiple times.
3152#Default:
3153# forward_max_tries 25
3154
3155# MEMORY CACHE OPTIONS
3156# -----------------------------------------------------------------------------
3157
3158# TAG: cache_mem (bytes)
3159# NOTE: THIS PARAMETER DOES NOT SPECIFY THE MAXIMUM PROCESS SIZE.
3160# IT ONLY PLACES A LIMIT ON HOW MUCH ADDITIONAL MEMORY SQUID WILL
3161# USE AS A MEMORY CACHE OF OBJECTS. SQUID USES MEMORY FOR OTHER
3162# THINGS AS WELL. SEE THE SQUID FAQ SECTION 8 FOR DETAILS.
3163#
3164# 'cache_mem' specifies the ideal amount of memory to be used
3165# for:
3166# * In-Transit objects
3167# * Hot Objects
3168# * Negative-Cached objects
3169#
3170# Data for these objects are stored in 4 KB blocks. This
3171# parameter specifies the ideal upper limit on the total size of
3172# 4 KB blocks allocated. In-Transit objects take the highest
3173# priority.
3174#
3175# In-transit objects have priority over the others. When
3176# additional space is needed for incoming data, negative-cached
3177# and hot objects will be released. In other words, the
3178# negative-cached and hot objects will fill up any unused space
3179# not needed for in-transit objects.
3180#
3181# If circumstances require, this limit will be exceeded.
3182# Specifically, if your incoming request rate requires more than
3183# 'cache_mem' of memory to hold in-transit objects, Squid will
3184# exceed this limit to satisfy the new requests. When the load
3185# decreases, blocks will be freed until the high-water mark is
3186# reached. Thereafter, blocks will be used to store hot
3187# objects.
3188#
3189# If shared memory caching is enabled, Squid does not use the shared
3190# cache space for in-transit objects, but they still consume as much
3191# local memory as they need. For more details about the shared memory
3192# cache, see memory_cache_shared.
3193#Default:
3194# cache_mem 256 MB
3195
3196# TAG: maximum_object_size_in_memory (bytes)
3197# Objects greater than this size will not be attempted to kept in
3198# the memory cache. This should be set high enough to keep objects
3199# accessed frequently in memory to improve performance whilst low
3200# enough to keep larger objects from hoarding cache_mem.
3201#Default:
3202# maximum_object_size_in_memory 512 KB
3203
3204# TAG: memory_cache_shared on|off
3205# Controls whether the memory cache is shared among SMP workers.
3206#
3207# The shared memory cache is meant to occupy cache_mem bytes and replace
3208# the non-shared memory cache, although some entities may still be
3209# cached locally by workers for now (e.g., internal and in-transit
3210# objects may be served from a local memory cache even if shared memory
3211# caching is enabled).
3212#
3213# By default, the memory cache is shared if and only if all of the
3214# following conditions are satisfied: Squid runs in SMP mode with
3215# multiple workers, cache_mem is positive, and Squid environment
3216# supports required IPC primitives (e.g., POSIX shared memory segments
3217# and GCC-style atomic operations).
3218#
3219# To avoid blocking locks, shared memory uses opportunistic algorithms
3220# that do not guarantee that every cachable entity that could have been
3221# shared among SMP workers will actually be shared.
3222#Default:
3223# "on" where supported if doing memory caching with multiple SMP workers.
3224
3225# TAG: memory_cache_mode
3226# Controls which objects to keep in the memory cache (cache_mem)
3227#
3228# always Keep most recently fetched objects in memory (default)
3229#
3230# disk Only disk cache hits are kept in memory, which means
3231# an object must first be cached on disk and then hit
3232# a second time before cached in memory.
3233#
3234# network Only objects fetched from network is kept in memory
3235#Default:
3236# Keep the most recently fetched objects in memory
3237
3238# TAG: memory_replacement_policy
3239# The memory replacement policy parameter determines which
3240# objects are purged from memory when memory space is needed.
3241#
3242# See cache_replacement_policy for details on algorithms.
3243#Default:
3244# memory_replacement_policy lru
3245
3246# DISK CACHE OPTIONS
3247# -----------------------------------------------------------------------------
3248
3249# TAG: cache_replacement_policy
3250# The cache replacement policy parameter determines which
3251# objects are evicted (replaced) when disk space is needed.
3252#
3253# lru : Squid's original list based LRU policy
3254# heap GDSF : Greedy-Dual Size Frequency
3255# heap LFUDA: Least Frequently Used with Dynamic Aging
3256# heap LRU : LRU policy implemented using a heap
3257#
3258# Applies to any cache_dir lines listed below this directive.
3259#
3260# The LRU policies keeps recently referenced objects.
3261#
3262# The heap GDSF policy optimizes object hit rate by keeping smaller
3263# popular objects in cache so it has a better chance of getting a
3264# hit. It achieves a lower byte hit rate than LFUDA though since
3265# it evicts larger (possibly popular) objects.
3266#
3267# The heap LFUDA policy keeps popular objects in cache regardless of
3268# their size and thus optimizes byte hit rate at the expense of
3269# hit rate since one large, popular object will prevent many
3270# smaller, slightly less popular objects from being cached.
3271#
3272# Both policies utilize a dynamic aging mechanism that prevents
3273# cache pollution that can otherwise occur with frequency-based
3274# replacement policies.
3275#
3276# NOTE: if using the LFUDA replacement policy you should increase
3277# the value of maximum_object_size above its default of 4 MB to
3278# to maximize the potential byte hit rate improvement of LFUDA.
3279#
3280# For more information about the GDSF and LFUDA cache replacement
3281# policies see http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html
3282# and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html.
3283#Default:
3284# cache_replacement_policy lru
3285
3286# TAG: minimum_object_size (bytes)
3287# Objects smaller than this size will NOT be saved on disk. The
3288# value is specified in bytes, and the default is 0 KB, which
3289# means all responses can be stored.
3290#Default:
3291# no limit
3292
3293# TAG: maximum_object_size (bytes)
3294# Set the default value for max-size parameter on any cache_dir.
3295# The value is specified in bytes, and the default is 4 MB.
3296#
3297# If you wish to get a high BYTES hit ratio, you should probably
3298# increase this (one 32 MB object hit counts for 3200 10KB
3299# hits).
3300#
3301# If you wish to increase hit ratio more than you want to
3302# save bandwidth you should leave this low.
3303#
3304# NOTE: if using the LFUDA replacement policy you should increase
3305# this value to maximize the byte hit rate improvement of LFUDA!
3306# See cache_replacement_policy for a discussion of this policy.
3307#Default:
3308# maximum_object_size 4 MB
3309
3310# TAG: cache_dir
3311# Format:
3312# cache_dir Type Directory-Name Fs-specific-data [options]
3313#
3314# You can specify multiple cache_dir lines to spread the
3315# cache among different disk partitions.
3316#
3317# Type specifies the kind of storage system to use. Only "ufs"
3318# is built by default. To enable any of the other storage systems
3319# see the --enable-storeio configure option.
3320#
3321# 'Directory' is a top-level directory where cache swap
3322# files will be stored. If you want to use an entire disk
3323# for caching, this can be the mount-point directory.
3324# The directory must exist and be writable by the Squid
3325# process. Squid will NOT create this directory for you.
3326#
3327# In SMP configurations, cache_dir must not precede the workers option
3328# and should use configuration macros or conditionals to give each
3329# worker interested in disk caching a dedicated cache directory.
3330#
3331#
3332# ==== The ufs store type ====
3333#
3334# "ufs" is the old well-known Squid storage format that has always
3335# been there.
3336#
3337# Usage:
3338# cache_dir ufs Directory-Name Mbytes L1 L2 [options]
3339#
3340# 'Mbytes' is the amount of disk space (MB) to use under this
3341# directory. The default is 100 MB. Change this to suit your
3342# configuration. Do NOT put the size of your disk drive here.
3343# Instead, if you want Squid to use the entire disk drive,
3344# subtract 20% and use that value.
3345#
3346# 'L1' is the number of first-level subdirectories which
3347# will be created under the 'Directory'. The default is 16.
3348#
3349# 'L2' is the number of second-level subdirectories which
3350# will be created under each first-level directory. The default
3351# is 256.
3352#
3353#
3354# ==== The aufs store type ====
3355#
3356# "aufs" uses the same storage format as "ufs", utilizing
3357# POSIX-threads to avoid blocking the main Squid process on
3358# disk-I/O. This was formerly known in Squid as async-io.
3359#
3360# Usage:
3361# cache_dir aufs Directory-Name Mbytes L1 L2 [options]
3362#
3363# see argument descriptions under ufs above
3364#
3365#
3366# ==== The diskd store type ====
3367#
3368# "diskd" uses the same storage format as "ufs", utilizing a
3369# separate process to avoid blocking the main Squid process on
3370# disk-I/O.
3371#
3372# Usage:
3373# cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]
3374#
3375# see argument descriptions under ufs above
3376#
3377# Q1 specifies the number of unacknowledged I/O requests when Squid
3378# stops opening new files. If this many messages are in the queues,
3379# Squid won't open new files. Default is 64
3380#
3381# Q2 specifies the number of unacknowledged messages when Squid
3382# starts blocking. If this many messages are in the queues,
3383# Squid blocks until it receives some replies. Default is 72
3384#
3385# When Q1 < Q2 (the default), the cache directory is optimized
3386# for lower response time at the expense of a decrease in hit
3387# ratio. If Q1 > Q2, the cache directory is optimized for
3388# higher hit ratio at the expense of an increase in response
3389# time.
3390#
3391#
3392# ==== The rock store type ====
3393#
3394# Usage:
3395# cache_dir rock Directory-Name Mbytes [options]
3396#
3397# The Rock Store type is a database-style storage. All cached
3398# entries are stored in a "database" file, using fixed-size slots.
3399# A single entry occupies one or more slots.
3400#
3401# If possible, Squid using Rock Store creates a dedicated kid
3402# process called "disker" to avoid blocking Squid worker(s) on disk
3403# I/O. One disker kid is created for each rock cache_dir. Diskers
3404# are created only when Squid, running in daemon mode, has support
3405# for the IpcIo disk I/O module.
3406#
3407# swap-timeout=msec: Squid will not start writing a miss to or
3408# reading a hit from disk if it estimates that the swap operation
3409# will take more than the specified number of milliseconds. By
3410# default and when set to zero, disables the disk I/O time limit
3411# enforcement. Ignored when using blocking I/O module because
3412# blocking synchronous I/O does not allow Squid to estimate the
3413# expected swap wait time.
3414#
3415# max-swap-rate=swaps/sec: Artificially limits disk access using
3416# the specified I/O rate limit. Swap out requests that
3417# would cause the average I/O rate to exceed the limit are
3418# delayed. Individual swap in requests (i.e., hits or reads) are
3419# not delayed, but they do contribute to measured swap rate and
3420# since they are placed in the same FIFO queue as swap out
3421# requests, they may wait longer if max-swap-rate is smaller.
3422# This is necessary on file systems that buffer "too
3423# many" writes and then start blocking Squid and other processes
3424# while committing those writes to disk. Usually used together
3425# with swap-timeout to avoid excessive delays and queue overflows
3426# when disk demand exceeds available disk "bandwidth". By default
3427# and when set to zero, disables the disk I/O rate limit
3428# enforcement. Currently supported by IpcIo module only.
3429#
3430# slot-size=bytes: The size of a database "record" used for
3431# storing cached responses. A cached response occupies at least
3432# one slot and all database I/O is done using individual slots so
3433# increasing this parameter leads to more disk space waste while
3434# decreasing it leads to more disk I/O overheads. Should be a
3435# multiple of your operating system I/O page size. Defaults to
3436# 16KBytes. A housekeeping header is stored with each slot and
3437# smaller slot-sizes will be rejected. The header is smaller than
3438# 100 bytes.
3439#
3440#
3441# ==== COMMON OPTIONS ====
3442#
3443# no-store no new objects should be stored to this cache_dir.
3444#
3445# min-size=n the minimum object size in bytes this cache_dir
3446# will accept. It's used to restrict a cache_dir
3447# to only store large objects (e.g. AUFS) while
3448# other stores are optimized for smaller objects
3449# (e.g. Rock).
3450# Defaults to 0.
3451#
3452# max-size=n the maximum object size in bytes this cache_dir
3453# supports.
3454# The value in maximum_object_size directive sets
3455# the default unless more specific details are
3456# available (ie a small store capacity).
3457#
3458# Note: To make optimal use of the max-size limits you should order
3459# the cache_dir lines with the smallest max-size value first.
3460#
3461#Default:
3462# No disk cache. Store cache ojects only in memory.
3463#
3464
3465# Uncomment and adjust the following to add a disk cache directory.
3466#cache_dir ufs /var/spool/squid 100 16 256
3467
3468# TAG: store_dir_select_algorithm
3469# How Squid selects which cache_dir to use when the response
3470# object will fit into more than one.
3471#
3472# Regardless of which algorithm is used the cache_dir min-size
3473# and max-size parameters are obeyed. As such they can affect
3474# the selection algorithm by limiting the set of considered
3475# cache_dir.
3476#
3477# Algorithms:
3478#
3479# least-load
3480#
3481# This algorithm is suited to caches with similar cache_dir
3482# sizes and disk speeds.
3483#
3484# The disk with the least I/O pending is selected.
3485# When there are multiple disks with the same I/O load ranking
3486# the cache_dir with most available capacity is selected.
3487#
3488# When a mix of cache_dir sizes are configured the faster disks
3489# have a naturally lower I/O loading and larger disks have more
3490# capacity. So space used to store objects and data throughput
3491# may be very unbalanced towards larger disks.
3492#
3493#
3494# round-robin
3495#
3496# This algorithm is suited to caches with unequal cache_dir
3497# disk sizes.
3498#
3499# Each cache_dir is selected in a rotation. The next suitable
3500# cache_dir is used.
3501#
3502# Available cache_dir capacity is only considered in relation
3503# to whether the object will fit and meets the min-size and
3504# max-size parameters.
3505#
3506# Disk I/O loading is only considered to prevent overload on slow
3507# disks. This algorithm does not spread objects by size, so any
3508# I/O loading per-disk may appear very unbalanced and volatile.
3509#
3510# If several cache_dirs use similar min-size, max-size, or other
3511# limits to to reject certain responses, then do not group such
3512# cache_dir lines together, to avoid round-robin selection bias
3513# towards the first cache_dir after the group. Instead, interleave
3514# cache_dir lines from different groups. For example:
3515#
3516# store_dir_select_algorithm round-robin
3517# cache_dir rock /hdd1 ... min-size=100000
3518# cache_dir rock /ssd1 ... max-size=99999
3519# cache_dir rock /hdd2 ... min-size=100000
3520# cache_dir rock /ssd2 ... max-size=99999
3521# cache_dir rock /hdd3 ... min-size=100000
3522# cache_dir rock /ssd3 ... max-size=99999
3523#Default:
3524# store_dir_select_algorithm least-load
3525
3526# TAG: max_open_disk_fds
3527# To avoid having disk as the I/O bottleneck Squid can optionally
3528# bypass the on-disk cache if more than this amount of disk file
3529# descriptors are open.
3530#
3531# A value of 0 indicates no limit.
3532#Default:
3533# no limit
3534
3535# TAG: cache_swap_low (percent, 0-100)
3536# The low-water mark for AUFS/UFS/diskd cache object eviction by
3537# the cache_replacement_policy algorithm.
3538#
3539# Removal begins when the swap (disk) usage of a cache_dir is
3540# above this low-water mark and attempts to maintain utilization
3541# near the low-water mark.
3542#
3543# As swap utilization increases towards the high-water mark set
3544# by cache_swap_high object eviction becomes more agressive.
3545#
3546# The value difference in percentages between low- and high-water
3547# marks represent an eviction rate of 300 objects per second and
3548# the rate continues to scale in agressiveness by multiples of
3549# this above the high-water mark.
3550#
3551# Defaults are 90% and 95%. If you have a large cache, 5% could be
3552# hundreds of MB. If this is the case you may wish to set these
3553# numbers closer together.
3554#
3555# See also cache_swap_high and cache_replacement_policy
3556#Default:
3557# cache_swap_low 90
3558
3559# TAG: cache_swap_high (percent, 0-100)
3560# The high-water mark for AUFS/UFS/diskd cache object eviction by
3561# the cache_replacement_policy algorithm.
3562#
3563# Removal begins when the swap (disk) usage of a cache_dir is
3564# above the low-water mark set by cache_swap_low and attempts to
3565# maintain utilization near the low-water mark.
3566#
3567# As swap utilization increases towards this high-water mark object
3568# eviction becomes more agressive.
3569#
3570# The value difference in percentages between low- and high-water
3571# marks represent an eviction rate of 300 objects per second and
3572# the rate continues to scale in agressiveness by multiples of
3573# this above the high-water mark.
3574#
3575# Defaults are 90% and 95%. If you have a large cache, 5% could be
3576# hundreds of MB. If this is the case you may wish to set these
3577# numbers closer together.
3578#
3579# See also cache_swap_low and cache_replacement_policy
3580#Default:
3581# cache_swap_high 95
3582
3583# LOGFILE OPTIONS
3584# -----------------------------------------------------------------------------
3585
3586# TAG: logformat
3587# Usage:
3588#
3589# logformat <name> <format specification>
3590#
3591# Defines an access log format.
3592#
3593# The <format specification> is a string with embedded % format codes
3594#
3595# % format codes all follow the same basic structure where all but
3596# the formatcode is optional. Output strings are automatically escaped
3597# as required according to their context and the output format
3598# modifiers are usually not needed, but can be specified if an explicit
3599# output format is desired.
3600#
3601# % ["|[|'|#] [-] [[0]width] [{argument}] formatcode
3602#
3603# " output in quoted string format
3604# [ output in squid text log format as used by log_mime_hdrs
3605# # output in URL quoted format
3606# ' output as-is
3607#
3608# - left aligned
3609#
3610# width minimum and/or maximum field width:
3611# [width_min][.width_max]
3612# When minimum starts with 0, the field is zero-padded.
3613# String values exceeding maximum width are truncated.
3614#
3615# {arg} argument such as header name etc
3616#
3617# Format codes:
3618#
3619# % a literal % character
3620# sn Unique sequence number per log line entry
3621# err_code The ID of an error response served by Squid or
3622# a similar internal error identifier.
3623# err_detail Additional err_code-dependent error information.
3624# note The annotation specified by the argument. Also
3625# logs the adaptation meta headers set by the
3626# adaptation_meta configuration parameter.
3627# If no argument given all annotations logged.
3628# The argument may include a separator to use with
3629# annotation values:
3630# name[:separator]
3631# By default, multiple note values are separated with ","
3632# and multiple notes are separated with "\r\n".
3633# When logging named notes with %{name}note, the
3634# explicitly configured separator is used between note
3635# values. When logging all notes with %note, the
3636# explicitly configured separator is used between
3637# individual notes. There is currently no way to
3638# specify both value and notes separators when logging
3639# all notes with %note.
3640#
3641# Connection related format codes:
3642#
3643# >a Client source IP address
3644# >A Client FQDN
3645# >p Client source port
3646# >eui Client source EUI (MAC address, EUI-48 or EUI-64 identifier)
3647# >la Local IP address the client connected to
3648# >lp Local port number the client connected to
3649# >qos Client connection TOS/DSCP value set by Squid
3650# >nfmark Client connection netfilter mark set by Squid
3651#
3652# la Local listening IP address the client connection was connected to.
3653# lp Local listening port number the client connection was connected to.
3654#
3655# <a Server IP address of the last server or peer connection
3656# <A Server FQDN or peer name
3657# <p Server port number of the last server or peer connection
3658# <la Local IP address of the last server or peer connection
3659# <lp Local port number of the last server or peer connection
3660# <qos Server connection TOS/DSCP value set by Squid
3661# <nfmark Server connection netfilter mark set by Squid
3662#
3663# Time related format codes:
3664#
3665# ts Seconds since epoch
3666# tu subsecond time (milliseconds)
3667# tl Local time. Optional strftime format argument
3668# default %d/%b/%Y:%H:%M:%S %z
3669# tg GMT time. Optional strftime format argument
3670# default %d/%b/%Y:%H:%M:%S %z
3671# tr Response time (milliseconds)
3672# dt Total time spent making DNS lookups (milliseconds)
3673# tS Approximate master transaction start time in
3674# <full seconds since epoch>.<fractional seconds> format.
3675# Currently, Squid considers the master transaction
3676# started when a complete HTTP request header initiating
3677# the transaction is received from the client. This is
3678# the same value that Squid uses to calculate transaction
3679# response time when logging %tr to access.log. Currently,
3680# Squid uses millisecond resolution for %tS values,
3681# similar to the default access.log "current time" field
3682# (%ts.%03tu).
3683#
3684# Access Control related format codes:
3685#
3686# et Tag returned by external acl
3687# ea Log string returned by external acl
3688# un User name (any available)
3689# ul User name from authentication
3690# ue User name from external acl helper
3691# ui User name from ident
3692# un A user name. Expands to the first available name
3693# from the following list of information sources:
3694# - authenticated user name, like %ul
3695# - user name supplied by an external ACL, like %ue
3696# - SSL client name, like %us
3697# - ident user name, like %ui
3698# credentials Client credentials. The exact meaning depends on
3699# the authentication scheme: For Basic authentication,
3700# it is the password; for Digest, the realm sent by the
3701# client; for NTLM and Negotiate, the client challenge
3702# or client credentials prefixed with "YR " or "KK ".
3703#
3704# HTTP related format codes:
3705#
3706# REQUEST
3707#
3708# [http::]rm Request method (GET/POST etc)
3709# [http::]>rm Request method from client
3710# [http::]<rm Request method sent to server or peer
3711# [http::]ru Request URL from client (historic, filtered for logging)
3712# [http::]>ru Request URL from client
3713# [http::]<ru Request URL sent to server or peer
3714# [http::]>rs Request URL scheme from client
3715# [http::]<rs Request URL scheme sent to server or peer
3716# [http::]>rd Request URL domain from client
3717# [http::]<rd Request URL domain sent to server or peer
3718# [http::]>rP Request URL port from client
3719# [http::]<rP Request URL port sent to server or peer
3720# [http::]rp Request URL path excluding hostname
3721# [http::]>rp Request URL path excluding hostname from client
3722# [http::]<rp Request URL path excluding hostname sent to server or peer
3723# [http::]rv Request protocol version
3724# [http::]>rv Request protocol version from client
3725# [http::]<rv Request protocol version sent to server or peer
3726#
3727# [http::]>h Original received request header.
3728# Usually differs from the request header sent by
3729# Squid, although most fields are often preserved.
3730# Accepts optional header field name/value filter
3731# argument using name[:[separator]element] format.
3732# [http::]>ha Received request header after adaptation and
3733# redirection (pre-cache REQMOD vectoring point).
3734# Usually differs from the request header sent by
3735# Squid, although most fields are often preserved.
3736# Optional header name argument as for >h
3737#
3738#
3739# RESPONSE
3740#
3741# [http::]<Hs HTTP status code received from the next hop
3742# [http::]>Hs HTTP status code sent to the client
3743#
3744# [http::]<h Reply header. Optional header name argument
3745# as for >h
3746#
3747# [http::]mt MIME content type
3748#
3749#
3750# SIZE COUNTERS
3751#
3752# [http::]st Total size of request + reply traffic with client
3753# [http::]>st Total size of request received from client.
3754# Excluding chunked encoding bytes.
3755# [http::]<st Total size of reply sent to client (after adaptation)
3756#
3757# [http::]>sh Size of request headers received from client
3758# [http::]<sh Size of reply headers sent to client (after adaptation)
3759#
3760# [http::]<sH Reply high offset sent
3761# [http::]<sS Upstream object size
3762#
3763# [http::]<bs Number of HTTP-equivalent message body bytes
3764# received from the next hop, excluding chunked
3765# transfer encoding and control messages.
3766# Generated FTP/Gopher listings are treated as
3767# received bodies.
3768#
3769#
3770# TIMING
3771#
3772# [http::]<pt Peer response time in milliseconds. The timer starts
3773# when the last request byte is sent to the next hop
3774# and stops when the last response byte is received.
3775# [http::]<tt Total time in milliseconds. The timer
3776# starts with the first connect request (or write I/O)
3777# sent to the first selected peer. The timer stops
3778# with the last I/O with the last peer.
3779#
3780# Squid handling related format codes:
3781#
3782# Ss Squid request status (TCP_MISS etc)
3783# Sh Squid hierarchy status (DEFAULT_PARENT etc)
3784#
3785# SSL-related format codes:
3786#
3787# ssl::bump_mode SslBump decision for the transaction:
3788#
3789# For CONNECT requests that initiated bumping of
3790# a connection and for any request received on
3791# an already bumped connection, Squid logs the
3792# corresponding SslBump mode ("server-first" or
3793# "client-first"). See the ssl_bump option for
3794# more information about these modes.
3795#
3796# A "none" token is logged for requests that
3797# triggered "ssl_bump" ACL evaluation matching
3798# either a "none" rule or no rules at all.
3799#
3800# In all other cases, a single dash ("-") is
3801# logged.
3802#
3803# ssl::>sni SSL client SNI sent to Squid. Available only
3804# after the peek, stare, or splice SSL bumping
3805# actions.
3806#
3807# If ICAP is enabled, the following code becomes available (as
3808# well as ICAP log codes documented with the icap_log option):
3809#
3810# icap::tt Total ICAP processing time for the HTTP
3811# transaction. The timer ticks when ICAP
3812# ACLs are checked and when ICAP
3813# transaction is in progress.
3814#
3815# If adaptation is enabled the following three codes become available:
3816#
3817# adapt::<last_h The header of the last ICAP response or
3818# meta-information from the last eCAP
3819# transaction related to the HTTP transaction.
3820# Like <h, accepts an optional header name
3821# argument.
3822#
3823# adapt::sum_trs Summed adaptation transaction response
3824# times recorded as a comma-separated list in
3825# the order of transaction start time. Each time
3826# value is recorded as an integer number,
3827# representing response time of one or more
3828# adaptation (ICAP or eCAP) transaction in
3829# milliseconds. When a failed transaction is
3830# being retried or repeated, its time is not
3831# logged individually but added to the
3832# replacement (next) transaction. See also:
3833# adapt::all_trs.
3834#
3835# adapt::all_trs All adaptation transaction response times.
3836# Same as adaptation_strs but response times of
3837# individual transactions are never added
3838# together. Instead, all transaction response
3839# times are recorded individually.
3840#
3841# You can prefix adapt::*_trs format codes with adaptation
3842# service name in curly braces to record response time(s) specific
3843# to that service. For example: %{my_service}adapt::sum_trs
3844#
3845# If SSL is enabled, the following formating codes become available:
3846#
3847# %ssl::>cert_subject The Subject field of the received client
3848# SSL certificate or a dash ('-') if Squid has
3849# received an invalid/malformed certificate or
3850# no certificate at all. Consider encoding the
3851# logged value because Subject often has spaces.
3852#
3853# %ssl::>cert_issuer The Issuer field of the received client
3854# SSL certificate or a dash ('-') if Squid has
3855# received an invalid/malformed certificate or
3856# no certificate at all. Consider encoding the
3857# logged value because Issuer often has spaces.
3858#
3859# The default formats available (which do not need re-defining) are:
3860#
3861#logformat squid %ts.%03tu %6tr %>a %Ss/%03>Hs %<st %rm %ru %[un %Sh/%<a %mt
3862#logformat common %>a %[ui %[un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st %Ss:%Sh
3863#logformat combined %>a %[ui %[un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st "%{Referer}>h" "%{User-Agent}>h" %Ss:%Sh
3864#logformat referrer %ts.%03tu %>a %{Referer}>h %ru
3865#logformat useragent %>a [%tl] "%{User-Agent}>h"
3866#
3867# NOTE: When the log_mime_hdrs directive is set to ON.
3868# The squid, common and combined formats have a safely encoded copy
3869# of the mime headers appended to each line within a pair of brackets.
3870#
3871# NOTE: The common and combined formats are not quite true to the Apache definition.
3872# The logs from Squid contain an extra status and hierarchy code appended.
3873#
3874#Default:
3875# The format definitions squid, common, combined, referrer, useragent are built in.
3876
3877# TAG: access_log
3878# Configures whether and how Squid logs HTTP and ICP transactions.
3879# If access logging is enabled, a single line is logged for every
3880# matching HTTP or ICP request. The recommended directive formats are:
3881#
3882# access_log <module>:<place> [option ...] [acl acl ...]
3883# access_log none [acl acl ...]
3884#
3885# The following directive format is accepted but may be deprecated:
3886# access_log <module>:<place> [<logformat name> [acl acl ...]]
3887#
3888# In most cases, the first ACL name must not contain the '=' character
3889# and should not be equal to an existing logformat name. You can always
3890# start with an 'all' ACL to work around those restrictions.
3891#
3892# Will log to the specified module:place using the specified format (which
3893# must be defined in a logformat directive) those entries which match
3894# ALL the acl's specified (which must be defined in acl clauses).
3895# If no acl is specified, all requests will be logged to this destination.
3896#
3897# ===== Available options for the recommended directive format =====
3898#
3899# logformat=name Names log line format (either built-in or
3900# defined by a logformat directive). Defaults
3901# to 'squid'.
3902#
3903# buffer-size=64KB Defines approximate buffering limit for log
3904# records (see buffered_logs). Squid should not
3905# keep more than the specified size and, hence,
3906# should flush records before the buffer becomes
3907# full to avoid overflows under normal
3908# conditions (the exact flushing algorithm is
3909# module-dependent though). The on-error option
3910# controls overflow handling.
3911#
3912# on-error=die|drop Defines action on unrecoverable errors. The
3913# 'drop' action ignores (i.e., does not log)
3914# affected log records. The default 'die' action
3915# kills the affected worker. The drop action
3916# support has not been tested for modules other
3917# than tcp.
3918#
3919# ===== Modules Currently available =====
3920#
3921# none Do not log any requests matching these ACL.
3922# Do not specify Place or logformat name.
3923#
3924# stdio Write each log line to disk immediately at the completion of
3925# each request.
3926# Place: the filename and path to be written.
3927#
3928# daemon Very similar to stdio. But instead of writing to disk the log
3929# line is passed to a daemon helper for asychronous handling instead.
3930# Place: varies depending on the daemon.
3931#
3932# log_file_daemon Place: the file name and path to be written.
3933#
3934# syslog To log each request via syslog facility.
3935# Place: The syslog facility and priority level for these entries.
3936# Place Format: facility.priority
3937#
3938# where facility could be any of:
3939# authpriv, daemon, local0 ... local7 or user.
3940#
3941# And priority could be any of:
3942# err, warning, notice, info, debug.
3943#
3944# udp To send each log line as text data to a UDP receiver.
3945# Place: The destination host name or IP and port.
3946# Place Format: //host:port
3947#
3948# tcp To send each log line as text data to a TCP receiver.
3949# Lines may be accumulated before sending (see buffered_logs).
3950# Place: The destination host name or IP and port.
3951# Place Format: //host:port
3952#
3953# Default:
3954# access_log daemon:/var/log/squid/access.log squid
3955#Default:
3956# access_log daemon:/var/log/squid/access.log squid
3957
3958# TAG: icap_log
3959# ICAP log files record ICAP transaction summaries, one line per
3960# transaction.
3961#
3962# The icap_log option format is:
3963# icap_log <filepath> [<logformat name> [acl acl ...]]
3964# icap_log none [acl acl ...]]
3965#
3966# Please see access_log option documentation for details. The two
3967# kinds of logs share the overall configuration approach and many
3968# features.
3969#
3970# ICAP processing of a single HTTP message or transaction may
3971# require multiple ICAP transactions. In such cases, multiple
3972# ICAP transaction log lines will correspond to a single access
3973# log line.
3974#
3975# ICAP log supports many access.log logformat %codes. In ICAP context,
3976# HTTP message-related %codes are applied to the HTTP message embedded
3977# in an ICAP message. Logformat "%http::>..." codes are used for HTTP
3978# messages embedded in ICAP requests while "%http::<..." codes are used
3979# for HTTP messages embedded in ICAP responses. For example:
3980#
3981# http::>h To-be-adapted HTTP message headers sent by Squid to
3982# the ICAP service. For REQMOD transactions, these are
3983# HTTP request headers. For RESPMOD, these are HTTP
3984# response headers, but Squid currently cannot log them
3985# (i.e., %http::>h will expand to "-" for RESPMOD).
3986#
3987# http::<h Adapted HTTP message headers sent by the ICAP
3988# service to Squid (i.e., HTTP request headers in regular
3989# REQMOD; HTTP response headers in RESPMOD and during
3990# request satisfaction in REQMOD).
3991#
3992# ICAP OPTIONS transactions do not embed HTTP messages.
3993#
3994# Several logformat codes below deal with ICAP message bodies. An ICAP
3995# message body, if any, typically includes a complete HTTP message
3996# (required HTTP headers plus optional HTTP message body). When
3997# computing HTTP message body size for these logformat codes, Squid
3998# either includes or excludes chunked encoding overheads; see
3999# code-specific documentation for details.
4000#
4001# For Secure ICAP services, all size-related information is currently
4002# computed before/after TLS encryption/decryption, as if TLS was not
4003# in use at all.
4004#
4005# The following format codes are also available for ICAP logs:
4006#
4007# icap::<A ICAP server IP address. Similar to <A.
4008#
4009# icap::<service_name ICAP service name from the icap_service
4010# option in Squid configuration file.
4011#
4012# icap::ru ICAP Request-URI. Similar to ru.
4013#
4014# icap::rm ICAP request method (REQMOD, RESPMOD, or
4015# OPTIONS). Similar to existing rm.
4016#
4017# icap::>st The total size of the ICAP request sent to the ICAP
4018# server (ICAP headers + ICAP body), including chunking
4019# metadata (if any).
4020#
4021# icap::<st The total size of the ICAP response received from the
4022# ICAP server (ICAP headers + ICAP body), including
4023# chunking metadata (if any).
4024#
4025# icap::<bs The size of the ICAP response body received from the
4026# ICAP server, excluding chunking metadata (if any).
4027#
4028# icap::tr Transaction response time (in
4029# milliseconds). The timer starts when
4030# the ICAP transaction is created and
4031# stops when the transaction is completed.
4032# Similar to tr.
4033#
4034# icap::tio Transaction I/O time (in milliseconds). The
4035# timer starts when the first ICAP request
4036# byte is scheduled for sending. The timers
4037# stops when the last byte of the ICAP response
4038# is received.
4039#
4040# icap::to Transaction outcome: ICAP_ERR* for all
4041# transaction errors, ICAP_OPT for OPTION
4042# transactions, ICAP_ECHO for 204
4043# responses, ICAP_MOD for message
4044# modification, and ICAP_SAT for request
4045# satisfaction. Similar to Ss.
4046#
4047# icap::Hs ICAP response status code. Similar to Hs.
4048#
4049# icap::>h ICAP request header(s). Similar to >h.
4050#
4051# icap::<h ICAP response header(s). Similar to <h.
4052#
4053# The default ICAP log format, which can be used without an explicit
4054# definition, is called icap_squid:
4055#
4056#logformat icap_squid %ts.%03tu %6icap::tr %>A %icap::to/%03icap::Hs %icap::<st %icap::rm %icap::ru %un -/%icap::<A -
4057#
4058# See also: logformat and %adapt::<last_h
4059#Default:
4060# none
4061
4062# TAG: logfile_daemon
4063# Specify the path to the logfile-writing daemon. This daemon is
4064# used to write the access and store logs, if configured.
4065#
4066# Squid sends a number of commands to the log daemon:
4067# L<data>\n - logfile data
4068# R\n - rotate file
4069# T\n - truncate file
4070# O\n - reopen file
4071# F\n - flush file
4072# r<n>\n - set rotate count to <n>
4073# b<n>\n - 1 = buffer output, 0 = don't buffer output
4074#
4075# No responses is expected.
4076#Default:
4077# logfile_daemon /usr/lib/squid/log_file_daemon
4078
4079# TAG: stats_collection allow|deny acl acl...
4080# This options allows you to control which requests gets accounted
4081# in performance counters.
4082#
4083# This clause only supports fast acl types.
4084# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4085#Default:
4086# Allow logging for all transactions.
4087
4088# TAG: cache_store_log
4089# Logs the activities of the storage manager. Shows which
4090# objects are ejected from the cache, and which objects are
4091# saved and for how long.
4092# There are not really utilities to analyze this data, so you can safely
4093# disable it (the default).
4094#
4095# Store log uses modular logging outputs. See access_log for the list
4096# of modules supported.
4097#
4098# Example:
4099# cache_store_log stdio:/var/log/squid/store.log
4100# cache_store_log daemon:/var/log/squid/store.log
4101#Default:
4102# none
4103
4104# TAG: cache_swap_state
4105# Location for the cache "swap.state" file. This index file holds
4106# the metadata of objects saved on disk. It is used to rebuild
4107# the cache during startup. Normally this file resides in each
4108# 'cache_dir' directory, but you may specify an alternate
4109# pathname here. Note you must give a full filename, not just
4110# a directory. Since this is the index for the whole object
4111# list you CANNOT periodically rotate it!
4112#
4113# If %s can be used in the file name it will be replaced with a
4114# a representation of the cache_dir name where each / is replaced
4115# with '.'. This is needed to allow adding/removing cache_dir
4116# lines when cache_swap_log is being used.
4117#
4118# If have more than one 'cache_dir', and %s is not used in the name
4119# these swap logs will have names such as:
4120#
4121# cache_swap_log.00
4122# cache_swap_log.01
4123# cache_swap_log.02
4124#
4125# The numbered extension (which is added automatically)
4126# corresponds to the order of the 'cache_dir' lines in this
4127# configuration file. If you change the order of the 'cache_dir'
4128# lines in this file, these index files will NOT correspond to
4129# the correct 'cache_dir' entry (unless you manually rename
4130# them). We recommend you do NOT use this option. It is
4131# better to keep these index files in each 'cache_dir' directory.
4132#Default:
4133# Store the journal inside its cache_dir
4134
4135# TAG: logfile_rotate
4136# Specifies the number of logfile rotations to make when you
4137# type 'squid -k rotate'. The default is 10, which will rotate
4138# with extensions 0 through 9. Setting logfile_rotate to 0 will
4139# disable the file name rotation, but the logfiles are still closed
4140# and re-opened. This will enable you to rename the logfiles
4141# yourself just before sending the rotate signal.
4142#
4143# Note, the 'squid -k rotate' command normally sends a USR1
4144# signal to the running squid process. In certain situations
4145# (e.g. on Linux with Async I/O), USR1 is used for other
4146# purposes, so -k rotate uses another signal. It is best to get
4147# in the habit of using 'squid -k rotate' instead of 'kill -USR1
4148# <pid>'.
4149#
4150# Note, from Squid-3.1 this option is only a default for cache.log,
4151# that log can be rotated separately by using debug_options.
4152#
4153# Note2, for Debian/Linux the default of logfile_rotate is
4154# zero, since it includes external logfile-rotation methods.
4155#Default:
4156# logfile_rotate 0
4157
4158# TAG: mime_table
4159# Path to Squid's icon configuration file.
4160#
4161# You shouldn't need to change this, but the default file contains
4162# examples and formatting information if you do.
4163#Default:
4164# mime_table /usr/share/squid/mime.conf
4165
4166# TAG: log_mime_hdrs on|off
4167# The Cache can record both the request and the response MIME
4168# headers for each HTTP transaction. The headers are encoded
4169# safely and will appear as two bracketed fields at the end of
4170# the access log (for either the native or httpd-emulated log
4171# formats). To enable this logging set log_mime_hdrs to 'on'.
4172#Default:
4173# log_mime_hdrs off
4174
4175# TAG: pid_filename
4176# A filename to write the process-id to. To disable, enter "none".
4177#Default:
4178# pid_filename /var/run/squid.pid
4179
4180# TAG: client_netmask
4181# A netmask for client addresses in logfiles and cachemgr output.
4182# Change this to protect the privacy of your cache clients.
4183# A netmask of 255.255.255.0 will log all IP's in that range with
4184# the last digit set to '0'.
4185#Default:
4186# Log full client IP address
4187
4188# TAG: strip_query_terms
4189# By default, Squid strips query terms from requested URLs before
4190# logging. This protects your user's privacy and reduces log size.
4191#
4192# When investigating HIT/MISS or other caching behaviour you
4193# will need to disable this to see the full URL used by Squid.
4194#Default:
4195# strip_query_terms on
4196
4197# TAG: buffered_logs on|off
4198# Whether to write/send access_log records ASAP or accumulate them and
4199# then write/send them in larger chunks. Buffering may improve
4200# performance because it decreases the number of I/Os. However,
4201# buffering increases the delay before log records become available to
4202# the final recipient (e.g., a disk file or logging daemon) and,
4203# hence, increases the risk of log records loss.
4204#
4205# Note that even when buffered_logs are off, Squid may have to buffer
4206# records if it cannot write/send them immediately due to pending I/Os
4207# (e.g., the I/O writing the previous log record) or connectivity loss.
4208#
4209# Currently honored by 'daemon' and 'tcp' access_log modules only.
4210#Default:
4211# buffered_logs off
4212
4213# TAG: netdb_filename
4214# Where Squid stores it's netdb journal.
4215# When enabled this journal preserves netdb state between restarts.
4216#
4217# To disable, enter "none".
4218#Default:
4219# netdb_filename stdio:/var/log/squid/netdb.state
4220
4221# OPTIONS FOR TROUBLESHOOTING
4222# -----------------------------------------------------------------------------
4223
4224# TAG: cache_log
4225# Squid administrative logging file.
4226#
4227# This is where general information about Squid behavior goes. You can
4228# increase the amount of data logged to this file and how often it is
4229# rotated with "debug_options"
4230#Default:
4231# cache_log /var/log/squid/cache.log
4232
4233# TAG: debug_options
4234# Logging options are set as section,level where each source file
4235# is assigned a unique section. Lower levels result in less
4236# output, Full debugging (level 9) can result in a very large
4237# log file, so be careful.
4238#
4239# The magic word "ALL" sets debugging levels for all sections.
4240# The default is to run with "ALL,1" to record important warnings.
4241#
4242# The rotate=N option can be used to keep more or less of these logs
4243# than would otherwise be kept by logfile_rotate.
4244# For most uses a single log should be enough to monitor current
4245# events affecting Squid.
4246#Default:
4247# Log all critical and important messages.
4248
4249# TAG: coredump_dir
4250# By default Squid leaves core files in the directory from where
4251# it was started. If you set 'coredump_dir' to a directory
4252# that exists, Squid will chdir() to that directory at startup
4253# and coredump files will be left there.
4254#
4255#Default:
4256# Use the directory from where Squid was started.
4257#
4258
4259# Leave coredumps in the first cache dir
4260coredump_dir /var/spool/squid
4261
4262# OPTIONS FOR FTP GATEWAYING
4263# -----------------------------------------------------------------------------
4264
4265# TAG: ftp_user
4266# If you want the anonymous login password to be more informative
4267# (and enable the use of picky FTP servers), set this to something
4268# reasonable for your domain, like wwwuser@somewhere.net
4269#
4270# The reason why this is domainless by default is the
4271# request can be made on the behalf of a user in any domain,
4272# depending on how the cache is used.
4273# Some FTP server also validate the email address is valid
4274# (for example perl.com).
4275#Default:
4276# ftp_user Squid@
4277
4278# TAG: ftp_passive
4279# If your firewall does not allow Squid to use passive
4280# connections, turn off this option.
4281#
4282# Use of ftp_epsv_all option requires this to be ON.
4283#Default:
4284# ftp_passive on
4285
4286# TAG: ftp_epsv_all
4287# FTP Protocol extensions permit the use of a special "EPSV ALL" command.
4288#
4289# NATs may be able to put the connection on a "fast path" through the
4290# translator, as the EPRT command will never be used and therefore,
4291# translation of the data portion of the segments will never be needed.
4292#
4293# When a client only expects to do two-way FTP transfers this may be
4294# useful.
4295# If squid finds that it must do a three-way FTP transfer after issuing
4296# an EPSV ALL command, the FTP session will fail.
4297#
4298# If you have any doubts about this option do not use it.
4299# Squid will nicely attempt all other connection methods.
4300#
4301# Requires ftp_passive to be ON (default) for any effect.
4302#Default:
4303# ftp_epsv_all off
4304
4305# TAG: ftp_epsv
4306# FTP Protocol extensions permit the use of a special "EPSV" command.
4307#
4308# NATs may be able to put the connection on a "fast path" through the
4309# translator using EPSV, as the EPRT command will never be used
4310# and therefore, translation of the data portion of the segments
4311# will never be needed.
4312#
4313# EPSV is often required to interoperate with FTP servers on IPv6
4314# networks. On the other hand, it may break some IPv4 servers.
4315#
4316# By default, EPSV may try EPSV with any FTP server. To fine tune
4317# that decision, you may restrict EPSV to certain clients or servers
4318# using ACLs:
4319#
4320# ftp_epsv allow|deny al1 acl2 ...
4321#
4322# WARNING: Disabling EPSV may cause problems with external NAT and IPv6.
4323#
4324# Only fast ACLs are supported.
4325# Requires ftp_passive to be ON (default) for any effect.
4326#Default:
4327# none
4328
4329# TAG: ftp_eprt
4330# FTP Protocol extensions permit the use of a special "EPRT" command.
4331#
4332# This extension provides a protocol neutral alternative to the
4333# IPv4-only PORT command. When supported it enables active FTP data
4334# channels over IPv6 and efficient NAT handling.
4335#
4336# Turning this OFF will prevent EPRT being attempted and will skip
4337# straight to using PORT for IPv4 servers.
4338#
4339# Some devices are known to not handle this extension correctly and
4340# may result in crashes. Devices which suport EPRT enough to fail
4341# cleanly will result in Squid attempting PORT anyway. This directive
4342# should only be disabled when EPRT results in device failures.
4343#
4344# WARNING: Doing so will convert Squid back to the old behavior with all
4345# the related problems with external NAT devices/layers and IPv4-only FTP.
4346#Default:
4347# ftp_eprt on
4348
4349# TAG: ftp_sanitycheck
4350# For security and data integrity reasons Squid by default performs
4351# sanity checks of the addresses of FTP data connections ensure the
4352# data connection is to the requested server. If you need to allow
4353# FTP connections to servers using another IP address for the data
4354# connection turn this off.
4355#Default:
4356# ftp_sanitycheck on
4357
4358# TAG: ftp_telnet_protocol
4359# The FTP protocol is officially defined to use the telnet protocol
4360# as transport channel for the control connection. However, many
4361# implementations are broken and does not respect this aspect of
4362# the FTP protocol.
4363#
4364# If you have trouble accessing files with ASCII code 255 in the
4365# path or similar problems involving this ASCII code you can
4366# try setting this directive to off. If that helps, report to the
4367# operator of the FTP server in question that their FTP server
4368# is broken and does not follow the FTP standard.
4369#Default:
4370# ftp_telnet_protocol on
4371
4372# OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
4373# -----------------------------------------------------------------------------
4374
4375# TAG: diskd_program
4376# Specify the location of the diskd executable.
4377# Note this is only useful if you have compiled in
4378# diskd as one of the store io modules.
4379#Default:
4380# diskd_program /usr/lib/squid/diskd
4381
4382# TAG: unlinkd_program
4383# Specify the location of the executable for file deletion process.
4384#Default:
4385# unlinkd_program /usr/lib/squid/unlinkd
4386
4387# TAG: pinger_program
4388# Specify the location of the executable for the pinger process.
4389#Default:
4390# pinger_program /usr/lib/squid/pinger
4391
4392# TAG: pinger_enable
4393# Control whether the pinger is active at run-time.
4394# Enables turning ICMP pinger on and off with a simple
4395# squid -k reconfigure.
4396#Default:
4397# pinger_enable on
4398
4399# OPTIONS FOR URL REWRITING
4400# -----------------------------------------------------------------------------
4401
4402# TAG: url_rewrite_program
4403# Specify the location of the executable URL rewriter to use.
4404# Since they can perform almost any function there isn't one included.
4405#
4406# For each requested URL, the rewriter will receive on line with the format
4407#
4408# [channel-ID <SP>] URL [<SP> extras]<NL>
4409#
4410# See url_rewrite_extras on how to send "extras" with optional values to
4411# the helper.
4412# After processing the request the helper must reply using the following format:
4413#
4414# [channel-ID <SP>] result [<SP> kv-pairs]
4415#
4416# The result code can be:
4417#
4418# OK status=30N url="..."
4419# Redirect the URL to the one supplied in 'url='.
4420# 'status=' is optional and contains the status code to send
4421# the client in Squids HTTP response. It must be one of the
4422# HTTP redirect status codes: 301, 302, 303, 307, 308.
4423# When no status is given Squid will use 302.
4424#
4425# OK rewrite-url="..."
4426# Rewrite the URL to the one supplied in 'rewrite-url='.
4427# The new URL is fetched directly by Squid and returned to
4428# the client as the response to its request.
4429#
4430# OK
4431# When neither of url= and rewrite-url= are sent Squid does
4432# not change the URL.
4433#
4434# ERR
4435# Do not change the URL.
4436#
4437# BH
4438# An internal error occurred in the helper, preventing
4439# a result being identified. The 'message=' key name is
4440# reserved for delivering a log message.
4441#
4442#
4443# In addition to the above kv-pairs Squid also understands the following
4444# optional kv-pairs received from URL rewriters:
4445# clt_conn_tag=TAG
4446# Associates a TAG with the client TCP connection.
4447# The TAG is treated as a regular annotation but persists across
4448# future requests on the client connection rather than just the
4449# current request. A helper may update the TAG during subsequent
4450# requests be returning a new kv-pair.
4451#
4452# When using the concurrency= option the protocol is changed by
4453# introducing a query channel tag in front of the request/response.
4454# The query channel tag is a number between 0 and concurrency-1.
4455# This value must be echoed back unchanged to Squid as the first part
4456# of the response relating to its request.
4457#
4458# WARNING: URL re-writing ability should be avoided whenever possible.
4459# Use the URL redirect form of response instead.
4460#
4461# Re-write creates a difference in the state held by the client
4462# and server. Possibly causing confusion when the server response
4463# contains snippets of its view state. Embeded URLs, response
4464# and content Location headers, etc. are not re-written by this
4465# interface.
4466#
4467# By default, a URL rewriter is not used.
4468#Default:
4469# none
4470
4471# TAG: url_rewrite_children
4472# The maximum number of redirector processes to spawn. If you limit
4473# it too few Squid will have to wait for them to process a backlog of
4474# URLs, slowing it down. If you allow too many they will use RAM
4475# and other system resources noticably.
4476#
4477# The startup= and idle= options allow some measure of skew in your
4478# tuning.
4479#
4480# startup=
4481#
4482# Sets a minimum of how many processes are to be spawned when Squid
4483# starts or reconfigures. When set to zero the first request will
4484# cause spawning of the first child process to handle it.
4485#
4486# Starting too few will cause an initial slowdown in traffic as Squid
4487# attempts to simultaneously spawn enough processes to cope.
4488#
4489# idle=
4490#
4491# Sets a minimum of how many processes Squid is to try and keep available
4492# at all times. When traffic begins to rise above what the existing
4493# processes can handle this many more will be spawned up to the maximum
4494# configured. A minimum setting of 1 is required.
4495#
4496# concurrency=
4497#
4498# The number of requests each redirector helper can handle in
4499# parallel. Defaults to 0 which indicates the redirector
4500# is a old-style single threaded redirector.
4501#
4502# When this directive is set to a value >= 1 then the protocol
4503# used to communicate with the helper is modified to include
4504# an ID in front of the request/response. The ID from the request
4505# must be echoed back with the response to that request.
4506#Default:
4507# url_rewrite_children 20 startup=0 idle=1 concurrency=0
4508
4509# TAG: url_rewrite_host_header
4510# To preserve same-origin security policies in browsers and
4511# prevent Host: header forgery by redirectors Squid rewrites
4512# any Host: header in redirected requests.
4513#
4514# If you are running an accelerator this may not be a wanted
4515# effect of a redirector. This directive enables you disable
4516# Host: alteration in reverse-proxy traffic.
4517#
4518# WARNING: Entries are cached on the result of the URL rewriting
4519# process, so be careful if you have domain-virtual hosts.
4520#
4521# WARNING: Squid and other software verifies the URL and Host
4522# are matching, so be careful not to relay through other proxies
4523# or inspecting firewalls with this disabled.
4524#Default:
4525# url_rewrite_host_header on
4526
4527# TAG: url_rewrite_access
4528# If defined, this access list specifies which requests are
4529# sent to the redirector processes.
4530#
4531# This clause supports both fast and slow acl types.
4532# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4533#Default:
4534# Allow, unless rules exist in squid.conf.
4535
4536# TAG: url_rewrite_bypass
4537# When this is 'on', a request will not go through the
4538# redirector if all the helpers are busy. If this is 'off'
4539# and the redirector queue grows too large, Squid will exit
4540# with a FATAL error and ask you to increase the number of
4541# redirectors. You should only enable this if the redirectors
4542# are not critical to your caching system. If you use
4543# redirectors for access control, and you enable this option,
4544# users may have access to pages they should not
4545# be allowed to request.
4546#Default:
4547# url_rewrite_bypass off
4548
4549# TAG: url_rewrite_extras
4550# Specifies a string to be append to request line format for the
4551# rewriter helper. "Quoted" format values may contain spaces and
4552# logformat %macros. In theory, any logformat %macro can be used.
4553# In practice, a %macro expands as a dash (-) if the helper request is
4554# sent before the required macro information is available to Squid.
4555#Default:
4556# url_rewrite_extras "%>a/%>A %un %>rm myip=%la myport=%lp"
4557
4558# OPTIONS FOR STORE ID
4559# -----------------------------------------------------------------------------
4560
4561# TAG: store_id_program
4562# Specify the location of the executable StoreID helper to use.
4563# Since they can perform almost any function there isn't one included.
4564#
4565# For each requested URL, the helper will receive one line with the format
4566#
4567# [channel-ID <SP>] URL [<SP> extras]<NL>
4568#
4569#
4570# After processing the request the helper must reply using the following format:
4571#
4572# [channel-ID <SP>] result [<SP> kv-pairs]
4573#
4574# The result code can be:
4575#
4576# OK store-id="..."
4577# Use the StoreID supplied in 'store-id='.
4578#
4579# ERR
4580# The default is to use HTTP request URL as the store ID.
4581#
4582# BH
4583# An internal error occured in the helper, preventing
4584# a result being identified.
4585#
4586# In addition to the above kv-pairs Squid also understands the following
4587# optional kv-pairs received from URL rewriters:
4588# clt_conn_tag=TAG
4589# Associates a TAG with the client TCP connection.
4590# Please see url_rewrite_program related documentation for this
4591# kv-pair
4592#
4593# Helper programs should be prepared to receive and possibly ignore
4594# additional whitespace-separated tokens on each input line.
4595#
4596# When using the concurrency= option the protocol is changed by
4597# introducing a query channel tag in front of the request/response.
4598# The query channel tag is a number between 0 and concurrency-1.
4599# This value must be echoed back unchanged to Squid as the first part
4600# of the response relating to its request.
4601#
4602# NOTE: when using StoreID refresh_pattern will apply to the StoreID
4603# returned from the helper and not the URL.
4604#
4605# WARNING: Wrong StoreID value returned by a careless helper may result
4606# in the wrong cached response returned to the user.
4607#
4608# By default, a StoreID helper is not used.
4609#Default:
4610# none
4611
4612# TAG: store_id_extras
4613# Specifies a string to be append to request line format for the
4614# StoreId helper. "Quoted" format values may contain spaces and
4615# logformat %macros. In theory, any logformat %macro can be used.
4616# In practice, a %macro expands as a dash (-) if the helper request is
4617# sent before the required macro information is available to Squid.
4618#Default:
4619# store_id_extras "%>a/%>A %un %>rm myip=%la myport=%lp"
4620
4621# TAG: store_id_children
4622# The maximum number of StoreID helper processes to spawn. If you limit
4623# it too few Squid will have to wait for them to process a backlog of
4624# requests, slowing it down. If you allow too many they will use RAM
4625# and other system resources noticably.
4626#
4627# The startup= and idle= options allow some measure of skew in your
4628# tuning.
4629#
4630# startup=
4631#
4632# Sets a minimum of how many processes are to be spawned when Squid
4633# starts or reconfigures. When set to zero the first request will
4634# cause spawning of the first child process to handle it.
4635#
4636# Starting too few will cause an initial slowdown in traffic as Squid
4637# attempts to simultaneously spawn enough processes to cope.
4638#
4639# idle=
4640#
4641# Sets a minimum of how many processes Squid is to try and keep available
4642# at all times. When traffic begins to rise above what the existing
4643# processes can handle this many more will be spawned up to the maximum
4644# configured. A minimum setting of 1 is required.
4645#
4646# concurrency=
4647#
4648# The number of requests each storeID helper can handle in
4649# parallel. Defaults to 0 which indicates the helper
4650# is a old-style single threaded program.
4651#
4652# When this directive is set to a value >= 1 then the protocol
4653# used to communicate with the helper is modified to include
4654# an ID in front of the request/response. The ID from the request
4655# must be echoed back with the response to that request.
4656#Default:
4657# store_id_children 20 startup=0 idle=1 concurrency=0
4658
4659# TAG: store_id_access
4660# If defined, this access list specifies which requests are
4661# sent to the StoreID processes. By default all requests
4662# are sent.
4663#
4664# This clause supports both fast and slow acl types.
4665# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4666#Default:
4667# Allow, unless rules exist in squid.conf.
4668
4669# TAG: store_id_bypass
4670# When this is 'on', a request will not go through the
4671# helper if all helpers are busy. If this is 'off'
4672# and the helper queue grows too large, Squid will exit
4673# with a FATAL error and ask you to increase the number of
4674# helpers. You should only enable this if the helperss
4675# are not critical to your caching system. If you use
4676# helpers for critical caching components, and you enable this
4677# option, users may not get objects from cache.
4678#Default:
4679# store_id_bypass on
4680
4681# OPTIONS FOR TUNING THE CACHE
4682# -----------------------------------------------------------------------------
4683
4684# TAG: cache
4685# Requests denied by this directive will not be served from the cache
4686# and their responses will not be stored in the cache. This directive
4687# has no effect on other transactions and on already cached responses.
4688#
4689# This clause supports both fast and slow acl types.
4690# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4691#
4692# This and the two other similar caching directives listed below are
4693# checked at different transaction processing stages, have different
4694# access to response information, affect different cache operations,
4695# and differ in slow ACLs support:
4696#
4697# * cache: Checked before Squid makes a hit/miss determination.
4698# No access to reply information!
4699# Denies both serving a hit and storing a miss.
4700# Supports both fast and slow ACLs.
4701# * send_hit: Checked after a hit was detected.
4702# Has access to reply (hit) information.
4703# Denies serving a hit only.
4704# Supports fast ACLs only.
4705# * store_miss: Checked before storing a cachable miss.
4706# Has access to reply (miss) information.
4707# Denies storing a miss only.
4708# Supports fast ACLs only.
4709#
4710# If you are not sure which of the three directives to use, apply the
4711# following decision logic:
4712#
4713# * If your ACL(s) are of slow type _and_ need response info, redesign.
4714# Squid does not support that particular combination at this time.
4715# Otherwise:
4716# * If your directive ACL(s) are of slow type, use "cache"; and/or
4717# * if your directive ACL(s) need no response info, use "cache".
4718# Otherwise:
4719# * If you do not want the response cached, use store_miss; and/or
4720# * if you do not want a hit on a cached response, use send_hit.
4721#Default:
4722# By default, this directive is unused and has no effect.
4723
4724# TAG: send_hit
4725# Responses denied by this directive will not be served from the cache
4726# (but may still be cached, see store_miss). This directive has no
4727# effect on the responses it allows and on the cached objects.
4728#
4729# Please see the "cache" directive for a summary of differences among
4730# store_miss, send_hit, and cache directives.
4731#
4732# Unlike the "cache" directive, send_hit only supports fast acl
4733# types. See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4734#
4735# For example:
4736#
4737# # apply custom Store ID mapping to some URLs
4738# acl MapMe dstdomain .c.example.com
4739# store_id_program ...
4740# store_id_access allow MapMe
4741#
4742# # but prevent caching of special responses
4743# # such as 302 redirects that cause StoreID loops
4744# acl Ordinary http_status 200-299
4745# store_miss deny MapMe !Ordinary
4746#
4747# # and do not serve any previously stored special responses
4748# # from the cache (in case they were already cached before
4749# # the above store_miss rule was in effect).
4750# send_hit deny MapMe !Ordinary
4751#Default:
4752# By default, this directive is unused and has no effect.
4753
4754# TAG: store_miss
4755# Responses denied by this directive will not be cached (but may still
4756# be served from the cache, see send_hit). This directive has no
4757# effect on the responses it allows and on the already cached responses.
4758#
4759# Please see the "cache" directive for a summary of differences among
4760# store_miss, send_hit, and cache directives. See the
4761# send_hit directive for a usage example.
4762#
4763# Unlike the "cache" directive, store_miss only supports fast acl
4764# types. See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4765#Default:
4766# By default, this directive is unused and has no effect.
4767
4768# TAG: max_stale time-units
4769# This option puts an upper limit on how stale content Squid
4770# will serve from the cache if cache validation fails.
4771# Can be overriden by the refresh_pattern max-stale option.
4772#Default:
4773# max_stale 1 week
4774
4775# TAG: refresh_pattern
4776# usage: refresh_pattern [-i] regex min percent max [options]
4777#
4778# By default, regular expressions are CASE-SENSITIVE. To make
4779# them case-insensitive, use the -i option.
4780#
4781# 'Min' is the time (in minutes) an object without an explicit
4782# expiry time should be considered fresh. The recommended
4783# value is 0, any higher values may cause dynamic applications
4784# to be erroneously cached unless the application designer
4785# has taken the appropriate actions.
4786#
4787# 'Percent' is a percentage of the objects age (time since last
4788# modification age) an object without explicit expiry time
4789# will be considered fresh.
4790#
4791# 'Max' is an upper limit on how long objects without an explicit
4792# expiry time will be considered fresh. The value is also used
4793# to form Cache-Control: max-age header for a request sent from
4794# Squid to origin/parent.
4795#
4796# options: override-expire
4797# override-lastmod
4798# reload-into-ims
4799# ignore-reload
4800# ignore-no-store
4801# ignore-must-revalidate
4802# ignore-private
4803# ignore-auth
4804# max-stale=NN
4805# refresh-ims
4806# store-stale
4807#
4808# override-expire enforces min age even if the server
4809# sent an explicit expiry time (e.g., with the
4810# Expires: header or Cache-Control: max-age). Doing this
4811# VIOLATES the HTTP standard. Enabling this feature
4812# could make you liable for problems which it causes.
4813#
4814# Note: override-expire does not enforce staleness - it only extends
4815# freshness / min. If the server returns a Expires time which
4816# is longer than your max time, Squid will still consider
4817# the object fresh for that period of time.
4818#
4819# override-lastmod enforces min age even on objects
4820# that were modified recently.
4821#
4822# reload-into-ims changes a client no-cache or ``reload''
4823# request for a cached entry into a conditional request using
4824# If-Modified-Since and/or If-None-Match headers, provided the
4825# cached entry has a Last-Modified and/or a strong ETag header.
4826# Doing this VIOLATES the HTTP standard. Enabling this feature
4827# could make you liable for problems which it causes.
4828#
4829# ignore-reload ignores a client no-cache or ``reload''
4830# header. Doing this VIOLATES the HTTP standard. Enabling
4831# this feature could make you liable for problems which
4832# it causes.
4833#
4834# ignore-no-store ignores any ``Cache-control: no-store''
4835# headers received from a server. Doing this VIOLATES
4836# the HTTP standard. Enabling this feature could make you
4837# liable for problems which it causes.
4838#
4839# ignore-must-revalidate ignores any ``Cache-Control: must-revalidate``
4840# headers received from a server. Doing this VIOLATES
4841# the HTTP standard. Enabling this feature could make you
4842# liable for problems which it causes.
4843#
4844# ignore-private ignores any ``Cache-control: private''
4845# headers received from a server. Doing this VIOLATES
4846# the HTTP standard. Enabling this feature could make you
4847# liable for problems which it causes.
4848#
4849# ignore-auth caches responses to requests with authorization,
4850# as if the originserver had sent ``Cache-control: public''
4851# in the response header. Doing this VIOLATES the HTTP standard.
4852# Enabling this feature could make you liable for problems which
4853# it causes.
4854#
4855# refresh-ims causes squid to contact the origin server
4856# when a client issues an If-Modified-Since request. This
4857# ensures that the client will receive an updated version
4858# if one is available.
4859#
4860# store-stale stores responses even if they don't have explicit
4861# freshness or a validator (i.e., Last-Modified or an ETag)
4862# present, or if they're already stale. By default, Squid will
4863# not cache such responses because they usually can't be
4864# reused. Note that such responses will be stale by default.
4865#
4866# max-stale=NN provide a maximum staleness factor. Squid won't
4867# serve objects more stale than this even if it failed to
4868# validate the object. Default: use the max_stale global limit.
4869#
4870# Basically a cached object is:
4871#
4872# FRESH if expire > now, else STALE
4873# STALE if age > max
4874# FRESH if lm-factor < percent, else STALE
4875# FRESH if age < min
4876# else STALE
4877#
4878# The refresh_pattern lines are checked in the order listed here.
4879# The first entry which matches is used. If none of the entries
4880# match the default will be used.
4881#
4882# Note, you must uncomment all the default lines if you want
4883# to change one. The default setting is only active if none is
4884# used.
4885#
4886#
4887
4888#
4889# Add any of your own refresh_pattern entries above these.
4890#
4891refresh_pattern ^ftp: 1440 20% 10080
4892refresh_pattern ^gopher: 1440 0% 1440
4893refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
4894refresh_pattern (Release|Packages(.gz)*)$ 0 20% 2880
4895# example lin deb packages
4896#refresh_pattern (\.deb|\.udeb)$ 129600 100% 129600
4897refresh_pattern . 0 20% 4320
4898
4899# TAG: quick_abort_min (KB)
4900#Default:
4901# quick_abort_min 16 KB
4902
4903# TAG: quick_abort_max (KB)
4904#Default:
4905# quick_abort_max 16 KB
4906
4907# TAG: quick_abort_pct (percent)
4908# The cache by default continues downloading aborted requests
4909# which are almost completed (less than 16 KB remaining). This
4910# may be undesirable on slow (e.g. SLIP) links and/or very busy
4911# caches. Impatient users may tie up file descriptors and
4912# bandwidth by repeatedly requesting and immediately aborting
4913# downloads.
4914#
4915# When the user aborts a request, Squid will check the
4916# quick_abort values to the amount of data transferred until
4917# then.
4918#
4919# If the transfer has less than 'quick_abort_min' KB remaining,
4920# it will finish the retrieval.
4921#
4922# If the transfer has more than 'quick_abort_max' KB remaining,
4923# it will abort the retrieval.
4924#
4925# If more than 'quick_abort_pct' of the transfer has completed,
4926# it will finish the retrieval.
4927#
4928# If you do not want any retrieval to continue after the client
4929# has aborted, set both 'quick_abort_min' and 'quick_abort_max'
4930# to '0 KB'.
4931#
4932# If you want retrievals to always continue if they are being
4933# cached set 'quick_abort_min' to '-1 KB'.
4934#Default:
4935# quick_abort_pct 95
4936
4937# TAG: read_ahead_gap buffer-size
4938# The amount of data the cache will buffer ahead of what has been
4939# sent to the client when retrieving an object from another server.
4940#Default:
4941# read_ahead_gap 16 KB
4942
4943# TAG: negative_ttl time-units
4944# Set the Default Time-to-Live (TTL) for failed requests.
4945# Certain types of failures (such as "connection refused" and
4946# "404 Not Found") are able to be negatively-cached for a short time.
4947# Modern web servers should provide Expires: header, however if they
4948# do not this can provide a minimum TTL.
4949# The default is not to cache errors with unknown expiry details.
4950#
4951# Note that this is different from negative caching of DNS lookups.
4952#
4953# WARNING: Doing this VIOLATES the HTTP standard. Enabling
4954# this feature could make you liable for problems which it
4955# causes.
4956#Default:
4957# negative_ttl 0 seconds
4958
4959# TAG: positive_dns_ttl time-units
4960# Upper limit on how long Squid will cache positive DNS responses.
4961# Default is 6 hours (360 minutes). This directive must be set
4962# larger than negative_dns_ttl.
4963#Default:
4964# positive_dns_ttl 6 hours
4965
4966# TAG: negative_dns_ttl time-units
4967# Time-to-Live (TTL) for negative caching of failed DNS lookups.
4968# This also sets the lower cache limit on positive lookups.
4969# Minimum value is 1 second, and it is not recommendable to go
4970# much below 10 seconds.
4971#Default:
4972# negative_dns_ttl 1 minutes
4973
4974# TAG: range_offset_limit size [acl acl...]
4975# usage: (size) [units] [[!]aclname]
4976#
4977# Sets an upper limit on how far (number of bytes) into the file
4978# a Range request may be to cause Squid to prefetch the whole file.
4979# If beyond this limit, Squid forwards the Range request as it is and
4980# the result is NOT cached.
4981#
4982# This is to stop a far ahead range request (lets say start at 17MB)
4983# from making Squid fetch the whole object up to that point before
4984# sending anything to the client.
4985#
4986# Multiple range_offset_limit lines may be specified, and they will
4987# be searched from top to bottom on each request until a match is found.
4988# The first match found will be used. If no line matches a request, the
4989# default limit of 0 bytes will be used.
4990#
4991# 'size' is the limit specified as a number of units.
4992#
4993# 'units' specifies whether to use bytes, KB, MB, etc.
4994# If no units are specified bytes are assumed.
4995#
4996# A size of 0 causes Squid to never fetch more than the
4997# client requested. (default)
4998#
4999# A size of 'none' causes Squid to always fetch the object from the
5000# beginning so it may cache the result. (2.0 style)
5001#
5002# 'aclname' is the name of a defined ACL.
5003#
5004# NP: Using 'none' as the byte value here will override any quick_abort settings
5005# that may otherwise apply to the range request. The range request will
5006# be fully fetched from start to finish regardless of the client
5007# actions. This affects bandwidth usage.
5008#Default:
5009# none
5010
5011# TAG: minimum_expiry_time (seconds)
5012# The minimum caching time according to (Expires - Date)
5013# headers Squid honors if the object can't be revalidated.
5014# The default is 60 seconds.
5015#
5016# In reverse proxy environments it might be desirable to honor
5017# shorter object lifetimes. It is most likely better to make
5018# your server return a meaningful Last-Modified header however.
5019#
5020# In ESI environments where page fragments often have short
5021# lifetimes, this will often be best set to 0.
5022#Default:
5023# minimum_expiry_time 60 seconds
5024
5025# TAG: store_avg_object_size (bytes)
5026# Average object size, used to estimate number of objects your
5027# cache can hold. The default is 13 KB.
5028#
5029# This is used to pre-seed the cache index memory allocation to
5030# reduce expensive reallocate operations while handling clients
5031# traffic. Too-large values may result in memory allocation during
5032# peak traffic, too-small values will result in wasted memory.
5033#
5034# Check the cache manager 'info' report metrics for the real
5035# object sizes seen by your Squid before tuning this.
5036#Default:
5037# store_avg_object_size 13 KB
5038
5039# TAG: store_objects_per_bucket
5040# Target number of objects per bucket in the store hash table.
5041# Lowering this value increases the total number of buckets and
5042# also the storage maintenance rate. The default is 20.
5043#Default:
5044# store_objects_per_bucket 20
5045
5046# HTTP OPTIONS
5047# -----------------------------------------------------------------------------
5048
5049# TAG: request_header_max_size (KB)
5050# This specifies the maximum size for HTTP headers in a request.
5051# Request headers are usually relatively small (about 512 bytes).
5052# Placing a limit on the request header size will catch certain
5053# bugs (for example with persistent connections) and possibly
5054# buffer-overflow or denial-of-service attacks.
5055#Default:
5056# request_header_max_size 64 KB
5057
5058# TAG: reply_header_max_size (KB)
5059# This specifies the maximum size for HTTP headers in a reply.
5060# Reply headers are usually relatively small (about 512 bytes).
5061# Placing a limit on the reply header size will catch certain
5062# bugs (for example with persistent connections) and possibly
5063# buffer-overflow or denial-of-service attacks.
5064#Default:
5065# reply_header_max_size 64 KB
5066
5067# TAG: request_body_max_size (bytes)
5068# This specifies the maximum size for an HTTP request body.
5069# In other words, the maximum size of a PUT/POST request.
5070# A user who attempts to send a request with a body larger
5071# than this limit receives an "Invalid Request" error message.
5072# If you set this parameter to a zero (the default), there will
5073# be no limit imposed.
5074#
5075# See also client_request_buffer_max_size for an alternative
5076# limitation on client uploads which can be configured.
5077#Default:
5078# No limit.
5079
5080# TAG: client_request_buffer_max_size (bytes)
5081# This specifies the maximum buffer size of a client request.
5082# It prevents squid eating too much memory when somebody uploads
5083# a large file.
5084#Default:
5085# client_request_buffer_max_size 512 KB
5086
5087# TAG: broken_posts
5088# A list of ACL elements which, if matched, causes Squid to send
5089# an extra CRLF pair after the body of a PUT/POST request.
5090#
5091# Some HTTP servers has broken implementations of PUT/POST,
5092# and rely on an extra CRLF pair sent by some WWW clients.
5093#
5094# Quote from RFC2616 section 4.1 on this matter:
5095#
5096# Note: certain buggy HTTP/1.0 client implementations generate an
5097# extra CRLF's after a POST request. To restate what is explicitly
5098# forbidden by the BNF, an HTTP/1.1 client must not preface or follow
5099# a request with an extra CRLF.
5100#
5101# This clause only supports fast acl types.
5102# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5103#
5104#Example:
5105# acl buggy_server url_regex ^http://....
5106# broken_posts allow buggy_server
5107#Default:
5108# Obey RFC 2616.
5109
5110# TAG: adaptation_uses_indirect_client on|off
5111# Controls whether the indirect client IP address (instead of the direct
5112# client IP address) is passed to adaptation services.
5113#
5114# See also: follow_x_forwarded_for adaptation_send_client_ip
5115#Default:
5116# adaptation_uses_indirect_client on
5117
5118# TAG: via on|off
5119# If set (default), Squid will include a Via header in requests and
5120# replies as required by RFC2616.
5121#Default:
5122# via on
5123
5124# TAG: ie_refresh on|off
5125# Microsoft Internet Explorer up until version 5.5 Service
5126# Pack 1 has an issue with transparent proxies, wherein it
5127# is impossible to force a refresh. Turning this on provides
5128# a partial fix to the problem, by causing all IMS-REFRESH
5129# requests from older IE versions to check the origin server
5130# for fresh content. This reduces hit ratio by some amount
5131# (~10% in my experience), but allows users to actually get
5132# fresh content when they want it. Note because Squid
5133# cannot tell if the user is using 5.5 or 5.5SP1, the behavior
5134# of 5.5 is unchanged from old versions of Squid (i.e. a
5135# forced refresh is impossible). Newer versions of IE will,
5136# hopefully, continue to have the new behavior and will be
5137# handled based on that assumption. This option defaults to
5138# the old Squid behavior, which is better for hit ratios but
5139# worse for clients using IE, if they need to be able to
5140# force fresh content.
5141#Default:
5142# ie_refresh off
5143
5144# TAG: vary_ignore_expire on|off
5145# Many HTTP servers supporting Vary gives such objects
5146# immediate expiry time with no cache-control header
5147# when requested by a HTTP/1.0 client. This option
5148# enables Squid to ignore such expiry times until
5149# HTTP/1.1 is fully implemented.
5150#
5151# WARNING: If turned on this may eventually cause some
5152# varying objects not intended for caching to get cached.
5153#Default:
5154# vary_ignore_expire off
5155
5156# TAG: request_entities
5157# Squid defaults to deny GET and HEAD requests with request entities,
5158# as the meaning of such requests are undefined in the HTTP standard
5159# even if not explicitly forbidden.
5160#
5161# Set this directive to on if you have clients which insists
5162# on sending request entities in GET or HEAD requests. But be warned
5163# that there is server software (both proxies and web servers) which
5164# can fail to properly process this kind of request which may make you
5165# vulnerable to cache pollution attacks if enabled.
5166#Default:
5167# request_entities off
5168
5169# TAG: request_header_access
5170# Usage: request_header_access header_name allow|deny [!]aclname ...
5171#
5172# WARNING: Doing this VIOLATES the HTTP standard. Enabling
5173# this feature could make you liable for problems which it
5174# causes.
5175#
5176# This option replaces the old 'anonymize_headers' and the
5177# older 'http_anonymizer' option with something that is much
5178# more configurable. A list of ACLs for each header name allows
5179# removal of specific header fields under specific conditions.
5180#
5181# This option only applies to outgoing HTTP request headers (i.e.,
5182# headers sent by Squid to the next HTTP hop such as a cache peer
5183# or an origin server). The option has no effect during cache hit
5184# detection. The equivalent adaptation vectoring point in ICAP
5185# terminology is post-cache REQMOD.
5186#
5187# The option is applied to individual outgoing request header
5188# fields. For each request header field F, Squid uses the first
5189# qualifying sets of request_header_access rules:
5190#
5191# 1. Rules with header_name equal to F's name.
5192# 2. Rules with header_name 'Other', provided F's name is not
5193# on the hard-coded list of commonly used HTTP header names.
5194# 3. Rules with header_name 'All'.
5195#
5196# Within that qualifying rule set, rule ACLs are checked as usual.
5197# If ACLs of an "allow" rule match, the header field is allowed to
5198# go through as is. If ACLs of a "deny" rule match, the header is
5199# removed and request_header_replace is then checked to identify
5200# if the removed header has a replacement. If no rules within the
5201# set have matching ACLs, the header field is left as is.
5202#
5203# For example, to achieve the same behavior as the old
5204# 'http_anonymizer standard' option, you should use:
5205#
5206# request_header_access From deny all
5207# request_header_access Referer deny all
5208# request_header_access User-Agent deny all
5209#
5210# Or, to reproduce the old 'http_anonymizer paranoid' feature
5211# you should use:
5212#
5213# request_header_access Authorization allow all
5214# request_header_access Proxy-Authorization allow all
5215# request_header_access Cache-Control allow all
5216# request_header_access Content-Length allow all
5217# request_header_access Content-Type allow all
5218# request_header_access Date allow all
5219# request_header_access Host allow all
5220# request_header_access If-Modified-Since allow all
5221# request_header_access Pragma allow all
5222# request_header_access Accept allow all
5223# request_header_access Accept-Charset allow all
5224# request_header_access Accept-Encoding allow all
5225# request_header_access Accept-Language allow all
5226# request_header_access Connection allow all
5227# request_header_access All deny all
5228#
5229# HTTP reply headers are controlled with the reply_header_access directive.
5230#
5231# By default, all headers are allowed (no anonymizing is performed).
5232#Default:
5233# No limits.
5234
5235# TAG: reply_header_access
5236# Usage: reply_header_access header_name allow|deny [!]aclname ...
5237#
5238# WARNING: Doing this VIOLATES the HTTP standard. Enabling
5239# this feature could make you liable for problems which it
5240# causes.
5241#
5242# This option only applies to reply headers, i.e., from the
5243# server to the client.
5244#
5245# This is the same as request_header_access, but in the other
5246# direction. Please see request_header_access for detailed
5247# documentation.
5248#
5249# For example, to achieve the same behavior as the old
5250# 'http_anonymizer standard' option, you should use:
5251#
5252# reply_header_access Server deny all
5253# reply_header_access WWW-Authenticate deny all
5254# reply_header_access Link deny all
5255#
5256# Or, to reproduce the old 'http_anonymizer paranoid' feature
5257# you should use:
5258#
5259# reply_header_access Allow allow all
5260# reply_header_access WWW-Authenticate allow all
5261# reply_header_access Proxy-Authenticate allow all
5262# reply_header_access Cache-Control allow all
5263# reply_header_access Content-Encoding allow all
5264# reply_header_access Content-Length allow all
5265# reply_header_access Content-Type allow all
5266# reply_header_access Date allow all
5267# reply_header_access Expires allow all
5268# reply_header_access Last-Modified allow all
5269# reply_header_access Location allow all
5270# reply_header_access Pragma allow all
5271# reply_header_access Content-Language allow all
5272# reply_header_access Retry-After allow all
5273# reply_header_access Title allow all
5274# reply_header_access Content-Disposition allow all
5275# reply_header_access Connection allow all
5276# reply_header_access All deny all
5277#
5278# HTTP request headers are controlled with the request_header_access directive.
5279#
5280# By default, all headers are allowed (no anonymizing is
5281# performed).
5282#Default:
5283# No limits.
5284
5285# TAG: request_header_replace
5286# Usage: request_header_replace header_name message
5287# Example: request_header_replace User-Agent Nutscrape/1.0 (CP/M; 8-bit)
5288#
5289# This option allows you to change the contents of headers
5290# denied with request_header_access above, by replacing them
5291# with some fixed string.
5292#
5293# This only applies to request headers, not reply headers.
5294#
5295# By default, headers are removed if denied.
5296#Default:
5297# none
5298
5299# TAG: reply_header_replace
5300# Usage: reply_header_replace header_name message
5301# Example: reply_header_replace Server Foo/1.0
5302#
5303# This option allows you to change the contents of headers
5304# denied with reply_header_access above, by replacing them
5305# with some fixed string.
5306#
5307# This only applies to reply headers, not request headers.
5308#
5309# By default, headers are removed if denied.
5310#Default:
5311# none
5312
5313# TAG: request_header_add
5314# Usage: request_header_add field-name field-value acl1 [acl2] ...
5315# Example: request_header_add X-Client-CA "CA=%ssl::>cert_issuer" all
5316#
5317# This option adds header fields to outgoing HTTP requests (i.e.,
5318# request headers sent by Squid to the next HTTP hop such as a
5319# cache peer or an origin server). The option has no effect during
5320# cache hit detection. The equivalent adaptation vectoring point
5321# in ICAP terminology is post-cache REQMOD.
5322#
5323# Field-name is a token specifying an HTTP header name. If a
5324# standard HTTP header name is used, Squid does not check whether
5325# the new header conflicts with any existing headers or violates
5326# HTTP rules. If the request to be modified already contains a
5327# field with the same name, the old field is preserved but the
5328# header field values are not merged.
5329#
5330# Field-value is either a token or a quoted string. If quoted
5331# string format is used, then the surrounding quotes are removed
5332# while escape sequences and %macros are processed.
5333#
5334# In theory, all of the logformat codes can be used as %macros.
5335# However, unlike logging (which happens at the very end of
5336# transaction lifetime), the transaction may not yet have enough
5337# information to expand a macro when the new header value is needed.
5338# And some information may already be available to Squid but not yet
5339# committed where the macro expansion code can access it (report
5340# such instances!). The macro will be expanded into a single dash
5341# ('-') in such cases. Not all macros have been tested.
5342#
5343# One or more Squid ACLs may be specified to restrict header
5344# injection to matching requests. As always in squid.conf, all
5345# ACLs in an option ACL list must be satisfied for the insertion
5346# to happen. The request_header_add option supports fast ACLs
5347# only.
5348#Default:
5349# none
5350
5351# TAG: note
5352# This option used to log custom information about the master
5353# transaction. For example, an admin may configure Squid to log
5354# which "user group" the transaction belongs to, where "user group"
5355# will be determined based on a set of ACLs and not [just]
5356# authentication information.
5357# Values of key/value pairs can be logged using %{key}note macros:
5358#
5359# note key value acl ...
5360# logformat myFormat ... %{key}note ...
5361#Default:
5362# none
5363
5364# TAG: relaxed_header_parser on|off|warn
5365# In the default "on" setting Squid accepts certain forms
5366# of non-compliant HTTP messages where it is unambiguous
5367# what the sending application intended even if the message
5368# is not correctly formatted. The messages is then normalized
5369# to the correct form when forwarded by Squid.
5370#
5371# If set to "warn" then a warning will be emitted in cache.log
5372# each time such HTTP error is encountered.
5373#
5374# If set to "off" then such HTTP errors will cause the request
5375# or response to be rejected.
5376#Default:
5377# relaxed_header_parser on
5378
5379# TAG: collapsed_forwarding (on|off)
5380# When enabled, instead of forwarding each concurrent request for
5381# the same URL, Squid just sends the first of them. The other, so
5382# called "collapsed" requests, wait for the response to the first
5383# request and, if it happens to be cachable, use that response.
5384# Here, "concurrent requests" means "received after the first
5385# request headers were parsed and before the corresponding response
5386# headers were parsed".
5387#
5388# This feature is disabled by default: enabling collapsed
5389# forwarding needlessly delays forwarding requests that look
5390# cachable (when they are collapsed) but then need to be forwarded
5391# individually anyway because they end up being for uncachable
5392# content. However, in some cases, such as acceleration of highly
5393# cachable content with periodic or grouped expiration times, the
5394# gains from collapsing [large volumes of simultaneous refresh
5395# requests] outweigh losses from such delays.
5396#
5397# Squid collapses two kinds of requests: regular client requests
5398# received on one of the listening ports and internal "cache
5399# revalidation" requests which are triggered by those regular
5400# requests hitting a stale cached object. Revalidation collapsing
5401# is currently disabled for Squid instances containing SMP-aware
5402# disk or memory caches and for Vary-controlled cached objects.
5403#Default:
5404# collapsed_forwarding off
5405
5406# TIMEOUTS
5407# -----------------------------------------------------------------------------
5408
5409# TAG: forward_timeout time-units
5410# This parameter specifies how long Squid should at most attempt in
5411# finding a forwarding path for the request before giving up.
5412#Default:
5413# forward_timeout 4 minutes
5414
5415# TAG: connect_timeout time-units
5416# This parameter specifies how long to wait for the TCP connect to
5417# the requested server or peer to complete before Squid should
5418# attempt to find another path where to forward the request.
5419#Default:
5420# connect_timeout 1 minute
5421
5422# TAG: peer_connect_timeout time-units
5423# This parameter specifies how long to wait for a pending TCP
5424# connection to a peer cache. The default is 30 seconds. You
5425# may also set different timeout values for individual neighbors
5426# with the 'connect-timeout' option on a 'cache_peer' line.
5427#Default:
5428# peer_connect_timeout 30 seconds
5429
5430# TAG: read_timeout time-units
5431# Applied on peer server connections.
5432#
5433# After each successful read(), the timeout will be extended by this
5434# amount. If no data is read again after this amount of time,
5435# the request is aborted and logged with ERR_READ_TIMEOUT.
5436#
5437# The default is 15 minutes.
5438#Default:
5439# read_timeout 15 minutes
5440
5441# TAG: write_timeout time-units
5442# This timeout is tracked for all connections that have data
5443# available for writing and are waiting for the socket to become
5444# ready. After each successful write, the timeout is extended by
5445# the configured amount. If Squid has data to write but the
5446# connection is not ready for the configured duration, the
5447# transaction associated with the connection is terminated. The
5448# default is 15 minutes.
5449#Default:
5450# write_timeout 15 minutes
5451
5452# TAG: request_timeout
5453# How long to wait for complete HTTP request headers after initial
5454# connection establishment.
5455#Default:
5456# request_timeout 5 minutes
5457
5458# TAG: client_idle_pconn_timeout
5459# How long to wait for the next HTTP request on a persistent
5460# client connection after the previous request completes.
5461#Default:
5462# client_idle_pconn_timeout 2 minutes
5463
5464# TAG: ftp_client_idle_timeout
5465# How long to wait for an FTP request on a connection to Squid ftp_port.
5466# Many FTP clients do not deal with idle connection closures well,
5467# necessitating a longer default timeout than client_idle_pconn_timeout
5468# used for incoming HTTP requests.
5469#Default:
5470# ftp_client_idle_timeout 30 minutes
5471
5472# TAG: client_lifetime time-units
5473# The maximum amount of time a client (browser) is allowed to
5474# remain connected to the cache process. This protects the Cache
5475# from having a lot of sockets (and hence file descriptors) tied up
5476# in a CLOSE_WAIT state from remote clients that go away without
5477# properly shutting down (either because of a network failure or
5478# because of a poor client implementation). The default is one
5479# day, 1440 minutes.
5480#
5481# NOTE: The default value is intended to be much larger than any
5482# client would ever need to be connected to your cache. You
5483# should probably change client_lifetime only as a last resort.
5484# If you seem to have many client connections tying up
5485# filedescriptors, we recommend first tuning the read_timeout,
5486# request_timeout, persistent_request_timeout and quick_abort values.
5487#Default:
5488# client_lifetime 1 day
5489
5490# TAG: half_closed_clients
5491# Some clients may shutdown the sending side of their TCP
5492# connections, while leaving their receiving sides open. Sometimes,
5493# Squid can not tell the difference between a half-closed and a
5494# fully-closed TCP connection.
5495#
5496# By default, Squid will immediately close client connections when
5497# read(2) returns "no more data to read."
5498#
5499# Change this option to 'on' and Squid will keep open connections
5500# until a read(2) or write(2) on the socket returns an error.
5501# This may show some benefits for reverse proxies. But if not
5502# it is recommended to leave OFF.
5503#Default:
5504# half_closed_clients off
5505
5506# TAG: server_idle_pconn_timeout
5507# Timeout for idle persistent connections to servers and other
5508# proxies.
5509#Default:
5510# server_idle_pconn_timeout 1 minute
5511
5512# TAG: ident_timeout
5513# Maximum time to wait for IDENT lookups to complete.
5514#
5515# If this is too high, and you enabled IDENT lookups from untrusted
5516# users, you might be susceptible to denial-of-service by having
5517# many ident requests going at once.
5518#Default:
5519# ident_timeout 10 seconds
5520
5521# TAG: shutdown_lifetime time-units
5522# When SIGTERM or SIGHUP is received, the cache is put into
5523# "shutdown pending" mode until all active sockets are closed.
5524# This value is the lifetime to set for all open descriptors
5525# during shutdown mode. Any active clients after this many
5526# seconds will receive a 'timeout' message.
5527#Default:
5528# shutdown_lifetime 30 seconds
5529
5530# ADMINISTRATIVE PARAMETERS
5531# -----------------------------------------------------------------------------
5532
5533# TAG: cache_mgr
5534# Email-address of local cache manager who will receive
5535# mail if the cache dies. The default is "webmaster".
5536#Default:
5537# cache_mgr webmaster
5538
5539# TAG: mail_from
5540# From: email-address for mail sent when the cache dies.
5541# The default is to use 'squid@unique_hostname'.
5542#
5543# See also: unique_hostname directive.
5544#Default:
5545# none
5546
5547# TAG: mail_program
5548# Email program used to send mail if the cache dies.
5549# The default is "mail". The specified program must comply
5550# with the standard Unix mail syntax:
5551# mail-program recipient < mailfile
5552#
5553# Optional command line options can be specified.
5554#Default:
5555# mail_program mail
5556
5557# TAG: cache_effective_user
5558# If you start Squid as root, it will change its effective/real
5559# UID/GID to the user specified below. The default is to change
5560# to UID of proxy.
5561# see also; cache_effective_group
5562#Default:
5563# cache_effective_user proxy
5564
5565# TAG: cache_effective_group
5566# Squid sets the GID to the effective user's default group ID
5567# (taken from the password file) and supplementary group list
5568# from the groups membership.
5569#
5570# If you want Squid to run with a specific GID regardless of
5571# the group memberships of the effective user then set this
5572# to the group (or GID) you want Squid to run as. When set
5573# all other group privileges of the effective user are ignored
5574# and only this GID is effective. If Squid is not started as
5575# root the user starting Squid MUST be member of the specified
5576# group.
5577#
5578# This option is not recommended by the Squid Team.
5579# Our preference is for administrators to configure a secure
5580# user account for squid with UID/GID matching system policies.
5581#Default:
5582# Use system group memberships of the cache_effective_user account
5583
5584# TAG: httpd_suppress_version_string on|off
5585# Suppress Squid version string info in HTTP headers and HTML error pages.
5586#Default:
5587# httpd_suppress_version_string off
5588
5589# TAG: visible_hostname
5590# If you want to present a special hostname in error messages, etc,
5591# define this. Otherwise, the return value of gethostname()
5592# will be used. If you have multiple caches in a cluster and
5593# get errors about IP-forwarding you must set them to have individual
5594# names with this setting.
5595#Default:
5596# Automatically detect the system host name
5597visible_hostname "sgp-303.asuscomm.com - p2"
5598
5599# TAG: unique_hostname
5600# If you want to have multiple machines with the same
5601# 'visible_hostname' you must give each machine a different
5602# 'unique_hostname' so forwarding loops can be detected.
5603#Default:
5604# Copy the value from visible_hostname
5605
5606# TAG: hostname_aliases
5607# A list of other DNS names your cache has.
5608#Default:
5609# none
5610
5611# TAG: umask
5612# Minimum umask which should be enforced while the proxy
5613# is running, in addition to the umask set at startup.
5614#
5615# For a traditional octal representation of umasks, start
5616# your value with 0.
5617#Default:
5618# umask 027
5619
5620# OPTIONS FOR THE CACHE REGISTRATION SERVICE
5621# -----------------------------------------------------------------------------
5622#
5623# This section contains parameters for the (optional) cache
5624# announcement service. This service is provided to help
5625# cache administrators locate one another in order to join or
5626# create cache hierarchies.
5627#
5628# An 'announcement' message is sent (via UDP) to the registration
5629# service by Squid. By default, the announcement message is NOT
5630# SENT unless you enable it with 'announce_period' below.
5631#
5632# The announcement message includes your hostname, plus the
5633# following information from this configuration file:
5634#
5635# http_port
5636# icp_port
5637# cache_mgr
5638#
5639# All current information is processed regularly and made
5640# available on the Web at http://www.ircache.net/Cache/Tracker/.
5641
5642# TAG: announce_period
5643# This is how frequently to send cache announcements.
5644#
5645# To enable announcing your cache, just set an announce period.
5646#
5647# Example:
5648# announce_period 1 day
5649#Default:
5650# Announcement messages disabled.
5651
5652# TAG: announce_host
5653# Set the hostname where announce registration messages will be sent.
5654#
5655# See also announce_port and announce_file
5656#Default:
5657# announce_host tracker.ircache.net
5658
5659# TAG: announce_file
5660# The contents of this file will be included in the announce
5661# registration messages.
5662#Default:
5663# none
5664
5665# TAG: announce_port
5666# Set the port where announce registration messages will be sent.
5667#
5668# See also announce_host and announce_file
5669#Default:
5670# announce_port 3131
5671
5672# HTTPD-ACCELERATOR OPTIONS
5673# -----------------------------------------------------------------------------
5674
5675# TAG: httpd_accel_surrogate_id
5676# Surrogates (http://www.esi.org/architecture_spec_1.0.html)
5677# need an identification token to allow control targeting. Because
5678# a farm of surrogates may all perform the same tasks, they may share
5679# an identification token.
5680#Default:
5681# visible_hostname is used if no specific ID is set.
5682
5683# TAG: http_accel_surrogate_remote on|off
5684# Remote surrogates (such as those in a CDN) honour the header
5685# "Surrogate-Control: no-store-remote".
5686#
5687# Set this to on to have squid behave as a remote surrogate.
5688#Default:
5689# http_accel_surrogate_remote off
5690
5691# TAG: esi_parser libxml2|expat|custom
5692# ESI markup is not strictly XML compatible. The custom ESI parser
5693# will give higher performance, but cannot handle non ASCII character
5694# encodings.
5695#Default:
5696# esi_parser custom
5697
5698# DELAY POOL PARAMETERS
5699# -----------------------------------------------------------------------------
5700
5701# TAG: delay_pools
5702# This represents the number of delay pools to be used. For example,
5703# if you have one class 2 delay pool and one class 3 delays pool, you
5704# have a total of 2 delay pools.
5705#
5706# See also delay_parameters, delay_class, delay_access for pool
5707# configuration details.
5708#Default:
5709# delay_pools 0
5710
5711# TAG: delay_class
5712# This defines the class of each delay pool. There must be exactly one
5713# delay_class line for each delay pool. For example, to define two
5714# delay pools, one of class 2 and one of class 3, the settings above
5715# and here would be:
5716#
5717# Example:
5718# delay_pools 4 # 4 delay pools
5719# delay_class 1 2 # pool 1 is a class 2 pool
5720# delay_class 2 3 # pool 2 is a class 3 pool
5721# delay_class 3 4 # pool 3 is a class 4 pool
5722# delay_class 4 5 # pool 4 is a class 5 pool
5723#
5724# The delay pool classes are:
5725#
5726# class 1 Everything is limited by a single aggregate
5727# bucket.
5728#
5729# class 2 Everything is limited by a single aggregate
5730# bucket as well as an "individual" bucket chosen
5731# from bits 25 through 32 of the IPv4 address.
5732#
5733# class 3 Everything is limited by a single aggregate
5734# bucket as well as a "network" bucket chosen
5735# from bits 17 through 24 of the IP address and a
5736# "individual" bucket chosen from bits 17 through
5737# 32 of the IPv4 address.
5738#
5739# class 4 Everything in a class 3 delay pool, with an
5740# additional limit on a per user basis. This
5741# only takes effect if the username is established
5742# in advance - by forcing authentication in your
5743# http_access rules.
5744#
5745# class 5 Requests are grouped according their tag (see
5746# external_acl's tag= reply).
5747#
5748#
5749# Each pool also requires a delay_parameters directive to configure the pool size
5750# and speed limits used whenever the pool is applied to a request. Along with
5751# a set of delay_access directives to determine when it is used.
5752#
5753# NOTE: If an IP address is a.b.c.d
5754# -> bits 25 through 32 are "d"
5755# -> bits 17 through 24 are "c"
5756# -> bits 17 through 32 are "c * 256 + d"
5757#
5758# NOTE-2: Due to the use of bitmasks in class 2,3,4 pools they only apply to
5759# IPv4 traffic. Class 1 and 5 pools may be used with IPv6 traffic.
5760#
5761# This clause only supports fast acl types.
5762# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5763#
5764# See also delay_parameters and delay_access.
5765#Default:
5766# none
5767
5768# TAG: delay_access
5769# This is used to determine which delay pool a request falls into.
5770#
5771# delay_access is sorted per pool and the matching starts with pool 1,
5772# then pool 2, ..., and finally pool N. The first delay pool where the
5773# request is allowed is selected for the request. If it does not allow
5774# the request to any pool then the request is not delayed (default).
5775#
5776# For example, if you want some_big_clients in delay
5777# pool 1 and lotsa_little_clients in delay pool 2:
5778#
5779# delay_access 1 allow some_big_clients
5780# delay_access 1 deny all
5781# delay_access 2 allow lotsa_little_clients
5782# delay_access 2 deny all
5783# delay_access 3 allow authenticated_clients
5784#
5785# See also delay_parameters and delay_class.
5786#
5787#Default:
5788# Deny using the pool, unless allow rules exist in squid.conf for the pool.
5789
5790# TAG: delay_parameters
5791# This defines the parameters for a delay pool. Each delay pool has
5792# a number of "buckets" associated with it, as explained in the
5793# description of delay_class.
5794#
5795# For a class 1 delay pool, the syntax is:
5796# delay_class pool 1
5797# delay_parameters pool aggregate
5798#
5799# For a class 2 delay pool:
5800# delay_class pool 2
5801# delay_parameters pool aggregate individual
5802#
5803# For a class 3 delay pool:
5804# delay_class pool 3
5805# delay_parameters pool aggregate network individual
5806#
5807# For a class 4 delay pool:
5808# delay_class pool 4
5809# delay_parameters pool aggregate network individual user
5810#
5811# For a class 5 delay pool:
5812# delay_class pool 5
5813# delay_parameters pool tagrate
5814#
5815# The option variables are:
5816#
5817# pool a pool number - ie, a number between 1 and the
5818# number specified in delay_pools as used in
5819# delay_class lines.
5820#
5821# aggregate the speed limit parameters for the aggregate bucket
5822# (class 1, 2, 3).
5823#
5824# individual the speed limit parameters for the individual
5825# buckets (class 2, 3).
5826#
5827# network the speed limit parameters for the network buckets
5828# (class 3).
5829#
5830# user the speed limit parameters for the user buckets
5831# (class 4).
5832#
5833# tagrate the speed limit parameters for the tag buckets
5834# (class 5).
5835#
5836# A pair of delay parameters is written restore/maximum, where restore is
5837# the number of bytes (not bits - modem and network speeds are usually
5838# quoted in bits) per second placed into the bucket, and maximum is the
5839# maximum number of bytes which can be in the bucket at any time.
5840#
5841# There must be one delay_parameters line for each delay pool.
5842#
5843#
5844# For example, if delay pool number 1 is a class 2 delay pool as in the
5845# above example, and is being used to strictly limit each host to 64Kbit/sec
5846# (plus overheads), with no overall limit, the line is:
5847#
5848# delay_parameters 1 none 8000/8000
5849#
5850# Note that 8 x 8K Byte/sec -> 64K bit/sec.
5851#
5852# Note that the word 'none' is used to represent no limit.
5853#
5854#
5855# And, if delay pool number 2 is a class 3 delay pool as in the above
5856# example, and you want to limit it to a total of 256Kbit/sec (strict limit)
5857# with each 8-bit network permitted 64Kbit/sec (strict limit) and each
5858# individual host permitted 4800bit/sec with a bucket maximum size of 64Kbits
5859# to permit a decent web page to be downloaded at a decent speed
5860# (if the network is not being limited due to overuse) but slow down
5861# large downloads more significantly:
5862#
5863# delay_parameters 2 32000/32000 8000/8000 600/8000
5864#
5865# Note that 8 x 32K Byte/sec -> 256K bit/sec.
5866# 8 x 8K Byte/sec -> 64K bit/sec.
5867# 8 x 600 Byte/sec -> 4800 bit/sec.
5868#
5869#
5870# Finally, for a class 4 delay pool as in the example - each user will
5871# be limited to 128Kbits/sec no matter how many workstations they are logged into.:
5872#
5873# delay_parameters 4 32000/32000 8000/8000 600/64000 16000/16000
5874#
5875#
5876# See also delay_class and delay_access.
5877#
5878#Default:
5879# none
5880
5881# TAG: delay_initial_bucket_level (percent, 0-100)
5882# The initial bucket percentage is used to determine how much is put
5883# in each bucket when squid starts, is reconfigured, or first notices
5884# a host accessing it (in class 2 and class 3, individual hosts and
5885# networks only have buckets associated with them once they have been
5886# "seen" by squid).
5887#Default:
5888# delay_initial_bucket_level 50
5889
5890# CLIENT DELAY POOL PARAMETERS
5891# -----------------------------------------------------------------------------
5892
5893# TAG: client_delay_pools
5894# This option specifies the number of client delay pools used. It must
5895# preceed other client_delay_* options.
5896#
5897# Example:
5898# client_delay_pools 2
5899#
5900# See also client_delay_parameters and client_delay_access.
5901#Default:
5902# client_delay_pools 0
5903
5904# TAG: client_delay_initial_bucket_level (percent, 0-no_limit)
5905# This option determines the initial bucket size as a percentage of
5906# max_bucket_size from client_delay_parameters. Buckets are created
5907# at the time of the "first" connection from the matching IP. Idle
5908# buckets are periodically deleted up.
5909#
5910# You can specify more than 100 percent but note that such "oversized"
5911# buckets are not refilled until their size goes down to max_bucket_size
5912# from client_delay_parameters.
5913#
5914# Example:
5915# client_delay_initial_bucket_level 50
5916#Default:
5917# client_delay_initial_bucket_level 50
5918
5919# TAG: client_delay_parameters
5920#
5921# This option configures client-side bandwidth limits using the
5922# following format:
5923#
5924# client_delay_parameters pool speed_limit max_bucket_size
5925#
5926# pool is an integer ID used for client_delay_access matching.
5927#
5928# speed_limit is bytes added to the bucket per second.
5929#
5930# max_bucket_size is the maximum size of a bucket, enforced after any
5931# speed_limit additions.
5932#
5933# Please see the delay_parameters option for more information and
5934# examples.
5935#
5936# Example:
5937# client_delay_parameters 1 1024 2048
5938# client_delay_parameters 2 51200 16384
5939#
5940# See also client_delay_access.
5941#
5942#Default:
5943# none
5944
5945# TAG: client_delay_access
5946# This option determines the client-side delay pool for the
5947# request:
5948#
5949# client_delay_access pool_ID allow|deny acl_name
5950#
5951# All client_delay_access options are checked in their pool ID
5952# order, starting with pool 1. The first checked pool with allowed
5953# request is selected for the request. If no ACL matches or there
5954# are no client_delay_access options, the request bandwidth is not
5955# limited.
5956#
5957# The ACL-selected pool is then used to find the
5958# client_delay_parameters for the request. Client-side pools are
5959# not used to aggregate clients. Clients are always aggregated
5960# based on their source IP addresses (one bucket per source IP).
5961#
5962# This clause only supports fast acl types.
5963# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5964# Additionally, only the client TCP connection details are available.
5965# ACLs testing HTTP properties will not work.
5966#
5967# Please see delay_access for more examples.
5968#
5969# Example:
5970# client_delay_access 1 allow low_rate_network
5971# client_delay_access 2 allow vips_network
5972#
5973#
5974# See also client_delay_parameters and client_delay_pools.
5975#Default:
5976# Deny use of the pool, unless allow rules exist in squid.conf for the pool.
5977
5978# WCCPv1 AND WCCPv2 CONFIGURATION OPTIONS
5979# -----------------------------------------------------------------------------
5980
5981# TAG: wccp_router
5982# Use this option to define your WCCP ``home'' router for
5983# Squid.
5984#
5985# wccp_router supports a single WCCP(v1) router
5986#
5987# wccp2_router supports multiple WCCPv2 routers
5988#
5989# only one of the two may be used at the same time and defines
5990# which version of WCCP to use.
5991#Default:
5992# WCCP disabled.
5993
5994# TAG: wccp2_router
5995# Use this option to define your WCCP ``home'' router for
5996# Squid.
5997#
5998# wccp_router supports a single WCCP(v1) router
5999#
6000# wccp2_router supports multiple WCCPv2 routers
6001#
6002# only one of the two may be used at the same time and defines
6003# which version of WCCP to use.
6004#Default:
6005# WCCPv2 disabled.
6006
6007# TAG: wccp_version
6008# This directive is only relevant if you need to set up WCCP(v1)
6009# to some very old and end-of-life Cisco routers. In all other
6010# setups it must be left unset or at the default setting.
6011# It defines an internal version in the WCCP(v1) protocol,
6012# with version 4 being the officially documented protocol.
6013#
6014# According to some users, Cisco IOS 11.2 and earlier only
6015# support WCCP version 3. If you're using that or an earlier
6016# version of IOS, you may need to change this value to 3, otherwise
6017# do not specify this parameter.
6018#Default:
6019# wccp_version 4
6020
6021# TAG: wccp2_rebuild_wait
6022# If this is enabled Squid will wait for the cache dir rebuild to finish
6023# before sending the first wccp2 HereIAm packet
6024#Default:
6025# wccp2_rebuild_wait on
6026
6027# TAG: wccp2_forwarding_method
6028# WCCP2 allows the setting of forwarding methods between the
6029# router/switch and the cache. Valid values are as follows:
6030#
6031# gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
6032# l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
6033#
6034# Currently (as of IOS 12.4) cisco routers only support GRE.
6035# Cisco switches only support the L2 redirect assignment method.
6036#Default:
6037# wccp2_forwarding_method gre
6038
6039# TAG: wccp2_return_method
6040# WCCP2 allows the setting of return methods between the
6041# router/switch and the cache for packets that the cache
6042# decides not to handle. Valid values are as follows:
6043#
6044# gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
6045# l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
6046#
6047# Currently (as of IOS 12.4) cisco routers only support GRE.
6048# Cisco switches only support the L2 redirect assignment.
6049#
6050# If the "ip wccp redirect exclude in" command has been
6051# enabled on the cache interface, then it is still safe for
6052# the proxy server to use a l2 redirect method even if this
6053# option is set to GRE.
6054#Default:
6055# wccp2_return_method gre
6056
6057# TAG: wccp2_assignment_method
6058# WCCP2 allows the setting of methods to assign the WCCP hash
6059# Valid values are as follows:
6060#
6061# hash - Hash assignment
6062# mask - Mask assignment
6063#
6064# As a general rule, cisco routers support the hash assignment method
6065# and cisco switches support the mask assignment method.
6066#Default:
6067# wccp2_assignment_method hash
6068
6069# TAG: wccp2_service
6070# WCCP2 allows for multiple traffic services. There are two
6071# types: "standard" and "dynamic". The standard type defines
6072# one service id - http (id 0). The dynamic service ids can be from
6073# 51 to 255 inclusive. In order to use a dynamic service id
6074# one must define the type of traffic to be redirected; this is done
6075# using the wccp2_service_info option.
6076#
6077# The "standard" type does not require a wccp2_service_info option,
6078# just specifying the service id will suffice.
6079#
6080# MD5 service authentication can be enabled by adding
6081# "password=<password>" to the end of this service declaration.
6082#
6083# Examples:
6084#
6085# wccp2_service standard 0 # for the 'web-cache' standard service
6086# wccp2_service dynamic 80 # a dynamic service type which will be
6087# # fleshed out with subsequent options.
6088# wccp2_service standard 0 password=foo
6089#Default:
6090# Use the 'web-cache' standard service.
6091
6092# TAG: wccp2_service_info
6093# Dynamic WCCPv2 services require further information to define the
6094# traffic you wish to have diverted.
6095#
6096# The format is:
6097#
6098# wccp2_service_info <id> protocol=<protocol> flags=<flag>,<flag>..
6099# priority=<priority> ports=<port>,<port>..
6100#
6101# The relevant WCCPv2 flags:
6102# + src_ip_hash, dst_ip_hash
6103# + source_port_hash, dst_port_hash
6104# + src_ip_alt_hash, dst_ip_alt_hash
6105# + src_port_alt_hash, dst_port_alt_hash
6106# + ports_source
6107#
6108# The port list can be one to eight entries.
6109#
6110# Example:
6111#
6112# wccp2_service_info 80 protocol=tcp flags=src_ip_hash,ports_source
6113# priority=240 ports=80
6114#
6115# Note: the service id must have been defined by a previous
6116# 'wccp2_service dynamic <id>' entry.
6117#Default:
6118# none
6119
6120# TAG: wccp2_weight
6121# Each cache server gets assigned a set of the destination
6122# hash proportional to their weight.
6123#Default:
6124# wccp2_weight 10000
6125
6126# TAG: wccp_address
6127# Use this option if you require WCCPv2 to use a specific
6128# interface address.
6129#
6130# The default behavior is to not bind to any specific address.
6131#Default:
6132# Address selected by the operating system.
6133
6134# TAG: wccp2_address
6135# Use this option if you require WCCP to use a specific
6136# interface address.
6137#
6138# The default behavior is to not bind to any specific address.
6139#Default:
6140# Address selected by the operating system.
6141
6142# PERSISTENT CONNECTION HANDLING
6143# -----------------------------------------------------------------------------
6144#
6145# Also see "pconn_timeout" in the TIMEOUTS section
6146
6147# TAG: client_persistent_connections
6148# Persistent connection support for clients.
6149# Squid uses persistent connections (when allowed). You can use
6150# this option to disable persistent connections with clients.
6151#Default:
6152# client_persistent_connections on
6153
6154# TAG: server_persistent_connections
6155# Persistent connection support for servers.
6156# Squid uses persistent connections (when allowed). You can use
6157# this option to disable persistent connections with servers.
6158#Default:
6159# server_persistent_connections on
6160
6161# TAG: persistent_connection_after_error
6162# With this directive the use of persistent connections after
6163# HTTP errors can be disabled. Useful if you have clients
6164# who fail to handle errors on persistent connections proper.
6165#Default:
6166# persistent_connection_after_error on
6167
6168# TAG: detect_broken_pconn
6169# Some servers have been found to incorrectly signal the use
6170# of HTTP/1.0 persistent connections even on replies not
6171# compatible, causing significant delays. This server problem
6172# has mostly been seen on redirects.
6173#
6174# By enabling this directive Squid attempts to detect such
6175# broken replies and automatically assume the reply is finished
6176# after 10 seconds timeout.
6177#Default:
6178# detect_broken_pconn off
6179
6180# CACHE DIGEST OPTIONS
6181# -----------------------------------------------------------------------------
6182
6183# TAG: digest_generation
6184# This controls whether the server will generate a Cache Digest
6185# of its contents. By default, Cache Digest generation is
6186# enabled if Squid is compiled with --enable-cache-digests defined.
6187#Default:
6188# digest_generation on
6189
6190# TAG: digest_bits_per_entry
6191# This is the number of bits of the server's Cache Digest which
6192# will be associated with the Digest entry for a given HTTP
6193# Method and URL (public key) combination. The default is 5.
6194#Default:
6195# digest_bits_per_entry 5
6196
6197# TAG: digest_rebuild_period (seconds)
6198# This is the wait time between Cache Digest rebuilds.
6199#Default:
6200# digest_rebuild_period 1 hour
6201
6202# TAG: digest_rewrite_period (seconds)
6203# This is the wait time between Cache Digest writes to
6204# disk.
6205#Default:
6206# digest_rewrite_period 1 hour
6207
6208# TAG: digest_swapout_chunk_size (bytes)
6209# This is the number of bytes of the Cache Digest to write to
6210# disk at a time. It defaults to 4096 bytes (4KB), the Squid
6211# default swap page.
6212#Default:
6213# digest_swapout_chunk_size 4096 bytes
6214
6215# TAG: digest_rebuild_chunk_percentage (percent, 0-100)
6216# This is the percentage of the Cache Digest to be scanned at a
6217# time. By default it is set to 10% of the Cache Digest.
6218#Default:
6219# digest_rebuild_chunk_percentage 10
6220
6221# SNMP OPTIONS
6222# -----------------------------------------------------------------------------
6223
6224# TAG: snmp_port
6225# The port number where Squid listens for SNMP requests. To enable
6226# SNMP support set this to a suitable port number. Port number
6227# 3401 is often used for the Squid SNMP agent. By default it's
6228# set to "0" (disabled)
6229#
6230# Example:
6231# snmp_port 3401
6232#Default:
6233# SNMP disabled.
6234
6235# TAG: snmp_access
6236# Allowing or denying access to the SNMP port.
6237#
6238# All access to the agent is denied by default.
6239# usage:
6240#
6241# snmp_access allow|deny [!]aclname ...
6242#
6243# This clause only supports fast acl types.
6244# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6245#
6246#Example:
6247# snmp_access allow snmppublic localhost
6248# snmp_access deny all
6249#Default:
6250# Deny, unless rules exist in squid.conf.
6251
6252# TAG: snmp_incoming_address
6253# Just like 'udp_incoming_address', but for the SNMP port.
6254#
6255# snmp_incoming_address is used for the SNMP socket receiving
6256# messages from SNMP agents.
6257#
6258# The default snmp_incoming_address is to listen on all
6259# available network interfaces.
6260#Default:
6261# Accept SNMP packets from all machine interfaces.
6262
6263# TAG: snmp_outgoing_address
6264# Just like 'udp_outgoing_address', but for the SNMP port.
6265#
6266# snmp_outgoing_address is used for SNMP packets returned to SNMP
6267# agents.
6268#
6269# If snmp_outgoing_address is not set it will use the same socket
6270# as snmp_incoming_address. Only change this if you want to have
6271# SNMP replies sent using another address than where this Squid
6272# listens for SNMP queries.
6273#
6274# NOTE, snmp_incoming_address and snmp_outgoing_address can not have
6275# the same value since they both use the same port.
6276#Default:
6277# Use snmp_incoming_address or an address selected by the operating system.
6278
6279# ICP OPTIONS
6280# -----------------------------------------------------------------------------
6281
6282# TAG: icp_port
6283# The port number where Squid sends and receives ICP queries to
6284# and from neighbor caches. The standard UDP port for ICP is 3130.
6285#
6286# Example:
6287# icp_port 3130
6288#Default:
6289# ICP disabled.
6290
6291# TAG: htcp_port
6292# The port number where Squid sends and receives HTCP queries to
6293# and from neighbor caches. To turn it on you want to set it to
6294# 4827.
6295#
6296# Example:
6297# htcp_port 4827
6298#Default:
6299# HTCP disabled.
6300
6301# TAG: log_icp_queries on|off
6302# If set, ICP queries are logged to access.log. You may wish
6303# do disable this if your ICP load is VERY high to speed things
6304# up or to simplify log analysis.
6305#Default:
6306# log_icp_queries on
6307
6308# TAG: udp_incoming_address
6309# udp_incoming_address is used for UDP packets received from other
6310# caches.
6311#
6312# The default behavior is to not bind to any specific address.
6313#
6314# Only change this if you want to have all UDP queries received on
6315# a specific interface/address.
6316#
6317# NOTE: udp_incoming_address is used by the ICP, HTCP, and DNS
6318# modules. Altering it will affect all of them in the same manner.
6319#
6320# see also; udp_outgoing_address
6321#
6322# NOTE, udp_incoming_address and udp_outgoing_address can not
6323# have the same value since they both use the same port.
6324#Default:
6325# Accept packets from all machine interfaces.
6326
6327# TAG: udp_outgoing_address
6328# udp_outgoing_address is used for UDP packets sent out to other
6329# caches.
6330#
6331# The default behavior is to not bind to any specific address.
6332#
6333# Instead it will use the same socket as udp_incoming_address.
6334# Only change this if you want to have UDP queries sent using another
6335# address than where this Squid listens for UDP queries from other
6336# caches.
6337#
6338# NOTE: udp_outgoing_address is used by the ICP, HTCP, and DNS
6339# modules. Altering it will affect all of them in the same manner.
6340#
6341# see also; udp_incoming_address
6342#
6343# NOTE, udp_incoming_address and udp_outgoing_address can not
6344# have the same value since they both use the same port.
6345#Default:
6346# Use udp_incoming_address or an address selected by the operating system.
6347
6348# TAG: icp_hit_stale on|off
6349# If you want to return ICP_HIT for stale cache objects, set this
6350# option to 'on'. If you have sibling relationships with caches
6351# in other administrative domains, this should be 'off'. If you only
6352# have sibling relationships with caches under your control,
6353# it is probably okay to set this to 'on'.
6354# If set to 'on', your siblings should use the option "allow-miss"
6355# on their cache_peer lines for connecting to you.
6356#Default:
6357# icp_hit_stale off
6358
6359# TAG: minimum_direct_hops
6360# If using the ICMP pinging stuff, do direct fetches for sites
6361# which are no more than this many hops away.
6362#Default:
6363# minimum_direct_hops 4
6364
6365# TAG: minimum_direct_rtt (msec)
6366# If using the ICMP pinging stuff, do direct fetches for sites
6367# which are no more than this many rtt milliseconds away.
6368#Default:
6369# minimum_direct_rtt 400
6370
6371# TAG: netdb_low
6372# The low water mark for the ICMP measurement database.
6373#
6374# Note: high watermark controlled by netdb_high directive.
6375#
6376# These watermarks are counts, not percents. The defaults are
6377# (low) 900 and (high) 1000. When the high water mark is
6378# reached, database entries will be deleted until the low
6379# mark is reached.
6380#Default:
6381# netdb_low 900
6382
6383# TAG: netdb_high
6384# The high water mark for the ICMP measurement database.
6385#
6386# Note: low watermark controlled by netdb_low directive.
6387#
6388# These watermarks are counts, not percents. The defaults are
6389# (low) 900 and (high) 1000. When the high water mark is
6390# reached, database entries will be deleted until the low
6391# mark is reached.
6392#Default:
6393# netdb_high 1000
6394
6395# TAG: netdb_ping_period
6396# The minimum period for measuring a site. There will be at
6397# least this much delay between successive pings to the same
6398# network. The default is five minutes.
6399#Default:
6400# netdb_ping_period 5 minutes
6401
6402# TAG: query_icmp on|off
6403# If you want to ask your peers to include ICMP data in their ICP
6404# replies, enable this option.
6405#
6406# If your peer has configured Squid (during compilation) with
6407# '--enable-icmp' that peer will send ICMP pings to origin server
6408# sites of the URLs it receives. If you enable this option the
6409# ICP replies from that peer will include the ICMP data (if available).
6410# Then, when choosing a parent cache, Squid will choose the parent with
6411# the minimal RTT to the origin server. When this happens, the
6412# hierarchy field of the access.log will be
6413# "CLOSEST_PARENT_MISS". This option is off by default.
6414#Default:
6415# query_icmp off
6416
6417# TAG: test_reachability on|off
6418# When this is 'on', ICP MISS replies will be ICP_MISS_NOFETCH
6419# instead of ICP_MISS if the target host is NOT in the ICMP
6420# database, or has a zero RTT.
6421#Default:
6422# test_reachability off
6423
6424# TAG: icp_query_timeout (msec)
6425# Normally Squid will automatically determine an optimal ICP
6426# query timeout value based on the round-trip-time of recent ICP
6427# queries. If you want to override the value determined by
6428# Squid, set this 'icp_query_timeout' to a non-zero value. This
6429# value is specified in MILLISECONDS, so, to use a 2-second
6430# timeout (the old default), you would write:
6431#
6432# icp_query_timeout 2000
6433#Default:
6434# Dynamic detection.
6435
6436# TAG: maximum_icp_query_timeout (msec)
6437# Normally the ICP query timeout is determined dynamically. But
6438# sometimes it can lead to very large values (say 5 seconds).
6439# Use this option to put an upper limit on the dynamic timeout
6440# value. Do NOT use this option to always use a fixed (instead
6441# of a dynamic) timeout value. To set a fixed timeout see the
6442# 'icp_query_timeout' directive.
6443#Default:
6444# maximum_icp_query_timeout 2000
6445
6446# TAG: minimum_icp_query_timeout (msec)
6447# Normally the ICP query timeout is determined dynamically. But
6448# sometimes it can lead to very small timeouts, even lower than
6449# the normal latency variance on your link due to traffic.
6450# Use this option to put an lower limit on the dynamic timeout
6451# value. Do NOT use this option to always use a fixed (instead
6452# of a dynamic) timeout value. To set a fixed timeout see the
6453# 'icp_query_timeout' directive.
6454#Default:
6455# minimum_icp_query_timeout 5
6456
6457# TAG: background_ping_rate time-units
6458# Controls how often the ICP pings are sent to siblings that
6459# have background-ping set.
6460#Default:
6461# background_ping_rate 10 seconds
6462
6463# MULTICAST ICP OPTIONS
6464# -----------------------------------------------------------------------------
6465
6466# TAG: mcast_groups
6467# This tag specifies a list of multicast groups which your server
6468# should join to receive multicasted ICP queries.
6469#
6470# NOTE! Be very careful what you put here! Be sure you
6471# understand the difference between an ICP _query_ and an ICP
6472# _reply_. This option is to be set only if you want to RECEIVE
6473# multicast queries. Do NOT set this option to SEND multicast
6474# ICP (use cache_peer for that). ICP replies are always sent via
6475# unicast, so this option does not affect whether or not you will
6476# receive replies from multicast group members.
6477#
6478# You must be very careful to NOT use a multicast address which
6479# is already in use by another group of caches.
6480#
6481# If you are unsure about multicast, please read the Multicast
6482# chapter in the Squid FAQ (http://www.squid-cache.org/FAQ/).
6483#
6484# Usage: mcast_groups 239.128.16.128 224.0.1.20
6485#
6486# By default, Squid doesn't listen on any multicast groups.
6487#Default:
6488# none
6489
6490# TAG: mcast_miss_addr
6491# Note: This option is only available if Squid is rebuilt with the
6492# -DMULTICAST_MISS_STREAM define
6493#
6494# If you enable this option, every "cache miss" URL will
6495# be sent out on the specified multicast address.
6496#
6497# Do not enable this option unless you are are absolutely
6498# certain you understand what you are doing.
6499#Default:
6500# disabled.
6501
6502# TAG: mcast_miss_ttl
6503# Note: This option is only available if Squid is rebuilt with the
6504# -DMULTICAST_MISS_STREAM define
6505#
6506# This is the time-to-live value for packets multicasted
6507# when multicasting off cache miss URLs is enabled. By
6508# default this is set to 'site scope', i.e. 16.
6509#Default:
6510# mcast_miss_ttl 16
6511
6512# TAG: mcast_miss_port
6513# Note: This option is only available if Squid is rebuilt with the
6514# -DMULTICAST_MISS_STREAM define
6515#
6516# This is the port number to be used in conjunction with
6517# 'mcast_miss_addr'.
6518#Default:
6519# mcast_miss_port 3135
6520
6521# TAG: mcast_miss_encode_key
6522# Note: This option is only available if Squid is rebuilt with the
6523# -DMULTICAST_MISS_STREAM define
6524#
6525# The URLs that are sent in the multicast miss stream are
6526# encrypted. This is the encryption key.
6527#Default:
6528# mcast_miss_encode_key XXXXXXXXXXXXXXXX
6529
6530# TAG: mcast_icp_query_timeout (msec)
6531# For multicast peers, Squid regularly sends out ICP "probes" to
6532# count how many other peers are listening on the given multicast
6533# address. This value specifies how long Squid should wait to
6534# count all the replies. The default is 2000 msec, or 2
6535# seconds.
6536#Default:
6537# mcast_icp_query_timeout 2000
6538
6539# INTERNAL ICON OPTIONS
6540# -----------------------------------------------------------------------------
6541
6542# TAG: icon_directory
6543# Where the icons are stored. These are normally kept in
6544# /usr/share/squid/icons
6545#Default:
6546# icon_directory /usr/share/squid/icons
6547
6548# TAG: global_internal_static
6549# This directive controls is Squid should intercept all requests for
6550# /squid-internal-static/ no matter which host the URL is requesting
6551# (default on setting), or if nothing special should be done for
6552# such URLs (off setting). The purpose of this directive is to make
6553# icons etc work better in complex cache hierarchies where it may
6554# not always be possible for all corners in the cache mesh to reach
6555# the server generating a directory listing.
6556#Default:
6557# global_internal_static on
6558
6559# TAG: short_icon_urls
6560# If this is enabled Squid will use short URLs for icons.
6561# If disabled it will revert to the old behavior of including
6562# it's own name and port in the URL.
6563#
6564# If you run a complex cache hierarchy with a mix of Squid and
6565# other proxies you may need to disable this directive.
6566#Default:
6567# short_icon_urls on
6568
6569# ERROR PAGE OPTIONS
6570# -----------------------------------------------------------------------------
6571
6572# TAG: error_directory
6573# If you wish to create your own versions of the default
6574# error files to customize them to suit your company copy
6575# the error/template files to another directory and point
6576# this tag at them.
6577#
6578# WARNING: This option will disable multi-language support
6579# on error pages if used.
6580#
6581# The squid developers are interested in making squid available in
6582# a wide variety of languages. If you are making translations for a
6583# language that Squid does not currently provide please consider
6584# contributing your translation back to the project.
6585# http://wiki.squid-cache.org/Translations
6586#
6587# The squid developers working on translations are happy to supply drop-in
6588# translated error files in exchange for any new language contributions.
6589#Default:
6590# Send error pages in the clients preferred language
6591
6592# TAG: error_default_language
6593# Set the default language which squid will send error pages in
6594# if no existing translation matches the clients language
6595# preferences.
6596#
6597# If unset (default) generic English will be used.
6598#
6599# The squid developers are interested in making squid available in
6600# a wide variety of languages. If you are interested in making
6601# translations for any language see the squid wiki for details.
6602# http://wiki.squid-cache.org/Translations
6603#Default:
6604# Generate English language pages.
6605
6606# TAG: error_log_languages
6607# Log to cache.log what languages users are attempting to
6608# auto-negotiate for translations.
6609#
6610# Successful negotiations are not logged. Only failures
6611# have meaning to indicate that Squid may need an upgrade
6612# of its error page translations.
6613#Default:
6614# error_log_languages on
6615
6616# TAG: err_page_stylesheet
6617# CSS Stylesheet to pattern the display of Squid default error pages.
6618#
6619# For information on CSS see http://www.w3.org/Style/CSS/
6620#Default:
6621# err_page_stylesheet /etc/squid/errorpage.css
6622
6623# TAG: err_html_text
6624# HTML text to include in error messages. Make this a "mailto"
6625# URL to your admin address, or maybe just a link to your
6626# organizations Web page.
6627#
6628# To include this in your error messages, you must rewrite
6629# the error template files (found in the "errors" directory).
6630# Wherever you want the 'err_html_text' line to appear,
6631# insert a %L tag in the error template file.
6632#Default:
6633# none
6634
6635# TAG: email_err_data on|off
6636# If enabled, information about the occurred error will be
6637# included in the mailto links of the ERR pages (if %W is set)
6638# so that the email body contains the data.
6639# Syntax is <A HREF="mailto:%w%W">%w</A>
6640#Default:
6641# email_err_data on
6642
6643# TAG: deny_info
6644# Usage: deny_info err_page_name acl
6645# or deny_info http://... acl
6646# or deny_info TCP_RESET acl
6647#
6648# This can be used to return a ERR_ page for requests which
6649# do not pass the 'http_access' rules. Squid remembers the last
6650# acl it evaluated in http_access, and if a 'deny_info' line exists
6651# for that ACL Squid returns a corresponding error page.
6652#
6653# The acl is typically the last acl on the http_access deny line which
6654# denied access. The exceptions to this rule are:
6655# - When Squid needs to request authentication credentials. It's then
6656# the first authentication related acl encountered
6657# - When none of the http_access lines matches. It's then the last
6658# acl processed on the last http_access line.
6659# - When the decision to deny access was made by an adaptation service,
6660# the acl name is the corresponding eCAP or ICAP service_name.
6661#
6662# NP: If providing your own custom error pages with error_directory
6663# you may also specify them by your custom file name:
6664# Example: deny_info ERR_CUSTOM_ACCESS_DENIED bad_guys
6665#
6666# By defaut Squid will send "403 Forbidden". A different 4xx or 5xx
6667# may be specified by prefixing the file name with the code and a colon.
6668# e.g. 404:ERR_CUSTOM_ACCESS_DENIED
6669#
6670# Alternatively you can tell Squid to reset the TCP connection
6671# by specifying TCP_RESET.
6672#
6673# Or you can specify an error URL or URL pattern. The browsers will
6674# get redirected to the specified URL after formatting tags have
6675# been replaced. Redirect will be done with 302 or 307 according to
6676# HTTP/1.1 specs. A different 3xx code may be specified by prefixing
6677# the URL. e.g. 303:http://example.com/
6678#
6679# URL FORMAT TAGS:
6680# %a - username (if available. Password NOT included)
6681# %B - FTP path URL
6682# %e - Error number
6683# %E - Error description
6684# %h - Squid hostname
6685# %H - Request domain name
6686# %i - Client IP Address
6687# %M - Request Method
6688# %o - Message result from external ACL helper
6689# %p - Request Port number
6690# %P - Request Protocol name
6691# %R - Request URL path
6692# %T - Timestamp in RFC 1123 format
6693# %U - Full canonical URL from client
6694# (HTTPS URLs terminate with *)
6695# %u - Full canonical URL from client
6696# %w - Admin email from squid.conf
6697# %x - Error name
6698# %% - Literal percent (%) code
6699#
6700#Default:
6701# none
6702
6703# OPTIONS INFLUENCING REQUEST FORWARDING
6704# -----------------------------------------------------------------------------
6705
6706# TAG: nonhierarchical_direct
6707# By default, Squid will send any non-hierarchical requests
6708# (not cacheable request type) direct to origin servers.
6709#
6710# When this is set to "off", Squid will prefer to send these
6711# requests to parents.
6712#
6713# Note that in most configurations, by turning this off you will only
6714# add latency to these request without any improvement in global hit
6715# ratio.
6716#
6717# This option only sets a preference. If the parent is unavailable a
6718# direct connection to the origin server may still be attempted. To
6719# completely prevent direct connections use never_direct.
6720#Default:
6721# nonhierarchical_direct on
6722
6723# TAG: prefer_direct
6724# Normally Squid tries to use parents for most requests. If you for some
6725# reason like it to first try going direct and only use a parent if
6726# going direct fails set this to on.
6727#
6728# By combining nonhierarchical_direct off and prefer_direct on you
6729# can set up Squid to use a parent as a backup path if going direct
6730# fails.
6731#
6732# Note: If you want Squid to use parents for all requests see
6733# the never_direct directive. prefer_direct only modifies how Squid
6734# acts on cacheable requests.
6735#Default:
6736# prefer_direct off
6737
6738# TAG: cache_miss_revalidate on|off
6739# RFC 7232 defines a conditional request mechanism to prevent
6740# response objects being unnecessarily transferred over the network.
6741# If that mechanism is used by the client and a cache MISS occurs
6742# it can prevent new cache entries being created.
6743#
6744# This option determines whether Squid on cache MISS will pass the
6745# client revalidation request to the server or tries to fetch new
6746# content for caching. It can be useful while the cache is mostly
6747# empty to more quickly have the cache populated by generating
6748# non-conditional GETs.
6749#
6750# When set to 'on' (default), Squid will pass all client If-* headers
6751# to the server. This permits server responses without a cacheable
6752# payload to be delivered and on MISS no new cache entry is created.
6753#
6754# When set to 'off' and if the request is cacheable, Squid will
6755# remove the clients If-Modified-Since and If-None-Match headers from
6756# the request sent to the server. This requests a 200 status response
6757# from the server to create a new cache entry with.
6758#Default:
6759# cache_miss_revalidate on
6760
6761# TAG: always_direct
6762# Usage: always_direct allow|deny [!]aclname ...
6763#
6764# Here you can use ACL elements to specify requests which should
6765# ALWAYS be forwarded by Squid to the origin servers without using
6766# any peers. For example, to always directly forward requests for
6767# local servers ignoring any parents or siblings you may have use
6768# something like:
6769#
6770# acl local-servers dstdomain my.domain.net
6771# always_direct allow local-servers
6772#
6773# To always forward FTP requests directly, use
6774#
6775# acl FTP proto FTP
6776# always_direct allow FTP
6777#
6778# NOTE: There is a similar, but opposite option named
6779# 'never_direct'. You need to be aware that "always_direct deny
6780# foo" is NOT the same thing as "never_direct allow foo". You
6781# may need to use a deny rule to exclude a more-specific case of
6782# some other rule. Example:
6783#
6784# acl local-external dstdomain external.foo.net
6785# acl local-servers dstdomain .foo.net
6786# always_direct deny local-external
6787# always_direct allow local-servers
6788#
6789# NOTE: If your goal is to make the client forward the request
6790# directly to the origin server bypassing Squid then this needs
6791# to be done in the client configuration. Squid configuration
6792# can only tell Squid how Squid should fetch the object.
6793#
6794# NOTE: This directive is not related to caching. The replies
6795# is cached as usual even if you use always_direct. To not cache
6796# the replies see the 'cache' directive.
6797#
6798# This clause supports both fast and slow acl types.
6799# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6800#Default:
6801# Prevent any cache_peer being used for this request.
6802
6803# TAG: never_direct
6804# Usage: never_direct allow|deny [!]aclname ...
6805#
6806# never_direct is the opposite of always_direct. Please read
6807# the description for always_direct if you have not already.
6808#
6809# With 'never_direct' you can use ACL elements to specify
6810# requests which should NEVER be forwarded directly to origin
6811# servers. For example, to force the use of a proxy for all
6812# requests, except those in your local domain use something like:
6813#
6814# acl local-servers dstdomain .foo.net
6815# never_direct deny local-servers
6816# never_direct allow all
6817#
6818# or if Squid is inside a firewall and there are local intranet
6819# servers inside the firewall use something like:
6820#
6821# acl local-intranet dstdomain .foo.net
6822# acl local-external dstdomain external.foo.net
6823# always_direct deny local-external
6824# always_direct allow local-intranet
6825# never_direct allow all
6826#
6827# This clause supports both fast and slow acl types.
6828# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6829#Default:
6830# Allow DNS results to be used for this request.
6831
6832# ADVANCED NETWORKING OPTIONS
6833# -----------------------------------------------------------------------------
6834
6835# TAG: incoming_udp_average
6836# Heavy voodoo here. I can't even believe you are reading this.
6837# Are you crazy? Don't even think about adjusting these unless
6838# you understand the algorithms in comm_select.c first!
6839#Default:
6840# incoming_udp_average 6
6841
6842# TAG: incoming_tcp_average
6843# Heavy voodoo here. I can't even believe you are reading this.
6844# Are you crazy? Don't even think about adjusting these unless
6845# you understand the algorithms in comm_select.c first!
6846#Default:
6847# incoming_tcp_average 4
6848
6849# TAG: incoming_dns_average
6850# Heavy voodoo here. I can't even believe you are reading this.
6851# Are you crazy? Don't even think about adjusting these unless
6852# you understand the algorithms in comm_select.c first!
6853#Default:
6854# incoming_dns_average 4
6855
6856# TAG: min_udp_poll_cnt
6857# Heavy voodoo here. I can't even believe you are reading this.
6858# Are you crazy? Don't even think about adjusting these unless
6859# you understand the algorithms in comm_select.c first!
6860#Default:
6861# min_udp_poll_cnt 8
6862
6863# TAG: min_dns_poll_cnt
6864# Heavy voodoo here. I can't even believe you are reading this.
6865# Are you crazy? Don't even think about adjusting these unless
6866# you understand the algorithms in comm_select.c first!
6867#Default:
6868# min_dns_poll_cnt 8
6869
6870# TAG: min_tcp_poll_cnt
6871# Heavy voodoo here. I can't even believe you are reading this.
6872# Are you crazy? Don't even think about adjusting these unless
6873# you understand the algorithms in comm_select.c first!
6874#Default:
6875# min_tcp_poll_cnt 8
6876
6877# TAG: accept_filter
6878# FreeBSD:
6879#
6880# The name of an accept(2) filter to install on Squid's
6881# listen socket(s). This feature is perhaps specific to
6882# FreeBSD and requires support in the kernel.
6883#
6884# The 'httpready' filter delays delivering new connections
6885# to Squid until a full HTTP request has been received.
6886# See the accf_http(9) man page for details.
6887#
6888# The 'dataready' filter delays delivering new connections
6889# to Squid until there is some data to process.
6890# See the accf_dataready(9) man page for details.
6891#
6892# Linux:
6893#
6894# The 'data' filter delays delivering of new connections
6895# to Squid until there is some data to process by TCP_ACCEPT_DEFER.
6896# You may optionally specify a number of seconds to wait by
6897# 'data=N' where N is the number of seconds. Defaults to 30
6898# if not specified. See the tcp(7) man page for details.
6899#EXAMPLE:
6900## FreeBSD
6901#accept_filter httpready
6902## Linux
6903#accept_filter data
6904#Default:
6905# none
6906
6907# TAG: client_ip_max_connections
6908# Set an absolute limit on the number of connections a single
6909# client IP can use. Any more than this and Squid will begin to drop
6910# new connections from the client until it closes some links.
6911#
6912# Note that this is a global limit. It affects all HTTP, HTCP, Gopher and FTP
6913# connections from the client. For finer control use the ACL access controls.
6914#
6915# Requires client_db to be enabled (the default).
6916#
6917# WARNING: This may noticably slow down traffic received via external proxies
6918# or NAT devices and cause them to rebound error messages back to their clients.
6919#Default:
6920# No limit.
6921
6922# TAG: tcp_recv_bufsize (bytes)
6923# Size of receive buffer to set for TCP sockets. Probably just
6924# as easy to change your kernel's default.
6925# Omit from squid.conf to use the default buffer size.
6926#Default:
6927# Use operating system TCP defaults.
6928
6929# ICAP OPTIONS
6930# -----------------------------------------------------------------------------
6931
6932# TAG: icap_enable on|off
6933# If you want to enable the ICAP module support, set this to on.
6934#Default:
6935# icap_enable off
6936
6937# TAG: icap_connect_timeout
6938# This parameter specifies how long to wait for the TCP connect to
6939# the requested ICAP server to complete before giving up and either
6940# terminating the HTTP transaction or bypassing the failure.
6941#
6942# The default for optional services is peer_connect_timeout.
6943# The default for essential services is connect_timeout.
6944# If this option is explicitly set, its value applies to all services.
6945#Default:
6946# none
6947
6948# TAG: icap_io_timeout time-units
6949# This parameter specifies how long to wait for an I/O activity on
6950# an established, active ICAP connection before giving up and
6951# either terminating the HTTP transaction or bypassing the
6952# failure.
6953#Default:
6954# Use read_timeout.
6955
6956# TAG: icap_service_failure_limit limit [in memory-depth time-units]
6957# The limit specifies the number of failures that Squid tolerates
6958# when establishing a new TCP connection with an ICAP service. If
6959# the number of failures exceeds the limit, the ICAP service is
6960# not used for new ICAP requests until it is time to refresh its
6961# OPTIONS.
6962#
6963# A negative value disables the limit. Without the limit, an ICAP
6964# service will not be considered down due to connectivity failures
6965# between ICAP OPTIONS requests.
6966#
6967# Squid forgets ICAP service failures older than the specified
6968# value of memory-depth. The memory fading algorithm
6969# is approximate because Squid does not remember individual
6970# errors but groups them instead, splitting the option
6971# value into ten time slots of equal length.
6972#
6973# When memory-depth is 0 and by default this option has no
6974# effect on service failure expiration.
6975#
6976# Squid always forgets failures when updating service settings
6977# using an ICAP OPTIONS transaction, regardless of this option
6978# setting.
6979#
6980# For example,
6981# # suspend service usage after 10 failures in 5 seconds:
6982# icap_service_failure_limit 10 in 5 seconds
6983#Default:
6984# icap_service_failure_limit 10
6985
6986# TAG: icap_service_revival_delay
6987# The delay specifies the number of seconds to wait after an ICAP
6988# OPTIONS request failure before requesting the options again. The
6989# failed ICAP service is considered "down" until fresh OPTIONS are
6990# fetched.
6991#
6992# The actual delay cannot be smaller than the hardcoded minimum
6993# delay of 30 seconds.
6994#Default:
6995# icap_service_revival_delay 180
6996
6997# TAG: icap_preview_enable on|off
6998# The ICAP Preview feature allows the ICAP server to handle the
6999# HTTP message by looking only at the beginning of the message body
7000# or even without receiving the body at all. In some environments,
7001# previews greatly speedup ICAP processing.
7002#
7003# During an ICAP OPTIONS transaction, the server may tell Squid what
7004# HTTP messages should be previewed and how big the preview should be.
7005# Squid will not use Preview if the server did not request one.
7006#
7007# To disable ICAP Preview for all ICAP services, regardless of
7008# individual ICAP server OPTIONS responses, set this option to "off".
7009#Example:
7010#icap_preview_enable off
7011#Default:
7012# icap_preview_enable on
7013
7014# TAG: icap_preview_size
7015# The default size of preview data to be sent to the ICAP server.
7016# This value might be overwritten on a per server basis by OPTIONS requests.
7017#Default:
7018# No preview sent.
7019
7020# TAG: icap_206_enable on|off
7021# 206 (Partial Content) responses is an ICAP extension that allows the
7022# ICAP agents to optionally combine adapted and original HTTP message
7023# content. The decision to combine is postponed until the end of the
7024# ICAP response. Squid supports Partial Content extension by default.
7025#
7026# Activation of the Partial Content extension is negotiated with each
7027# ICAP service during OPTIONS exchange. Most ICAP servers should handle
7028# negotation correctly even if they do not support the extension, but
7029# some might fail. To disable Partial Content support for all ICAP
7030# services and to avoid any negotiation, set this option to "off".
7031#
7032# Example:
7033# icap_206_enable off
7034#Default:
7035# icap_206_enable on
7036
7037# TAG: icap_default_options_ttl
7038# The default TTL value for ICAP OPTIONS responses that don't have
7039# an Options-TTL header.
7040#Default:
7041# icap_default_options_ttl 60
7042
7043# TAG: icap_persistent_connections on|off
7044# Whether or not Squid should use persistent connections to
7045# an ICAP server.
7046#Default:
7047# icap_persistent_connections on
7048
7049# TAG: adaptation_send_client_ip on|off
7050# If enabled, Squid shares HTTP client IP information with adaptation
7051# services. For ICAP, Squid adds the X-Client-IP header to ICAP requests.
7052# For eCAP, Squid sets the libecap::metaClientIp transaction option.
7053#
7054# See also: adaptation_uses_indirect_client
7055#Default:
7056# adaptation_send_client_ip off
7057
7058# TAG: adaptation_send_username on|off
7059# This sends authenticated HTTP client username (if available) to
7060# the adaptation service.
7061#
7062# For ICAP, the username value is encoded based on the
7063# icap_client_username_encode option and is sent using the header
7064# specified by the icap_client_username_header option.
7065#Default:
7066# adaptation_send_username off
7067
7068# TAG: icap_client_username_header
7069# ICAP request header name to use for adaptation_send_username.
7070#Default:
7071# icap_client_username_header X-Client-Username
7072
7073# TAG: icap_client_username_encode on|off
7074# Whether to base64 encode the authenticated client username.
7075#Default:
7076# icap_client_username_encode off
7077
7078# TAG: icap_service
7079# Defines a single ICAP service using the following format:
7080#
7081# icap_service id vectoring_point uri [option ...]
7082#
7083# id: ID
7084# an opaque identifier or name which is used to direct traffic to
7085# this specific service. Must be unique among all adaptation
7086# services in squid.conf.
7087#
7088# vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
7089# This specifies at which point of transaction processing the
7090# ICAP service should be activated. *_postcache vectoring points
7091# are not yet supported.
7092#
7093# uri: icap://servername:port/servicepath
7094# ICAP server and service location.
7095#
7096# ICAP does not allow a single service to handle both REQMOD and RESPMOD
7097# transactions. Squid does not enforce that requirement. You can specify
7098# services with the same service_url and different vectoring_points. You
7099# can even specify multiple identical services as long as their
7100# service_names differ.
7101#
7102# To activate a service, use the adaptation_access directive. To group
7103# services, use adaptation_service_chain and adaptation_service_set.
7104#
7105# Service options are separated by white space. ICAP services support
7106# the following name=value options:
7107#
7108# bypass=on|off|1|0
7109# If set to 'on' or '1', the ICAP service is treated as
7110# optional. If the service cannot be reached or malfunctions,
7111# Squid will try to ignore any errors and process the message as
7112# if the service was not enabled. No all ICAP errors can be
7113# bypassed. If set to 0, the ICAP service is treated as
7114# essential and all ICAP errors will result in an error page
7115# returned to the HTTP client.
7116#
7117# Bypass is off by default: services are treated as essential.
7118#
7119# routing=on|off|1|0
7120# If set to 'on' or '1', the ICAP service is allowed to
7121# dynamically change the current message adaptation plan by
7122# returning a chain of services to be used next. The services
7123# are specified using the X-Next-Services ICAP response header
7124# value, formatted as a comma-separated list of service names.
7125# Each named service should be configured in squid.conf. Other
7126# services are ignored. An empty X-Next-Services value results
7127# in an empty plan which ends the current adaptation.
7128#
7129# Dynamic adaptation plan may cross or cover multiple supported
7130# vectoring points in their natural processing order.
7131#
7132# Routing is not allowed by default: the ICAP X-Next-Services
7133# response header is ignored.
7134#
7135# ipv6=on|off
7136# Only has effect on split-stack systems. The default on those systems
7137# is to use IPv4-only connections. When set to 'on' this option will
7138# make Squid use IPv6-only connections to contact this ICAP service.
7139#
7140# on-overload=block|bypass|wait|force
7141# If the service Max-Connections limit has been reached, do
7142# one of the following for each new ICAP transaction:
7143# * block: send an HTTP error response to the client
7144# * bypass: ignore the "over-connected" ICAP service
7145# * wait: wait (in a FIFO queue) for an ICAP connection slot
7146# * force: proceed, ignoring the Max-Connections limit
7147#
7148# In SMP mode with N workers, each worker assumes the service
7149# connection limit is Max-Connections/N, even though not all
7150# workers may use a given service.
7151#
7152# The default value is "bypass" if service is bypassable,
7153# otherwise it is set to "wait".
7154#
7155#
7156# max-conn=number
7157# Use the given number as the Max-Connections limit, regardless
7158# of the Max-Connections value given by the service, if any.
7159#
7160# Older icap_service format without optional named parameters is
7161# deprecated but supported for backward compatibility.
7162#
7163#Example:
7164#icap_service svcBlocker reqmod_precache icap://icap1.mydomain.net:1344/reqmod bypass=0
7165#icap_service svcLogger reqmod_precache icap://icap2.mydomain.net:1344/respmod routing=on
7166#Default:
7167# none
7168
7169# TAG: icap_class
7170# This deprecated option was documented to define an ICAP service
7171# chain, even though it actually defined a set of similar, redundant
7172# services, and the chains were not supported.
7173#
7174# To define a set of redundant services, please use the
7175# adaptation_service_set directive. For service chains, use
7176# adaptation_service_chain.
7177#Default:
7178# none
7179
7180# TAG: icap_access
7181# This option is deprecated. Please use adaptation_access, which
7182# has the same ICAP functionality, but comes with better
7183# documentation, and eCAP support.
7184#Default:
7185# none
7186
7187# eCAP OPTIONS
7188# -----------------------------------------------------------------------------
7189
7190# TAG: ecap_enable on|off
7191# Controls whether eCAP support is enabled.
7192#Default:
7193# ecap_enable off
7194
7195# TAG: ecap_service
7196# Defines a single eCAP service
7197#
7198# ecap_service id vectoring_point uri [option ...]
7199#
7200# id: ID
7201# an opaque identifier or name which is used to direct traffic to
7202# this specific service. Must be unique among all adaptation
7203# services in squid.conf.
7204#
7205# vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
7206# This specifies at which point of transaction processing the
7207# eCAP service should be activated. *_postcache vectoring points
7208# are not yet supported.
7209#
7210# uri: ecap://vendor/service_name?custom&cgi=style¶meters=optional
7211# Squid uses the eCAP service URI to match this configuration
7212# line with one of the dynamically loaded services. Each loaded
7213# eCAP service must have a unique URI. Obtain the right URI from
7214# the service provider.
7215#
7216# To activate a service, use the adaptation_access directive. To group
7217# services, use adaptation_service_chain and adaptation_service_set.
7218#
7219# Service options are separated by white space. eCAP services support
7220# the following name=value options:
7221#
7222# bypass=on|off|1|0
7223# If set to 'on' or '1', the eCAP service is treated as optional.
7224# If the service cannot be reached or malfunctions, Squid will try
7225# to ignore any errors and process the message as if the service
7226# was not enabled. No all eCAP errors can be bypassed.
7227# If set to 'off' or '0', the eCAP service is treated as essential
7228# and all eCAP errors will result in an error page returned to the
7229# HTTP client.
7230#
7231# Bypass is off by default: services are treated as essential.
7232#
7233# routing=on|off|1|0
7234# If set to 'on' or '1', the eCAP service is allowed to
7235# dynamically change the current message adaptation plan by
7236# returning a chain of services to be used next.
7237#
7238# Dynamic adaptation plan may cross or cover multiple supported
7239# vectoring points in their natural processing order.
7240#
7241# Routing is not allowed by default.
7242#
7243# Older ecap_service format without optional named parameters is
7244# deprecated but supported for backward compatibility.
7245#
7246#
7247#Example:
7248#ecap_service s1 reqmod_precache ecap://filters.R.us/leakDetector?on_error=block bypass=off
7249#ecap_service s2 respmod_precache ecap://filters.R.us/virusFilter config=/etc/vf.cfg bypass=on
7250#Default:
7251# none
7252
7253# TAG: loadable_modules
7254# Instructs Squid to load the specified dynamic module(s) or activate
7255# preloaded module(s).
7256#Example:
7257#loadable_modules /usr/lib/MinimalAdapter.so
7258#Default:
7259# none
7260
7261# MESSAGE ADAPTATION OPTIONS
7262# -----------------------------------------------------------------------------
7263
7264# TAG: adaptation_service_set
7265#
7266# Configures an ordered set of similar, redundant services. This is
7267# useful when hot standby or backup adaptation servers are available.
7268#
7269# adaptation_service_set set_name service_name1 service_name2 ...
7270#
7271# The named services are used in the set declaration order. The first
7272# applicable adaptation service from the set is used first. The next
7273# applicable service is tried if and only if the transaction with the
7274# previous service fails and the message waiting to be adapted is still
7275# intact.
7276#
7277# When adaptation starts, broken services are ignored as if they were
7278# not a part of the set. A broken service is a down optional service.
7279#
7280# The services in a set must be attached to the same vectoring point
7281# (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
7282#
7283# If all services in a set are optional then adaptation failures are
7284# bypassable. If all services in the set are essential, then a
7285# transaction failure with one service may still be retried using
7286# another service from the set, but when all services fail, the master
7287# transaction fails as well.
7288#
7289# A set may contain a mix of optional and essential services, but that
7290# is likely to lead to surprising results because broken services become
7291# ignored (see above), making previously bypassable failures fatal.
7292# Technically, it is the bypassability of the last failed service that
7293# matters.
7294#
7295# See also: adaptation_access adaptation_service_chain
7296#
7297#Example:
7298#adaptation_service_set svcBlocker urlFilterPrimary urlFilterBackup
7299#adaptation service_set svcLogger loggerLocal loggerRemote
7300#Default:
7301# none
7302
7303# TAG: adaptation_service_chain
7304#
7305# Configures a list of complementary services that will be applied
7306# one-by-one, forming an adaptation chain or pipeline. This is useful
7307# when Squid must perform different adaptations on the same message.
7308#
7309# adaptation_service_chain chain_name service_name1 svc_name2 ...
7310#
7311# The named services are used in the chain declaration order. The first
7312# applicable adaptation service from the chain is used first. The next
7313# applicable service is applied to the successful adaptation results of
7314# the previous service in the chain.
7315#
7316# When adaptation starts, broken services are ignored as if they were
7317# not a part of the chain. A broken service is a down optional service.
7318#
7319# Request satisfaction terminates the adaptation chain because Squid
7320# does not currently allow declaration of RESPMOD services at the
7321# "reqmod_precache" vectoring point (see icap_service or ecap_service).
7322#
7323# The services in a chain must be attached to the same vectoring point
7324# (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
7325#
7326# A chain may contain a mix of optional and essential services. If an
7327# essential adaptation fails (or the failure cannot be bypassed for
7328# other reasons), the master transaction fails. Otherwise, the failure
7329# is bypassed as if the failed adaptation service was not in the chain.
7330#
7331# See also: adaptation_access adaptation_service_set
7332#
7333#Example:
7334#adaptation_service_chain svcRequest requestLogger urlFilter leakDetector
7335#Default:
7336# none
7337
7338# TAG: adaptation_access
7339# Sends an HTTP transaction to an ICAP or eCAP adaptation service.
7340#
7341# adaptation_access service_name allow|deny [!]aclname...
7342# adaptation_access set_name allow|deny [!]aclname...
7343#
7344# At each supported vectoring point, the adaptation_access
7345# statements are processed in the order they appear in this
7346# configuration file. Statements pointing to the following services
7347# are ignored (i.e., skipped without checking their ACL):
7348#
7349# - services serving different vectoring points
7350# - "broken-but-bypassable" services
7351# - "up" services configured to ignore such transactions
7352# (e.g., based on the ICAP Transfer-Ignore header).
7353#
7354# When a set_name is used, all services in the set are checked
7355# using the same rules, to find the first applicable one. See
7356# adaptation_service_set for details.
7357#
7358# If an access list is checked and there is a match, the
7359# processing stops: For an "allow" rule, the corresponding
7360# adaptation service is used for the transaction. For a "deny"
7361# rule, no adaptation service is activated.
7362#
7363# It is currently not possible to apply more than one adaptation
7364# service at the same vectoring point to the same HTTP transaction.
7365#
7366# See also: icap_service and ecap_service
7367#
7368#Example:
7369#adaptation_access service_1 allow all
7370#Default:
7371# Allow, unless rules exist in squid.conf.
7372
7373# TAG: adaptation_service_iteration_limit
7374# Limits the number of iterations allowed when applying adaptation
7375# services to a message. If your longest adaptation set or chain
7376# may have more than 16 services, increase the limit beyond its
7377# default value of 16. If detecting infinite iteration loops sooner
7378# is critical, make the iteration limit match the actual number
7379# of services in your longest adaptation set or chain.
7380#
7381# Infinite adaptation loops are most likely with routing services.
7382#
7383# See also: icap_service routing=1
7384#Default:
7385# adaptation_service_iteration_limit 16
7386
7387# TAG: adaptation_masterx_shared_names
7388# For each master transaction (i.e., the HTTP request and response
7389# sequence, including all related ICAP and eCAP exchanges), Squid
7390# maintains a table of metadata. The table entries are (name, value)
7391# pairs shared among eCAP and ICAP exchanges. The table is destroyed
7392# with the master transaction.
7393#
7394# This option specifies the table entry names that Squid must accept
7395# from and forward to the adaptation transactions.
7396#
7397# An ICAP REQMOD or RESPMOD transaction may set an entry in the
7398# shared table by returning an ICAP header field with a name
7399# specified in adaptation_masterx_shared_names.
7400#
7401# An eCAP REQMOD or RESPMOD transaction may set an entry in the
7402# shared table by implementing the libecap::visitEachOption() API
7403# to provide an option with a name specified in
7404# adaptation_masterx_shared_names.
7405#
7406# Squid will store and forward the set entry to subsequent adaptation
7407# transactions within the same master transaction scope.
7408#
7409# Only one shared entry name is supported at this time.
7410#
7411#Example:
7412## share authentication information among ICAP services
7413#adaptation_masterx_shared_names X-Subscriber-ID
7414#Default:
7415# none
7416
7417# TAG: adaptation_meta
7418# This option allows Squid administrator to add custom ICAP request
7419# headers or eCAP options to Squid ICAP requests or eCAP transactions.
7420# Use it to pass custom authentication tokens and other
7421# transaction-state related meta information to an ICAP/eCAP service.
7422#
7423# The addition of a meta header is ACL-driven:
7424# adaptation_meta name value [!]aclname ...
7425#
7426# Processing for a given header name stops after the first ACL list match.
7427# Thus, it is impossible to add two headers with the same name. If no ACL
7428# lists match for a given header name, no such header is added. For
7429# example:
7430#
7431# # do not debug transactions except for those that need debugging
7432# adaptation_meta X-Debug 1 needs_debugging
7433#
7434# # log all transactions except for those that must remain secret
7435# adaptation_meta X-Log 1 !keep_secret
7436#
7437# # mark transactions from users in the "G 1" group
7438# adaptation_meta X-Authenticated-Groups "G 1" authed_as_G1
7439#
7440# The "value" parameter may be a regular squid.conf token or a "double
7441# quoted string". Within the quoted string, use backslash (\) to escape
7442# any character, which is currently only useful for escaping backslashes
7443# and double quotes. For example,
7444# "this string has one backslash (\\) and two \"quotes\""
7445#
7446# Used adaptation_meta header values may be logged via %note
7447# logformat code. If multiple adaptation_meta headers with the same name
7448# are used during master transaction lifetime, the header values are
7449# logged in the order they were used and duplicate values are ignored
7450# (only the first repeated value will be logged).
7451#Default:
7452# none
7453
7454# TAG: icap_retry
7455# This ACL determines which retriable ICAP transactions are
7456# retried. Transactions that received a complete ICAP response
7457# and did not have to consume or produce HTTP bodies to receive
7458# that response are usually retriable.
7459#
7460# icap_retry allow|deny [!]aclname ...
7461#
7462# Squid automatically retries some ICAP I/O timeouts and errors
7463# due to persistent connection race conditions.
7464#
7465# See also: icap_retry_limit
7466#Default:
7467# icap_retry deny all
7468
7469# TAG: icap_retry_limit
7470# Limits the number of retries allowed.
7471#
7472# Communication errors due to persistent connection race
7473# conditions are unavoidable, automatically retried, and do not
7474# count against this limit.
7475#
7476# See also: icap_retry
7477#Default:
7478# No retries are allowed.
7479
7480# DNS OPTIONS
7481# -----------------------------------------------------------------------------
7482
7483# TAG: check_hostnames
7484# For security and stability reasons Squid can check
7485# hostnames for Internet standard RFC compliance. If you want
7486# Squid to perform these checks turn this directive on.
7487#Default:
7488# check_hostnames off
7489
7490# TAG: allow_underscore
7491# Underscore characters is not strictly allowed in Internet hostnames
7492# but nevertheless used by many sites. Set this to off if you want
7493# Squid to be strict about the standard.
7494# This check is performed only when check_hostnames is set to on.
7495#Default:
7496# allow_underscore on
7497
7498# TAG: dns_retransmit_interval
7499# Initial retransmit interval for DNS queries. The interval is
7500# doubled each time all configured DNS servers have been tried.
7501#Default:
7502# dns_retransmit_interval 5 seconds
7503
7504# TAG: dns_timeout
7505# DNS Query timeout. If no response is received to a DNS query
7506# within this time all DNS servers for the queried domain
7507# are assumed to be unavailable.
7508#Default:
7509# dns_timeout 30 seconds
7510
7511# TAG: dns_packet_max
7512# Maximum number of bytes packet size to advertise via EDNS.
7513# Set to "none" to disable EDNS large packet support.
7514#
7515# For legacy reasons DNS UDP replies will default to 512 bytes which
7516# is too small for many responses. EDNS provides a means for Squid to
7517# negotiate receiving larger responses back immediately without having
7518# to failover with repeat requests. Responses larger than this limit
7519# will retain the old behaviour of failover to TCP DNS.
7520#
7521# Squid has no real fixed limit internally, but allowing packet sizes
7522# over 1500 bytes requires network jumbogram support and is usually not
7523# necessary.
7524#
7525# WARNING: The RFC also indicates that some older resolvers will reply
7526# with failure of the whole request if the extension is added. Some
7527# resolvers have already been identified which will reply with mangled
7528# EDNS response on occasion. Usually in response to many-KB jumbogram
7529# sizes being advertised by Squid.
7530# Squid will currently treat these both as an unable-to-resolve domain
7531# even if it would be resolvable without EDNS.
7532#Default:
7533# EDNS disabled
7534
7535# TAG: dns_defnames on|off
7536# Normally the RES_DEFNAMES resolver option is disabled
7537# (see res_init(3)). This prevents caches in a hierarchy
7538# from interpreting single-component hostnames locally. To allow
7539# Squid to handle single-component names, enable this option.
7540#Default:
7541# Search for single-label domain names is disabled.
7542
7543# TAG: dns_multicast_local on|off
7544# When set to on, Squid sends multicast DNS lookups on the local
7545# network for domains ending in .local and .arpa.
7546# This enables local servers and devices to be contacted in an
7547# ad-hoc or zero-configuration network environment.
7548#Default:
7549# Search for .local and .arpa names is disabled.
7550
7551# TAG: dns_nameservers
7552# Use this if you want to specify a list of DNS name servers
7553# (IP addresses) to use instead of those given in your
7554# /etc/resolv.conf file.
7555#
7556# On Windows platforms, if no value is specified here or in
7557# the /etc/resolv.conf file, the list of DNS name servers are
7558# taken from the Windows registry, both static and dynamic DHCP
7559# configurations are supported.
7560#
7561# Example: dns_nameservers 10.0.0.1 192.172.0.4
7562#Default:
7563# Use operating system definitions
7564
7565# TAG: hosts_file
7566# Location of the host-local IP name-address associations
7567# database. Most Operating Systems have such a file on different
7568# default locations:
7569# - Un*X & Linux: /etc/hosts
7570# - Windows NT/2000: %SystemRoot%\system32\drivers\etc\hosts
7571# (%SystemRoot% value install default is c:\winnt)
7572# - Windows XP/2003: %SystemRoot%\system32\drivers\etc\hosts
7573# (%SystemRoot% value install default is c:\windows)
7574# - Windows 9x/Me: %windir%\hosts
7575# (%windir% value is usually c:\windows)
7576# - Cygwin: /etc/hosts
7577#
7578# The file contains newline-separated definitions, in the
7579# form ip_address_in_dotted_form name [name ...] names are
7580# whitespace-separated. Lines beginning with an hash (#)
7581# character are comments.
7582#
7583# The file is checked at startup and upon configuration.
7584# If set to 'none', it won't be checked.
7585# If append_domain is used, that domain will be added to
7586# domain-local (i.e. not containing any dot character) host
7587# definitions.
7588#Default:
7589# hosts_file /etc/hosts
7590
7591# TAG: append_domain
7592# Appends local domain name to hostnames without any dots in
7593# them. append_domain must begin with a period.
7594#
7595# Be warned there are now Internet names with no dots in
7596# them using only top-domain names, so setting this may
7597# cause some Internet sites to become unavailable.
7598#
7599#Example:
7600# append_domain .yourdomain.com
7601#Default:
7602# Use operating system definitions
7603
7604# TAG: ignore_unknown_nameservers
7605# By default Squid checks that DNS responses are received
7606# from the same IP addresses they are sent to. If they
7607# don't match, Squid ignores the response and writes a warning
7608# message to cache.log. You can allow responses from unknown
7609# nameservers by setting this option to 'off'.
7610#Default:
7611# ignore_unknown_nameservers on
7612
7613# TAG: dns_v4_first
7614# With the IPv6 Internet being as fast or faster than IPv4 Internet
7615# for most networks Squid prefers to contact websites over IPv6.
7616#
7617# This option reverses the order of preference to make Squid contact
7618# dual-stack websites over IPv4 first. Squid will still perform both
7619# IPv6 and IPv4 DNS lookups before connecting.
7620#
7621# WARNING:
7622# This option will restrict the situations under which IPv6
7623# connectivity is used (and tested), potentially hiding network
7624# problems which would otherwise be detected and warned about.
7625#Default:
7626# dns_v4_first off
7627
7628# TAG: ipcache_size (number of entries)
7629# Maximum number of DNS IP cache entries.
7630#Default:
7631# ipcache_size 1024
7632
7633# TAG: ipcache_low (percent)
7634#Default:
7635# ipcache_low 90
7636
7637# TAG: ipcache_high (percent)
7638# The size, low-, and high-water marks for the IP cache.
7639#Default:
7640# ipcache_high 95
7641
7642# TAG: fqdncache_size (number of entries)
7643# Maximum number of FQDN cache entries.
7644#Default:
7645# fqdncache_size 1024
7646
7647# MISCELLANEOUS
7648# -----------------------------------------------------------------------------
7649
7650# TAG: configuration_includes_quoted_values on|off
7651# If set, Squid will recognize each "quoted string" after a configuration
7652# directive as a single parameter. The quotes are stripped before the
7653# parameter value is interpreted or used.
7654# See "Values with spaces, quotes, and other special characters"
7655# section for more details.
7656#Default:
7657# configuration_includes_quoted_values off
7658
7659# TAG: memory_pools on|off
7660# If set, Squid will keep pools of allocated (but unused) memory
7661# available for future use. If memory is a premium on your
7662# system and you believe your malloc library outperforms Squid
7663# routines, disable this.
7664#Default:
7665# memory_pools on
7666
7667# TAG: memory_pools_limit (bytes)
7668# Used only with memory_pools on:
7669# memory_pools_limit 50 MB
7670#
7671# If set to a non-zero value, Squid will keep at most the specified
7672# limit of allocated (but unused) memory in memory pools. All free()
7673# requests that exceed this limit will be handled by your malloc
7674# library. Squid does not pre-allocate any memory, just safe-keeps
7675# objects that otherwise would be free()d. Thus, it is safe to set
7676# memory_pools_limit to a reasonably high value even if your
7677# configuration will use less memory.
7678#
7679# If set to none, Squid will keep all memory it can. That is, there
7680# will be no limit on the total amount of memory used for safe-keeping.
7681#
7682# To disable memory allocation optimization, do not set
7683# memory_pools_limit to 0 or none. Set memory_pools to "off" instead.
7684#
7685# An overhead for maintaining memory pools is not taken into account
7686# when the limit is checked. This overhead is close to four bytes per
7687# object kept. However, pools may actually _save_ memory because of
7688# reduced memory thrashing in your malloc library.
7689#Default:
7690# memory_pools_limit 5 MB
7691
7692# TAG: forwarded_for on|off|transparent|truncate|delete
7693# If set to "on", Squid will append your client's IP address
7694# in the HTTP requests it forwards. By default it looks like:
7695#
7696# X-Forwarded-For: 192.1.2.3
7697#
7698# If set to "off", it will appear as
7699#
7700# X-Forwarded-For: unknown
7701#
7702# If set to "transparent", Squid will not alter the
7703# X-Forwarded-For header in any way.
7704#
7705# If set to "delete", Squid will delete the entire
7706# X-Forwarded-For header.
7707#
7708# If set to "truncate", Squid will remove all existing
7709# X-Forwarded-For entries, and place the client IP as the sole entry.
7710#Default:
7711# forwarded_for on
7712
7713# TAG: cachemgr_passwd
7714# Specify passwords for cachemgr operations.
7715#
7716# Usage: cachemgr_passwd password action action ...
7717#
7718# Some valid actions are (see cache manager menu for a full list):
7719# 5min
7720# 60min
7721# asndb
7722# authenticator
7723# cbdata
7724# client_list
7725# comm_incoming
7726# config *
7727# counters
7728# delay
7729# digest_stats
7730# dns
7731# events
7732# filedescriptors
7733# fqdncache
7734# histograms
7735# http_headers
7736# info
7737# io
7738# ipcache
7739# mem
7740# menu
7741# netdb
7742# non_peers
7743# objects
7744# offline_toggle *
7745# pconn
7746# peer_select
7747# reconfigure *
7748# redirector
7749# refresh
7750# server_list
7751# shutdown *
7752# store_digest
7753# storedir
7754# utilization
7755# via_headers
7756# vm_objects
7757#
7758# * Indicates actions which will not be performed without a
7759# valid password, others can be performed if not listed here.
7760#
7761# To disable an action, set the password to "disable".
7762# To allow performing an action without a password, set the
7763# password to "none".
7764#
7765# Use the keyword "all" to set the same password for all actions.
7766#
7767#Example:
7768# cachemgr_passwd secret shutdown
7769# cachemgr_passwd lesssssssecret info stats/objects
7770# cachemgr_passwd disable all
7771#Default:
7772# No password. Actions which require password are denied.
7773
7774# TAG: client_db on|off
7775# If you want to disable collecting per-client statistics,
7776# turn off client_db here.
7777#Default:
7778# client_db on
7779
7780# TAG: refresh_all_ims on|off
7781# When you enable this option, squid will always check
7782# the origin server for an update when a client sends an
7783# If-Modified-Since request. Many browsers use IMS
7784# requests when the user requests a reload, and this
7785# ensures those clients receive the latest version.
7786#
7787# By default (off), squid may return a Not Modified response
7788# based on the age of the cached version.
7789#Default:
7790# refresh_all_ims off
7791
7792# TAG: reload_into_ims on|off
7793# When you enable this option, client no-cache or ``reload''
7794# requests will be changed to If-Modified-Since requests.
7795# Doing this VIOLATES the HTTP standard. Enabling this
7796# feature could make you liable for problems which it
7797# causes.
7798#
7799# see also refresh_pattern for a more selective approach.
7800#Default:
7801# reload_into_ims off
7802
7803# TAG: connect_retries
7804# This sets the maximum number of connection attempts made for each
7805# TCP connection. The connect_retries attempts must all still
7806# complete within the connection timeout period.
7807#
7808# The default is not to re-try if the first connection attempt fails.
7809# The (not recommended) maximum is 10 tries.
7810#
7811# A warning message will be generated if it is set to a too-high
7812# value and the configured value will be over-ridden.
7813#
7814# Note: These re-tries are in addition to forward_max_tries
7815# which limit how many different addresses may be tried to find
7816# a useful server.
7817#Default:
7818# Do not retry failed connections.
7819
7820# TAG: retry_on_error
7821# If set to ON Squid will automatically retry requests when
7822# receiving an error response with status 403 (Forbidden),
7823# 500 (Internal Error), 501 or 503 (Service not available).
7824# Status 502 and 504 (Gateway errors) are always retried.
7825#
7826# This is mainly useful if you are in a complex cache hierarchy to
7827# work around access control errors.
7828#
7829# NOTE: This retry will attempt to find another working destination.
7830# Which is different from the server which just failed.
7831#Default:
7832# retry_on_error off
7833
7834# TAG: as_whois_server
7835# WHOIS server to query for AS numbers. NOTE: AS numbers are
7836# queried only when Squid starts up, not for every request.
7837#Default:
7838# as_whois_server whois.ra.net
7839
7840# TAG: offline_mode
7841# Enable this option and Squid will never try to validate cached
7842# objects.
7843#Default:
7844# offline_mode off
7845
7846# TAG: uri_whitespace
7847# What to do with requests that have whitespace characters in the
7848# URI. Options:
7849#
7850# strip: The whitespace characters are stripped out of the URL.
7851# This is the behavior recommended by RFC2396 and RFC3986
7852# for tolerant handling of generic URI.
7853# NOTE: This is one difference between generic URI and HTTP URLs.
7854#
7855# deny: The request is denied. The user receives an "Invalid
7856# Request" message.
7857# This is the behaviour recommended by RFC2616 for safe
7858# handling of HTTP request URL.
7859#
7860# allow: The request is allowed and the URI is not changed. The
7861# whitespace characters remain in the URI. Note the
7862# whitespace is passed to redirector processes if they
7863# are in use.
7864# Note this may be considered a violation of RFC2616
7865# request parsing where whitespace is prohibited in the
7866# URL field.
7867#
7868# encode: The request is allowed and the whitespace characters are
7869# encoded according to RFC1738.
7870#
7871# chop: The request is allowed and the URI is chopped at the
7872# first whitespace.
7873#
7874#
7875# NOTE the current Squid implementation of encode and chop violates
7876# RFC2616 by not using a 301 redirect after altering the URL.
7877#Default:
7878# uri_whitespace strip
7879
7880# TAG: chroot
7881# Specifies a directory where Squid should do a chroot() while
7882# initializing. This also causes Squid to fully drop root
7883# privileges after initializing. This means, for example, if you
7884# use a HTTP port less than 1024 and try to reconfigure, you may
7885# get an error saying that Squid can not open the port.
7886#Default:
7887# none
7888
7889# TAG: balance_on_multiple_ip
7890# Modern IP resolvers in squid sort lookup results by preferred access.
7891# By default squid will use these IP in order and only rotates to
7892# the next listed when the most preffered fails.
7893#
7894# Some load balancing servers based on round robin DNS have been
7895# found not to preserve user session state across requests
7896# to different IP addresses.
7897#
7898# Enabling this directive Squid rotates IP's per request.
7899#Default:
7900# balance_on_multiple_ip off
7901
7902# TAG: pipeline_prefetch
7903# HTTP clients may send a pipeline of 1+N requests to Squid using a
7904# single connection, without waiting for Squid to respond to the first
7905# of those requests. This option limits the number of concurrent
7906# requests Squid will try to handle in parallel. If set to N, Squid
7907# will try to receive and process up to 1+N requests on the same
7908# connection concurrently.
7909#
7910# Defaults to 0 (off) for bandwidth management and access logging
7911# reasons.
7912#
7913# NOTE: pipelining requires persistent connections to clients.
7914#
7915# WARNING: pipelining breaks NTLM and Negotiate/Kerberos authentication.
7916#Default:
7917# Do not pre-parse pipelined requests.
7918
7919# TAG: high_response_time_warning (msec)
7920# If the one-minute median response time exceeds this value,
7921# Squid prints a WARNING with debug level 0 to get the
7922# administrators attention. The value is in milliseconds.
7923#Default:
7924# disabled.
7925
7926# TAG: high_page_fault_warning
7927# If the one-minute average page fault rate exceeds this
7928# value, Squid prints a WARNING with debug level 0 to get
7929# the administrators attention. The value is in page faults
7930# per second.
7931#Default:
7932# disabled.
7933
7934# TAG: high_memory_warning
7935# Note: This option is only available if Squid is rebuilt with the
7936# GNU Malloc with mstats()
7937#
7938# If the memory usage (as determined by gnumalloc, if available and used)
7939# exceeds this amount, Squid prints a WARNING with debug level 0 to get
7940# the administrators attention.
7941#Default:
7942# disabled.
7943
7944# TAG: sleep_after_fork (microseconds)
7945# When this is set to a non-zero value, the main Squid process
7946# sleeps the specified number of microseconds after a fork()
7947# system call. This sleep may help the situation where your
7948# system reports fork() failures due to lack of (virtual)
7949# memory. Note, however, if you have a lot of child
7950# processes, these sleep delays will add up and your
7951# Squid will not service requests for some amount of time
7952# until all the child processes have been started.
7953# On Windows value less then 1000 (1 milliseconds) are
7954# rounded to 1000.
7955#Default:
7956# sleep_after_fork 0
7957
7958# TAG: windows_ipaddrchangemonitor on|off
7959# Note: This option is only available if Squid is rebuilt with the
7960# MS Windows
7961#
7962# On Windows Squid by default will monitor IP address changes and will
7963# reconfigure itself after any detected event. This is very useful for
7964# proxies connected to internet with dial-up interfaces.
7965# In some cases (a Proxy server acting as VPN gateway is one) it could be
7966# desiderable to disable this behaviour setting this to 'off'.
7967# Note: after changing this, Squid service must be restarted.
7968#Default:
7969# windows_ipaddrchangemonitor on
7970
7971# TAG: eui_lookup
7972# Whether to lookup the EUI or MAC address of a connected client.
7973#Default:
7974# eui_lookup on
7975
7976# TAG: max_filedescriptors
7977# Reduce the maximum number of filedescriptors supported below
7978# the usual operating system defaults.
7979#
7980# Remove from squid.conf to inherit the current ulimit setting.
7981#
7982# Note: Changing this requires a restart of Squid. Also
7983# not all I/O types supports large values (eg on Windows).
7984#Default:
7985# Use operating system limits set by ulimit.