· 6 years ago · Sep 27, 2019, 07:30 AM
1# WELCOME TO SQUID 4.6
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# Logformat Macros
103#
104# Logformat macros can be used in many places outside of the logformat
105# directive. In theory, all of the logformat codes can be used as %macros,
106# where they are supported. In practice, a %macro expands as a dash (-) when
107# the transaction does not yet have enough information and a value is needed.
108#
109# There is no definitive list of what tokens are available at the various
110# stages of the transaction.
111#
112# And some information may already be available to Squid but not yet
113# committed where the macro expansion code can access it (report
114# such instances!). The macro will be expanded into a single dash
115# ('-') in such cases. Not all macros have been tested.
116#
117
118# TAG: broken_vary_encoding
119# This option is not yet supported by Squid-3.
120#Default:
121# none
122
123# TAG: cache_vary
124# This option is not yet supported by Squid-3.
125#Default:
126# none
127
128# TAG: error_map
129# This option is not yet supported by Squid-3.
130#Default:
131# none
132
133# TAG: external_refresh_check
134# This option is not yet supported by Squid-3.
135#Default:
136# none
137
138# TAG: location_rewrite_program
139# This option is not yet supported by Squid-3.
140#Default:
141# none
142
143# TAG: refresh_stale_hit
144# This option is not yet supported by Squid-3.
145#Default:
146# none
147
148# TAG: cache_peer_domain
149# Replace with dstdomain ACLs and cache_peer_access.
150#Default:
151# none
152
153# TAG: ie_refresh
154# Remove this line. The behaviour enabled by this is no longer needed.
155#Default:
156# none
157
158# TAG: sslproxy_cafile
159# Remove this line. Use tls_outgoing_options cafile= instead.
160#Default:
161# none
162
163# TAG: sslproxy_capath
164# Remove this line. Use tls_outgoing_options capath= instead.
165#Default:
166# none
167
168# TAG: sslproxy_cipher
169# Remove this line. Use tls_outgoing_options cipher= instead.
170#Default:
171# none
172
173# TAG: sslproxy_client_certificate
174# Remove this line. Use tls_outgoing_options cert= instead.
175#Default:
176# none
177
178# TAG: sslproxy_client_key
179# Remove this line. Use tls_outgoing_options key= instead.
180#Default:
181# none
182
183# TAG: sslproxy_flags
184# Remove this line. Use tls_outgoing_options flags= instead.
185#Default:
186# none
187
188# TAG: sslproxy_options
189# Remove this line. Use tls_outgoing_options options= instead.
190#Default:
191# none
192
193# TAG: sslproxy_version
194# Remove this line. Use tls_outgoing_options options= instead.
195#Default:
196# none
197
198# TAG: hierarchy_stoplist
199# Remove this line. Use always_direct or cache_peer_access ACLs instead if you need to prevent cache_peer use.
200#Default:
201# none
202
203# TAG: log_access
204# Remove this line. Use acls with access_log directives to control access logging
205#Default:
206# none
207
208# TAG: log_icap
209# Remove this line. Use acls with icap_log directives to control icap logging
210#Default:
211# none
212
213# TAG: ignore_ims_on_miss
214# Remove this line. The HTTP/1.1 feature is now configured by 'cache_miss_revalidate'.
215#Default:
216# none
217
218# TAG: balance_on_multiple_ip
219# Remove this line. Squid performs a 'Happy Eyeballs' algorithm, this multiple-IP algorithm is not longer relevant.
220#Default:
221# none
222
223# TAG: chunked_request_body_max_size
224# Remove this line. Squid is now HTTP/1.1 compliant.
225#Default:
226# none
227
228# TAG: dns_v4_fallback
229# Remove this line. Squid performs a 'Happy Eyeballs' algorithm, the 'fallback' algorithm is no longer relevant.
230#Default:
231# none
232
233# TAG: emulate_httpd_log
234# Replace this with an access_log directive using the format 'common' or 'combined'.
235#Default:
236# none
237
238# TAG: forward_log
239# Use a regular access.log with ACL limiting it to MISS events.
240#Default:
241# none
242
243# TAG: ftp_list_width
244# Remove this line. Configure FTP page display using the CSS controls in errorpages.css instead.
245#Default:
246# none
247
248# TAG: ignore_expect_100
249# Remove this line. The HTTP/1.1 feature is now fully supported by default.
250#Default:
251# none
252
253# TAG: log_fqdn
254# Remove this option from your config. To log FQDN use %>A in the log format.
255#Default:
256# none
257
258# TAG: log_ip_on_direct
259# Remove this option from your config. To log server or peer names use %<A in the log format.
260#Default:
261# none
262
263# TAG: maximum_single_addr_tries
264# Replaced by connect_retries. The behaviour has changed, please read the documentation before altering.
265#Default:
266# none
267
268# TAG: referer_log
269# Replace this with an access_log directive using the format 'referrer'.
270#Default:
271# none
272
273# TAG: update_headers
274# Remove this line. The feature is supported by default in storage types where update is implemented.
275#Default:
276# none
277
278# TAG: url_rewrite_concurrency
279# Remove this line. Set the 'concurrency=' option of url_rewrite_children instead.
280#Default:
281# none
282
283# TAG: useragent_log
284# Replace this with an access_log directive using the format 'useragent'.
285#Default:
286# none
287
288# TAG: dns_testnames
289# Remove this line. DNS is no longer tested on startup.
290#Default:
291# none
292
293# TAG: extension_methods
294# Remove this line. All valid methods for HTTP are accepted by default.
295#Default:
296# none
297
298# TAG: zero_buffers
299#Default:
300# none
301
302# TAG: incoming_rate
303#Default:
304# none
305
306# TAG: server_http11
307# Remove this line. HTTP/1.1 is supported by default.
308#Default:
309# none
310
311# TAG: upgrade_http0.9
312# Remove this line. ICY/1.0 streaming protocol is supported by default.
313#Default:
314# none
315
316# TAG: zph_local
317# Alter these entries. Use the qos_flows directive instead.
318#Default:
319# none
320
321# TAG: header_access
322# Since squid-3.0 replace with request_header_access or reply_header_access
323# depending on whether you wish to match client requests or server replies.
324#Default:
325# none
326
327# TAG: httpd_accel_no_pmtu_disc
328# Since squid-3.0 use the 'disable-pmtu-discovery' flag on http_port instead.
329#Default:
330# none
331
332# TAG: wais_relay_host
333# Replace this line with 'cache_peer' configuration.
334#Default:
335# none
336
337# TAG: wais_relay_port
338# Replace this line with 'cache_peer' configuration.
339#Default:
340# none
341
342# OPTIONS FOR SMP
343# -----------------------------------------------------------------------------
344
345# TAG: workers
346# Number of main Squid processes or "workers" to fork and maintain.
347# 0: "no daemon" mode, like running "squid -N ..."
348# 1: "no SMP" mode, start one main Squid process daemon (default)
349# N: start N main Squid process daemons (i.e., SMP mode)
350#
351# In SMP mode, each worker does nearly all what a single Squid daemon
352# does (e.g., listen on http_port and forward HTTP requests).
353#Default:
354# SMP support disabled.
355
356# TAG: cpu_affinity_map
357# Usage: cpu_affinity_map process_numbers=P1,P2,... cores=C1,C2,...
358#
359# Sets 1:1 mapping between Squid processes and CPU cores. For example,
360#
361# cpu_affinity_map process_numbers=1,2,3,4 cores=1,3,5,7
362#
363# affects processes 1 through 4 only and places them on the first
364# four even cores, starting with core #1.
365#
366# CPU cores are numbered starting from 1. Requires support for
367# sched_getaffinity(2) and sched_setaffinity(2) system calls.
368#
369# Multiple cpu_affinity_map options are merged.
370#
371# See also: workers
372#Default:
373# Let operating system decide.
374
375# TAG: shared_memory_locking on|off
376# Whether to ensure that all required shared memory is available by
377# "locking" that shared memory into RAM when Squid starts. The
378# alternative is faster startup time followed by slightly slower
379# performance and, if not enough RAM is actually available during
380# runtime, mysterious crashes.
381#
382# SMP Squid uses many shared memory segments. These segments are
383# brought into Squid memory space using an mmap(2) system call. During
384# Squid startup, the mmap() call often succeeds regardless of whether
385# the system has enough RAM. In general, Squid cannot tell whether the
386# kernel applies this "optimistic" memory allocation policy (but
387# popular modern kernels usually use it).
388#
389# Later, if Squid attempts to actually access the mapped memory
390# regions beyond what the kernel is willing to allocate, the
391# "optimistic" kernel simply kills Squid kid with a SIGBUS signal.
392# Some of the memory limits enforced by the kernel are currently
393# poorly understood: We do not know how to detect and check them. This
394# option ensures that the mapped memory will be available.
395#
396# This option may have a positive performance side-effect: Locking
397# memory at start avoids runtime paging I/O. Paging slows Squid down.
398#
399# Locking memory may require a large enough RLIMIT_MEMLOCK OS limit,
400# CAP_IPC_LOCK capability, or equivalent.
401#Default:
402# shared_memory_locking off
403
404# TAG: hopeless_kid_revival_delay time-units
405# Normally, when a kid process dies, Squid immediately restarts the
406# kid. A kid experiencing frequent deaths is marked as "hopeless" for
407# the duration specified by this directive. Hopeless kids are not
408# automatically restarted.
409#
410# Currently, zero values are not supported because they result in
411# misconfigured SMP Squid instances running forever, endlessly
412# restarting each dying kid. To effectively disable hopeless kids
413# revival, set the delay to a huge value (e.g., 1 year).
414#
415# Reconfiguration also clears all hopeless kids designations, allowing
416# for manual revival of hopeless kids.
417#Default:
418# hopeless_kid_revival_delay 1 hour
419
420# OPTIONS FOR AUTHENTICATION
421# -----------------------------------------------------------------------------
422
423# TAG: auth_param
424# This is used to define parameters for the various authentication
425# schemes supported by Squid.
426#
427# format: auth_param scheme parameter [setting]
428#
429# The order in which authentication schemes are presented to the client is
430# dependent on the order the scheme first appears in config file. IE
431# has a bug (it's not RFC 2617 compliant) in that it will use the basic
432# scheme if basic is the first entry presented, even if more secure
433# schemes are presented. For now use the order in the recommended
434# settings section below. If other browsers have difficulties (don't
435# recognize the schemes offered even if you are using basic) either
436# put basic first, or disable the other schemes (by commenting out their
437# program entry).
438#
439# Once an authentication scheme is fully configured, it can only be
440# shutdown by shutting squid down and restarting. Changes can be made on
441# the fly and activated with a reconfigure. I.E. You can change to a
442# different helper, but not unconfigure the helper completely.
443#
444# Please note that while this directive defines how Squid processes
445# authentication it does not automatically activate authentication.
446# To use authentication you must in addition make use of ACLs based
447# on login name in http_access (proxy_auth, proxy_auth_regex or
448# external with %LOGIN used in the format tag). The browser will be
449# challenged for authentication on the first such acl encountered
450# in http_access processing and will also be re-challenged for new
451# login credentials if the request is being denied by a proxy_auth
452# type acl.
453#
454# WARNING: authentication can't be used in a transparently intercepting
455# proxy as the client then thinks it is talking to an origin server and
456# not the proxy. This is a limitation of bending the TCP/IP protocol to
457# transparently intercepting port 80, not a limitation in Squid.
458# Ports flagged 'transparent', 'intercept', or 'tproxy' have
459# authentication disabled.
460#
461# === Parameters common to all schemes. ===
462#
463# "program" cmdline
464# Specifies the command for the external authenticator.
465#
466# By default, each authentication scheme is not used unless a
467# program is specified.
468#
469# See http://wiki.squid-cache.org/Features/AddonHelpers for
470# more details on helper operations and creating your own.
471#
472# "key_extras" format
473# Specifies a string to be append to request line format for
474# the authentication helper. "Quoted" format values may contain
475# spaces and logformat %macros. In theory, any logformat %macro
476# can be used. In practice, a %macro expands as a dash (-) if
477# the helper request is sent before the required macro
478# information is available to Squid.
479#
480# By default, Squid uses request formats provided in
481# scheme-specific examples below (search for %credentials).
482#
483# The expanded key_extras value is added to the Squid credentials
484# cache and, hence, will affect authentication. It can be used to
485# autenticate different users with identical user names (e.g.,
486# when user authentication depends on http_port).
487#
488# Avoid adding frequently changing information to key_extras. For
489# example, if you add user source IP, and it changes frequently
490# in your environment, then max_user_ip ACL is going to treat
491# every user+IP combination as a unique "user", breaking the ACL
492# and wasting a lot of memory on those user records. It will also
493# force users to authenticate from scratch whenever their IP
494# changes.
495#
496# "realm" string
497# Specifies the protection scope (aka realm name) which is to be
498# reported to the client for the authentication scheme. It is
499# commonly part of the text the user will see when prompted for
500# their username and password.
501#
502# For Basic the default is "Squid proxy-caching web server".
503# For Digest there is no default, this parameter is mandatory.
504# For NTLM and Negotiate this parameter is ignored.
505#
506# "children" numberofchildren [startup=N] [idle=N] [concurrency=N]
507# [queue-size=N] [on-persistent-overload=action]
508#
509# The maximum number of authenticator processes to spawn. If
510# you start too few Squid will have to wait for them to process
511# a backlog of credential verifications, slowing it down. When
512# password verifications are done via a (slow) network you are
513# likely to need lots of authenticator processes.
514#
515# The startup= and idle= options permit some skew in the exact
516# amount run. A minimum of startup=N will begin during startup
517# and reconfigure. Squid will start more in groups of up to
518# idle=N in an attempt to meet traffic needs and to keep idle=N
519# free above those traffic needs up to the maximum.
520#
521# The concurrency= option sets the number of concurrent requests
522# the helper can process. The default of 0 is used for helpers
523# who only supports one request at a time. Setting this to a
524# number greater than 0 changes the protocol used to include a
525# channel ID field first on the request/response line, allowing
526# multiple requests to be sent to the same helper in parallel
527# without waiting for the response.
528#
529# Concurrency must not be set unless it's known the helper
530# supports the input format with channel-ID fields.
531#
532# The queue-size option sets the maximum number of queued
533# requests. A request is queued when no existing child can
534# accept it due to concurrency limit and no new child can be
535# started due to numberofchildren limit. The default maximum is
536# 2*numberofchildren. Squid is allowed to temporarily exceed the
537# configured maximum, marking the affected helper as
538# "overloaded". If the helper overload lasts more than 3
539# minutes, the action prescribed by the on-persistent-overload
540# option applies.
541#
542# The on-persistent-overload=action option specifies Squid
543# reaction to a new helper request arriving when the helper
544# has been overloaded for more that 3 minutes already. The number
545# of queued requests determines whether the helper is overloaded
546# (see the queue-size option).
547#
548# Two actions are supported:
549#
550# die Squid worker quits. This is the default behavior.
551#
552# ERR Squid treats the helper request as if it was
553# immediately submitted, and the helper immediately
554# replied with an ERR response. This action has no effect
555# on the already queued and in-progress helper requests.
556#
557# NOTE: NTLM and Negotiate schemes do not support concurrency
558# in the Squid code module even though some helpers can.
559#
560#
561#
562# === Example Configuration ===
563#
564# This configuration displays the recommended authentication scheme
565# order from most to least secure with recommended minimum configuration
566# settings for each scheme:
567#
568##auth_param negotiate program <uncomment and complete this line to activate>
569##auth_param negotiate children 20 startup=0 idle=1
570##auth_param negotiate keep_alive on
571##
572##auth_param digest program <uncomment and complete this line to activate>
573##auth_param digest children 20 startup=0 idle=1
574##auth_param digest realm Squid proxy-caching web server
575##auth_param digest nonce_garbage_interval 5 minutes
576##auth_param digest nonce_max_duration 30 minutes
577##auth_param digest nonce_max_count 50
578##
579##auth_param ntlm program <uncomment and complete this line to activate>
580##auth_param ntlm children 20 startup=0 idle=1
581##auth_param ntlm keep_alive on
582##
583##auth_param basic program <uncomment and complete this line>
584##auth_param basic children 5 startup=5 idle=1
585##auth_param basic realm Squid proxy-caching web server
586##auth_param basic credentialsttl 2 hours
587#Default:
588# none
589
590# TAG: authenticate_cache_garbage_interval
591# The time period between garbage collection across the username cache.
592# This is a trade-off between memory utilization (long intervals - say
593# 2 days) and CPU (short intervals - say 1 minute). Only change if you
594# have good reason to.
595#Default:
596# authenticate_cache_garbage_interval 1 hour
597
598# TAG: authenticate_ttl
599# The time a user & their credentials stay in the logged in
600# user cache since their last request. When the garbage
601# interval passes, all user credentials that have passed their
602# TTL are removed from memory.
603#Default:
604# authenticate_ttl 1 hour
605
606# TAG: authenticate_ip_ttl
607# If you use proxy authentication and the 'max_user_ip' ACL,
608# this directive controls how long Squid remembers the IP
609# addresses associated with each user. Use a small value
610# (e.g., 60 seconds) if your users might change addresses
611# quickly, as is the case with dialup. You might be safe
612# using a larger value (e.g., 2 hours) in a corporate LAN
613# environment with relatively static address assignments.
614#Default:
615# authenticate_ip_ttl 1 second
616
617# ACCESS CONTROLS
618# -----------------------------------------------------------------------------
619
620# TAG: external_acl_type
621# This option defines external acl classes using a helper program
622# to look up the status
623#
624# external_acl_type name [options] FORMAT /path/to/helper [helper arguments]
625#
626# Options:
627#
628# ttl=n TTL in seconds for cached results (defaults to 3600
629# for 1 hour)
630#
631# negative_ttl=n
632# TTL for cached negative lookups (default same
633# as ttl)
634#
635# grace=n Percentage remaining of TTL where a refresh of a
636# cached entry should be initiated without needing to
637# wait for a new reply. (default is for no grace period)
638#
639# cache=n The maximum number of entries in the result cache. The
640# default limit is 262144 entries. Each cache entry usually
641# consumes at least 256 bytes. Squid currently does not remove
642# expired cache entries until the limit is reached, so a proxy
643# will sooner or later reach the limit. The expanded FORMAT
644# value is used as the cache key, so if the details in FORMAT
645# are highly variable, a larger cache may be needed to produce
646# reduction in helper load.
647#
648# children-max=n
649# Maximum number of acl helper processes spawned to service
650# external acl lookups of this type. (default 5)
651#
652# children-startup=n
653# Minimum number of acl helper processes to spawn during
654# startup and reconfigure to service external acl lookups
655# of this type. (default 0)
656#
657# children-idle=n
658# Number of acl helper processes to keep ahead of traffic
659# loads. Squid will spawn this many at once whenever load
660# rises above the capabilities of existing processes.
661# Up to the value of children-max. (default 1)
662#
663# concurrency=n concurrency level per process. Only used with helpers
664# capable of processing more than one query at a time.
665#
666# queue-size=N The queue-size option sets the maximum number of
667# queued requests. A request is queued when no existing
668# helper can accept it due to concurrency limit and no
669# new helper can be started due to children-max limit.
670# If the queued requests exceed queue size, the acl is
671# ignored. The default value is set to 2*children-max.
672#
673# protocol=2.5 Compatibility mode for Squid-2.5 external acl helpers.
674#
675# ipv4 / ipv6 IP protocol used to communicate with this helper.
676# The default is to auto-detect IPv6 and use it when available.
677#
678#
679# FORMAT is a series of %macro codes. See logformat directive for a full list
680# of the accepted codes. Although note that at the time of any external ACL
681# being tested data may not be available and thus some %macro expand to '-'.
682#
683# In addition to the logformat codes; when processing external ACLs these
684# additional macros are made available:
685#
686# %ACL The name of the ACL being tested.
687#
688# %DATA The ACL arguments specified in the referencing config
689# 'acl ... external' line, separated by spaces (an
690# "argument string"). see acl external.
691#
692# If there are no ACL arguments %DATA expands to '-'.
693#
694# If you do not specify a DATA macro inside FORMAT,
695# Squid automatically appends %DATA to your FORMAT.
696# Note that Squid-3.x may expand %DATA to whitespace
697# or nothing in this case.
698#
699# By default, Squid applies URL-encoding to each ACL
700# argument inside the argument string. If an explicit
701# encoding modifier is used (e.g., %#DATA), then Squid
702# encodes the whole argument string as a single token
703# (e.g., with %#DATA, spaces between arguments become
704# %20).
705#
706# If SSL is enabled, the following formating codes become available:
707#
708# %USER_CERT SSL User certificate in PEM format
709# %USER_CERTCHAIN SSL User certificate chain in PEM format
710# %USER_CERT_xx SSL User certificate subject attribute xx
711# %USER_CA_CERT_xx SSL User certificate issuer attribute xx
712#
713#
714# NOTE: all other format codes accepted by older Squid versions
715# are deprecated.
716#
717#
718# General request syntax:
719#
720# [channel-ID] FORMAT-values
721#
722#
723# FORMAT-values consists of transaction details expanded with
724# whitespace separation per the config file FORMAT specification
725# using the FORMAT macros listed above.
726#
727# Request values sent to the helper are URL escaped to protect
728# each value in requests against whitespaces.
729#
730# If using protocol=2.5 then the request sent to the helper is not
731# URL escaped to protect against whitespace.
732#
733# NOTE: protocol=3.0 is deprecated as no longer necessary.
734#
735# When using the concurrency= option the protocol is changed by
736# introducing a query channel tag in front of the request/response.
737# The query channel tag is a number between 0 and concurrency-1.
738# This value must be echoed back unchanged to Squid as the first part
739# of the response relating to its request.
740#
741#
742# The helper receives lines expanded per the above format specification
743# and for each input line returns 1 line starting with OK/ERR/BH result
744# code and optionally followed by additional keywords with more details.
745#
746#
747# General result syntax:
748#
749# [channel-ID] result keyword=value ...
750#
751# Result consists of one of the codes:
752#
753# OK
754# the ACL test produced a match.
755#
756# ERR
757# the ACL test does not produce a match.
758#
759# BH
760# An internal error occurred in the helper, preventing
761# a result being identified.
762#
763# The meaning of 'a match' is determined by your squid.conf
764# access control configuration. See the Squid wiki for details.
765#
766# Defined keywords:
767#
768# user= The users name (login)
769#
770# password= The users password (for login= cache_peer option)
771#
772# message= Message describing the reason for this response.
773# Available as %o in error pages.
774# Useful on (ERR and BH results).
775#
776# tag= Apply a tag to a request. Only sets a tag once,
777# does not alter existing tags.
778#
779# log= String to be logged in access.log. Available as
780# %ea in logformat specifications.
781#
782# clt_conn_tag= Associates a TAG with the client TCP connection.
783# Please see url_rewrite_program related documentation
784# for this kv-pair.
785#
786# Any keywords may be sent on any response whether OK, ERR or BH.
787#
788# All response keyword values need to be a single token with URL
789# escaping, or enclosed in double quotes (") and escaped using \ on
790# any double quotes or \ characters within the value. The wrapping
791# double quotes are removed before the value is interpreted by Squid.
792# \r and \n are also replace by CR and LF.
793#
794# Some example key values:
795#
796# user=John%20Smith
797# user="John Smith"
798# user="J. \"Bob\" Smith"
799#Default:
800# none
801
802# TAG: acl
803# Defining an Access List
804#
805# Every access list definition must begin with an aclname and acltype,
806# followed by either type-specific arguments or a quoted filename that
807# they are read from.
808#
809# acl aclname acltype argument ...
810# acl aclname acltype "file" ...
811#
812# When using "file", the file should contain one item per line.
813#
814#
815# ACL Options
816#
817# Some acl types supports options which changes their default behaviour:
818#
819# -i,+i By default, regular expressions are CASE-SENSITIVE. To make them
820# case-insensitive, use the -i option. To return case-sensitive
821# use the +i option between patterns, or make a new ACL line
822# without -i.
823#
824# -n Disable lookups and address type conversions. If lookup or
825# conversion is required because the parameter type (IP or
826# domain name) does not match the message address type (domain
827# name or IP), then the ACL would immediately declare a mismatch
828# without any warnings or lookups.
829#
830# -m[=delimiters]
831# Perform a list membership test, interpreting values as
832# comma-separated token lists and matching against individual
833# tokens instead of whole values.
834# The optional "delimiters" parameter specifies one or more
835# alternative non-alphanumeric delimiter characters.
836# non-alphanumeric delimiter characters.
837#
838# -- Used to stop processing all options, in the case the first acl
839# value has '-' character as first character (for example the '-'
840# is a valid domain name)
841#
842# Some acl types require suspending the current request in order
843# to access some external data source.
844# Those which do are marked with the tag [slow], those which
845# don't are marked as [fast].
846# See http://wiki.squid-cache.org/SquidFaq/SquidAcl
847# for further information
848#
849# ***** ACL TYPES AVAILABLE *****
850#
851# acl aclname src ip-address/mask ... # clients IP address [fast]
852# acl aclname src addr1-addr2/mask ... # range of addresses [fast]
853# acl aclname dst [-n] ip-address/mask ... # URL host's IP address [slow]
854# acl aclname localip ip-address/mask ... # IP address the client connected to [fast]
855#
856#if USE_SQUID_EUI
857# acl aclname arp mac-address ...
858# acl aclname eui64 eui64-address ...
859# # [fast]
860# # MAC (EUI-48) and EUI-64 addresses use xx:xx:xx:xx:xx:xx notation.
861# #
862# # The 'arp' ACL code is not portable to all operating systems.
863# # It works on Linux, Solaris, Windows, FreeBSD, and some other
864# # BSD variants.
865# #
866# # The eui_lookup directive is required to be 'on' (the default)
867# # and Squid built with --enable-eui for MAC/EUI addresses to be
868# # available for this ACL.
869# #
870# # Squid can only determine the MAC/EUI address for IPv4
871# # clients that are on the same subnet. If the client is on a
872# # different subnet, then Squid cannot find out its address.
873# #
874# # IPv6 protocol does not contain ARP. MAC/EUI is either
875# # encoded directly in the IPv6 address or not available.
876#endif
877# acl aclname clientside_mark mark[/mask] ...
878# # matches CONNMARK of an accepted connection [fast]
879# #
880# # mark and mask are unsigned integers (hex, octal, or decimal).
881# # If multiple marks are given, then the ACL matches if at least
882# # one mark matches.
883# #
884# # Uses netfilter-conntrack library.
885# # Requires building Squid with --enable-linux-netfilter.
886# #
887# # The client, various intermediaries, and Squid itself may set
888# # CONNMARK at various times. The last CONNMARK set wins. This ACL
889# # checks the mark present on an accepted connection or set by
890# # Squid afterwards, depending on the ACL check timing. This ACL
891# # effectively ignores any mark set by other agents after Squid has
892# # accepted the connection.
893#
894# acl aclname srcdomain .foo.com ...
895# # reverse lookup, from client IP [slow]
896# acl aclname dstdomain [-n] .foo.com ...
897# # Destination server from URL [fast]
898# acl aclname srcdom_regex [-i] \.foo\.com ...
899# # regex matching client name [slow]
900# acl aclname dstdom_regex [-n] [-i] \.foo\.com ...
901# # regex matching server [fast]
902# #
903# # For dstdomain and dstdom_regex a reverse lookup is tried if a IP
904# # based URL is used and no match is found. The name "none" is used
905# # if the reverse lookup fails.
906#
907# acl aclname src_as number ...
908# acl aclname dst_as number ...
909# # [fast]
910# # Except for access control, AS numbers can be used for
911# # routing of requests to specific caches. Here's an
912# # example for routing all requests for AS#1241 and only
913# # those to mycache.mydomain.net:
914# # acl asexample dst_as 1241
915# # cache_peer_access mycache.mydomain.net allow asexample
916# # cache_peer_access mycache_mydomain.net deny all
917#
918# acl aclname peername myPeer ...
919# acl aclname peername_regex [-i] regex-pattern ...
920# # [fast]
921# # match against a named cache_peer entry
922# # set unique name= on cache_peer lines for reliable use.
923#
924# acl aclname time [day-abbrevs] [h1:m1-h2:m2]
925# # [fast]
926# # day-abbrevs:
927# # S - Sunday
928# # M - Monday
929# # T - Tuesday
930# # W - Wednesday
931# # H - Thursday
932# # F - Friday
933# # A - Saturday
934# # h1:m1 must be less than h2:m2
935#
936# acl aclname url_regex [-i] ^http:// ...
937# # regex matching on whole URL [fast]
938# acl aclname urllogin [-i] [^a-zA-Z0-9] ...
939# # regex matching on URL login field
940# acl aclname urlpath_regex [-i] \.gif$ ...
941# # regex matching on URL path [fast]
942#
943# acl aclname port 80 70 21 0-1024... # destination TCP port [fast]
944# # ranges are alloed
945# acl aclname localport 3128 ... # TCP port the client connected to [fast]
946# # NP: for interception mode this is usually '80'
947#
948# acl aclname myportname 3128 ... # *_port name [fast]
949#
950# acl aclname proto HTTP FTP ... # request protocol [fast]
951#
952# acl aclname method GET POST ... # HTTP request method [fast]
953#
954# acl aclname http_status 200 301 500- 400-403 ...
955# # status code in reply [fast]
956#
957# acl aclname browser [-i] regexp ...
958# # pattern match on User-Agent header (see also req_header below) [fast]
959#
960# acl aclname referer_regex [-i] regexp ...
961# # pattern match on Referer header [fast]
962# # Referer is highly unreliable, so use with care
963#
964# acl aclname ident [-i] username ...
965# acl aclname ident_regex [-i] pattern ...
966# # string match on ident output [slow]
967# # use REQUIRED to accept any non-null ident.
968#
969# acl aclname proxy_auth [-i] username ...
970# acl aclname proxy_auth_regex [-i] pattern ...
971# # perform http authentication challenge to the client and match against
972# # supplied credentials [slow]
973# #
974# # takes a list of allowed usernames.
975# # use REQUIRED to accept any valid username.
976# #
977# # Will use proxy authentication in forward-proxy scenarios, and plain
978# # http authenticaiton in reverse-proxy scenarios
979# #
980# # NOTE: when a Proxy-Authentication header is sent but it is not
981# # needed during ACL checking the username is NOT logged
982# # in access.log.
983# #
984# # NOTE: proxy_auth requires a EXTERNAL authentication program
985# # to check username/password combinations (see
986# # auth_param directive).
987# #
988# # NOTE: proxy_auth can't be used in a transparent/intercepting proxy
989# # as the browser needs to be configured for using a proxy in order
990# # to respond to proxy authentication.
991#
992# acl aclname snmp_community string ...
993# # A community string to limit access to your SNMP Agent [fast]
994# # Example:
995# #
996# # acl snmppublic snmp_community public
997#
998# acl aclname maxconn number
999# # This will be matched when the client's IP address has
1000# # more than <number> TCP connections established. [fast]
1001# # NOTE: This only measures direct TCP links so X-Forwarded-For
1002# # indirect clients are not counted.
1003#
1004# acl aclname max_user_ip [-s] number
1005# # This will be matched when the user attempts to log in from more
1006# # than <number> different ip addresses. The authenticate_ip_ttl
1007# # parameter controls the timeout on the ip entries. [fast]
1008# # If -s is specified the limit is strict, denying browsing
1009# # from any further IP addresses until the ttl has expired. Without
1010# # -s Squid will just annoy the user by "randomly" denying requests.
1011# # (the counter is reset each time the limit is reached and a
1012# # request is denied)
1013# # NOTE: in acceleration mode or where there is mesh of child proxies,
1014# # clients may appear to come from multiple addresses if they are
1015# # going through proxy farms, so a limit of 1 may cause user problems.
1016#
1017# acl aclname random probability
1018# # Pseudo-randomly match requests. Based on the probability given.
1019# # Probability may be written as a decimal (0.333), fraction (1/3)
1020# # or ratio of matches:non-matches (3:5).
1021#
1022# acl aclname req_mime_type [-i] mime-type ...
1023# # regex match against the mime type of the request generated
1024# # by the client. Can be used to detect file upload or some
1025# # types HTTP tunneling requests [fast]
1026# # NOTE: This does NOT match the reply. You cannot use this
1027# # to match the returned file type.
1028#
1029# acl aclname req_header header-name [-i] any\.regex\.here
1030# # regex match against any of the known request headers. May be
1031# # thought of as a superset of "browser", "referer" and "mime-type"
1032# # ACL [fast]
1033#
1034# acl aclname rep_mime_type [-i] mime-type ...
1035# # regex match against the mime type of the reply received by
1036# # squid. Can be used to detect file download or some
1037# # types HTTP tunneling requests. [fast]
1038# # NOTE: This has no effect in http_access rules. It only has
1039# # effect in rules that affect the reply data stream such as
1040# # http_reply_access.
1041#
1042# acl aclname rep_header header-name [-i] any\.regex\.here
1043# # regex match against any of the known reply headers. May be
1044# # thought of as a superset of "browser", "referer" and "mime-type"
1045# # ACLs [fast]
1046#
1047# acl aclname external class_name [arguments...]
1048# # external ACL lookup via a helper class defined by the
1049# # external_acl_type directive [slow]
1050#
1051# acl aclname user_cert attribute values...
1052# # match against attributes in a user SSL certificate
1053# # attribute is one of DN/C/O/CN/L/ST or a numerical OID [fast]
1054#
1055# acl aclname ca_cert attribute values...
1056# # match against attributes a users issuing CA SSL certificate
1057# # attribute is one of DN/C/O/CN/L/ST or a numerical OID [fast]
1058#
1059# acl aclname ext_user [-i] username ...
1060# acl aclname ext_user_regex [-i] pattern ...
1061# # string match on username returned by external acl helper [slow]
1062# # use REQUIRED to accept any non-null user name.
1063#
1064# acl aclname tag tagvalue ...
1065# # string match on tag returned by external acl helper [fast]
1066# # DEPRECATED. Only the first tag will match with this ACL.
1067# # Use the 'note' ACL instead for handling multiple tag values.
1068#
1069# acl aclname hier_code codename ...
1070# # string match against squid hierarchy code(s); [fast]
1071# # e.g., DIRECT, PARENT_HIT, NONE, etc.
1072# #
1073# # NOTE: This has no effect in http_access rules. It only has
1074# # effect in rules that affect the reply data stream such as
1075# # http_reply_access.
1076#
1077# acl aclname note [-m[=delimiters]] name [value ...]
1078# # match transaction annotation [fast]
1079# # Without values, matches any annotation with a given name.
1080# # With value(s), matches any annotation with a given name that
1081# # also has one of the given values.
1082# # If the -m flag is used, then the value of the named
1083# # annotation is interpreted as a list of tokens, and the ACL
1084# # matches individual name=token pairs rather than whole
1085# # name=value pairs. See "ACL Options" above for more info.
1086# # Annotation sources include note and adaptation_meta directives
1087# # as well as helper and eCAP responses.
1088#
1089# acl aclname adaptation_service service ...
1090# # Matches the name of any icap_service, ecap_service,
1091# # adaptation_service_set, or adaptation_service_chain that Squid
1092# # has used (or attempted to use) for the master transaction.
1093# # This ACL must be defined after the corresponding adaptation
1094# # service is named in squid.conf. This ACL is usable with
1095# # adaptation_meta because it starts matching immediately after
1096# # the service has been selected for adaptation.
1097#
1098# acl aclname transaction_initiator initiator ...
1099# # Matches transaction's initiator [fast]
1100# #
1101# # Supported initiators are:
1102# # esi: matches transactions fetching ESI resources
1103# # certificate-fetching: matches transactions fetching
1104# # a missing intermediate TLS certificate
1105# # cache-digest: matches transactions fetching Cache Digests
1106# # from a cache_peer
1107# # htcp: matches HTCP requests from peers
1108# # icp: matches ICP requests to peers
1109# # icmp: matches ICMP RTT database (NetDB) requests to peers
1110# # asn: matches asns db requests
1111# # internal: matches any of the above
1112# # client: matches transactions containing an HTTP or FTP
1113# # client request received at a Squid *_port
1114# # all: matches any transaction, including internal transactions
1115# # without a configurable initiator and hopefully rare
1116# # transactions without a known-to-Squid initiator
1117# #
1118# # Multiple initiators are ORed.
1119#
1120# acl aclname has component
1121# # matches a transaction "component" [fast]
1122# #
1123# # Supported transaction components are:
1124# # request: transaction has a request header (at least)
1125# # response: transaction has a response header (at least)
1126# # ALE: transaction has an internally-generated Access Log Entry
1127# # structure; bugs notwithstanding, all transaction have it
1128# #
1129# # For example, the following configuration helps when dealing with HTTP
1130# # clients that close connections without sending a request header:
1131# #
1132# # acl hasRequest has request
1133# # acl logMe note important_transaction
1134# # # avoid "logMe ACL is used in context without an HTTP request" warnings
1135# # access_log ... logformat=detailed hasRequest logMe
1136# # # log request-less transactions, instead of ignoring them
1137# # access_log ... logformat=brief !hasRequest
1138# #
1139# # Multiple components are not supported for one "acl" rule, but
1140# # can be specified (and are ORed) using multiple same-name rules:
1141# #
1142# # # OK, this strange logging daemon needs request or response,
1143# # # but can work without either a request or a response:
1144# # acl hasWhatMyLoggingDaemonNeeds has request
1145# # acl hasWhatMyLoggingDaemonNeeds has response
1146#
1147# acl aclname any-of acl1 acl2 ...
1148# # match any one of the acls [fast or slow]
1149# # The first matching ACL stops further ACL evaluation.
1150# #
1151# # ACLs from multiple any-of lines with the same name are ORed.
1152# # For example, A = (a1 or a2) or (a3 or a4) can be written as
1153# # acl A any-of a1 a2
1154# # acl A any-of a3 a4
1155# #
1156# # This group ACL is fast if all evaluated ACLs in the group are fast
1157# # and slow otherwise.
1158#
1159# acl aclname all-of acl1 acl2 ...
1160# # match all of the acls [fast or slow]
1161# # The first mismatching ACL stops further ACL evaluation.
1162# #
1163# # ACLs from multiple all-of lines with the same name are ORed.
1164# # For example, B = (b1 and b2) or (b3 and b4) can be written as
1165# # acl B all-of b1 b2
1166# # acl B all-of b3 b4
1167# #
1168# # This group ACL is fast if all evaluated ACLs in the group are fast
1169# # and slow otherwise.
1170#
1171# Examples:
1172# acl macaddress arp 09:00:2b:23:45:67
1173# acl myexample dst_as 1241
1174# acl password proxy_auth REQUIRED
1175# acl fileupload req_mime_type -i ^multipart/form-data$
1176# acl javascript rep_mime_type -i ^application/x-javascript$
1177#
1178#Default:
1179# ACLs all, manager, localhost, and to_localhost are predefined.
1180#
1181#
1182# Recommended minimum configuration:
1183#
1184
1185# Example rule allowing access from your local networks.
1186# Adapt to list your (internal) IP networks from where browsing
1187# should be allowed
1188acl localnet src 0.0.0.1-0.255.255.255 # RFC 1122 "this" network (LAN)
1189acl localnet src 10.0.0.0/8 # RFC 1918 local private network (LAN)
1190acl localnet src 100.64.0.0/10 # RFC 6598 shared address space (CGN)
1191acl localnet src 169.254.0.0/16 # RFC 3927 link-local (directly plugged) machines
1192acl localnet src 172.16.0.0/12 # RFC 1918 local private network (LAN)
1193acl localnet src 192.168.0.0/16 # RFC 1918 local private network (LAN)
1194acl localnet src fc00::/7 # RFC 4193 local private network range
1195acl localnet src fe80::/10 # RFC 4291 link-local (directly plugged) machines
1196
1197acl SSL_ports port 443
1198acl Safe_ports port 80 # http
1199acl Safe_ports port 21 # ftp
1200acl Safe_ports port 443 # https
1201acl Safe_ports port 70 # gopher
1202acl Safe_ports port 210 # wais
1203acl Safe_ports port 1025-65535 # unregistered ports
1204acl Safe_ports port 280 # http-mgmt
1205acl Safe_ports port 488 # gss-http
1206acl Safe_ports port 591 # filemaker
1207acl Safe_ports port 777 # multiling http
1208acl CONNECT method CONNECT
1209
1210# TAG: proxy_protocol_access
1211# Determine which client proxies can be trusted to provide correct
1212# information regarding real client IP address using PROXY protocol.
1213#
1214# Requests may pass through a chain of several other proxies
1215# before reaching us. The original source details may by sent in:
1216# * HTTP message Forwarded header, or
1217# * HTTP message X-Forwarded-For header, or
1218# * PROXY protocol connection header.
1219#
1220# This directive is solely for validating new PROXY protocol
1221# connections received from a port flagged with require-proxy-header.
1222# It is checked only once after TCP connection setup.
1223#
1224# A deny match results in TCP connection closure.
1225#
1226# An allow match is required for Squid to permit the corresponding
1227# TCP connection, before Squid even looks for HTTP request headers.
1228# If there is an allow match, Squid starts using PROXY header information
1229# to determine the source address of the connection for all future ACL
1230# checks, logging, etc.
1231#
1232# SECURITY CONSIDERATIONS:
1233#
1234# Any host from which we accept client IP details can place
1235# incorrect information in the relevant header, and Squid
1236# will use the incorrect information as if it were the
1237# source address of the request. This may enable remote
1238# hosts to bypass any access control restrictions that are
1239# based on the client's source addresses.
1240#
1241# This clause only supports fast acl types.
1242# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1243#Default:
1244# all TCP connections to ports with require-proxy-header will be denied
1245
1246# TAG: follow_x_forwarded_for
1247# Determine which client proxies can be trusted to provide correct
1248# information regarding real client IP address.
1249#
1250# Requests may pass through a chain of several other proxies
1251# before reaching us. The original source details may by sent in:
1252# * HTTP message Forwarded header, or
1253# * HTTP message X-Forwarded-For header, or
1254# * PROXY protocol connection header.
1255#
1256# PROXY protocol connections are controlled by the proxy_protocol_access
1257# directive which is checked before this.
1258#
1259# If a request reaches us from a source that is allowed by this
1260# directive, then we trust the information it provides regarding
1261# the IP of the client it received from (if any).
1262#
1263# For the purpose of ACLs used in this directive the src ACL type always
1264# matches the address we are testing and srcdomain matches its rDNS.
1265#
1266# On each HTTP request Squid checks for X-Forwarded-For header fields.
1267# If found the header values are iterated in reverse order and an allow
1268# match is required for Squid to continue on to the next value.
1269# The verification ends when a value receives a deny match, cannot be
1270# tested, or there are no more values to test.
1271# NOTE: Squid does not yet follow the Forwarded HTTP header.
1272#
1273# The end result of this process is an IP address that we will
1274# refer to as the indirect client address. This address may
1275# be treated as the client address for access control, ICAP, delay
1276# pools and logging, depending on the acl_uses_indirect_client,
1277# icap_uses_indirect_client, delay_pool_uses_indirect_client,
1278# log_uses_indirect_client and tproxy_uses_indirect_client options.
1279#
1280# This clause only supports fast acl types.
1281# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1282#
1283# SECURITY CONSIDERATIONS:
1284#
1285# Any host from which we accept client IP details can place
1286# incorrect information in the relevant header, and Squid
1287# will use the incorrect information as if it were the
1288# source address of the request. This may enable remote
1289# hosts to bypass any access control restrictions that are
1290# based on the client's source addresses.
1291#
1292# For example:
1293#
1294# acl localhost src 127.0.0.1
1295# acl my_other_proxy srcdomain .proxy.example.com
1296# follow_x_forwarded_for allow localhost
1297# follow_x_forwarded_for allow my_other_proxy
1298#Default:
1299# X-Forwarded-For header will be ignored.
1300
1301# TAG: acl_uses_indirect_client on|off
1302# Controls whether the indirect client address
1303# (see follow_x_forwarded_for) is used instead of the
1304# direct client address in acl matching.
1305#
1306# NOTE: maxconn ACL considers direct TCP links and indirect
1307# clients will always have zero. So no match.
1308#Default:
1309# acl_uses_indirect_client on
1310
1311# TAG: delay_pool_uses_indirect_client on|off
1312# Controls whether the indirect client address
1313# (see follow_x_forwarded_for) is used instead of the
1314# direct client address in delay pools.
1315#Default:
1316# delay_pool_uses_indirect_client on
1317
1318# TAG: log_uses_indirect_client on|off
1319# Controls whether the indirect client address
1320# (see follow_x_forwarded_for) is used instead of the
1321# direct client address in the access log.
1322#Default:
1323# log_uses_indirect_client on
1324
1325# TAG: tproxy_uses_indirect_client on|off
1326# Controls whether the indirect client address
1327# (see follow_x_forwarded_for) is used instead of the
1328# direct client address when spoofing the outgoing client.
1329#
1330# This has no effect on requests arriving in non-tproxy
1331# mode ports.
1332#
1333# SECURITY WARNING: Usage of this option is dangerous
1334# and should not be used trivially. Correct configuration
1335# of follow_x_forwarded_for with a limited set of trusted
1336# sources is required to prevent abuse of your proxy.
1337#Default:
1338# tproxy_uses_indirect_client off
1339
1340# TAG: spoof_client_ip
1341# Control client IP address spoofing of TPROXY traffic based on
1342# defined access lists.
1343#
1344# spoof_client_ip allow|deny [!]aclname ...
1345#
1346# If there are no "spoof_client_ip" lines present, the default
1347# is to "allow" spoofing of any suitable request.
1348#
1349# Note that the cache_peer "no-tproxy" option overrides this ACL.
1350#
1351# This clause supports fast acl types.
1352# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1353#Default:
1354# Allow spoofing on all TPROXY traffic.
1355
1356# TAG: http_access
1357# Allowing or Denying access based on defined access lists
1358#
1359# To allow or deny a message received on an HTTP, HTTPS, or FTP port:
1360# http_access allow|deny [!]aclname ...
1361#
1362# NOTE on default values:
1363#
1364# If there are no "access" lines present, the default is to deny
1365# the request.
1366#
1367# If none of the "access" lines cause a match, the default is the
1368# opposite of the last line in the list. If the last line was
1369# deny, the default is allow. Conversely, if the last line
1370# is allow, the default will be deny. For these reasons, it is a
1371# good idea to have an "deny all" entry at the end of your access
1372# lists to avoid potential confusion.
1373#
1374# This clause supports both fast and slow acl types.
1375# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1376#
1377#Default:
1378# Deny, unless rules exist in squid.conf.
1379#
1380
1381#
1382# Recommended minimum Access Permission configuration:
1383#
1384# Deny requests to certain unsafe ports
1385http_access deny !Safe_ports
1386
1387# Deny CONNECT to other than secure SSL ports
1388http_access deny CONNECT !SSL_ports
1389
1390# Only allow cachemgr access from localhost
1391http_access allow localhost manager
1392http_access deny manager
1393
1394# We strongly recommend the following be uncommented to protect innocent
1395# web applications running on the proxy server who think the only
1396# one who can access services on "localhost" is a local user
1397#http_access deny to_localhost
1398
1399#
1400# INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS
1401#
1402include /etc/squid/conf.d/*
1403
1404# Example rule allowing access from your local networks.
1405# Adapt localnet in the ACL section to list your (internal) IP networks
1406# from where browsing should be allowed
1407#http_access allow localnet
1408http_access allow localhost
1409
1410# And finally deny all other access to this proxy
1411http_access deny all
1412
1413# TAG: adapted_http_access
1414# Allowing or Denying access based on defined access lists
1415#
1416# Essentially identical to http_access, but runs after redirectors
1417# and ICAP/eCAP adaptation. Allowing access control based on their
1418# output.
1419#
1420# If not set then only http_access is used.
1421#Default:
1422# Allow, unless rules exist in squid.conf.
1423
1424# TAG: http_reply_access
1425# Allow replies to client requests. This is complementary to http_access.
1426#
1427# http_reply_access allow|deny [!] aclname ...
1428#
1429# NOTE: if there are no access lines present, the default is to allow
1430# all replies.
1431#
1432# If none of the access lines cause a match the opposite of the
1433# last line will apply. Thus it is good practice to end the rules
1434# with an "allow all" or "deny all" entry.
1435#
1436# This clause supports both fast and slow acl types.
1437# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1438#Default:
1439# Allow, unless rules exist in squid.conf.
1440
1441# TAG: icp_access
1442# Allowing or Denying access to the ICP port based on defined
1443# access lists
1444#
1445# icp_access allow|deny [!]aclname ...
1446#
1447# NOTE: The default if no icp_access lines are present is to
1448# deny all traffic. This default may cause problems with peers
1449# using ICP.
1450#
1451# This clause only supports fast acl types.
1452# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1453#
1454## Allow ICP queries from local networks only
1455##icp_access allow localnet
1456##icp_access deny all
1457#Default:
1458# Deny, unless rules exist in squid.conf.
1459
1460# TAG: htcp_access
1461# Allowing or Denying access to the HTCP port based on defined
1462# access lists
1463#
1464# htcp_access allow|deny [!]aclname ...
1465#
1466# See also htcp_clr_access for details on access control for
1467# cache purge (CLR) HTCP messages.
1468#
1469# NOTE: The default if no htcp_access lines are present is to
1470# deny all traffic. This default may cause problems with peers
1471# using the htcp option.
1472#
1473# This clause only supports fast acl types.
1474# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1475#
1476## Allow HTCP queries from local networks only
1477##htcp_access allow localnet
1478##htcp_access deny all
1479#Default:
1480# Deny, unless rules exist in squid.conf.
1481
1482# TAG: htcp_clr_access
1483# Allowing or Denying access to purge content using HTCP based
1484# on defined access lists.
1485# See htcp_access for details on general HTCP access control.
1486#
1487# htcp_clr_access allow|deny [!]aclname ...
1488#
1489# This clause only supports fast acl types.
1490# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1491#
1492## Allow HTCP CLR requests from trusted peers
1493#acl htcp_clr_peer src 192.0.2.2 2001:DB8::2
1494#htcp_clr_access allow htcp_clr_peer
1495#htcp_clr_access deny all
1496#Default:
1497# Deny, unless rules exist in squid.conf.
1498
1499# TAG: miss_access
1500# Determines whether network access is permitted when satisfying a request.
1501#
1502# For example;
1503# to force your neighbors to use you as a sibling instead of
1504# a parent.
1505#
1506# acl localclients src 192.0.2.0/24 2001:DB8::a:0/64
1507# miss_access deny !localclients
1508# miss_access allow all
1509#
1510# This means only your local clients are allowed to fetch relayed/MISS
1511# replies from the network and all other clients can only fetch cached
1512# objects (HITs).
1513#
1514# The default for this setting allows all clients who passed the
1515# http_access rules to relay via this proxy.
1516#
1517# This clause only supports fast acl types.
1518# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1519#Default:
1520# Allow, unless rules exist in squid.conf.
1521
1522# TAG: ident_lookup_access
1523# A list of ACL elements which, if matched, cause an ident
1524# (RFC 931) lookup to be performed for this request. For
1525# example, you might choose to always perform ident lookups
1526# for your main multi-user Unix boxes, but not for your Macs
1527# and PCs. By default, ident lookups are not performed for
1528# any requests.
1529#
1530# To enable ident lookups for specific client addresses, you
1531# can follow this example:
1532#
1533# acl ident_aware_hosts src 198.168.1.0/24
1534# ident_lookup_access allow ident_aware_hosts
1535# ident_lookup_access deny all
1536#
1537# Only src type ACL checks are fully supported. A srcdomain
1538# ACL might work at times, but it will not always provide
1539# the correct result.
1540#
1541# This clause only supports fast acl types.
1542# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1543#Default:
1544# Unless rules exist in squid.conf, IDENT is not fetched.
1545
1546# TAG: reply_body_max_size size [acl acl...]
1547# This option specifies the maximum size of a reply body. It can be
1548# used to prevent users from downloading very large files, such as
1549# MP3's and movies. When the reply headers are received, the
1550# reply_body_max_size lines are processed, and the first line where
1551# all (if any) listed ACLs are true is used as the maximum body size
1552# for this reply.
1553#
1554# This size is checked twice. First when we get the reply headers,
1555# we check the content-length value. If the content length value exists
1556# and is larger than the allowed size, the request is denied and the
1557# user receives an error message that says "the request or reply
1558# is too large." If there is no content-length, and the reply
1559# size exceeds this limit, the client's connection is just closed
1560# and they will receive a partial reply.
1561#
1562# WARNING: downstream caches probably can not detect a partial reply
1563# if there is no content-length header, so they will cache
1564# partial responses and give them out as hits. You should NOT
1565# use this option if you have downstream caches.
1566#
1567# WARNING: A maximum size smaller than the size of squid's error messages
1568# will cause an infinite loop and crash squid. Ensure that the smallest
1569# non-zero value you use is greater that the maximum header size plus
1570# the size of your largest error page.
1571#
1572# If you set this parameter none (the default), there will be
1573# no limit imposed.
1574#
1575# Configuration Format is:
1576# reply_body_max_size SIZE UNITS [acl ...]
1577# ie.
1578# reply_body_max_size 10 MB
1579#
1580#Default:
1581# No limit is applied.
1582
1583# TAG: on_unsupported_protocol
1584# Determines Squid behavior when encountering strange requests at the
1585# beginning of an accepted TCP connection or the beginning of a bumped
1586# CONNECT tunnel. Controlling Squid reaction to unexpected traffic is
1587# especially useful in interception environments where Squid is likely
1588# to see connections for unsupported protocols that Squid should either
1589# terminate or tunnel at TCP level.
1590#
1591# on_unsupported_protocol <action> [!]acl ...
1592#
1593# The first matching action wins. Only fast ACLs are supported.
1594#
1595# Supported actions are:
1596#
1597# tunnel: Establish a TCP connection with the intended server and
1598# blindly shovel TCP packets between the client and server.
1599#
1600# respond: Respond with an error message, using the transfer protocol
1601# for the Squid port that received the request (e.g., HTTP
1602# for connections intercepted at the http_port). This is the
1603# default.
1604#
1605# Squid expects the following traffic patterns:
1606#
1607# http_port: a plain HTTP request
1608# https_port: SSL/TLS handshake followed by an [encrypted] HTTP request
1609# ftp_port: a plain FTP command (no on_unsupported_protocol support yet!)
1610# CONNECT tunnel on http_port: same as https_port
1611# CONNECT tunnel on https_port: same as https_port
1612#
1613# Currently, this directive has effect on intercepted connections and
1614# bumped tunnels only. Other cases are not supported because Squid
1615# cannot know the intended destination of other traffic.
1616#
1617# For example:
1618# # define what Squid errors indicate receiving non-HTTP traffic:
1619# acl foreignProtocol squid_error ERR_PROTOCOL_UNKNOWN ERR_TOO_BIG
1620# # define what Squid errors indicate receiving nothing:
1621# acl serverTalksFirstProtocol squid_error ERR_REQUEST_START_TIMEOUT
1622# # tunnel everything that does not look like HTTP:
1623# on_unsupported_protocol tunnel foreignProtocol
1624# # tunnel if we think the client waits for the server to talk first:
1625# on_unsupported_protocol tunnel serverTalksFirstProtocol
1626# # in all other error cases, just send an HTTP "error page" response:
1627# on_unsupported_protocol respond all
1628#
1629# See also: squid_error ACL
1630#Default:
1631# Respond with an error message to unidentifiable traffic
1632
1633# NETWORK OPTIONS
1634# -----------------------------------------------------------------------------
1635
1636# TAG: http_port
1637# Usage: port [mode] [options]
1638# hostname:port [mode] [options]
1639# 1.2.3.4:port [mode] [options]
1640#
1641# The socket addresses where Squid will listen for HTTP client
1642# requests. You may specify multiple socket addresses.
1643# There are three forms: port alone, hostname with port, and
1644# IP address with port. If you specify a hostname or IP
1645# address, Squid binds the socket to that specific
1646# address. Most likely, you do not need to bind to a specific
1647# address, so you can use the port number alone.
1648#
1649# If you are running Squid in accelerator mode, you
1650# probably want to listen on port 80 also, or instead.
1651#
1652# The -a command line option may be used to specify additional
1653# port(s) where Squid listens for proxy request. Such ports will
1654# be plain proxy ports with no options.
1655#
1656# You may specify multiple socket addresses on multiple lines.
1657#
1658# Modes:
1659#
1660# intercept Support for IP-Layer NAT interception delivering
1661# traffic to this Squid port.
1662# NP: disables authentication on the port.
1663#
1664# tproxy Support Linux TPROXY (or BSD divert-to) with spoofing
1665# of outgoing connections using the client IP address.
1666# NP: disables authentication on the port.
1667#
1668# accel Accelerator / reverse proxy mode
1669#
1670# ssl-bump For each CONNECT request allowed by ssl_bump ACLs,
1671# establish secure connection with the client and with
1672# the server, decrypt HTTPS messages as they pass through
1673# Squid, and treat them as unencrypted HTTP messages,
1674# becoming the man-in-the-middle.
1675#
1676# The ssl_bump option is required to fully enable
1677# bumping of CONNECT requests.
1678#
1679# Omitting the mode flag causes default forward proxy mode to be used.
1680#
1681#
1682# Accelerator Mode Options:
1683#
1684# defaultsite=domainname
1685# What to use for the Host: header if it is not present
1686# in a request. Determines what site (not origin server)
1687# accelerators should consider the default.
1688#
1689# no-vhost Disable using HTTP/1.1 Host header for virtual domain support.
1690#
1691# protocol= Protocol to reconstruct accelerated and intercepted
1692# requests with. Defaults to HTTP/1.1 for http_port and
1693# HTTPS/1.1 for https_port.
1694# When an unsupported value is configured Squid will
1695# produce a FATAL error.
1696# Values: HTTP or HTTP/1.1, HTTPS or HTTPS/1.1
1697#
1698# vport Virtual host port support. Using the http_port number
1699# instead of the port passed on Host: headers.
1700#
1701# vport=NN Virtual host port support. Using the specified port
1702# number instead of the port passed on Host: headers.
1703#
1704# act-as-origin
1705# Act as if this Squid is the origin server.
1706# This currently means generate new Date: and Expires:
1707# headers on HIT instead of adding Age:.
1708#
1709# ignore-cc Ignore request Cache-Control headers.
1710#
1711# WARNING: This option violates HTTP specifications if
1712# used in non-accelerator setups.
1713#
1714# allow-direct Allow direct forwarding in accelerator mode. Normally
1715# accelerated requests are denied direct forwarding as if
1716# never_direct was used.
1717#
1718# WARNING: this option opens accelerator mode to security
1719# vulnerabilities usually only affecting in interception
1720# mode. Make sure to protect forwarding with suitable
1721# http_access rules when using this.
1722#
1723#
1724# SSL Bump Mode Options:
1725# In addition to these options ssl-bump requires TLS/SSL options.
1726#
1727# generate-host-certificates[=<on|off>]
1728# Dynamically create SSL server certificates for the
1729# destination hosts of bumped CONNECT requests.When
1730# enabled, the cert and key options are used to sign
1731# generated certificates. Otherwise generated
1732# certificate will be selfsigned.
1733# If there is a CA certificate lifetime of the generated
1734# certificate equals lifetime of the CA certificate. If
1735# generated certificate is selfsigned lifetime is three
1736# years.
1737# This option is enabled by default when ssl-bump is used.
1738# See the ssl-bump option above for more information.
1739#
1740# dynamic_cert_mem_cache_size=SIZE
1741# Approximate total RAM size spent on cached generated
1742# certificates. If set to zero, caching is disabled. The
1743# default value is 4MB.
1744#
1745# TLS / SSL Options:
1746#
1747# tls-cert= Path to file containing an X.509 certificate (PEM format)
1748# to be used in the TLS handshake ServerHello.
1749#
1750# If this certificate is constrained by KeyUsage TLS
1751# feature it must allow HTTP server usage, along with
1752# any additional restrictions imposed by your choice
1753# of options= settings.
1754#
1755# When OpenSSL is used this file may also contain a
1756# chain of intermediate CA certificates to send in the
1757# TLS handshake.
1758#
1759# When GnuTLS is used this option (and any paired
1760# tls-key= option) may be repeated to load multiple
1761# certificates for different domains.
1762#
1763# Also, when generate-host-certificates=on is configured
1764# the first tls-cert= option must be a CA certificate
1765# capable of signing the automatically generated
1766# certificates.
1767#
1768# tls-key= Path to a file containing private key file (PEM format)
1769# for the previous tls-cert= option.
1770#
1771# If tls-key= is not specified tls-cert= is assumed to
1772# reference a PEM file containing both the certificate
1773# and private key.
1774#
1775# cipher= Colon separated list of supported ciphers.
1776# NOTE: some ciphers such as EDH ciphers depend on
1777# additional settings. If those settings are
1778# omitted the ciphers may be silently ignored
1779# by the OpenSSL library.
1780#
1781# options= Various SSL implementation options. The most important
1782# being:
1783#
1784# NO_SSLv3 Disallow the use of SSLv3
1785#
1786# NO_TLSv1 Disallow the use of TLSv1.0
1787#
1788# NO_TLSv1_1 Disallow the use of TLSv1.1
1789#
1790# NO_TLSv1_2 Disallow the use of TLSv1.2
1791#
1792# SINGLE_DH_USE
1793# Always create a new key when using
1794# temporary/ephemeral DH key exchanges
1795#
1796# SINGLE_ECDH_USE
1797# Enable ephemeral ECDH key exchange.
1798# The adopted curve should be specified
1799# using the tls-dh option.
1800#
1801# NO_TICKET
1802# Disable use of RFC5077 session tickets.
1803# Some servers may have problems
1804# understanding the TLS extension due
1805# to ambiguous specification in RFC4507.
1806#
1807# ALL Enable various bug workarounds
1808# suggested as "harmless" by OpenSSL
1809# Be warned that this reduces SSL/TLS
1810# strength to some attacks.
1811#
1812# See the OpenSSL SSL_CTX_set_options documentation for a
1813# more complete list.
1814#
1815# clientca= File containing the list of CAs to use when
1816# requesting a client certificate.
1817#
1818# tls-cafile= PEM file containing CA certificates to use when verifying
1819# client certificates. If not configured clientca will be
1820# used. May be repeated to load multiple files.
1821#
1822# capath= Directory containing additional CA certificates
1823# and CRL lists to use when verifying client certificates.
1824# Requires OpenSSL or LibreSSL.
1825#
1826# crlfile= File of additional CRL lists to use when verifying
1827# the client certificate, in addition to CRLs stored in
1828# the capath. Implies VERIFY_CRL flag below.
1829#
1830# tls-dh=[curve:]file
1831# File containing DH parameters for temporary/ephemeral DH key
1832# exchanges, optionally prefixed by a curve for ephemeral ECDH
1833# key exchanges.
1834# See OpenSSL documentation for details on how to create the
1835# DH parameter file. Supported curves for ECDH can be listed
1836# using the "openssl ecparam -list_curves" command.
1837# WARNING: EDH and EECDH ciphers will be silently disabled if
1838# this option is not set.
1839#
1840# sslflags= Various flags modifying the use of SSL:
1841# DELAYED_AUTH
1842# Don't request client certificates
1843# immediately, but wait until acl processing
1844# requires a certificate (not yet implemented).
1845# NO_SESSION_REUSE
1846# Don't allow for session reuse. Each connection
1847# will result in a new SSL session.
1848# VERIFY_CRL
1849# Verify CRL lists when accepting client
1850# certificates.
1851# VERIFY_CRL_ALL
1852# Verify CRL lists for all certificates in the
1853# client certificate chain.
1854#
1855# tls-default-ca[=off]
1856# Whether to use the system Trusted CAs. Default is OFF.
1857#
1858# tls-no-npn Do not use the TLS NPN extension to advertise HTTP/1.1.
1859#
1860# sslcontext= SSL session ID context identifier.
1861#
1862# Other Options:
1863#
1864# connection-auth[=on|off]
1865# use connection-auth=off to tell Squid to prevent
1866# forwarding Microsoft connection oriented authentication
1867# (NTLM, Negotiate and Kerberos)
1868#
1869# disable-pmtu-discovery=
1870# Control Path-MTU discovery usage:
1871# off lets OS decide on what to do (default).
1872# transparent disable PMTU discovery when transparent
1873# support is enabled.
1874# always disable always PMTU discovery.
1875#
1876# In many setups of transparently intercepting proxies
1877# Path-MTU discovery can not work on traffic towards the
1878# clients. This is the case when the intercepting device
1879# does not fully track connections and fails to forward
1880# ICMP must fragment messages to the cache server. If you
1881# have such setup and experience that certain clients
1882# sporadically hang or never complete requests set
1883# disable-pmtu-discovery option to 'transparent'.
1884#
1885# name= Specifies a internal name for the port. Defaults to
1886# the port specification (port or addr:port)
1887#
1888# tcpkeepalive[=idle,interval,timeout]
1889# Enable TCP keepalive probes of idle connections.
1890# In seconds; idle is the initial time before TCP starts
1891# probing the connection, interval how often to probe, and
1892# timeout the time before giving up.
1893#
1894# require-proxy-header
1895# Require PROXY protocol version 1 or 2 connections.
1896# The proxy_protocol_access is required to whitelist
1897# downstream proxies which can be trusted.
1898#
1899# If you run Squid on a dual-homed machine with an internal
1900# and an external interface we recommend you to specify the
1901# internal address:port in http_port. This way Squid will only be
1902# visible on the internal address.
1903#
1904#
1905
1906# Squid normally listens to port 3128
1907http_port 3128
1908
1909# TAG: https_port
1910# Usage: [ip:]port [mode] tls-cert=certificate.pem [options]
1911#
1912# The socket address where Squid will listen for client requests made
1913# over TLS or SSL connections. Commonly referred to as HTTPS.
1914#
1915# This is most useful for situations where you are running squid in
1916# accelerator mode and you want to do the TLS work at the accelerator
1917# level.
1918#
1919# You may specify multiple socket addresses on multiple lines,
1920# each with their own certificate and/or options.
1921#
1922# The tls-cert= option is mandatory on HTTPS ports.
1923#
1924# See http_port for a list of modes and options.
1925#Default:
1926# none
1927
1928# TAG: ftp_port
1929# Enables Native FTP proxy by specifying the socket address where Squid
1930# listens for FTP client requests. See http_port directive for various
1931# ways to specify the listening address and mode.
1932#
1933# Usage: ftp_port address [mode] [options]
1934#
1935# WARNING: This is a new, experimental, complex feature that has seen
1936# limited production exposure. Some Squid modules (e.g., caching) do not
1937# currently work with native FTP proxying, and many features have not
1938# even been tested for compatibility. Test well before deploying!
1939#
1940# Native FTP proxying differs substantially from proxying HTTP requests
1941# with ftp:// URIs because Squid works as an FTP server and receives
1942# actual FTP commands (rather than HTTP requests with FTP URLs).
1943#
1944# Native FTP commands accepted at ftp_port are internally converted or
1945# wrapped into HTTP-like messages. The same happens to Native FTP
1946# responses received from FTP origin servers. Those HTTP-like messages
1947# are shoveled through regular access control and adaptation layers
1948# between the FTP client and the FTP origin server. This allows Squid to
1949# examine, adapt, block, and log FTP exchanges. Squid reuses most HTTP
1950# mechanisms when shoveling wrapped FTP messages. For example,
1951# http_access and adaptation_access directives are used.
1952#
1953# Modes:
1954#
1955# intercept Same as http_port intercept. The FTP origin address is
1956# determined based on the intended destination of the
1957# intercepted connection.
1958#
1959# tproxy Support Linux TPROXY for spoofing outgoing
1960# connections using the client IP address.
1961# NP: disables authentication and maybe IPv6 on the port.
1962#
1963# By default (i.e., without an explicit mode option), Squid extracts the
1964# FTP origin address from the login@origin parameter of the FTP USER
1965# command. Many popular FTP clients support such native FTP proxying.
1966#
1967# Options:
1968#
1969# name=token Specifies an internal name for the port. Defaults to
1970# the port address. Usable with myportname ACL.
1971#
1972# ftp-track-dirs
1973# Enables tracking of FTP directories by injecting extra
1974# PWD commands and adjusting Request-URI (in wrapping
1975# HTTP requests) to reflect the current FTP server
1976# directory. Tracking is disabled by default.
1977#
1978# protocol=FTP Protocol to reconstruct accelerated and intercepted
1979# requests with. Defaults to FTP. No other accepted
1980# values have been tested with. An unsupported value
1981# results in a FATAL error. Accepted values are FTP,
1982# HTTP (or HTTP/1.1), and HTTPS (or HTTPS/1.1).
1983#
1984# Other http_port modes and options that are not specific to HTTP and
1985# HTTPS may also work.
1986#Default:
1987# none
1988
1989# TAG: tcp_outgoing_tos
1990# Allows you to select a TOS/Diffserv value for packets outgoing
1991# on the server side, based on an ACL.
1992#
1993# tcp_outgoing_tos ds-field [!]aclname ...
1994#
1995# Example where normal_service_net uses the TOS value 0x00
1996# and good_service_net uses 0x20
1997#
1998# acl normal_service_net src 10.0.0.0/24
1999# acl good_service_net src 10.0.1.0/24
2000# tcp_outgoing_tos 0x00 normal_service_net
2001# tcp_outgoing_tos 0x20 good_service_net
2002#
2003# TOS/DSCP values really only have local significance - so you should
2004# know what you're specifying. For more information, see RFC2474,
2005# RFC2475, and RFC3260.
2006#
2007# The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
2008# "default" to use whatever default your host has.
2009# Note that only multiples of 4 are usable as the two rightmost bits have
2010# been redefined for use by ECN (RFC 3168 section 23.1).
2011# The squid parser will enforce this by masking away the ECN bits.
2012#
2013# Processing proceeds in the order specified, and stops at first fully
2014# matching line.
2015#
2016# Only fast ACLs are supported.
2017#Default:
2018# none
2019
2020# TAG: clientside_tos
2021# Allows you to select a TOS/DSCP value for packets being transmitted
2022# on the client-side, based on an ACL.
2023#
2024# clientside_tos ds-field [!]aclname ...
2025#
2026# Example where normal_service_net uses the TOS value 0x00
2027# and good_service_net uses 0x20
2028#
2029# acl normal_service_net src 10.0.0.0/24
2030# acl good_service_net src 10.0.1.0/24
2031# clientside_tos 0x00 normal_service_net
2032# clientside_tos 0x20 good_service_net
2033#
2034# Note: This feature is incompatible with qos_flows. Any TOS values set here
2035# will be overwritten by TOS values in qos_flows.
2036#
2037# The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
2038# "default" to use whatever default your host has.
2039# Note that only multiples of 4 are usable as the two rightmost bits have
2040# been redefined for use by ECN (RFC 3168 section 23.1).
2041# The squid parser will enforce this by masking away the ECN bits.
2042#
2043# This clause only supports fast acl types.
2044# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2045#Default:
2046# none
2047
2048# TAG: tcp_outgoing_mark
2049# Note: This option is only available if Squid is rebuilt with the
2050# Packet MARK (Linux)
2051#
2052# Allows you to apply a Netfilter mark value to outgoing packets
2053# on the server side, based on an ACL.
2054#
2055# tcp_outgoing_mark mark-value [!]aclname ...
2056#
2057# Example where normal_service_net uses the mark value 0x00
2058# and good_service_net uses 0x20
2059#
2060# acl normal_service_net src 10.0.0.0/24
2061# acl good_service_net src 10.0.1.0/24
2062# tcp_outgoing_mark 0x00 normal_service_net
2063# tcp_outgoing_mark 0x20 good_service_net
2064#
2065# Only fast ACLs are supported.
2066#Default:
2067# none
2068
2069# TAG: clientside_mark
2070# Note: This option is only available if Squid is rebuilt with the
2071# Packet MARK (Linux)
2072#
2073# Allows you to apply a Netfilter mark value to packets being transmitted
2074# on the client-side, based on an ACL.
2075#
2076# clientside_mark mark-value [!]aclname ...
2077#
2078# Example where normal_service_net uses the mark value 0x00
2079# and good_service_net uses 0x20
2080#
2081# acl normal_service_net src 10.0.0.0/24
2082# acl good_service_net src 10.0.1.0/24
2083# clientside_mark 0x00 normal_service_net
2084# clientside_mark 0x20 good_service_net
2085#
2086# Note: This feature is incompatible with qos_flows. Any mark values set here
2087# will be overwritten by mark values in qos_flows.
2088#
2089# This clause only supports fast acl types.
2090# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2091#Default:
2092# none
2093
2094# TAG: qos_flows
2095# Allows you to select a TOS/DSCP value to mark outgoing
2096# connections to the client, based on where the reply was sourced.
2097# For platforms using netfilter, allows you to set a netfilter mark
2098# value instead of, or in addition to, a TOS value.
2099#
2100# By default this functionality is disabled. To enable it with the default
2101# settings simply use "qos_flows mark" or "qos_flows tos". Default
2102# settings will result in the netfilter mark or TOS value being copied
2103# from the upstream connection to the client. Note that it is the connection
2104# CONNMARK value not the packet MARK value that is copied.
2105#
2106# It is not currently possible to copy the mark or TOS value from the
2107# client to the upstream connection request.
2108#
2109# TOS values really only have local significance - so you should
2110# know what you're specifying. For more information, see RFC2474,
2111# RFC2475, and RFC3260.
2112#
2113# The TOS/DSCP byte must be exactly that - a octet value 0 - 255.
2114# Note that only multiples of 4 are usable as the two rightmost bits have
2115# been redefined for use by ECN (RFC 3168 section 23.1).
2116# The squid parser will enforce this by masking away the ECN bits.
2117#
2118# Mark values can be any unsigned 32-bit integer value.
2119#
2120# This setting is configured by setting the following values:
2121#
2122# tos|mark Whether to set TOS or netfilter mark values
2123#
2124# local-hit=0xFF Value to mark local cache hits.
2125#
2126# sibling-hit=0xFF Value to mark hits from sibling peers.
2127#
2128# parent-hit=0xFF Value to mark hits from parent peers.
2129#
2130# miss=0xFF[/mask] Value to mark cache misses. Takes precedence
2131# over the preserve-miss feature (see below), unless
2132# mask is specified, in which case only the bits
2133# specified in the mask are written.
2134#
2135# The TOS variant of the following features are only possible on Linux
2136# and require your kernel to be patched with the TOS preserving ZPH
2137# patch, available from http://zph.bratcheda.org
2138# No patch is needed to preserve the netfilter mark, which will work
2139# with all variants of netfilter.
2140#
2141# disable-preserve-miss
2142# This option disables the preservation of the TOS or netfilter
2143# mark. By default, the existing TOS or netfilter mark value of
2144# the response coming from the remote server will be retained
2145# and masked with miss-mark.
2146# NOTE: in the case of a netfilter mark, the mark must be set on
2147# the connection (using the CONNMARK target) not on the packet
2148# (MARK target).
2149#
2150# miss-mask=0xFF
2151# Allows you to mask certain bits in the TOS or mark value
2152# received from the remote server, before copying the value to
2153# the TOS sent towards clients.
2154# Default for tos: 0xFF (TOS from server is not changed).
2155# Default for mark: 0xFFFFFFFF (mark from server is not changed).
2156#
2157# All of these features require the --enable-zph-qos compilation flag
2158# (enabled by default). Netfilter marking also requires the
2159# libnetfilter_conntrack libraries (--with-netfilter-conntrack) and
2160# libcap 2.09+ (--with-libcap).
2161#
2162#Default:
2163# none
2164
2165# TAG: tcp_outgoing_address
2166# Allows you to map requests to different outgoing IP addresses
2167# based on the username or source address of the user making
2168# the request.
2169#
2170# tcp_outgoing_address ipaddr [[!]aclname] ...
2171#
2172# For example;
2173# Forwarding clients with dedicated IPs for certain subnets.
2174#
2175# acl normal_service_net src 10.0.0.0/24
2176# acl good_service_net src 10.0.2.0/24
2177#
2178# tcp_outgoing_address 2001:db8::c001 good_service_net
2179# tcp_outgoing_address 10.1.0.2 good_service_net
2180#
2181# tcp_outgoing_address 2001:db8::beef normal_service_net
2182# tcp_outgoing_address 10.1.0.1 normal_service_net
2183#
2184# tcp_outgoing_address 2001:db8::1
2185# tcp_outgoing_address 10.1.0.3
2186#
2187# Processing proceeds in the order specified, and stops at first fully
2188# matching line.
2189#
2190# Squid will add an implicit IP version test to each line.
2191# Requests going to IPv4 websites will use the outgoing 10.1.0.* addresses.
2192# Requests going to IPv6 websites will use the outgoing 2001:db8:* addresses.
2193#
2194#
2195# NOTE: The use of this directive using client dependent ACLs is
2196# incompatible with the use of server side persistent connections. To
2197# ensure correct results it is best to set server_persistent_connections
2198# to off when using this directive in such configurations.
2199#
2200# NOTE: The use of this directive to set a local IP on outgoing TCP links
2201# is incompatible with using TPROXY to set client IP out outbound TCP links.
2202# When needing to contact peers use the no-tproxy cache_peer option and the
2203# client_dst_passthru directive re-enable normal forwarding such as this.
2204#
2205# This clause only supports fast acl types.
2206# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2207#Default:
2208# Address selection is performed by the operating system.
2209
2210# TAG: host_verify_strict
2211# Regardless of this option setting, when dealing with intercepted
2212# traffic, Squid always verifies that the destination IP address matches
2213# the Host header domain or IP (called 'authority form URL').
2214#
2215# This enforcement is performed to satisfy a MUST-level requirement in
2216# RFC 2616 section 14.23: "The Host field value MUST represent the naming
2217# authority of the origin server or gateway given by the original URL".
2218#
2219# When set to ON:
2220# Squid always responds with an HTTP 409 (Conflict) error
2221# page and logs a security warning if there is no match.
2222#
2223# Squid verifies that the destination IP address matches
2224# the Host header for forward-proxy and reverse-proxy traffic
2225# as well. For those traffic types, Squid also enables the
2226# following checks, comparing the corresponding Host header
2227# and Request-URI components:
2228#
2229# * The host names (domain or IP) must be identical,
2230# but valueless or missing Host header disables all checks.
2231# For the two host names to match, both must be either IP
2232# or FQDN.
2233#
2234# * Port numbers must be identical, but if a port is missing
2235# the scheme-default port is assumed.
2236#
2237#
2238# When set to OFF (the default):
2239# Squid allows suspicious requests to continue but logs a
2240# security warning and blocks caching of the response.
2241#
2242# * Forward-proxy traffic is not checked at all.
2243#
2244# * Reverse-proxy traffic is not checked at all.
2245#
2246# * Intercepted traffic which passes verification is handled
2247# according to client_dst_passthru.
2248#
2249# * Intercepted requests which fail verification are sent
2250# to the client original destination instead of DIRECT.
2251# This overrides 'client_dst_passthru off'.
2252#
2253# For now suspicious intercepted CONNECT requests are always
2254# responded to with an HTTP 409 (Conflict) error page.
2255#
2256#
2257# SECURITY NOTE:
2258#
2259# As described in CVE-2009-0801 when the Host: header alone is used
2260# to determine the destination of a request it becomes trivial for
2261# malicious scripts on remote websites to bypass browser same-origin
2262# security policy and sandboxing protections.
2263#
2264# The cause of this is that such applets are allowed to perform their
2265# own HTTP stack, in which case the same-origin policy of the browser
2266# sandbox only verifies that the applet tries to contact the same IP
2267# as from where it was loaded at the IP level. The Host: header may
2268# be different from the connected IP and approved origin.
2269#
2270#Default:
2271# host_verify_strict off
2272
2273# TAG: client_dst_passthru
2274# With NAT or TPROXY intercepted traffic Squid may pass the request
2275# directly to the original client destination IP or seek a faster
2276# source using the HTTP Host header.
2277#
2278# Using Host to locate alternative servers can provide faster
2279# connectivity with a range of failure recovery options.
2280# But can also lead to connectivity trouble when the client and
2281# server are attempting stateful interactions unaware of the proxy.
2282#
2283# This option (on by default) prevents alternative DNS entries being
2284# located to send intercepted traffic DIRECT to an origin server.
2285# The clients original destination IP and port will be used instead.
2286#
2287# Regardless of this option setting, when dealing with intercepted
2288# traffic Squid will verify the Host: header and any traffic which
2289# fails Host verification will be treated as if this option were ON.
2290#
2291# see host_verify_strict for details on the verification process.
2292#Default:
2293# client_dst_passthru on
2294
2295# TLS OPTIONS
2296# -----------------------------------------------------------------------------
2297
2298# TAG: tls_outgoing_options
2299# disable Do not support https:// URLs.
2300#
2301# cert=/path/to/client/certificate
2302# A client X.509 certificate to use when connecting.
2303#
2304# key=/path/to/client/private_key
2305# The private key corresponding to the cert= above.
2306#
2307# If key= is not specified cert= is assumed to
2308# reference a PEM file containing both the certificate
2309# and private key.
2310#
2311# cipher=... The list of valid TLS ciphers to use.
2312#
2313# min-version=1.N
2314# The minimum TLS protocol version to permit.
2315# To control SSLv3 use the options= parameter.
2316# Supported Values: 1.0 (default), 1.1, 1.2
2317#
2318# options=... Specify various TLS/SSL implementation options.
2319#
2320# OpenSSL options most important are:
2321#
2322# NO_SSLv3 Disallow the use of SSLv3
2323#
2324# SINGLE_DH_USE
2325# Always create a new key when using
2326# temporary/ephemeral DH key exchanges
2327#
2328# NO_TICKET
2329# Disable use of RFC5077 session tickets.
2330# Some servers may have problems
2331# understanding the TLS extension due
2332# to ambiguous specification in RFC4507.
2333#
2334# ALL Enable various bug workarounds
2335# suggested as "harmless" by OpenSSL
2336# Be warned that this reduces SSL/TLS
2337# strength to some attacks.
2338#
2339# See the OpenSSL SSL_CTX_set_options documentation
2340# for a more complete list.
2341#
2342# GnuTLS options most important are:
2343#
2344# %NO_TICKETS
2345# Disable use of RFC5077 session tickets.
2346# Some servers may have problems
2347# understanding the TLS extension due
2348# to ambiguous specification in RFC4507.
2349#
2350# See the GnuTLS Priority Strings documentation
2351# for a more complete list.
2352# http://www.gnutls.org/manual/gnutls.html#Priority-Strings
2353#
2354#
2355# cafile= PEM file containing CA certificates to use when verifying
2356# the peer certificate. May be repeated to load multiple files.
2357#
2358# capath= A directory containing additional CA certificates to
2359# use when verifying the peer certificate.
2360# Requires OpenSSL or LibreSSL.
2361#
2362# crlfile=... A certificate revocation list file to use when
2363# verifying the peer certificate.
2364#
2365# flags=... Specify various flags modifying the TLS implementation:
2366#
2367# DONT_VERIFY_PEER
2368# Accept certificates even if they fail to
2369# verify.
2370# DONT_VERIFY_DOMAIN
2371# Don't verify the peer certificate
2372# matches the server name
2373#
2374# default-ca[=off]
2375# Whether to use the system Trusted CAs. Default is ON.
2376#
2377# domain= The peer name as advertised in its certificate.
2378# Used for verifying the correctness of the received peer
2379# certificate. If not specified the peer hostname will be
2380# used.
2381#Default:
2382# tls_outgoing_options min-version=1.0
2383
2384# SSL OPTIONS
2385# -----------------------------------------------------------------------------
2386
2387# TAG: ssl_unclean_shutdown
2388# Note: This option is only available if Squid is rebuilt with the
2389# --with-openssl
2390#
2391# Some browsers (especially MSIE) bugs out on SSL shutdown
2392# messages.
2393#Default:
2394# ssl_unclean_shutdown off
2395
2396# TAG: ssl_engine
2397# Note: This option is only available if Squid is rebuilt with the
2398# --with-openssl
2399#
2400# The OpenSSL engine to use. You will need to set this if you
2401# would like to use hardware SSL acceleration for example.
2402#Default:
2403# none
2404
2405# TAG: sslproxy_session_ttl
2406# Note: This option is only available if Squid is rebuilt with the
2407# --with-openssl
2408#
2409# Sets the timeout value for SSL sessions
2410#Default:
2411# sslproxy_session_ttl 300
2412
2413# TAG: sslproxy_session_cache_size
2414# Note: This option is only available if Squid is rebuilt with the
2415# --with-openssl
2416#
2417# Sets the cache size to use for ssl session
2418#Default:
2419# sslproxy_session_cache_size 2 MB
2420
2421# TAG: sslproxy_foreign_intermediate_certs
2422# Note: This option is only available if Squid is rebuilt with the
2423# --with-openssl
2424#
2425# Many origin servers fail to send their full server certificate
2426# chain for verification, assuming the client already has or can
2427# easily locate any missing intermediate certificates.
2428#
2429# Squid uses the certificates from the specified file to fill in
2430# these missing chains when trying to validate origin server
2431# certificate chains.
2432#
2433# The file is expected to contain zero or more PEM-encoded
2434# intermediate certificates. These certificates are not treated
2435# as trusted root certificates, and any self-signed certificate in
2436# this file will be ignored.
2437#Default:
2438# none
2439
2440# TAG: sslproxy_cert_sign_hash
2441# Note: This option is only available if Squid is rebuilt with the
2442# --with-openssl
2443#
2444# Sets the hashing algorithm to use when signing generated certificates.
2445# Valid algorithm names depend on the OpenSSL library used. The following
2446# names are usually available: sha1, sha256, sha512, and md5. Please see
2447# your OpenSSL library manual for the available hashes. By default, Squids
2448# that support this option use sha256 hashes.
2449#
2450# Squid does not forcefully purge cached certificates that were generated
2451# with an algorithm other than the currently configured one. They remain
2452# in the cache, subject to the regular cache eviction policy, and become
2453# useful if the algorithm changes again.
2454#Default:
2455# none
2456
2457# TAG: ssl_bump
2458# Note: This option is only available if Squid is rebuilt with the
2459# --with-openssl
2460#
2461# This option is consulted when a CONNECT request is received on
2462# an http_port (or a new connection is intercepted at an
2463# https_port), provided that port was configured with an ssl-bump
2464# flag. The subsequent data on the connection is either treated as
2465# HTTPS and decrypted OR tunneled at TCP level without decryption,
2466# depending on the first matching bumping "action".
2467#
2468# ssl_bump <action> [!]acl ...
2469#
2470# The following bumping actions are currently supported:
2471#
2472# splice
2473# Become a TCP tunnel without decrypting proxied traffic.
2474# This is the default action.
2475#
2476# bump
2477# When used on step SslBump1, establishes a secure connection
2478# with the client first, then connect to the server.
2479# When used on step SslBump2 or SslBump3, establishes a secure
2480# connection with the server and, using a mimicked server
2481# certificate, with the client.
2482#
2483# peek
2484# Receive client (step SslBump1) or server (step SslBump2)
2485# certificate while preserving the possibility of splicing the
2486# connection. Peeking at the server certificate (during step 2)
2487# usually precludes bumping of the connection at step 3.
2488#
2489# stare
2490# Receive client (step SslBump1) or server (step SslBump2)
2491# certificate while preserving the possibility of bumping the
2492# connection. Staring at the server certificate (during step 2)
2493# usually precludes splicing of the connection at step 3.
2494#
2495# terminate
2496# Close client and server connections.
2497#
2498# Backward compatibility actions available at step SslBump1:
2499#
2500# client-first
2501# Bump the connection. Establish a secure connection with the
2502# client first, then connect to the server. This old mode does
2503# not allow Squid to mimic server SSL certificate and does not
2504# work with intercepted SSL connections.
2505#
2506# server-first
2507# Bump the connection. Establish a secure connection with the
2508# server first, then establish a secure connection with the
2509# client, using a mimicked server certificate. Works with both
2510# CONNECT requests and intercepted SSL connections, but does
2511# not allow to make decisions based on SSL handshake info.
2512#
2513# peek-and-splice
2514# Decide whether to bump or splice the connection based on
2515# client-to-squid and server-to-squid SSL hello messages.
2516# XXX: Remove.
2517#
2518# none
2519# Same as the "splice" action.
2520#
2521# All ssl_bump rules are evaluated at each of the supported bumping
2522# steps. Rules with actions that are impossible at the current step are
2523# ignored. The first matching ssl_bump action wins and is applied at the
2524# end of the current step. If no rules match, the splice action is used.
2525# See the at_step ACL for a list of the supported SslBump steps.
2526#
2527# This clause supports both fast and slow acl types.
2528# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2529#
2530# See also: http_port ssl-bump, https_port ssl-bump, and acl at_step.
2531#
2532#
2533# # Example: Bump all TLS connections except those originating from
2534# # localhost or those going to example.com.
2535#
2536# acl broken_sites ssl::server_name .example.com
2537# ssl_bump splice localhost
2538# ssl_bump splice broken_sites
2539# ssl_bump bump all
2540#Default:
2541# Become a TCP tunnel without decrypting proxied traffic.
2542
2543# TAG: sslproxy_cert_error
2544# Note: This option is only available if Squid is rebuilt with the
2545# --with-openssl
2546#
2547# Use this ACL to bypass server certificate validation errors.
2548#
2549# For example, the following lines will bypass all validation errors
2550# when talking to servers for example.com. All other
2551# validation errors will result in ERR_SECURE_CONNECT_FAIL error.
2552#
2553# acl BrokenButTrustedServers dstdomain example.com
2554# sslproxy_cert_error allow BrokenButTrustedServers
2555# sslproxy_cert_error deny all
2556#
2557# This clause only supports fast acl types.
2558# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2559# Using slow acl types may result in server crashes
2560#
2561# Without this option, all server certificate validation errors
2562# terminate the transaction to protect Squid and the client.
2563#
2564# SQUID_X509_V_ERR_INFINITE_VALIDATION error cannot be bypassed
2565# but should not happen unless your OpenSSL library is buggy.
2566#
2567# SECURITY WARNING:
2568# Bypassing validation errors is dangerous because an
2569# error usually implies that the server cannot be trusted
2570# and the connection may be insecure.
2571#
2572# See also: sslproxy_flags and DONT_VERIFY_PEER.
2573#Default:
2574# Server certificate errors terminate the transaction.
2575
2576# TAG: sslproxy_cert_sign
2577# Note: This option is only available if Squid is rebuilt with the
2578# --with-openssl
2579#
2580#
2581# sslproxy_cert_sign <signing algorithm> acl ...
2582#
2583# The following certificate signing algorithms are supported:
2584#
2585# signTrusted
2586# Sign using the configured CA certificate which is usually
2587# placed in and trusted by end-user browsers. This is the
2588# default for trusted origin server certificates.
2589#
2590# signUntrusted
2591# Sign to guarantee an X509_V_ERR_CERT_UNTRUSTED browser error.
2592# This is the default for untrusted origin server certificates
2593# that are not self-signed (see ssl::certUntrusted).
2594#
2595# signSelf
2596# Sign using a self-signed certificate with the right CN to
2597# generate a X509_V_ERR_DEPTH_ZERO_SELF_SIGNED_CERT error in the
2598# browser. This is the default for self-signed origin server
2599# certificates (see ssl::certSelfSigned).
2600#
2601# This clause only supports fast acl types.
2602#
2603# When sslproxy_cert_sign acl(s) match, Squid uses the corresponding
2604# signing algorithm to generate the certificate and ignores all
2605# subsequent sslproxy_cert_sign options (the first match wins). If no
2606# acl(s) match, the default signing algorithm is determined by errors
2607# detected when obtaining and validating the origin server certificate.
2608#
2609# WARNING: SQUID_X509_V_ERR_DOMAIN_MISMATCH and ssl:certDomainMismatch can
2610# be used with sslproxy_cert_adapt, but if and only if Squid is bumping a
2611# CONNECT request that carries a domain name. In all other cases (CONNECT
2612# to an IP address or an intercepted SSL connection), Squid cannot detect
2613# the domain mismatch at certificate generation time when
2614# bump-server-first is used.
2615#Default:
2616# none
2617
2618# TAG: sslproxy_cert_adapt
2619# Note: This option is only available if Squid is rebuilt with the
2620# --with-openssl
2621#
2622#
2623# sslproxy_cert_adapt <adaptation algorithm> acl ...
2624#
2625# The following certificate adaptation algorithms are supported:
2626#
2627# setValidAfter
2628# Sets the "Not After" property to the "Not After" property of
2629# the CA certificate used to sign generated certificates.
2630#
2631# setValidBefore
2632# Sets the "Not Before" property to the "Not Before" property of
2633# the CA certificate used to sign generated certificates.
2634#
2635# setCommonName or setCommonName{CN}
2636# Sets Subject.CN property to the host name specified as a
2637# CN parameter or, if no explicit CN parameter was specified,
2638# extracted from the CONNECT request. It is a misconfiguration
2639# to use setCommonName without an explicit parameter for
2640# intercepted or tproxied SSL connections.
2641#
2642# This clause only supports fast acl types.
2643#
2644# Squid first groups sslproxy_cert_adapt options by adaptation algorithm.
2645# Within a group, when sslproxy_cert_adapt acl(s) match, Squid uses the
2646# corresponding adaptation algorithm to generate the certificate and
2647# ignores all subsequent sslproxy_cert_adapt options in that algorithm's
2648# group (i.e., the first match wins within each algorithm group). If no
2649# acl(s) match, the default mimicking action takes place.
2650#
2651# WARNING: SQUID_X509_V_ERR_DOMAIN_MISMATCH and ssl:certDomainMismatch can
2652# be used with sslproxy_cert_adapt, but if and only if Squid is bumping a
2653# CONNECT request that carries a domain name. In all other cases (CONNECT
2654# to an IP address or an intercepted SSL connection), Squid cannot detect
2655# the domain mismatch at certificate generation time when
2656# bump-server-first is used.
2657#Default:
2658# none
2659
2660# TAG: sslpassword_program
2661# Note: This option is only available if Squid is rebuilt with the
2662# --with-openssl
2663#
2664# Specify a program used for entering SSL key passphrases
2665# when using encrypted SSL certificate keys. If not specified
2666# keys must either be unencrypted, or Squid started with the -N
2667# option to allow it to query interactively for the passphrase.
2668#
2669# The key file name is given as argument to the program allowing
2670# selection of the right password if you have multiple encrypted
2671# keys.
2672#Default:
2673# none
2674
2675# OPTIONS RELATING TO EXTERNAL SSL_CRTD
2676# -----------------------------------------------------------------------------
2677
2678# TAG: sslcrtd_program
2679# Note: This option is only available if Squid is rebuilt with the
2680# --enable-ssl-crtd
2681#
2682# Specify the location and options of the executable for certificate
2683# generator.
2684#
2685# /usr/lib/squid/security_file_certgen program can use a disk cache to improve response
2686# times on repeated requests. To enable caching, specify -s and -M
2687# parameters. If those parameters are not given, the program generates
2688# a new certificate on every request.
2689#
2690# For more information use:
2691# /usr/lib/squid/security_file_certgen -h
2692#Default:
2693# sslcrtd_program /usr/lib/squid/security_file_certgen -s /var/spool/squid/ssl_db -M 4MB
2694
2695# TAG: sslcrtd_children
2696# Note: This option is only available if Squid is rebuilt with the
2697# --enable-ssl-crtd
2698#
2699# Specifies the maximum number of certificate generation processes that
2700# Squid may spawn (numberofchildren) and several related options. Using
2701# too few of these helper processes (a.k.a. "helpers") creates request
2702# queues. Using too many helpers wastes your system resources. Squid
2703# does not support spawning more than 32 helpers.
2704#
2705# Usage: numberofchildren [option]...
2706#
2707# The startup= and idle= options allow some measure of skew in your
2708# tuning.
2709#
2710# startup=N
2711#
2712# Sets the minimum number of processes to spawn when Squid
2713# starts or reconfigures. When set to zero the first request will
2714# cause spawning of the first child process to handle it.
2715#
2716# Starting too few children temporary slows Squid under load while it
2717# tries to spawn enough additional processes to cope with traffic.
2718#
2719# idle=N
2720#
2721# Sets a minimum of how many processes Squid is to try and keep available
2722# at all times. When traffic begins to rise above what the existing
2723# processes can handle this many more will be spawned up to the maximum
2724# configured. A minimum setting of 1 is required.
2725#
2726# queue-size=N
2727#
2728# Sets the maximum number of queued requests. A request is queued when
2729# no existing child is idle and no new child can be started due to
2730# numberofchildren limit. If the queued requests exceed queue size for
2731# more than 3 minutes squid aborts its operation. The default value is
2732# set to 2*numberofchildren.
2733#
2734# You must have at least one ssl_crtd process.
2735#Default:
2736# sslcrtd_children 32 startup=5 idle=1
2737
2738# TAG: sslcrtvalidator_program
2739# Note: This option is only available if Squid is rebuilt with the
2740# --with-openssl
2741#
2742# Specify the location and options of the executable for ssl_crt_validator
2743# process.
2744#
2745# Usage: sslcrtvalidator_program [ttl=n] [cache=n] path ...
2746#
2747# Options:
2748# ttl=n TTL in seconds for cached results. The default is 60 secs
2749# cache=n limit the result cache size. The default value is 2048
2750#Default:
2751# none
2752
2753# TAG: sslcrtvalidator_children
2754# Note: This option is only available if Squid is rebuilt with the
2755# --with-openssl
2756#
2757# Specifies the maximum number of certificate validation processes that
2758# Squid may spawn (numberofchildren) and several related options. Using
2759# too few of these helper processes (a.k.a. "helpers") creates request
2760# queues. Using too many helpers wastes your system resources. Squid
2761# does not support spawning more than 32 helpers.
2762#
2763# Usage: numberofchildren [option]...
2764#
2765# The startup= and idle= options allow some measure of skew in your
2766# tuning.
2767#
2768# startup=N
2769#
2770# Sets the minimum number of processes to spawn when Squid
2771# starts or reconfigures. When set to zero the first request will
2772# cause spawning of the first child process to handle it.
2773#
2774# Starting too few children temporary slows Squid under load while it
2775# tries to spawn enough additional processes to cope with traffic.
2776#
2777# idle=N
2778#
2779# Sets a minimum of how many processes Squid is to try and keep available
2780# at all times. When traffic begins to rise above what the existing
2781# processes can handle this many more will be spawned up to the maximum
2782# configured. A minimum setting of 1 is required.
2783#
2784# concurrency=
2785#
2786# The number of requests each certificate validator helper can handle in
2787# parallel. A value of 0 indicates the certficate validator does not
2788# support concurrency. Defaults to 1.
2789#
2790# When this directive is set to a value >= 1 then the protocol
2791# used to communicate with the helper is modified to include
2792# a request ID in front of the request/response. The request
2793# ID from the request must be echoed back with the response
2794# to that request.
2795#
2796# queue-size=N
2797#
2798# Sets the maximum number of queued requests. A request is queued when
2799# no existing child can accept it due to concurrency limit and no new
2800# child can be started due to numberofchildren limit. If the queued
2801# requests exceed queue size for more than 3 minutes squid aborts its
2802# operation. The default value is set to 2*numberofchildren.
2803#
2804# You must have at least one ssl_crt_validator process.
2805#Default:
2806# sslcrtvalidator_children 32 startup=5 idle=1 concurrency=1
2807
2808# OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
2809# -----------------------------------------------------------------------------
2810
2811# TAG: cache_peer
2812# To specify other caches in a hierarchy, use the format:
2813#
2814# cache_peer hostname type http-port icp-port [options]
2815#
2816# For example,
2817#
2818# # proxy icp
2819# # hostname type port port options
2820# # -------------------- -------- ----- ----- -----------
2821# cache_peer parent.foo.net parent 3128 3130 default
2822# cache_peer sib1.foo.net sibling 3128 3130 proxy-only
2823# cache_peer sib2.foo.net sibling 3128 3130 proxy-only
2824# cache_peer example.com parent 80 0 default
2825# cache_peer cdn.example.com sibling 3128 0
2826#
2827# type: either 'parent', 'sibling', or 'multicast'.
2828#
2829# proxy-port: The port number where the peer accept HTTP requests.
2830# For other Squid proxies this is usually 3128
2831# For web servers this is usually 80
2832#
2833# icp-port: Used for querying neighbor caches about objects.
2834# Set to 0 if the peer does not support ICP or HTCP.
2835# See ICP and HTCP options below for additional details.
2836#
2837#
2838# ==== ICP OPTIONS ====
2839#
2840# You MUST also set icp_port and icp_access explicitly when using these options.
2841# The defaults will prevent peer traffic using ICP.
2842#
2843#
2844# no-query Disable ICP queries to this neighbor.
2845#
2846# multicast-responder
2847# Indicates the named peer is a member of a multicast group.
2848# ICP queries will not be sent directly to the peer, but ICP
2849# replies will be accepted from it.
2850#
2851# closest-only Indicates that, for ICP_OP_MISS replies, we'll only forward
2852# CLOSEST_PARENT_MISSes and never FIRST_PARENT_MISSes.
2853#
2854# background-ping
2855# To only send ICP queries to this neighbor infrequently.
2856# This is used to keep the neighbor round trip time updated
2857# and is usually used in conjunction with weighted-round-robin.
2858#
2859#
2860# ==== HTCP OPTIONS ====
2861#
2862# You MUST also set htcp_port and htcp_access explicitly when using these options.
2863# The defaults will prevent peer traffic using HTCP.
2864#
2865#
2866# htcp Send HTCP, instead of ICP, queries to the neighbor.
2867# You probably also want to set the "icp-port" to 4827
2868# instead of 3130. This directive accepts a comma separated
2869# list of options described below.
2870#
2871# htcp=oldsquid Send HTCP to old Squid versions (2.5 or earlier).
2872#
2873# htcp=no-clr Send HTCP to the neighbor but without
2874# sending any CLR requests. This cannot be used with
2875# only-clr.
2876#
2877# htcp=only-clr Send HTCP to the neighbor but ONLY CLR requests.
2878# This cannot be used with no-clr.
2879#
2880# htcp=no-purge-clr
2881# Send HTCP to the neighbor including CLRs but only when
2882# they do not result from PURGE requests.
2883#
2884# htcp=forward-clr
2885# Forward any HTCP CLR requests this proxy receives to the peer.
2886#
2887#
2888# ==== PEER SELECTION METHODS ====
2889#
2890# The default peer selection method is ICP, with the first responding peer
2891# being used as source. These options can be used for better load balancing.
2892#
2893#
2894# default This is a parent cache which can be used as a "last-resort"
2895# if a peer cannot be located by any of the peer-selection methods.
2896# If specified more than once, only the first is used.
2897#
2898# round-robin Load-Balance parents which should be used in a round-robin
2899# fashion in the absence of any ICP queries.
2900# weight=N can be used to add bias.
2901#
2902# weighted-round-robin
2903# Load-Balance parents which should be used in a round-robin
2904# fashion with the frequency of each parent being based on the
2905# round trip time. Closer parents are used more often.
2906# Usually used for background-ping parents.
2907# weight=N can be used to add bias.
2908#
2909# carp Load-Balance parents which should be used as a CARP array.
2910# The requests will be distributed among the parents based on the
2911# CARP load balancing hash function based on their weight.
2912#
2913# userhash Load-balance parents based on the client proxy_auth or ident username.
2914#
2915# sourcehash Load-balance parents based on the client source IP.
2916#
2917# multicast-siblings
2918# To be used only for cache peers of type "multicast".
2919# ALL members of this multicast group have "sibling"
2920# relationship with it, not "parent". This is to a multicast
2921# group when the requested object would be fetched only from
2922# a "parent" cache, anyway. It's useful, e.g., when
2923# configuring a pool of redundant Squid proxies, being
2924# members of the same multicast group.
2925#
2926#
2927# ==== PEER SELECTION OPTIONS ====
2928#
2929# weight=N use to affect the selection of a peer during any weighted
2930# peer-selection mechanisms.
2931# The weight must be an integer; default is 1,
2932# larger weights are favored more.
2933# This option does not affect parent selection if a peering
2934# protocol is not in use.
2935#
2936# basetime=N Specify a base amount to be subtracted from round trip
2937# times of parents.
2938# It is subtracted before division by weight in calculating
2939# which parent to fectch from. If the rtt is less than the
2940# base time the rtt is set to a minimal value.
2941#
2942# ttl=N Specify a TTL to use when sending multicast ICP queries
2943# to this address.
2944# Only useful when sending to a multicast group.
2945# Because we don't accept ICP replies from random
2946# hosts, you must configure other group members as
2947# peers with the 'multicast-responder' option.
2948#
2949# no-delay To prevent access to this neighbor from influencing the
2950# delay pools.
2951#
2952# digest-url=URL Tell Squid to fetch the cache digest (if digests are
2953# enabled) for this host from the specified URL rather
2954# than the Squid default location.
2955#
2956#
2957# ==== CARP OPTIONS ====
2958#
2959# carp-key=key-specification
2960# use a different key than the full URL to hash against the peer.
2961# the key-specification is a comma-separated list of the keywords
2962# scheme, host, port, path, params
2963# Order is not important.
2964#
2965# ==== ACCELERATOR / REVERSE-PROXY OPTIONS ====
2966#
2967# originserver Causes this parent to be contacted as an origin server.
2968# Meant to be used in accelerator setups when the peer
2969# is a web server.
2970#
2971# forceddomain=name
2972# Set the Host header of requests forwarded to this peer.
2973# Useful in accelerator setups where the server (peer)
2974# expects a certain domain name but clients may request
2975# others. ie example.com or www.example.com
2976#
2977# no-digest Disable request of cache digests.
2978#
2979# no-netdb-exchange
2980# Disables requesting ICMP RTT database (NetDB).
2981#
2982#
2983# ==== AUTHENTICATION OPTIONS ====
2984#
2985# login=user:password
2986# If this is a personal/workgroup proxy and your parent
2987# requires proxy authentication.
2988#
2989# Note: The string can include URL escapes (i.e. %20 for
2990# spaces). This also means % must be written as %%.
2991#
2992# login=PASSTHRU
2993# Send login details received from client to this peer.
2994# Both Proxy- and WWW-Authorization headers are passed
2995# without alteration to the peer.
2996# Authentication is not required by Squid for this to work.
2997#
2998# Note: This will pass any form of authentication but
2999# only Basic auth will work through a proxy unless the
3000# connection-auth options are also used.
3001#
3002# login=PASS Send login details received from client to this peer.
3003# Authentication is not required by this option.
3004#
3005# If there are no client-provided authentication headers
3006# to pass on, but username and password are available
3007# from an external ACL user= and password= result tags
3008# they may be sent instead.
3009#
3010# Note: To combine this with proxy_auth both proxies must
3011# share the same user database as HTTP only allows for
3012# a single login (one for proxy, one for origin server).
3013# Also be warned this will expose your users proxy
3014# password to the peer. USE WITH CAUTION
3015#
3016# login=*:password
3017# Send the username to the upstream cache, but with a
3018# fixed password. This is meant to be used when the peer
3019# is in another administrative domain, but it is still
3020# needed to identify each user.
3021# The star can optionally be followed by some extra
3022# information which is added to the username. This can
3023# be used to identify this proxy to the peer, similar to
3024# the login=username:password option above.
3025#
3026# login=NEGOTIATE
3027# If this is a personal/workgroup proxy and your parent
3028# requires a secure proxy authentication.
3029# The first principal from the default keytab or defined by
3030# the environment variable KRB5_KTNAME will be used.
3031#
3032# WARNING: The connection may transmit requests from multiple
3033# clients. Negotiate often assumes end-to-end authentication
3034# and a single-client. Which is not strictly true here.
3035#
3036# login=NEGOTIATE:principal_name
3037# If this is a personal/workgroup proxy and your parent
3038# requires a secure proxy authentication.
3039# The principal principal_name from the default keytab or
3040# defined by the environment variable KRB5_KTNAME will be
3041# used.
3042#
3043# WARNING: The connection may transmit requests from multiple
3044# clients. Negotiate often assumes end-to-end authentication
3045# and a single-client. Which is not strictly true here.
3046#
3047# connection-auth=on|off
3048# Tell Squid that this peer does or not support Microsoft
3049# connection oriented authentication, and any such
3050# challenges received from there should be ignored.
3051# Default is auto to automatically determine the status
3052# of the peer.
3053#
3054# auth-no-keytab
3055# Do not use a keytab to authenticate to a peer when
3056# login=NEGOTIATE is specified. Let the GSSAPI
3057# implementation determine which already existing
3058# credentials cache to use instead.
3059#
3060#
3061# ==== SSL / HTTPS / TLS OPTIONS ====
3062#
3063# tls Encrypt connections to this peer with TLS.
3064#
3065# sslcert=/path/to/ssl/certificate
3066# A client X.509 certificate to use when connecting to
3067# this peer.
3068#
3069# sslkey=/path/to/ssl/key
3070# The private key corresponding to sslcert above.
3071#
3072# If sslkey= is not specified sslcert= is assumed to
3073# reference a PEM file containing both the certificate
3074# and private key.
3075#
3076# sslcipher=... The list of valid SSL ciphers to use when connecting
3077# to this peer.
3078#
3079# tls-min-version=1.N
3080# The minimum TLS protocol version to permit. To control
3081# SSLv3 use the tls-options= parameter.
3082# Supported Values: 1.0 (default), 1.1, 1.2
3083#
3084# tls-options=... Specify various TLS implementation options.
3085#
3086# OpenSSL options most important are:
3087#
3088# NO_SSLv3 Disallow the use of SSLv3
3089#
3090# SINGLE_DH_USE
3091# Always create a new key when using
3092# temporary/ephemeral DH key exchanges
3093#
3094# NO_TICKET
3095# Disable use of RFC5077 session tickets.
3096# Some servers may have problems
3097# understanding the TLS extension due
3098# to ambiguous specification in RFC4507.
3099#
3100# ALL Enable various bug workarounds
3101# suggested as "harmless" by OpenSSL
3102# Be warned that this reduces SSL/TLS
3103# strength to some attacks.
3104#
3105# See the OpenSSL SSL_CTX_set_options documentation for a
3106# more complete list.
3107#
3108# GnuTLS options most important are:
3109#
3110# %NO_TICKETS
3111# Disable use of RFC5077 session tickets.
3112# Some servers may have problems
3113# understanding the TLS extension due
3114# to ambiguous specification in RFC4507.
3115#
3116# See the GnuTLS Priority Strings documentation
3117# for a more complete list.
3118# http://www.gnutls.org/manual/gnutls.html#Priority-Strings
3119#
3120# tls-cafile= PEM file containing CA certificates to use when verifying
3121# the peer certificate. May be repeated to load multiple files.
3122#
3123# sslcapath=... A directory containing additional CA certificates to
3124# use when verifying the peer certificate.
3125# Requires OpenSSL or LibreSSL.
3126#
3127# sslcrlfile=... A certificate revocation list file to use when
3128# verifying the peer certificate.
3129#
3130# sslflags=... Specify various flags modifying the SSL implementation:
3131#
3132# DONT_VERIFY_PEER
3133# Accept certificates even if they fail to
3134# verify.
3135#
3136# DONT_VERIFY_DOMAIN
3137# Don't verify the peer certificate
3138# matches the server name
3139#
3140# ssldomain= The peer name as advertised in it's certificate.
3141# Used for verifying the correctness of the received peer
3142# certificate. If not specified the peer hostname will be
3143# used.
3144#
3145# front-end-https[=off|on|auto]
3146# Enable the "Front-End-Https: On" header needed when
3147# using Squid as a SSL frontend in front of Microsoft OWA.
3148# See MS KB document Q307347 for details on this header.
3149# If set to auto the header will only be added if the
3150# request is forwarded as a https:// URL.
3151#
3152# tls-default-ca[=off]
3153# Whether to use the system Trusted CAs. Default is ON.
3154#
3155# tls-no-npn Do not use the TLS NPN extension to advertise HTTP/1.1.
3156#
3157# ==== GENERAL OPTIONS ====
3158#
3159# connect-timeout=N
3160# A peer-specific connect timeout.
3161# Also see the peer_connect_timeout directive.
3162#
3163# connect-fail-limit=N
3164# How many times connecting to a peer must fail before
3165# it is marked as down. Standby connection failures
3166# count towards this limit. Default is 10.
3167#
3168# allow-miss Disable Squid's use of only-if-cached when forwarding
3169# requests to siblings. This is primarily useful when
3170# icp_hit_stale is used by the sibling. Excessive use
3171# of this option may result in forwarding loops. One way
3172# to prevent peering loops when using this option, is to
3173# deny cache peer usage on requests from a peer:
3174# acl fromPeer ...
3175# cache_peer_access peerName deny fromPeer
3176#
3177# max-conn=N Limit the number of concurrent connections the Squid
3178# may open to this peer, including already opened idle
3179# and standby connections. There is no peer-specific
3180# connection limit by default.
3181#
3182# A peer exceeding the limit is not used for new
3183# requests unless a standby connection is available.
3184#
3185# max-conn currently works poorly with idle persistent
3186# connections: When a peer reaches its max-conn limit,
3187# and there are idle persistent connections to the peer,
3188# the peer may not be selected because the limiting code
3189# does not know whether Squid can reuse those idle
3190# connections.
3191#
3192# standby=N Maintain a pool of N "hot standby" connections to an
3193# UP peer, available for requests when no idle
3194# persistent connection is available (or safe) to use.
3195# By default and with zero N, no such pool is maintained.
3196# N must not exceed the max-conn limit (if any).
3197#
3198# At start or after reconfiguration, Squid opens new TCP
3199# standby connections until there are N connections
3200# available and then replenishes the standby pool as
3201# opened connections are used up for requests. A used
3202# connection never goes back to the standby pool, but
3203# may go to the regular idle persistent connection pool
3204# shared by all peers and origin servers.
3205#
3206# Squid never opens multiple new standby connections
3207# concurrently. This one-at-a-time approach minimizes
3208# flooding-like effect on peers. Furthermore, just a few
3209# standby connections should be sufficient in most cases
3210# to supply most new requests with a ready-to-use
3211# connection.
3212#
3213# Standby connections obey server_idle_pconn_timeout.
3214# For the feature to work as intended, the peer must be
3215# configured to accept and keep them open longer than
3216# the idle timeout at the connecting Squid, to minimize
3217# race conditions typical to idle used persistent
3218# connections. Default request_timeout and
3219# server_idle_pconn_timeout values ensure such a
3220# configuration.
3221#
3222# name=xxx Unique name for the peer.
3223# Required if you have multiple peers on the same host
3224# but different ports.
3225# This name can be used in cache_peer_access and similar
3226# directives to identify the peer.
3227# Can be used by outgoing access controls through the
3228# peername ACL type.
3229#
3230# no-tproxy Do not use the client-spoof TPROXY support when forwarding
3231# requests to this peer. Use normal address selection instead.
3232# This overrides the spoof_client_ip ACL.
3233#
3234# proxy-only objects fetched from the peer will not be stored locally.
3235#
3236#Default:
3237# none
3238
3239# TAG: cache_peer_access
3240# Restricts usage of cache_peer proxies.
3241#
3242# Usage:
3243# cache_peer_access peer-name allow|deny [!]aclname ...
3244#
3245# For the required peer-name parameter, use either the value of the
3246# cache_peer name=value parameter or, if name=value is missing, the
3247# cache_peer hostname parameter.
3248#
3249# This directive narrows down the selection of peering candidates, but
3250# does not determine the order in which the selected candidates are
3251# contacted. That order is determined by the peer selection algorithms
3252# (see PEER SELECTION sections in the cache_peer documentation).
3253#
3254# If a deny rule matches, the corresponding peer will not be contacted
3255# for the current transaction -- Squid will not send ICP queries and
3256# will not forward HTTP requests to that peer. An allow match leaves
3257# the corresponding peer in the selection. The first match for a given
3258# peer wins for that peer.
3259#
3260# The relative order of cache_peer_access directives for the same peer
3261# matters. The relative order of any two cache_peer_access directives
3262# for different peers does not matter. To ease interpretation, it is a
3263# good idea to group cache_peer_access directives for the same peer
3264# together.
3265#
3266# A single cache_peer_access directive may be evaluated multiple times
3267# for a given transaction because individual peer selection algorithms
3268# may check it independently from each other. These redundant checks
3269# may be optimized away in future Squid versions.
3270#
3271# This clause only supports fast acl types.
3272# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
3273#
3274#Default:
3275# No peer usage restrictions.
3276
3277# TAG: neighbor_type_domain
3278# Modify the cache_peer neighbor type when passing requests
3279# about specific domains to the peer.
3280#
3281# Usage:
3282# neighbor_type_domain neighbor parent|sibling domain domain ...
3283#
3284# For example:
3285# cache_peer foo.example.com parent 3128 3130
3286# neighbor_type_domain foo.example.com sibling .au .de
3287#
3288# The above configuration treats all requests to foo.example.com as a
3289# parent proxy unless the request is for a .au or .de ccTLD domain name.
3290#Default:
3291# The peer type from cache_peer directive is used for all requests to that peer.
3292
3293# TAG: dead_peer_timeout (seconds)
3294# This controls how long Squid waits to declare a peer cache
3295# as "dead." If there are no ICP replies received in this
3296# amount of time, Squid will declare the peer dead and not
3297# expect to receive any further ICP replies. However, it
3298# continues to send ICP queries, and will mark the peer as
3299# alive upon receipt of the first subsequent ICP reply.
3300#
3301# This timeout also affects when Squid expects to receive ICP
3302# replies from peers. If more than 'dead_peer' seconds have
3303# passed since the last ICP reply was received, Squid will not
3304# expect to receive an ICP reply on the next query. Thus, if
3305# your time between requests is greater than this timeout, you
3306# will see a lot of requests sent DIRECT to origin servers
3307# instead of to your parents.
3308#Default:
3309# dead_peer_timeout 10 seconds
3310
3311# TAG: forward_max_tries
3312# Limits the number of attempts to forward the request.
3313#
3314# For the purpose of this limit, Squid counts all high-level request
3315# forwarding attempts, including any same-destination retries after
3316# certain persistent connection failures and any attempts to use a
3317# different peer. However, low-level connection reopening attempts
3318# (enabled using connect_retries) are not counted.
3319#
3320# See also: forward_timeout and connect_retries.
3321#Default:
3322# forward_max_tries 25
3323
3324# MEMORY CACHE OPTIONS
3325# -----------------------------------------------------------------------------
3326
3327# TAG: cache_mem (bytes)
3328# NOTE: THIS PARAMETER DOES NOT SPECIFY THE MAXIMUM PROCESS SIZE.
3329# IT ONLY PLACES A LIMIT ON HOW MUCH ADDITIONAL MEMORY SQUID WILL
3330# USE AS A MEMORY CACHE OF OBJECTS. SQUID USES MEMORY FOR OTHER
3331# THINGS AS WELL. SEE THE SQUID FAQ SECTION 8 FOR DETAILS.
3332#
3333# 'cache_mem' specifies the ideal amount of memory to be used
3334# for:
3335# * In-Transit objects
3336# * Hot Objects
3337# * Negative-Cached objects
3338#
3339# Data for these objects are stored in 4 KB blocks. This
3340# parameter specifies the ideal upper limit on the total size of
3341# 4 KB blocks allocated. In-Transit objects take the highest
3342# priority.
3343#
3344# In-transit objects have priority over the others. When
3345# additional space is needed for incoming data, negative-cached
3346# and hot objects will be released. In other words, the
3347# negative-cached and hot objects will fill up any unused space
3348# not needed for in-transit objects.
3349#
3350# If circumstances require, this limit will be exceeded.
3351# Specifically, if your incoming request rate requires more than
3352# 'cache_mem' of memory to hold in-transit objects, Squid will
3353# exceed this limit to satisfy the new requests. When the load
3354# decreases, blocks will be freed until the high-water mark is
3355# reached. Thereafter, blocks will be used to store hot
3356# objects.
3357#
3358# If shared memory caching is enabled, Squid does not use the shared
3359# cache space for in-transit objects, but they still consume as much
3360# local memory as they need. For more details about the shared memory
3361# cache, see memory_cache_shared.
3362#Default:
3363# cache_mem 256 MB
3364
3365# TAG: maximum_object_size_in_memory (bytes)
3366# Objects greater than this size will not be attempted to kept in
3367# the memory cache. This should be set high enough to keep objects
3368# accessed frequently in memory to improve performance whilst low
3369# enough to keep larger objects from hoarding cache_mem.
3370#Default:
3371# maximum_object_size_in_memory 512 KB
3372
3373# TAG: memory_cache_shared on|off
3374# Controls whether the memory cache is shared among SMP workers.
3375#
3376# The shared memory cache is meant to occupy cache_mem bytes and replace
3377# the non-shared memory cache, although some entities may still be
3378# cached locally by workers for now (e.g., internal and in-transit
3379# objects may be served from a local memory cache even if shared memory
3380# caching is enabled).
3381#
3382# By default, the memory cache is shared if and only if all of the
3383# following conditions are satisfied: Squid runs in SMP mode with
3384# multiple workers, cache_mem is positive, and Squid environment
3385# supports required IPC primitives (e.g., POSIX shared memory segments
3386# and GCC-style atomic operations).
3387#
3388# To avoid blocking locks, shared memory uses opportunistic algorithms
3389# that do not guarantee that every cachable entity that could have been
3390# shared among SMP workers will actually be shared.
3391#Default:
3392# "on" where supported if doing memory caching with multiple SMP workers.
3393
3394# TAG: memory_cache_mode
3395# Controls which objects to keep in the memory cache (cache_mem)
3396#
3397# always Keep most recently fetched objects in memory (default)
3398#
3399# disk Only disk cache hits are kept in memory, which means
3400# an object must first be cached on disk and then hit
3401# a second time before cached in memory.
3402#
3403# network Only objects fetched from network is kept in memory
3404#Default:
3405# Keep the most recently fetched objects in memory
3406
3407# TAG: memory_replacement_policy
3408# The memory replacement policy parameter determines which
3409# objects are purged from memory when memory space is needed.
3410#
3411# See cache_replacement_policy for details on algorithms.
3412#Default:
3413# memory_replacement_policy lru
3414
3415# DISK CACHE OPTIONS
3416# -----------------------------------------------------------------------------
3417
3418# TAG: cache_replacement_policy
3419# The cache replacement policy parameter determines which
3420# objects are evicted (replaced) when disk space is needed.
3421#
3422# lru : Squid's original list based LRU policy
3423# heap GDSF : Greedy-Dual Size Frequency
3424# heap LFUDA: Least Frequently Used with Dynamic Aging
3425# heap LRU : LRU policy implemented using a heap
3426#
3427# Applies to any cache_dir lines listed below this directive.
3428#
3429# The LRU policies keeps recently referenced objects.
3430#
3431# The heap GDSF policy optimizes object hit rate by keeping smaller
3432# popular objects in cache so it has a better chance of getting a
3433# hit. It achieves a lower byte hit rate than LFUDA though since
3434# it evicts larger (possibly popular) objects.
3435#
3436# The heap LFUDA policy keeps popular objects in cache regardless of
3437# their size and thus optimizes byte hit rate at the expense of
3438# hit rate since one large, popular object will prevent many
3439# smaller, slightly less popular objects from being cached.
3440#
3441# Both policies utilize a dynamic aging mechanism that prevents
3442# cache pollution that can otherwise occur with frequency-based
3443# replacement policies.
3444#
3445# NOTE: if using the LFUDA replacement policy you should increase
3446# the value of maximum_object_size above its default of 4 MB to
3447# to maximize the potential byte hit rate improvement of LFUDA.
3448#
3449# For more information about the GDSF and LFUDA cache replacement
3450# policies see http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html
3451# and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html.
3452#Default:
3453# cache_replacement_policy lru
3454
3455# TAG: minimum_object_size (bytes)
3456# Objects smaller than this size will NOT be saved on disk. The
3457# value is specified in bytes, and the default is 0 KB, which
3458# means all responses can be stored.
3459#Default:
3460# no limit
3461
3462# TAG: maximum_object_size (bytes)
3463# Set the default value for max-size parameter on any cache_dir.
3464# The value is specified in bytes, and the default is 4 MB.
3465#
3466# If you wish to get a high BYTES hit ratio, you should probably
3467# increase this (one 32 MB object hit counts for 3200 10KB
3468# hits).
3469#
3470# If you wish to increase hit ratio more than you want to
3471# save bandwidth you should leave this low.
3472#
3473# NOTE: if using the LFUDA replacement policy you should increase
3474# this value to maximize the byte hit rate improvement of LFUDA!
3475# See cache_replacement_policy for a discussion of this policy.
3476#Default:
3477# maximum_object_size 4 MB
3478
3479# TAG: cache_dir
3480# Format:
3481# cache_dir Type Directory-Name Fs-specific-data [options]
3482#
3483# You can specify multiple cache_dir lines to spread the
3484# cache among different disk partitions.
3485#
3486# Type specifies the kind of storage system to use. Only "ufs"
3487# is built by default. To enable any of the other storage systems
3488# see the --enable-storeio configure option.
3489#
3490# 'Directory' is a top-level directory where cache swap
3491# files will be stored. If you want to use an entire disk
3492# for caching, this can be the mount-point directory.
3493# The directory must exist and be writable by the Squid
3494# process. Squid will NOT create this directory for you.
3495#
3496# In SMP configurations, cache_dir must not precede the workers option
3497# and should use configuration macros or conditionals to give each
3498# worker interested in disk caching a dedicated cache directory.
3499#
3500#
3501# ==== The ufs store type ====
3502#
3503# "ufs" is the old well-known Squid storage format that has always
3504# been there.
3505#
3506# Usage:
3507# cache_dir ufs Directory-Name Mbytes L1 L2 [options]
3508#
3509# 'Mbytes' is the amount of disk space (MB) to use under this
3510# directory. The default is 100 MB. Change this to suit your
3511# configuration. Do NOT put the size of your disk drive here.
3512# Instead, if you want Squid to use the entire disk drive,
3513# subtract 20% and use that value.
3514#
3515# 'L1' is the number of first-level subdirectories which
3516# will be created under the 'Directory'. The default is 16.
3517#
3518# 'L2' is the number of second-level subdirectories which
3519# will be created under each first-level directory. The default
3520# is 256.
3521#
3522#
3523# ==== The aufs store type ====
3524#
3525# "aufs" uses the same storage format as "ufs", utilizing
3526# POSIX-threads to avoid blocking the main Squid process on
3527# disk-I/O. This was formerly known in Squid as async-io.
3528#
3529# Usage:
3530# cache_dir aufs Directory-Name Mbytes L1 L2 [options]
3531#
3532# see argument descriptions under ufs above
3533#
3534#
3535# ==== The diskd store type ====
3536#
3537# "diskd" uses the same storage format as "ufs", utilizing a
3538# separate process to avoid blocking the main Squid process on
3539# disk-I/O.
3540#
3541# Usage:
3542# cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]
3543#
3544# see argument descriptions under ufs above
3545#
3546# Q1 specifies the number of unacknowledged I/O requests when Squid
3547# stops opening new files. If this many messages are in the queues,
3548# Squid won't open new files. Default is 64
3549#
3550# Q2 specifies the number of unacknowledged messages when Squid
3551# starts blocking. If this many messages are in the queues,
3552# Squid blocks until it receives some replies. Default is 72
3553#
3554# When Q1 < Q2 (the default), the cache directory is optimized
3555# for lower response time at the expense of a decrease in hit
3556# ratio. If Q1 > Q2, the cache directory is optimized for
3557# higher hit ratio at the expense of an increase in response
3558# time.
3559#
3560#
3561# ==== The rock store type ====
3562#
3563# Usage:
3564# cache_dir rock Directory-Name Mbytes [options]
3565#
3566# The Rock Store type is a database-style storage. All cached
3567# entries are stored in a "database" file, using fixed-size slots.
3568# A single entry occupies one or more slots.
3569#
3570# If possible, Squid using Rock Store creates a dedicated kid
3571# process called "disker" to avoid blocking Squid worker(s) on disk
3572# I/O. One disker kid is created for each rock cache_dir. Diskers
3573# are created only when Squid, running in daemon mode, has support
3574# for the IpcIo disk I/O module.
3575#
3576# swap-timeout=msec: Squid will not start writing a miss to or
3577# reading a hit from disk if it estimates that the swap operation
3578# will take more than the specified number of milliseconds. By
3579# default and when set to zero, disables the disk I/O time limit
3580# enforcement. Ignored when using blocking I/O module because
3581# blocking synchronous I/O does not allow Squid to estimate the
3582# expected swap wait time.
3583#
3584# max-swap-rate=swaps/sec: Artificially limits disk access using
3585# the specified I/O rate limit. Swap out requests that
3586# would cause the average I/O rate to exceed the limit are
3587# delayed. Individual swap in requests (i.e., hits or reads) are
3588# not delayed, but they do contribute to measured swap rate and
3589# since they are placed in the same FIFO queue as swap out
3590# requests, they may wait longer if max-swap-rate is smaller.
3591# This is necessary on file systems that buffer "too
3592# many" writes and then start blocking Squid and other processes
3593# while committing those writes to disk. Usually used together
3594# with swap-timeout to avoid excessive delays and queue overflows
3595# when disk demand exceeds available disk "bandwidth". By default
3596# and when set to zero, disables the disk I/O rate limit
3597# enforcement. Currently supported by IpcIo module only.
3598#
3599# slot-size=bytes: The size of a database "record" used for
3600# storing cached responses. A cached response occupies at least
3601# one slot and all database I/O is done using individual slots so
3602# increasing this parameter leads to more disk space waste while
3603# decreasing it leads to more disk I/O overheads. Should be a
3604# multiple of your operating system I/O page size. Defaults to
3605# 16KBytes. A housekeeping header is stored with each slot and
3606# smaller slot-sizes will be rejected. The header is smaller than
3607# 100 bytes.
3608#
3609#
3610# ==== COMMON OPTIONS ====
3611#
3612# no-store no new objects should be stored to this cache_dir.
3613#
3614# min-size=n the minimum object size in bytes this cache_dir
3615# will accept. It's used to restrict a cache_dir
3616# to only store large objects (e.g. AUFS) while
3617# other stores are optimized for smaller objects
3618# (e.g. Rock).
3619# Defaults to 0.
3620#
3621# max-size=n the maximum object size in bytes this cache_dir
3622# supports.
3623# The value in maximum_object_size directive sets
3624# the default unless more specific details are
3625# available (ie a small store capacity).
3626#
3627# Note: To make optimal use of the max-size limits you should order
3628# the cache_dir lines with the smallest max-size value first.
3629#
3630#Default:
3631# No disk cache. Store cache ojects only in memory.
3632#
3633
3634# Uncomment and adjust the following to add a disk cache directory.
3635#cache_dir ufs /var/spool/squid 100 16 256
3636
3637# TAG: store_dir_select_algorithm
3638# How Squid selects which cache_dir to use when the response
3639# object will fit into more than one.
3640#
3641# Regardless of which algorithm is used the cache_dir min-size
3642# and max-size parameters are obeyed. As such they can affect
3643# the selection algorithm by limiting the set of considered
3644# cache_dir.
3645#
3646# Algorithms:
3647#
3648# least-load
3649#
3650# This algorithm is suited to caches with similar cache_dir
3651# sizes and disk speeds.
3652#
3653# The disk with the least I/O pending is selected.
3654# When there are multiple disks with the same I/O load ranking
3655# the cache_dir with most available capacity is selected.
3656#
3657# When a mix of cache_dir sizes are configured the faster disks
3658# have a naturally lower I/O loading and larger disks have more
3659# capacity. So space used to store objects and data throughput
3660# may be very unbalanced towards larger disks.
3661#
3662#
3663# round-robin
3664#
3665# This algorithm is suited to caches with unequal cache_dir
3666# disk sizes.
3667#
3668# Each cache_dir is selected in a rotation. The next suitable
3669# cache_dir is used.
3670#
3671# Available cache_dir capacity is only considered in relation
3672# to whether the object will fit and meets the min-size and
3673# max-size parameters.
3674#
3675# Disk I/O loading is only considered to prevent overload on slow
3676# disks. This algorithm does not spread objects by size, so any
3677# I/O loading per-disk may appear very unbalanced and volatile.
3678#
3679# If several cache_dirs use similar min-size, max-size, or other
3680# limits to to reject certain responses, then do not group such
3681# cache_dir lines together, to avoid round-robin selection bias
3682# towards the first cache_dir after the group. Instead, interleave
3683# cache_dir lines from different groups. For example:
3684#
3685# store_dir_select_algorithm round-robin
3686# cache_dir rock /hdd1 ... min-size=100000
3687# cache_dir rock /ssd1 ... max-size=99999
3688# cache_dir rock /hdd2 ... min-size=100000
3689# cache_dir rock /ssd2 ... max-size=99999
3690# cache_dir rock /hdd3 ... min-size=100000
3691# cache_dir rock /ssd3 ... max-size=99999
3692cache_dir ufs /var/spool/squid 100 16 256
3693#Default:
3694# store_dir_select_algorithm least-load
3695
3696# TAG: max_open_disk_fds
3697# To avoid having disk as the I/O bottleneck Squid can optionally
3698# bypass the on-disk cache if more than this amount of disk file
3699# descriptors are open.
3700#
3701# A value of 0 indicates no limit.
3702#Default:
3703# no limit
3704
3705# TAG: cache_swap_low (percent, 0-100)
3706# The low-water mark for AUFS/UFS/diskd cache object eviction by
3707# the cache_replacement_policy algorithm.
3708#
3709# Removal begins when the swap (disk) usage of a cache_dir is
3710# above this low-water mark and attempts to maintain utilization
3711# near the low-water mark.
3712#
3713# As swap utilization increases towards the high-water mark set
3714# by cache_swap_high object eviction becomes more agressive.
3715#
3716# The value difference in percentages between low- and high-water
3717# marks represent an eviction rate of 300 objects per second and
3718# the rate continues to scale in agressiveness by multiples of
3719# this above the high-water mark.
3720#
3721# Defaults are 90% and 95%. If you have a large cache, 5% could be
3722# hundreds of MB. If this is the case you may wish to set these
3723# numbers closer together.
3724#
3725# See also cache_swap_high and cache_replacement_policy
3726#Default:
3727# cache_swap_low 90
3728
3729# TAG: cache_swap_high (percent, 0-100)
3730# The high-water mark for AUFS/UFS/diskd cache object eviction by
3731# the cache_replacement_policy algorithm.
3732#
3733# Removal begins when the swap (disk) usage of a cache_dir is
3734# above the low-water mark set by cache_swap_low and attempts to
3735# maintain utilization near the low-water mark.
3736#
3737# As swap utilization increases towards this high-water mark object
3738# eviction becomes more agressive.
3739#
3740# The value difference in percentages between low- and high-water
3741# marks represent an eviction rate of 300 objects per second and
3742# the rate continues to scale in agressiveness by multiples of
3743# this above the high-water mark.
3744#
3745# Defaults are 90% and 95%. If you have a large cache, 5% could be
3746# hundreds of MB. If this is the case you may wish to set these
3747# numbers closer together.
3748#
3749# See also cache_swap_low and cache_replacement_policy
3750#Default:
3751# cache_swap_high 95
3752
3753# LOGFILE OPTIONS
3754# -----------------------------------------------------------------------------
3755
3756# TAG: logformat
3757# Usage:
3758#
3759# logformat <name> <format specification>
3760#
3761# Defines an access log format.
3762#
3763# The <format specification> is a string with embedded % format codes
3764#
3765# % format codes all follow the same basic structure where all
3766# components but the formatcode are optional and usually unnecessary,
3767# especially when dealing with common codes.
3768#
3769# % [encoding] [-] [[0]width] [{arg}] formatcode [{arg}]
3770#
3771# encoding escapes or otherwise protects "special" characters:
3772#
3773# " Quoted string encoding where quote(") and
3774# backslash(\) characters are \-escaped while
3775# CR, LF, and TAB characters are encoded as \r,
3776# \n, and \t two-character sequences.
3777#
3778# [ Custom Squid encoding where percent(%), square
3779# brackets([]), backslash(\) and characters with
3780# codes outside of [32,126] range are %-encoded.
3781# SP is not encoded. Used by log_mime_hdrs.
3782#
3783# # URL encoding (a.k.a. percent-encoding) where
3784# all URL unsafe and control characters (per RFC
3785# 1738) are %-encoded.
3786#
3787# / Shell-like encoding where quote(") and
3788# backslash(\) characters are \-escaped while CR
3789# and LF characters are encoded as \r and \n
3790# two-character sequences. Values containing SP
3791# character(s) are surrounded by quotes(").
3792#
3793# ' Raw/as-is encoding with no escaping/quoting.
3794#
3795# Default encoding: When no explicit encoding is
3796# specified, each %code determines its own encoding.
3797# Most %codes use raw/as-is encoding, but some codes use
3798# a so called "pass-through URL encoding" where all URL
3799# unsafe and control characters (per RFC 1738) are
3800# %-encoded, but the percent character(%) is left as is.
3801#
3802# - left aligned
3803#
3804# width minimum and/or maximum field width:
3805# [width_min][.width_max]
3806# When minimum starts with 0, the field is zero-padded.
3807# String values exceeding maximum width are truncated.
3808#
3809# {arg} argument such as header name etc. This field may be
3810# placed before or after the token, but not both at once.
3811#
3812# Format codes:
3813#
3814# % a literal % character
3815# sn Unique sequence number per log line entry
3816# err_code The ID of an error response served by Squid or
3817# a similar internal error identifier.
3818# err_detail Additional err_code-dependent error information.
3819# note The annotation specified by the argument. Also
3820# logs the adaptation meta headers set by the
3821# adaptation_meta configuration parameter.
3822# If no argument given all annotations logged.
3823# The argument may include a separator to use with
3824# annotation values:
3825# name[:separator]
3826# By default, multiple note values are separated with ","
3827# and multiple notes are separated with "\r\n".
3828# When logging named notes with %{name}note, the
3829# explicitly configured separator is used between note
3830# values. When logging all notes with %note, the
3831# explicitly configured separator is used between
3832# individual notes. There is currently no way to
3833# specify both value and notes separators when logging
3834# all notes with %note.
3835#
3836# Connection related format codes:
3837#
3838# >a Client source IP address
3839# >A Client FQDN
3840# >p Client source port
3841# >eui Client source EUI (MAC address, EUI-48 or EUI-64 identifier)
3842# >la Local IP address the client connected to
3843# >lp Local port number the client connected to
3844# >qos Client connection TOS/DSCP value set by Squid
3845# >nfmark Client connection netfilter mark set by Squid
3846#
3847# la Local listening IP address the client connection was connected to.
3848# lp Local listening port number the client connection was connected to.
3849#
3850# <a Server IP address of the last server or peer connection
3851# <A Server FQDN or peer name
3852# <p Server port number of the last server or peer connection
3853# <la Local IP address of the last server or peer connection
3854# <lp Local port number of the last server or peer connection
3855# <qos Server connection TOS/DSCP value set by Squid
3856# <nfmark Server connection netfilter mark set by Squid
3857#
3858# >handshake Raw client handshake
3859# Initial client bytes received by Squid on a newly
3860# accepted TCP connection or inside a just established
3861# CONNECT tunnel. Squid stops accumulating handshake
3862# bytes as soon as the handshake parser succeeds or
3863# fails (determining whether the client is using the
3864# expected protocol).
3865#
3866# For HTTP clients, the handshake is the request line.
3867# For TLS clients, the handshake consists of all TLS
3868# records up to and including the TLS record that
3869# contains the last byte of the first ClientHello
3870# message. For clients using an unsupported protocol,
3871# this field contains the bytes received by Squid at the
3872# time of the handshake parsing failure.
3873#
3874# See the on_unsupported_protocol directive for more
3875# information on Squid handshake traffic expectations.
3876#
3877# Current support is limited to these contexts:
3878# - http_port connections, but only when the
3879# on_unsupported_protocol directive is in use.
3880# - https_port connections (and CONNECT tunnels) that
3881# are subject to the ssl_bump peek or stare action.
3882#
3883# To protect binary handshake data, this field is always
3884# base64-encoded (RFC 4648 Section 4). If logformat
3885# field encoding is configured, that encoding is applied
3886# on top of base64. Otherwise, the computed base64 value
3887# is recorded as is.
3888#
3889# Time related format codes:
3890#
3891# ts Seconds since epoch
3892# tu subsecond time (milliseconds)
3893# tl Local time. Optional strftime format argument
3894# default %d/%b/%Y:%H:%M:%S %z
3895# tg GMT time. Optional strftime format argument
3896# default %d/%b/%Y:%H:%M:%S %z
3897# tr Response time (milliseconds)
3898# dt Total time spent making DNS lookups (milliseconds)
3899# tS Approximate master transaction start time in
3900# <full seconds since epoch>.<fractional seconds> format.
3901# Currently, Squid considers the master transaction
3902# started when a complete HTTP request header initiating
3903# the transaction is received from the client. This is
3904# the same value that Squid uses to calculate transaction
3905# response time when logging %tr to access.log. Currently,
3906# Squid uses millisecond resolution for %tS values,
3907# similar to the default access.log "current time" field
3908# (%ts.%03tu).
3909#
3910# Access Control related format codes:
3911#
3912# et Tag returned by external acl
3913# ea Log string returned by external acl
3914# un User name (any available)
3915# ul User name from authentication
3916# ue User name from external acl helper
3917# ui User name from ident
3918# un A user name. Expands to the first available name
3919# from the following list of information sources:
3920# - authenticated user name, like %ul
3921# - user name supplied by an external ACL, like %ue
3922# - SSL client name, like %us
3923# - ident user name, like %ui
3924# credentials Client credentials. The exact meaning depends on
3925# the authentication scheme: For Basic authentication,
3926# it is the password; for Digest, the realm sent by the
3927# client; for NTLM and Negotiate, the client challenge
3928# or client credentials prefixed with "YR " or "KK ".
3929#
3930# HTTP related format codes:
3931#
3932# REQUEST
3933#
3934# [http::]rm Request method (GET/POST etc)
3935# [http::]>rm Request method from client
3936# [http::]<rm Request method sent to server or peer
3937#
3938# [http::]ru Request URL received (or computed) and sanitized
3939#
3940# Logs request URI received from the client, a
3941# request adaptation service, or a request
3942# redirector (whichever was applied last).
3943#
3944# Computed URLs are URIs of internally generated
3945# requests and various "error:..." URIs.
3946#
3947# Honors strip_query_terms and uri_whitespace.
3948#
3949# This field is not encoded by default. Encoding
3950# this field using variants of %-encoding will
3951# clash with uri_whitespace modifications that
3952# also use %-encoding.
3953#
3954# [http::]>ru Request URL received from the client (or computed)
3955#
3956# Computed URLs are URIs of internally generated
3957# requests and various "error:..." URIs.
3958#
3959# Unlike %ru, this request URI is not affected
3960# by request adaptation, URL rewriting services,
3961# and strip_query_terms.
3962#
3963# Honors uri_whitespace.
3964#
3965# This field is using pass-through URL encoding
3966# by default. Encoding this field using other
3967# variants of %-encoding will clash with
3968# uri_whitespace modifications that also use
3969# %-encoding.
3970#
3971# [http::]<ru Request URL sent to server or peer
3972# [http::]>rs Request URL scheme from client
3973# [http::]<rs Request URL scheme sent to server or peer
3974# [http::]>rd Request URL domain from client
3975# [http::]<rd Request URL domain sent to server or peer
3976# [http::]>rP Request URL port from client
3977# [http::]<rP Request URL port sent to server or peer
3978# [http::]rp Request URL path excluding hostname
3979# [http::]>rp Request URL path excluding hostname from client
3980# [http::]<rp Request URL path excluding hostname sent to server or peer
3981# [http::]rv Request protocol version
3982# [http::]>rv Request protocol version from client
3983# [http::]<rv Request protocol version sent to server or peer
3984#
3985# [http::]>h Original received request header.
3986# Usually differs from the request header sent by
3987# Squid, although most fields are often preserved.
3988# Accepts optional header field name/value filter
3989# argument using name[:[separator]element] format.
3990# [http::]>ha Received request header after adaptation and
3991# redirection (pre-cache REQMOD vectoring point).
3992# Usually differs from the request header sent by
3993# Squid, although most fields are often preserved.
3994# Optional header name argument as for >h
3995#
3996# RESPONSE
3997#
3998# [http::]<Hs HTTP status code received from the next hop
3999# [http::]>Hs HTTP status code sent to the client
4000#
4001# [http::]<h Reply header. Optional header name argument
4002# as for >h
4003#
4004# [http::]mt MIME content type
4005#
4006#
4007# SIZE COUNTERS
4008#
4009# [http::]st Total size of request + reply traffic with client
4010# [http::]>st Total size of request received from client.
4011# Excluding chunked encoding bytes.
4012# [http::]<st Total size of reply sent to client (after adaptation)
4013#
4014# [http::]>sh Size of request headers received from client
4015# [http::]<sh Size of reply headers sent to client (after adaptation)
4016#
4017# [http::]<sH Reply high offset sent
4018# [http::]<sS Upstream object size
4019#
4020# [http::]<bs Number of HTTP-equivalent message body bytes
4021# received from the next hop, excluding chunked
4022# transfer encoding and control messages.
4023# Generated FTP/Gopher listings are treated as
4024# received bodies.
4025#
4026# TIMING
4027#
4028# [http::]<pt Peer response time in milliseconds. The timer starts
4029# when the last request byte is sent to the next hop
4030# and stops when the last response byte is received.
4031# [http::]<tt Total time in milliseconds. The timer
4032# starts with the first connect request (or write I/O)
4033# sent to the first selected peer. The timer stops
4034# with the last I/O with the last peer.
4035#
4036# Squid handling related format codes:
4037#
4038# Ss Squid request status (TCP_MISS etc)
4039# Sh Squid hierarchy status (DEFAULT_PARENT etc)
4040#
4041# SSL-related format codes:
4042#
4043# ssl::bump_mode SslBump decision for the transaction:
4044#
4045# For CONNECT requests that initiated bumping of
4046# a connection and for any request received on
4047# an already bumped connection, Squid logs the
4048# corresponding SslBump mode ("splice", "bump",
4049# "peek", "stare", "terminate", "server-first"
4050# or "client-first"). See the ssl_bump option
4051# for more information about these modes.
4052#
4053# A "none" token is logged for requests that
4054# triggered "ssl_bump" ACL evaluation matching
4055# a "none" rule.
4056#
4057# In all other cases, a single dash ("-") is
4058# logged.
4059#
4060# ssl::>sni SSL client SNI sent to Squid.
4061#
4062# ssl::>cert_subject
4063# The Subject field of the received client
4064# SSL certificate or a dash ('-') if Squid has
4065# received an invalid/malformed certificate or
4066# no certificate at all. Consider encoding the
4067# logged value because Subject often has spaces.
4068#
4069# ssl::>cert_issuer
4070# The Issuer field of the received client
4071# SSL certificate or a dash ('-') if Squid has
4072# received an invalid/malformed certificate or
4073# no certificate at all. Consider encoding the
4074# logged value because Issuer often has spaces.
4075#
4076# ssl::<cert_subject
4077# The Subject field of the received server
4078# TLS certificate or a dash ('-') if this is
4079# not available. Consider encoding the logged
4080# value because Subject often has spaces.
4081#
4082# ssl::<cert_issuer
4083# The Issuer field of the received server
4084# TLS certificate or a dash ('-') if this is
4085# not available. Consider encoding the logged
4086# value because Issuer often has spaces.
4087#
4088# ssl::<cert_errors
4089# The list of certificate validation errors
4090# detected by Squid (including OpenSSL and
4091# certificate validation helper components). The
4092# errors are listed in the discovery order. By
4093# default, the error codes are separated by ':'.
4094# Accepts an optional separator argument.
4095#
4096# %ssl::>negotiated_version The negotiated TLS version of the
4097# client connection.
4098#
4099# %ssl::<negotiated_version The negotiated TLS version of the
4100# last server or peer connection.
4101#
4102# %ssl::>received_hello_version The TLS version of the Hello
4103# message received from TLS client.
4104#
4105# %ssl::<received_hello_version The TLS version of the Hello
4106# message received from TLS server.
4107#
4108# %ssl::>received_supported_version The maximum TLS version
4109# supported by the TLS client.
4110#
4111# %ssl::<received_supported_version The maximum TLS version
4112# supported by the TLS server.
4113#
4114# %ssl::>negotiated_cipher The negotiated cipher of the
4115# client connection.
4116#
4117# %ssl::<negotiated_cipher The negotiated cipher of the
4118# last server or peer connection.
4119#
4120# If ICAP is enabled, the following code becomes available (as
4121# well as ICAP log codes documented with the icap_log option):
4122#
4123# icap::tt Total ICAP processing time for the HTTP
4124# transaction. The timer ticks when ICAP
4125# ACLs are checked and when ICAP
4126# transaction is in progress.
4127#
4128# If adaptation is enabled the following codes become available:
4129#
4130# adapt::<last_h The header of the last ICAP response or
4131# meta-information from the last eCAP
4132# transaction related to the HTTP transaction.
4133# Like <h, accepts an optional header name
4134# argument.
4135#
4136# adapt::sum_trs Summed adaptation transaction response
4137# times recorded as a comma-separated list in
4138# the order of transaction start time. Each time
4139# value is recorded as an integer number,
4140# representing response time of one or more
4141# adaptation (ICAP or eCAP) transaction in
4142# milliseconds. When a failed transaction is
4143# being retried or repeated, its time is not
4144# logged individually but added to the
4145# replacement (next) transaction. See also:
4146# adapt::all_trs.
4147#
4148# adapt::all_trs All adaptation transaction response times.
4149# Same as adaptation_strs but response times of
4150# individual transactions are never added
4151# together. Instead, all transaction response
4152# times are recorded individually.
4153#
4154# You can prefix adapt::*_trs format codes with adaptation
4155# service name in curly braces to record response time(s) specific
4156# to that service. For example: %{my_service}adapt::sum_trs
4157#
4158# The default formats available (which do not need re-defining) are:
4159#
4160#logformat squid %ts.%03tu %6tr %>a %Ss/%03>Hs %<st %rm %ru %[un %Sh/%<a %mt
4161#logformat common %>a %[ui %[un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st %Ss:%Sh
4162#logformat combined %>a %[ui %[un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st "%{Referer}>h" "%{User-Agent}>h" %Ss:%Sh
4163#logformat referrer %ts.%03tu %>a %{Referer}>h %ru
4164#logformat useragent %>a [%tl] "%{User-Agent}>h"
4165#
4166# NOTE: When the log_mime_hdrs directive is set to ON.
4167# The squid, common and combined formats have a safely encoded copy
4168# of the mime headers appended to each line within a pair of brackets.
4169#
4170# NOTE: The common and combined formats are not quite true to the Apache definition.
4171# The logs from Squid contain an extra status and hierarchy code appended.
4172#
4173#Default:
4174# The format definitions squid, common, combined, referrer, useragent are built in.
4175
4176# TAG: access_log
4177# Configures whether and how Squid logs HTTP and ICP transactions.
4178# If access logging is enabled, a single line is logged for every
4179# matching HTTP or ICP request. The recommended directive formats are:
4180#
4181# access_log <module>:<place> [option ...] [acl acl ...]
4182# access_log none [acl acl ...]
4183#
4184# The following directive format is accepted but may be deprecated:
4185# access_log <module>:<place> [<logformat name> [acl acl ...]]
4186#
4187# In most cases, the first ACL name must not contain the '=' character
4188# and should not be equal to an existing logformat name. You can always
4189# start with an 'all' ACL to work around those restrictions.
4190#
4191# Will log to the specified module:place using the specified format (which
4192# must be defined in a logformat directive) those entries which match
4193# ALL the acl's specified (which must be defined in acl clauses).
4194# If no acl is specified, all requests will be logged to this destination.
4195#
4196# ===== Available options for the recommended directive format =====
4197#
4198# logformat=name Names log line format (either built-in or
4199# defined by a logformat directive). Defaults
4200# to 'squid'.
4201#
4202# buffer-size=64KB Defines approximate buffering limit for log
4203# records (see buffered_logs). Squid should not
4204# keep more than the specified size and, hence,
4205# should flush records before the buffer becomes
4206# full to avoid overflows under normal
4207# conditions (the exact flushing algorithm is
4208# module-dependent though). The on-error option
4209# controls overflow handling.
4210#
4211# on-error=die|drop Defines action on unrecoverable errors. The
4212# 'drop' action ignores (i.e., does not log)
4213# affected log records. The default 'die' action
4214# kills the affected worker. The drop action
4215# support has not been tested for modules other
4216# than tcp.
4217#
4218# rotate=N Specifies the number of log file rotations to
4219# make when you run 'squid -k rotate'. The default
4220# is to obey the logfile_rotate directive. Setting
4221# rotate=0 will disable the file name rotation,
4222# but the log files are still closed and re-opened.
4223# This will enable you to rename the logfiles
4224# yourself just before sending the rotate signal.
4225# Only supported by the stdio module.
4226#
4227# ===== Modules Currently available =====
4228#
4229# none Do not log any requests matching these ACL.
4230# Do not specify Place or logformat name.
4231#
4232# stdio Write each log line to disk immediately at the completion of
4233# each request.
4234# Place: the filename and path to be written.
4235#
4236# daemon Very similar to stdio. But instead of writing to disk the log
4237# line is passed to a daemon helper for asychronous handling instead.
4238# Place: varies depending on the daemon.
4239#
4240# log_file_daemon Place: the file name and path to be written.
4241#
4242# syslog To log each request via syslog facility.
4243# Place: The syslog facility and priority level for these entries.
4244# Place Format: facility.priority
4245#
4246# where facility could be any of:
4247# authpriv, daemon, local0 ... local7 or user.
4248#
4249# And priority could be any of:
4250# err, warning, notice, info, debug.
4251#
4252# udp To send each log line as text data to a UDP receiver.
4253# Place: The destination host name or IP and port.
4254# Place Format: //host:port
4255#
4256# tcp To send each log line as text data to a TCP receiver.
4257# Lines may be accumulated before sending (see buffered_logs).
4258# Place: The destination host name or IP and port.
4259# Place Format: //host:port
4260#
4261# Default:
4262# access_log daemon:/var/log/squid/access.log squid
4263#Default:
4264# access_log daemon:/var/log/squid/access.log squid
4265
4266# TAG: icap_log
4267# ICAP log files record ICAP transaction summaries, one line per
4268# transaction.
4269#
4270# The icap_log option format is:
4271# icap_log <filepath> [<logformat name> [acl acl ...]]
4272# icap_log none [acl acl ...]]
4273#
4274# Please see access_log option documentation for details. The two
4275# kinds of logs share the overall configuration approach and many
4276# features.
4277#
4278# ICAP processing of a single HTTP message or transaction may
4279# require multiple ICAP transactions. In such cases, multiple
4280# ICAP transaction log lines will correspond to a single access
4281# log line.
4282#
4283# ICAP log supports many access.log logformat %codes. In ICAP context,
4284# HTTP message-related %codes are applied to the HTTP message embedded
4285# in an ICAP message. Logformat "%http::>..." codes are used for HTTP
4286# messages embedded in ICAP requests while "%http::<..." codes are used
4287# for HTTP messages embedded in ICAP responses. For example:
4288#
4289# http::>h To-be-adapted HTTP message headers sent by Squid to
4290# the ICAP service. For REQMOD transactions, these are
4291# HTTP request headers. For RESPMOD, these are HTTP
4292# response headers, but Squid currently cannot log them
4293# (i.e., %http::>h will expand to "-" for RESPMOD).
4294#
4295# http::<h Adapted HTTP message headers sent by the ICAP
4296# service to Squid (i.e., HTTP request headers in regular
4297# REQMOD; HTTP response headers in RESPMOD and during
4298# request satisfaction in REQMOD).
4299#
4300# ICAP OPTIONS transactions do not embed HTTP messages.
4301#
4302# Several logformat codes below deal with ICAP message bodies. An ICAP
4303# message body, if any, typically includes a complete HTTP message
4304# (required HTTP headers plus optional HTTP message body). When
4305# computing HTTP message body size for these logformat codes, Squid
4306# either includes or excludes chunked encoding overheads; see
4307# code-specific documentation for details.
4308#
4309# For Secure ICAP services, all size-related information is currently
4310# computed before/after TLS encryption/decryption, as if TLS was not
4311# in use at all.
4312#
4313# The following format codes are also available for ICAP logs:
4314#
4315# icap::<A ICAP server IP address. Similar to <A.
4316#
4317# icap::<service_name ICAP service name from the icap_service
4318# option in Squid configuration file.
4319#
4320# icap::ru ICAP Request-URI. Similar to ru.
4321#
4322# icap::rm ICAP request method (REQMOD, RESPMOD, or
4323# OPTIONS). Similar to existing rm.
4324#
4325# icap::>st The total size of the ICAP request sent to the ICAP
4326# server (ICAP headers + ICAP body), including chunking
4327# metadata (if any).
4328#
4329# icap::<st The total size of the ICAP response received from the
4330# ICAP server (ICAP headers + ICAP body), including
4331# chunking metadata (if any).
4332#
4333# icap::<bs The size of the ICAP response body received from the
4334# ICAP server, excluding chunking metadata (if any).
4335#
4336# icap::tr Transaction response time (in
4337# milliseconds). The timer starts when
4338# the ICAP transaction is created and
4339# stops when the transaction is completed.
4340# Similar to tr.
4341#
4342# icap::tio Transaction I/O time (in milliseconds). The
4343# timer starts when the first ICAP request
4344# byte is scheduled for sending. The timers
4345# stops when the last byte of the ICAP response
4346# is received.
4347#
4348# icap::to Transaction outcome: ICAP_ERR* for all
4349# transaction errors, ICAP_OPT for OPTION
4350# transactions, ICAP_ECHO for 204
4351# responses, ICAP_MOD for message
4352# modification, and ICAP_SAT for request
4353# satisfaction. Similar to Ss.
4354#
4355# icap::Hs ICAP response status code. Similar to Hs.
4356#
4357# icap::>h ICAP request header(s). Similar to >h.
4358#
4359# icap::<h ICAP response header(s). Similar to <h.
4360#
4361# The default ICAP log format, which can be used without an explicit
4362# definition, is called icap_squid:
4363#
4364#logformat icap_squid %ts.%03tu %6icap::tr %>A %icap::to/%03icap::Hs %icap::<st %icap::rm %icap::ru %un -/%icap::<A -
4365#
4366# See also: logformat and %adapt::<last_h
4367#Default:
4368# none
4369
4370# TAG: logfile_daemon
4371# Specify the path to the logfile-writing daemon. This daemon is
4372# used to write the access and store logs, if configured.
4373#
4374# Squid sends a number of commands to the log daemon:
4375# L<data>\n - logfile data
4376# R\n - rotate file
4377# T\n - truncate file
4378# O\n - reopen file
4379# F\n - flush file
4380# r<n>\n - set rotate count to <n>
4381# b<n>\n - 1 = buffer output, 0 = don't buffer output
4382#
4383# No responses is expected.
4384#Default:
4385# logfile_daemon /usr/lib/squid/log_file_daemon
4386
4387# TAG: stats_collection allow|deny acl acl...
4388# This options allows you to control which requests gets accounted
4389# in performance counters.
4390#
4391# This clause only supports fast acl types.
4392# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4393#Default:
4394# Allow logging for all transactions.
4395
4396# TAG: cache_store_log
4397# Logs the activities of the storage manager. Shows which
4398# objects are ejected from the cache, and which objects are
4399# saved and for how long.
4400# There are not really utilities to analyze this data, so you can safely
4401# disable it (the default).
4402#
4403# Store log uses modular logging outputs. See access_log for the list
4404# of modules supported.
4405#
4406# Example:
4407# cache_store_log stdio:/var/log/squid/store.log
4408# cache_store_log daemon:/var/log/squid/store.log
4409#Default:
4410# none
4411
4412# TAG: cache_swap_state
4413# Location for the cache "swap.state" file. This index file holds
4414# the metadata of objects saved on disk. It is used to rebuild
4415# the cache during startup. Normally this file resides in each
4416# 'cache_dir' directory, but you may specify an alternate
4417# pathname here. Note you must give a full filename, not just
4418# a directory. Since this is the index for the whole object
4419# list you CANNOT periodically rotate it!
4420#
4421# If %s can be used in the file name it will be replaced with a
4422# a representation of the cache_dir name where each / is replaced
4423# with '.'. This is needed to allow adding/removing cache_dir
4424# lines when cache_swap_log is being used.
4425#
4426# If have more than one 'cache_dir', and %s is not used in the name
4427# these swap logs will have names such as:
4428#
4429# cache_swap_log.00
4430# cache_swap_log.01
4431# cache_swap_log.02
4432#
4433# The numbered extension (which is added automatically)
4434# corresponds to the order of the 'cache_dir' lines in this
4435# configuration file. If you change the order of the 'cache_dir'
4436# lines in this file, these index files will NOT correspond to
4437# the correct 'cache_dir' entry (unless you manually rename
4438# them). We recommend you do NOT use this option. It is
4439# better to keep these index files in each 'cache_dir' directory.
4440#Default:
4441# Store the journal inside its cache_dir
4442
4443# TAG: logfile_rotate
4444# Specifies the default number of logfile rotations to make when you
4445# type 'squid -k rotate'. The default is 10, which will rotate
4446# with extensions 0 through 9. Setting logfile_rotate to 0 will
4447# disable the file name rotation, but the logfiles are still closed
4448# and re-opened. This will enable you to rename the logfiles
4449# yourself just before sending the rotate signal.
4450#
4451# Note, from Squid-3.1 this option is only a default for cache.log,
4452# that log can be rotated separately by using debug_options.
4453#
4454# Note, from Squid-4 this option is only a default for access.log
4455# recorded by stdio: module. Those logs can be rotated separately by
4456# using the rotate=N option on their access_log directive.
4457#
4458# Note, the 'squid -k rotate' command normally sends a USR1
4459# signal to the running squid process. In certain situations
4460# (e.g. on Linux with Async I/O), USR1 is used for other
4461# purposes, so -k rotate uses another signal. It is best to get
4462# in the habit of using 'squid -k rotate' instead of 'kill -USR1
4463# <pid>'.
4464#
4465# Note, for Debian/Linux the default of logfile_rotate is
4466# zero, since it includes external logfile-rotation methods.
4467#Default:
4468# logfile_rotate 0
4469
4470# TAG: mime_table
4471# Path to Squid's icon configuration file.
4472#
4473# You shouldn't need to change this, but the default file contains
4474# examples and formatting information if you do.
4475#Default:
4476# mime_table /usr/share/squid/mime.conf
4477
4478# TAG: log_mime_hdrs on|off
4479# The Cache can record both the request and the response MIME
4480# headers for each HTTP transaction. The headers are encoded
4481# safely and will appear as two bracketed fields at the end of
4482# the access log (for either the native or httpd-emulated log
4483# formats). To enable this logging set log_mime_hdrs to 'on'.
4484#Default:
4485# log_mime_hdrs off
4486
4487# TAG: pid_filename
4488# A filename to write the process-id to. To disable, enter "none".
4489#Default:
4490# pid_filename /var/run/squid.pid
4491
4492# TAG: client_netmask
4493# A netmask for client addresses in logfiles and cachemgr output.
4494# Change this to protect the privacy of your cache clients.
4495# A netmask of 255.255.255.0 will log all IP's in that range with
4496# the last digit set to '0'.
4497#Default:
4498# Log full client IP address
4499
4500# TAG: strip_query_terms
4501# By default, Squid strips query terms from requested URLs before
4502# logging. This protects your user's privacy and reduces log size.
4503#
4504# When investigating HIT/MISS or other caching behaviour you
4505# will need to disable this to see the full URL used by Squid.
4506#Default:
4507# strip_query_terms on
4508
4509# TAG: buffered_logs on|off
4510# Whether to write/send access_log records ASAP or accumulate them and
4511# then write/send them in larger chunks. Buffering may improve
4512# performance because it decreases the number of I/Os. However,
4513# buffering increases the delay before log records become available to
4514# the final recipient (e.g., a disk file or logging daemon) and,
4515# hence, increases the risk of log records loss.
4516#
4517# Note that even when buffered_logs are off, Squid may have to buffer
4518# records if it cannot write/send them immediately due to pending I/Os
4519# (e.g., the I/O writing the previous log record) or connectivity loss.
4520#
4521# Currently honored by 'daemon' and 'tcp' access_log modules only.
4522#Default:
4523# buffered_logs off
4524
4525# TAG: netdb_filename
4526# Where Squid stores it's netdb journal.
4527# When enabled this journal preserves netdb state between restarts.
4528#
4529# To disable, enter "none".
4530#Default:
4531# netdb_filename stdio:/var/spool/squid/netdb.state
4532
4533# OPTIONS FOR TROUBLESHOOTING
4534# -----------------------------------------------------------------------------
4535
4536# TAG: cache_log
4537# Squid administrative logging file.
4538#
4539# This is where general information about Squid behavior goes. You can
4540# increase the amount of data logged to this file and how often it is
4541# rotated with "debug_options"
4542#Default:
4543# cache_log /var/log/squid/cache.log
4544
4545# TAG: debug_options
4546# Logging options are set as section,level where each source file
4547# is assigned a unique section. Lower levels result in less
4548# output, Full debugging (level 9) can result in a very large
4549# log file, so be careful.
4550#
4551# The magic word "ALL" sets debugging levels for all sections.
4552# The default is to run with "ALL,1" to record important warnings.
4553#
4554# The rotate=N option can be used to keep more or less of these logs
4555# than would otherwise be kept by logfile_rotate.
4556# For most uses a single log should be enough to monitor current
4557# events affecting Squid.
4558#Default:
4559# Log all critical and important messages.
4560
4561# TAG: coredump_dir
4562# By default Squid leaves core files in the directory from where
4563# it was started. If you set 'coredump_dir' to a directory
4564# that exists, Squid will chdir() to that directory at startup
4565# and coredump files will be left there.
4566#
4567#Default:
4568# Use the directory from where Squid was started.
4569#
4570
4571# Leave coredumps in the first cache dir
4572coredump_dir /var/spool/squid
4573
4574# OPTIONS FOR FTP GATEWAYING
4575# -----------------------------------------------------------------------------
4576
4577# TAG: ftp_user
4578# If you want the anonymous login password to be more informative
4579# (and enable the use of picky FTP servers), set this to something
4580# reasonable for your domain, like wwwuser@somewhere.net
4581#
4582# The reason why this is domainless by default is the
4583# request can be made on the behalf of a user in any domain,
4584# depending on how the cache is used.
4585# Some FTP server also validate the email address is valid
4586# (for example perl.com).
4587#Default:
4588# ftp_user Squid@
4589
4590# TAG: ftp_passive
4591# If your firewall does not allow Squid to use passive
4592# connections, turn off this option.
4593#
4594# Use of ftp_epsv_all option requires this to be ON.
4595#Default:
4596# ftp_passive on
4597
4598# TAG: ftp_epsv_all
4599# FTP Protocol extensions permit the use of a special "EPSV ALL" command.
4600#
4601# NATs may be able to put the connection on a "fast path" through the
4602# translator, as the EPRT command will never be used and therefore,
4603# translation of the data portion of the segments will never be needed.
4604#
4605# When a client only expects to do two-way FTP transfers this may be
4606# useful.
4607# If squid finds that it must do a three-way FTP transfer after issuing
4608# an EPSV ALL command, the FTP session will fail.
4609#
4610# If you have any doubts about this option do not use it.
4611# Squid will nicely attempt all other connection methods.
4612#
4613# Requires ftp_passive to be ON (default) for any effect.
4614#Default:
4615# ftp_epsv_all off
4616
4617# TAG: ftp_epsv
4618# FTP Protocol extensions permit the use of a special "EPSV" command.
4619#
4620# NATs may be able to put the connection on a "fast path" through the
4621# translator using EPSV, as the EPRT command will never be used
4622# and therefore, translation of the data portion of the segments
4623# will never be needed.
4624#
4625# EPSV is often required to interoperate with FTP servers on IPv6
4626# networks. On the other hand, it may break some IPv4 servers.
4627#
4628# By default, EPSV may try EPSV with any FTP server. To fine tune
4629# that decision, you may restrict EPSV to certain clients or servers
4630# using ACLs:
4631#
4632# ftp_epsv allow|deny al1 acl2 ...
4633#
4634# WARNING: Disabling EPSV may cause problems with external NAT and IPv6.
4635#
4636# Only fast ACLs are supported.
4637# Requires ftp_passive to be ON (default) for any effect.
4638#Default:
4639# none
4640
4641# TAG: ftp_eprt
4642# FTP Protocol extensions permit the use of a special "EPRT" command.
4643#
4644# This extension provides a protocol neutral alternative to the
4645# IPv4-only PORT command. When supported it enables active FTP data
4646# channels over IPv6 and efficient NAT handling.
4647#
4648# Turning this OFF will prevent EPRT being attempted and will skip
4649# straight to using PORT for IPv4 servers.
4650#
4651# Some devices are known to not handle this extension correctly and
4652# may result in crashes. Devices which suport EPRT enough to fail
4653# cleanly will result in Squid attempting PORT anyway. This directive
4654# should only be disabled when EPRT results in device failures.
4655#
4656# WARNING: Doing so will convert Squid back to the old behavior with all
4657# the related problems with external NAT devices/layers and IPv4-only FTP.
4658#Default:
4659# ftp_eprt on
4660
4661# TAG: ftp_sanitycheck
4662# For security and data integrity reasons Squid by default performs
4663# sanity checks of the addresses of FTP data connections ensure the
4664# data connection is to the requested server. If you need to allow
4665# FTP connections to servers using another IP address for the data
4666# connection turn this off.
4667#Default:
4668# ftp_sanitycheck on
4669
4670# TAG: ftp_telnet_protocol
4671# The FTP protocol is officially defined to use the telnet protocol
4672# as transport channel for the control connection. However, many
4673# implementations are broken and does not respect this aspect of
4674# the FTP protocol.
4675#
4676# If you have trouble accessing files with ASCII code 255 in the
4677# path or similar problems involving this ASCII code you can
4678# try setting this directive to off. If that helps, report to the
4679# operator of the FTP server in question that their FTP server
4680# is broken and does not follow the FTP standard.
4681#Default:
4682# ftp_telnet_protocol on
4683
4684# OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
4685# -----------------------------------------------------------------------------
4686
4687# TAG: diskd_program
4688# Specify the location of the diskd executable.
4689# Note this is only useful if you have compiled in
4690# diskd as one of the store io modules.
4691#Default:
4692# diskd_program /usr/lib/squid/diskd
4693
4694# TAG: unlinkd_program
4695# Specify the location of the executable for file deletion process.
4696#Default:
4697# unlinkd_program /usr/lib/squid/unlinkd
4698
4699# TAG: pinger_program
4700# Specify the location of the executable for the pinger process.
4701#Default:
4702# pinger_program /usr/lib/squid/pinger
4703
4704# TAG: pinger_enable
4705# Control whether the pinger is active at run-time.
4706# Enables turning ICMP pinger on and off with a simple
4707# squid -k reconfigure.
4708#Default:
4709# pinger_enable on
4710
4711# OPTIONS FOR URL REWRITING
4712# -----------------------------------------------------------------------------
4713
4714# TAG: url_rewrite_program
4715# Specify the location of the executable URL rewriter to use.
4716# Since they can perform almost any function there isn't one included.
4717#
4718# For each requested URL, the rewriter will receive on line with the format
4719#
4720# [channel-ID <SP>] URL [<SP> extras]<NL>
4721#
4722# See url_rewrite_extras on how to send "extras" with optional values to
4723# the helper.
4724# After processing the request the helper must reply using the following format:
4725#
4726# [channel-ID <SP>] result [<SP> kv-pairs]
4727#
4728# The result code can be:
4729#
4730# OK status=30N url="..."
4731# Redirect the URL to the one supplied in 'url='.
4732# 'status=' is optional and contains the status code to send
4733# the client in Squids HTTP response. It must be one of the
4734# HTTP redirect status codes: 301, 302, 303, 307, 308.
4735# When no status is given Squid will use 302.
4736#
4737# OK rewrite-url="..."
4738# Rewrite the URL to the one supplied in 'rewrite-url='.
4739# The new URL is fetched directly by Squid and returned to
4740# the client as the response to its request.
4741#
4742# OK
4743# When neither of url= and rewrite-url= are sent Squid does
4744# not change the URL.
4745#
4746# ERR
4747# Do not change the URL.
4748#
4749# BH
4750# An internal error occurred in the helper, preventing
4751# a result being identified. The 'message=' key name is
4752# reserved for delivering a log message.
4753#
4754#
4755# In addition to the above kv-pairs Squid also understands the following
4756# optional kv-pairs received from URL rewriters:
4757# clt_conn_tag=TAG
4758# Associates a TAG with the client TCP connection.
4759# The TAG is treated as a regular annotation but persists across
4760# future requests on the client connection rather than just the
4761# current request. A helper may update the TAG during subsequent
4762# requests be returning a new kv-pair.
4763#
4764# When using the concurrency= option the protocol is changed by
4765# introducing a query channel tag in front of the request/response.
4766# The query channel tag is a number between 0 and concurrency-1.
4767# This value must be echoed back unchanged to Squid as the first part
4768# of the response relating to its request.
4769#
4770# WARNING: URL re-writing ability should be avoided whenever possible.
4771# Use the URL redirect form of response instead.
4772#
4773# Re-write creates a difference in the state held by the client
4774# and server. Possibly causing confusion when the server response
4775# contains snippets of its view state. Embeded URLs, response
4776# and content Location headers, etc. are not re-written by this
4777# interface.
4778#
4779# By default, a URL rewriter is not used.
4780#Default:
4781# none
4782
4783# TAG: url_rewrite_children
4784# Specifies the maximum number of redirector processes that Squid may
4785# spawn (numberofchildren) and several related options. Using too few of
4786# these helper processes (a.k.a. "helpers") creates request queues.
4787# Using too many helpers wastes your system resources.
4788#
4789# Usage: numberofchildren [option]...
4790#
4791# The startup= and idle= options allow some measure of skew in your
4792# tuning.
4793#
4794# startup=
4795#
4796# Sets a minimum of how many processes are to be spawned when Squid
4797# starts or reconfigures. When set to zero the first request will
4798# cause spawning of the first child process to handle it.
4799#
4800# Starting too few will cause an initial slowdown in traffic as Squid
4801# attempts to simultaneously spawn enough processes to cope.
4802#
4803# idle=
4804#
4805# Sets a minimum of how many processes Squid is to try and keep available
4806# at all times. When traffic begins to rise above what the existing
4807# processes can handle this many more will be spawned up to the maximum
4808# configured. A minimum setting of 1 is required.
4809#
4810# concurrency=
4811#
4812# The number of requests each redirector helper can handle in
4813# parallel. Defaults to 0 which indicates the redirector
4814# is a old-style single threaded redirector.
4815#
4816# When this directive is set to a value >= 1 then the protocol
4817# used to communicate with the helper is modified to include
4818# an ID in front of the request/response. The ID from the request
4819# must be echoed back with the response to that request.
4820#
4821# queue-size=N
4822#
4823# Sets the maximum number of queued requests. A request is queued when
4824# no existing child can accept it due to concurrency limit and no new
4825# child can be started due to numberofchildren limit. The default
4826# maximum is zero if url_rewrite_bypass is enabled and
4827# 2*numberofchildren otherwise. If the queued requests exceed queue size
4828# and redirector_bypass configuration option is set, then redirector is
4829# bypassed. Otherwise, Squid is allowed to temporarily exceed the
4830# configured maximum, marking the affected helper as "overloaded". If
4831# the helper overload lasts more than 3 minutes, the action prescribed
4832# by the on-persistent-overload option applies.
4833#
4834# on-persistent-overload=action
4835#
4836# Specifies Squid reaction to a new helper request arriving when the helper
4837# has been overloaded for more that 3 minutes already. The number of queued
4838# requests determines whether the helper is overloaded (see the queue-size
4839# option).
4840#
4841# Two actions are supported:
4842#
4843# die Squid worker quits. This is the default behavior.
4844#
4845# ERR Squid treats the helper request as if it was
4846# immediately submitted, and the helper immediately
4847# replied with an ERR response. This action has no effect
4848# on the already queued and in-progress helper requests.
4849#Default:
4850# url_rewrite_children 20 startup=0 idle=1 concurrency=0
4851
4852# TAG: url_rewrite_host_header
4853# To preserve same-origin security policies in browsers and
4854# prevent Host: header forgery by redirectors Squid rewrites
4855# any Host: header in redirected requests.
4856#
4857# If you are running an accelerator this may not be a wanted
4858# effect of a redirector. This directive enables you disable
4859# Host: alteration in reverse-proxy traffic.
4860#
4861# WARNING: Entries are cached on the result of the URL rewriting
4862# process, so be careful if you have domain-virtual hosts.
4863#
4864# WARNING: Squid and other software verifies the URL and Host
4865# are matching, so be careful not to relay through other proxies
4866# or inspecting firewalls with this disabled.
4867#Default:
4868# url_rewrite_host_header on
4869
4870# TAG: url_rewrite_access
4871# If defined, this access list specifies which requests are
4872# sent to the redirector processes.
4873#
4874# This clause supports both fast and slow acl types.
4875# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4876#Default:
4877# Allow, unless rules exist in squid.conf.
4878
4879# TAG: url_rewrite_bypass
4880# When this is 'on', a request will not go through the
4881# redirector if all the helpers are busy. If this is 'off' and the
4882# redirector queue grows too large, the action is prescribed by the
4883# on-persistent-overload option. You should only enable this if the
4884# redirectors are not critical to your caching system. If you use
4885# redirectors for access control, and you enable this option,
4886# users may have access to pages they should not
4887# be allowed to request.
4888#
4889# Enabling this option sets the default url_rewrite_children queue-size
4890# option value to 0.
4891#Default:
4892# url_rewrite_bypass off
4893
4894# TAG: url_rewrite_extras
4895# Specifies a string to be append to request line format for the
4896# rewriter helper. "Quoted" format values may contain spaces and
4897# logformat %macros. In theory, any logformat %macro can be used.
4898# In practice, a %macro expands as a dash (-) if the helper request is
4899# sent before the required macro information is available to Squid.
4900#Default:
4901# url_rewrite_extras "%>a/%>A %un %>rm myip=%la myport=%lp"
4902
4903# TAG: url_rewrite_timeout
4904# Squid times active requests to redirector. The timeout value and Squid
4905# reaction to a timed out request are configurable using the following
4906# format:
4907#
4908# url_rewrite_timeout timeout time-units on_timeout=<action> [response=<quoted-response>]
4909#
4910# supported timeout actions:
4911# fail Squid return a ERR_GATEWAY_FAILURE error page
4912#
4913# bypass Do not re-write the URL
4914#
4915# retry Send the lookup to the helper again
4916#
4917# use_configured_response
4918# Use the <quoted-response> as helper response
4919#Default:
4920# Squid waits for the helper response forever
4921
4922# OPTIONS FOR STORE ID
4923# -----------------------------------------------------------------------------
4924
4925# TAG: store_id_program
4926# Specify the location of the executable StoreID helper to use.
4927# Since they can perform almost any function there isn't one included.
4928#
4929# For each requested URL, the helper will receive one line with the format
4930#
4931# [channel-ID <SP>] URL [<SP> extras]<NL>
4932#
4933#
4934# After processing the request the helper must reply using the following format:
4935#
4936# [channel-ID <SP>] result [<SP> kv-pairs]
4937#
4938# The result code can be:
4939#
4940# OK store-id="..."
4941# Use the StoreID supplied in 'store-id='.
4942#
4943# ERR
4944# The default is to use HTTP request URL as the store ID.
4945#
4946# BH
4947# An internal error occurred in the helper, preventing
4948# a result being identified.
4949#
4950# In addition to the above kv-pairs Squid also understands the following
4951# optional kv-pairs received from URL rewriters:
4952# clt_conn_tag=TAG
4953# Associates a TAG with the client TCP connection.
4954# Please see url_rewrite_program related documentation for this
4955# kv-pair
4956#
4957# Helper programs should be prepared to receive and possibly ignore
4958# additional whitespace-separated tokens on each input line.
4959#
4960# When using the concurrency= option the protocol is changed by
4961# introducing a query channel tag in front of the request/response.
4962# The query channel tag is a number between 0 and concurrency-1.
4963# This value must be echoed back unchanged to Squid as the first part
4964# of the response relating to its request.
4965#
4966# NOTE: when using StoreID refresh_pattern will apply to the StoreID
4967# returned from the helper and not the URL.
4968#
4969# WARNING: Wrong StoreID value returned by a careless helper may result
4970# in the wrong cached response returned to the user.
4971#
4972# By default, a StoreID helper is not used.
4973#Default:
4974# none
4975
4976# TAG: store_id_extras
4977# Specifies a string to be append to request line format for the
4978# StoreId helper. "Quoted" format values may contain spaces and
4979# logformat %macros. In theory, any logformat %macro can be used.
4980# In practice, a %macro expands as a dash (-) if the helper request is
4981# sent before the required macro information is available to Squid.
4982#Default:
4983# store_id_extras "%>a/%>A %un %>rm myip=%la myport=%lp"
4984
4985# TAG: store_id_children
4986# Specifies the maximum number of StoreID helper processes that Squid
4987# may spawn (numberofchildren) and several related options. Using
4988# too few of these helper processes (a.k.a. "helpers") creates request
4989# queues. Using too many helpers wastes your system resources.
4990#
4991# Usage: numberofchildren [option]...
4992#
4993# The startup= and idle= options allow some measure of skew in your
4994# tuning.
4995#
4996# startup=
4997#
4998# Sets a minimum of how many processes are to be spawned when Squid
4999# starts or reconfigures. When set to zero the first request will
5000# cause spawning of the first child process to handle it.
5001#
5002# Starting too few will cause an initial slowdown in traffic as Squid
5003# attempts to simultaneously spawn enough processes to cope.
5004#
5005# idle=
5006#
5007# Sets a minimum of how many processes Squid is to try and keep available
5008# at all times. When traffic begins to rise above what the existing
5009# processes can handle this many more will be spawned up to the maximum
5010# configured. A minimum setting of 1 is required.
5011#
5012# concurrency=
5013#
5014# The number of requests each storeID helper can handle in
5015# parallel. Defaults to 0 which indicates the helper
5016# is a old-style single threaded program.
5017#
5018# When this directive is set to a value >= 1 then the protocol
5019# used to communicate with the helper is modified to include
5020# an ID in front of the request/response. The ID from the request
5021# must be echoed back with the response to that request.
5022#
5023# queue-size=N
5024#
5025# Sets the maximum number of queued requests to N. A request is queued
5026# when no existing child can accept it due to concurrency limit and no
5027# new child can be started due to numberofchildren limit. The default
5028# maximum is 2*numberofchildren. If the queued requests exceed queue
5029# size and redirector_bypass configuration option is set, then
5030# redirector is bypassed. Otherwise, Squid is allowed to temporarily
5031# exceed the configured maximum, marking the affected helper as
5032# "overloaded". If the helper overload lasts more than 3 minutes, the
5033# action prescribed by the on-persistent-overload option applies.
5034#
5035# on-persistent-overload=action
5036#
5037# Specifies Squid reaction to a new helper request arriving when the helper
5038# has been overloaded for more that 3 minutes already. The number of queued
5039# requests determines whether the helper is overloaded (see the queue-size
5040# option).
5041#
5042# Two actions are supported:
5043#
5044# die Squid worker quits. This is the default behavior.
5045#
5046# ERR Squid treats the helper request as if it was
5047# immediately submitted, and the helper immediately
5048# replied with an ERR response. This action has no effect
5049# on the already queued and in-progress helper requests.
5050#Default:
5051# store_id_children 20 startup=0 idle=1 concurrency=0
5052
5053# TAG: store_id_access
5054# If defined, this access list specifies which requests are
5055# sent to the StoreID processes. By default all requests
5056# are sent.
5057#
5058# This clause supports both fast and slow acl types.
5059# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5060#Default:
5061# Allow, unless rules exist in squid.conf.
5062
5063# TAG: store_id_bypass
5064# When this is 'on', a request will not go through the
5065# helper if all helpers are busy. If this is 'off' and the helper
5066# queue grows too large, the action is prescribed by the
5067# on-persistent-overload option. You should only enable this if the
5068# helpers are not critical to your caching system. If you use
5069# helpers for critical caching components, and you enable this
5070# option, users may not get objects from cache.
5071# This options sets default queue-size option of the store_id_children
5072# to 0.
5073#Default:
5074# store_id_bypass on
5075
5076# OPTIONS FOR TUNING THE CACHE
5077# -----------------------------------------------------------------------------
5078
5079# TAG: cache
5080# Requests denied by this directive will not be served from the cache
5081# and their responses will not be stored in the cache. This directive
5082# has no effect on other transactions and on already cached responses.
5083#
5084# This clause supports both fast and slow acl types.
5085# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5086#
5087# This and the two other similar caching directives listed below are
5088# checked at different transaction processing stages, have different
5089# access to response information, affect different cache operations,
5090# and differ in slow ACLs support:
5091#
5092# * cache: Checked before Squid makes a hit/miss determination.
5093# No access to reply information!
5094# Denies both serving a hit and storing a miss.
5095# Supports both fast and slow ACLs.
5096# * send_hit: Checked after a hit was detected.
5097# Has access to reply (hit) information.
5098# Denies serving a hit only.
5099# Supports fast ACLs only.
5100# * store_miss: Checked before storing a cachable miss.
5101# Has access to reply (miss) information.
5102# Denies storing a miss only.
5103# Supports fast ACLs only.
5104#
5105# If you are not sure which of the three directives to use, apply the
5106# following decision logic:
5107#
5108# * If your ACL(s) are of slow type _and_ need response info, redesign.
5109# Squid does not support that particular combination at this time.
5110# Otherwise:
5111# * If your directive ACL(s) are of slow type, use "cache"; and/or
5112# * if your directive ACL(s) need no response info, use "cache".
5113# Otherwise:
5114# * If you do not want the response cached, use store_miss; and/or
5115# * if you do not want a hit on a cached response, use send_hit.
5116#Default:
5117# By default, this directive is unused and has no effect.
5118
5119# TAG: send_hit
5120# Responses denied by this directive will not be served from the cache
5121# (but may still be cached, see store_miss). This directive has no
5122# effect on the responses it allows and on the cached objects.
5123#
5124# Please see the "cache" directive for a summary of differences among
5125# store_miss, send_hit, and cache directives.
5126#
5127# Unlike the "cache" directive, send_hit only supports fast acl
5128# types. See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5129#
5130# For example:
5131#
5132# # apply custom Store ID mapping to some URLs
5133# acl MapMe dstdomain .c.example.com
5134# store_id_program ...
5135# store_id_access allow MapMe
5136#
5137# # but prevent caching of special responses
5138# # such as 302 redirects that cause StoreID loops
5139# acl Ordinary http_status 200-299
5140# store_miss deny MapMe !Ordinary
5141#
5142# # and do not serve any previously stored special responses
5143# # from the cache (in case they were already cached before
5144# # the above store_miss rule was in effect).
5145# send_hit deny MapMe !Ordinary
5146#Default:
5147# By default, this directive is unused and has no effect.
5148
5149# TAG: store_miss
5150# Responses denied by this directive will not be cached (but may still
5151# be served from the cache, see send_hit). This directive has no
5152# effect on the responses it allows and on the already cached responses.
5153#
5154# Please see the "cache" directive for a summary of differences among
5155# store_miss, send_hit, and cache directives. See the
5156# send_hit directive for a usage example.
5157#
5158# Unlike the "cache" directive, store_miss only supports fast acl
5159# types. See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5160#Default:
5161# By default, this directive is unused and has no effect.
5162
5163# TAG: max_stale time-units
5164# This option puts an upper limit on how stale content Squid
5165# will serve from the cache if cache validation fails.
5166# Can be overriden by the refresh_pattern max-stale option.
5167#Default:
5168# max_stale 1 week
5169
5170# TAG: refresh_pattern
5171# usage: refresh_pattern [-i] regex min percent max [options]
5172#
5173# By default, regular expressions are CASE-SENSITIVE. To make
5174# them case-insensitive, use the -i option.
5175#
5176# 'Min' is the time (in minutes) an object without an explicit
5177# expiry time should be considered fresh. The recommended
5178# value is 0, any higher values may cause dynamic applications
5179# to be erroneously cached unless the application designer
5180# has taken the appropriate actions.
5181#
5182# 'Percent' is a percentage of the objects age (time since last
5183# modification age) an object without explicit expiry time
5184# will be considered fresh.
5185#
5186# 'Max' is an upper limit on how long objects without an explicit
5187# expiry time will be considered fresh. The value is also used
5188# to form Cache-Control: max-age header for a request sent from
5189# Squid to origin/parent.
5190#
5191# options: override-expire
5192# override-lastmod
5193# reload-into-ims
5194# ignore-reload
5195# ignore-no-store
5196# ignore-private
5197# max-stale=NN
5198# refresh-ims
5199# store-stale
5200#
5201# override-expire enforces min age even if the server
5202# sent an explicit expiry time (e.g., with the
5203# Expires: header or Cache-Control: max-age). Doing this
5204# VIOLATES the HTTP standard. Enabling this feature
5205# could make you liable for problems which it causes.
5206#
5207# Note: override-expire does not enforce staleness - it only extends
5208# freshness / min. If the server returns a Expires time which
5209# is longer than your max time, Squid will still consider
5210# the object fresh for that period of time.
5211#
5212# override-lastmod enforces min age even on objects
5213# that were modified recently.
5214#
5215# reload-into-ims changes a client no-cache or ``reload''
5216# request for a cached entry into a conditional request using
5217# If-Modified-Since and/or If-None-Match headers, provided the
5218# cached entry has a Last-Modified and/or a strong ETag header.
5219# Doing this VIOLATES the HTTP standard. Enabling this feature
5220# could make you liable for problems which it causes.
5221#
5222# ignore-reload ignores a client no-cache or ``reload''
5223# header. Doing this VIOLATES the HTTP standard. Enabling
5224# this feature could make you liable for problems which
5225# it causes.
5226#
5227# ignore-no-store ignores any ``Cache-control: no-store''
5228# headers received from a server. Doing this VIOLATES
5229# the HTTP standard. Enabling this feature could make you
5230# liable for problems which it causes.
5231#
5232# ignore-private ignores any ``Cache-control: private''
5233# headers received from a server. Doing this VIOLATES
5234# the HTTP standard. Enabling this feature could make you
5235# liable for problems which it causes.
5236#
5237# refresh-ims causes squid to contact the origin server
5238# when a client issues an If-Modified-Since request. This
5239# ensures that the client will receive an updated version
5240# if one is available.
5241#
5242# store-stale stores responses even if they don't have explicit
5243# freshness or a validator (i.e., Last-Modified or an ETag)
5244# present, or if they're already stale. By default, Squid will
5245# not cache such responses because they usually can't be
5246# reused. Note that such responses will be stale by default.
5247#
5248# max-stale=NN provide a maximum staleness factor. Squid won't
5249# serve objects more stale than this even if it failed to
5250# validate the object. Default: use the max_stale global limit.
5251#
5252# Basically a cached object is:
5253#
5254# FRESH if expire > now, else STALE
5255# STALE if age > max
5256# FRESH if lm-factor < percent, else STALE
5257# FRESH if age < min
5258# else STALE
5259#
5260# The refresh_pattern lines are checked in the order listed here.
5261# The first entry which matches is used. If none of the entries
5262# match the default will be used.
5263#
5264# Note, you must uncomment all the default lines if you want
5265# to change one. The default setting is only active if none is
5266# used.
5267#
5268#
5269
5270#
5271# Add any of your own refresh_pattern entries above these.
5272#
5273refresh_pattern ^ftp: 1440 20% 10080
5274refresh_pattern ^gopher: 1440 0% 1440
5275refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
5276refresh_pattern . 0 20% 4320
5277
5278# TAG: quick_abort_min (KB)
5279#Default:
5280# quick_abort_min 16 KB
5281
5282# TAG: quick_abort_max (KB)
5283#Default:
5284# quick_abort_max 16 KB
5285
5286# TAG: quick_abort_pct (percent)
5287# The cache by default continues downloading aborted requests
5288# which are almost completed (less than 16 KB remaining). This
5289# may be undesirable on slow (e.g. SLIP) links and/or very busy
5290# caches. Impatient users may tie up file descriptors and
5291# bandwidth by repeatedly requesting and immediately aborting
5292# downloads.
5293#
5294# When the user aborts a request, Squid will check the
5295# quick_abort values to the amount of data transferred until
5296# then.
5297#
5298# If the transfer has less than 'quick_abort_min' KB remaining,
5299# it will finish the retrieval.
5300#
5301# If the transfer has more than 'quick_abort_max' KB remaining,
5302# it will abort the retrieval.
5303#
5304# If more than 'quick_abort_pct' of the transfer has completed,
5305# it will finish the retrieval.
5306#
5307# If you do not want any retrieval to continue after the client
5308# has aborted, set both 'quick_abort_min' and 'quick_abort_max'
5309# to '0 KB'.
5310#
5311# If you want retrievals to always continue if they are being
5312# cached set 'quick_abort_min' to '-1 KB'.
5313#Default:
5314# quick_abort_pct 95
5315
5316# TAG: read_ahead_gap buffer-size
5317# The amount of data the cache will buffer ahead of what has been
5318# sent to the client when retrieving an object from another server.
5319#Default:
5320# read_ahead_gap 16 KB
5321
5322# TAG: negative_ttl time-units
5323# Set the Default Time-to-Live (TTL) for failed requests.
5324# Certain types of failures (such as "connection refused" and
5325# "404 Not Found") are able to be negatively-cached for a short time.
5326# Modern web servers should provide Expires: header, however if they
5327# do not this can provide a minimum TTL.
5328# The default is not to cache errors with unknown expiry details.
5329#
5330# Note that this is different from negative caching of DNS lookups.
5331#
5332# WARNING: Doing this VIOLATES the HTTP standard. Enabling
5333# this feature could make you liable for problems which it
5334# causes.
5335#Default:
5336# negative_ttl 0 seconds
5337
5338# TAG: positive_dns_ttl time-units
5339# Upper limit on how long Squid will cache positive DNS responses.
5340# Default is 6 hours (360 minutes). This directive must be set
5341# larger than negative_dns_ttl.
5342#Default:
5343# positive_dns_ttl 6 hours
5344
5345# TAG: negative_dns_ttl time-units
5346# Time-to-Live (TTL) for negative caching of failed DNS lookups.
5347# This also sets the lower cache limit on positive lookups.
5348# Minimum value is 1 second, and it is not recommendable to go
5349# much below 10 seconds.
5350#Default:
5351# negative_dns_ttl 1 minutes
5352
5353# TAG: range_offset_limit size [acl acl...]
5354# usage: (size) [units] [[!]aclname]
5355#
5356# Sets an upper limit on how far (number of bytes) into the file
5357# a Range request may be to cause Squid to prefetch the whole file.
5358# If beyond this limit, Squid forwards the Range request as it is and
5359# the result is NOT cached.
5360#
5361# This is to stop a far ahead range request (lets say start at 17MB)
5362# from making Squid fetch the whole object up to that point before
5363# sending anything to the client.
5364#
5365# Multiple range_offset_limit lines may be specified, and they will
5366# be searched from top to bottom on each request until a match is found.
5367# The first match found will be used. If no line matches a request, the
5368# default limit of 0 bytes will be used.
5369#
5370# 'size' is the limit specified as a number of units.
5371#
5372# 'units' specifies whether to use bytes, KB, MB, etc.
5373# If no units are specified bytes are assumed.
5374#
5375# A size of 0 causes Squid to never fetch more than the
5376# client requested. (default)
5377#
5378# A size of 'none' causes Squid to always fetch the object from the
5379# beginning so it may cache the result. (2.0 style)
5380#
5381# 'aclname' is the name of a defined ACL.
5382#
5383# NP: Using 'none' as the byte value here will override any quick_abort settings
5384# that may otherwise apply to the range request. The range request will
5385# be fully fetched from start to finish regardless of the client
5386# actions. This affects bandwidth usage.
5387#Default:
5388# none
5389
5390# TAG: minimum_expiry_time (seconds)
5391# The minimum caching time according to (Expires - Date)
5392# headers Squid honors if the object can't be revalidated.
5393# The default is 60 seconds.
5394#
5395# In reverse proxy environments it might be desirable to honor
5396# shorter object lifetimes. It is most likely better to make
5397# your server return a meaningful Last-Modified header however.
5398#
5399# In ESI environments where page fragments often have short
5400# lifetimes, this will often be best set to 0.
5401#Default:
5402# minimum_expiry_time 60 seconds
5403
5404# TAG: store_avg_object_size (bytes)
5405# Average object size, used to estimate number of objects your
5406# cache can hold. The default is 13 KB.
5407#
5408# This is used to pre-seed the cache index memory allocation to
5409# reduce expensive reallocate operations while handling clients
5410# traffic. Too-large values may result in memory allocation during
5411# peak traffic, too-small values will result in wasted memory.
5412#
5413# Check the cache manager 'info' report metrics for the real
5414# object sizes seen by your Squid before tuning this.
5415#Default:
5416# store_avg_object_size 13 KB
5417
5418# TAG: store_objects_per_bucket
5419# Target number of objects per bucket in the store hash table.
5420# Lowering this value increases the total number of buckets and
5421# also the storage maintenance rate. The default is 20.
5422#Default:
5423# store_objects_per_bucket 20
5424
5425# HTTP OPTIONS
5426# -----------------------------------------------------------------------------
5427
5428# TAG: request_header_max_size (KB)
5429# This specifies the maximum size for HTTP headers in a request.
5430# Request headers are usually relatively small (about 512 bytes).
5431# Placing a limit on the request header size will catch certain
5432# bugs (for example with persistent connections) and possibly
5433# buffer-overflow or denial-of-service attacks.
5434#Default:
5435# request_header_max_size 64 KB
5436
5437# TAG: reply_header_max_size (KB)
5438# This specifies the maximum size for HTTP headers in a reply.
5439# Reply headers are usually relatively small (about 512 bytes).
5440# Placing a limit on the reply header size will catch certain
5441# bugs (for example with persistent connections) and possibly
5442# buffer-overflow or denial-of-service attacks.
5443#Default:
5444# reply_header_max_size 64 KB
5445
5446# TAG: request_body_max_size (bytes)
5447# This specifies the maximum size for an HTTP request body.
5448# In other words, the maximum size of a PUT/POST request.
5449# A user who attempts to send a request with a body larger
5450# than this limit receives an "Invalid Request" error message.
5451# If you set this parameter to a zero (the default), there will
5452# be no limit imposed.
5453#
5454# See also client_request_buffer_max_size for an alternative
5455# limitation on client uploads which can be configured.
5456#Default:
5457# No limit.
5458
5459# TAG: client_request_buffer_max_size (bytes)
5460# This specifies the maximum buffer size of a client request.
5461# It prevents squid eating too much memory when somebody uploads
5462# a large file.
5463#Default:
5464# client_request_buffer_max_size 512 KB
5465
5466# TAG: broken_posts
5467# A list of ACL elements which, if matched, causes Squid to send
5468# an extra CRLF pair after the body of a PUT/POST request.
5469#
5470# Some HTTP servers has broken implementations of PUT/POST,
5471# and rely on an extra CRLF pair sent by some WWW clients.
5472#
5473# Quote from RFC2616 section 4.1 on this matter:
5474#
5475# Note: certain buggy HTTP/1.0 client implementations generate an
5476# extra CRLF's after a POST request. To restate what is explicitly
5477# forbidden by the BNF, an HTTP/1.1 client must not preface or follow
5478# a request with an extra CRLF.
5479#
5480# This clause only supports fast acl types.
5481# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5482#
5483#Example:
5484# acl buggy_server url_regex ^http://....
5485# broken_posts allow buggy_server
5486#Default:
5487# Obey RFC 2616.
5488
5489# TAG: adaptation_uses_indirect_client on|off
5490# Controls whether the indirect client IP address (instead of the direct
5491# client IP address) is passed to adaptation services.
5492#
5493# See also: follow_x_forwarded_for adaptation_send_client_ip
5494#Default:
5495# adaptation_uses_indirect_client on
5496
5497# TAG: via on|off
5498# If set (default), Squid will include a Via header in requests and
5499# replies as required by RFC2616.
5500#Default:
5501# via on
5502
5503# TAG: vary_ignore_expire on|off
5504# Many HTTP servers supporting Vary gives such objects
5505# immediate expiry time with no cache-control header
5506# when requested by a HTTP/1.0 client. This option
5507# enables Squid to ignore such expiry times until
5508# HTTP/1.1 is fully implemented.
5509#
5510# WARNING: If turned on this may eventually cause some
5511# varying objects not intended for caching to get cached.
5512#Default:
5513# vary_ignore_expire off
5514
5515# TAG: request_entities
5516# Squid defaults to deny GET and HEAD requests with request entities,
5517# as the meaning of such requests are undefined in the HTTP standard
5518# even if not explicitly forbidden.
5519#
5520# Set this directive to on if you have clients which insists
5521# on sending request entities in GET or HEAD requests. But be warned
5522# that there is server software (both proxies and web servers) which
5523# can fail to properly process this kind of request which may make you
5524# vulnerable to cache pollution attacks if enabled.
5525#Default:
5526# request_entities off
5527
5528# TAG: request_header_access
5529# Usage: request_header_access header_name allow|deny [!]aclname ...
5530#
5531# WARNING: Doing this VIOLATES the HTTP standard. Enabling
5532# this feature could make you liable for problems which it
5533# causes.
5534#
5535# This option replaces the old 'anonymize_headers' and the
5536# older 'http_anonymizer' option with something that is much
5537# more configurable. A list of ACLs for each header name allows
5538# removal of specific header fields under specific conditions.
5539#
5540# This option only applies to outgoing HTTP request headers (i.e.,
5541# headers sent by Squid to the next HTTP hop such as a cache peer
5542# or an origin server). The option has no effect during cache hit
5543# detection. The equivalent adaptation vectoring point in ICAP
5544# terminology is post-cache REQMOD.
5545#
5546# The option is applied to individual outgoing request header
5547# fields. For each request header field F, Squid uses the first
5548# qualifying sets of request_header_access rules:
5549#
5550# 1. Rules with header_name equal to F's name.
5551# 2. Rules with header_name 'Other', provided F's name is not
5552# on the hard-coded list of commonly used HTTP header names.
5553# 3. Rules with header_name 'All'.
5554#
5555# Within that qualifying rule set, rule ACLs are checked as usual.
5556# If ACLs of an "allow" rule match, the header field is allowed to
5557# go through as is. If ACLs of a "deny" rule match, the header is
5558# removed and request_header_replace is then checked to identify
5559# if the removed header has a replacement. If no rules within the
5560# set have matching ACLs, the header field is left as is.
5561#
5562# For example, to achieve the same behavior as the old
5563# 'http_anonymizer standard' option, you should use:
5564#
5565# request_header_access From deny all
5566# request_header_access Referer deny all
5567# request_header_access User-Agent deny all
5568#
5569# Or, to reproduce the old 'http_anonymizer paranoid' feature
5570# you should use:
5571#
5572# request_header_access Authorization allow all
5573# request_header_access Proxy-Authorization allow all
5574# request_header_access Cache-Control allow all
5575# request_header_access Content-Length allow all
5576# request_header_access Content-Type allow all
5577# request_header_access Date allow all
5578# request_header_access Host allow all
5579# request_header_access If-Modified-Since allow all
5580# request_header_access Pragma allow all
5581# request_header_access Accept allow all
5582# request_header_access Accept-Charset allow all
5583# request_header_access Accept-Encoding allow all
5584# request_header_access Accept-Language allow all
5585# request_header_access Connection allow all
5586# request_header_access All deny all
5587#
5588# HTTP reply headers are controlled with the reply_header_access directive.
5589#
5590# By default, all headers are allowed (no anonymizing is performed).
5591#Default:
5592# No limits.
5593
5594# TAG: reply_header_access
5595# Usage: reply_header_access header_name allow|deny [!]aclname ...
5596#
5597# WARNING: Doing this VIOLATES the HTTP standard. Enabling
5598# this feature could make you liable for problems which it
5599# causes.
5600#
5601# This option only applies to reply headers, i.e., from the
5602# server to the client.
5603#
5604# This is the same as request_header_access, but in the other
5605# direction. Please see request_header_access for detailed
5606# documentation.
5607#
5608# For example, to achieve the same behavior as the old
5609# 'http_anonymizer standard' option, you should use:
5610#
5611# reply_header_access Server deny all
5612# reply_header_access WWW-Authenticate deny all
5613# reply_header_access Link deny all
5614#
5615# Or, to reproduce the old 'http_anonymizer paranoid' feature
5616# you should use:
5617#
5618# reply_header_access Allow allow all
5619# reply_header_access WWW-Authenticate allow all
5620# reply_header_access Proxy-Authenticate allow all
5621# reply_header_access Cache-Control allow all
5622# reply_header_access Content-Encoding allow all
5623# reply_header_access Content-Length allow all
5624# reply_header_access Content-Type allow all
5625# reply_header_access Date allow all
5626# reply_header_access Expires allow all
5627# reply_header_access Last-Modified allow all
5628# reply_header_access Location allow all
5629# reply_header_access Pragma allow all
5630# reply_header_access Content-Language allow all
5631# reply_header_access Retry-After allow all
5632# reply_header_access Title allow all
5633# reply_header_access Content-Disposition allow all
5634# reply_header_access Connection allow all
5635# reply_header_access All deny all
5636#
5637# HTTP request headers are controlled with the request_header_access directive.
5638#
5639# By default, all headers are allowed (no anonymizing is
5640# performed).
5641#Default:
5642# No limits.
5643
5644# TAG: request_header_replace
5645# Usage: request_header_replace header_name message
5646# Example: request_header_replace User-Agent Nutscrape/1.0 (CP/M; 8-bit)
5647#
5648# This option allows you to change the contents of headers
5649# denied with request_header_access above, by replacing them
5650# with some fixed string.
5651#
5652# This only applies to request headers, not reply headers.
5653#
5654# By default, headers are removed if denied.
5655#Default:
5656# none
5657
5658# TAG: reply_header_replace
5659# Usage: reply_header_replace header_name message
5660# Example: reply_header_replace Server Foo/1.0
5661#
5662# This option allows you to change the contents of headers
5663# denied with reply_header_access above, by replacing them
5664# with some fixed string.
5665#
5666# This only applies to reply headers, not request headers.
5667#
5668# By default, headers are removed if denied.
5669#Default:
5670# none
5671
5672# TAG: request_header_add
5673# Usage: request_header_add field-name field-value [ acl ... ]
5674# Example: request_header_add X-Client-CA "CA=%ssl::>cert_issuer" all
5675#
5676# This option adds header fields to outgoing HTTP requests (i.e.,
5677# request headers sent by Squid to the next HTTP hop such as a
5678# cache peer or an origin server). The option has no effect during
5679# cache hit detection. The equivalent adaptation vectoring point
5680# in ICAP terminology is post-cache REQMOD.
5681#
5682# Field-name is a token specifying an HTTP header name. If a
5683# standard HTTP header name is used, Squid does not check whether
5684# the new header conflicts with any existing headers or violates
5685# HTTP rules. If the request to be modified already contains a
5686# field with the same name, the old field is preserved but the
5687# header field values are not merged.
5688#
5689# Field-value is either a token or a quoted string. If quoted
5690# string format is used, then the surrounding quotes are removed
5691# while escape sequences and %macros are processed.
5692#
5693# One or more Squid ACLs may be specified to restrict header
5694# injection to matching requests. As always in squid.conf, all
5695# ACLs in the ACL list must be satisfied for the insertion to
5696# happen. The request_header_add supports fast ACLs only.
5697#
5698# See also: reply_header_add.
5699#Default:
5700# none
5701
5702# TAG: reply_header_add
5703# Usage: reply_header_add field-name field-value [ acl ... ]
5704# Example: reply_header_add X-Client-CA "CA=%ssl::>cert_issuer" all
5705#
5706# This option adds header fields to outgoing HTTP responses (i.e., response
5707# headers delivered by Squid to the client). This option has no effect on
5708# cache hit detection. The equivalent adaptation vectoring point in
5709# ICAP terminology is post-cache RESPMOD. This option does not apply to
5710# successful CONNECT replies.
5711#
5712# Field-name is a token specifying an HTTP header name. If a
5713# standard HTTP header name is used, Squid does not check whether
5714# the new header conflicts with any existing headers or violates
5715# HTTP rules. If the response to be modified already contains a
5716# field with the same name, the old field is preserved but the
5717# header field values are not merged.
5718#
5719# Field-value is either a token or a quoted string. If quoted
5720# string format is used, then the surrounding quotes are removed
5721# while escape sequences and %macros are processed.
5722#
5723# One or more Squid ACLs may be specified to restrict header
5724# injection to matching responses. As always in squid.conf, all
5725# ACLs in the ACL list must be satisfied for the insertion to
5726# happen. The reply_header_add option supports fast ACLs only.
5727#
5728# See also: request_header_add.
5729#Default:
5730# none
5731
5732# TAG: note
5733# This option used to log custom information about the master
5734# transaction. For example, an admin may configure Squid to log
5735# which "user group" the transaction belongs to, where "user group"
5736# will be determined based on a set of ACLs and not [just]
5737# authentication information.
5738# Values of key/value pairs can be logged using %{key}note macros:
5739#
5740# note key value acl ...
5741# logformat myFormat ... %{key}note ...
5742#Default:
5743# none
5744
5745# TAG: relaxed_header_parser on|off|warn
5746# In the default "on" setting Squid accepts certain forms
5747# of non-compliant HTTP messages where it is unambiguous
5748# what the sending application intended even if the message
5749# is not correctly formatted. The messages is then normalized
5750# to the correct form when forwarded by Squid.
5751#
5752# If set to "warn" then a warning will be emitted in cache.log
5753# each time such HTTP error is encountered.
5754#
5755# If set to "off" then such HTTP errors will cause the request
5756# or response to be rejected.
5757#Default:
5758# relaxed_header_parser on
5759
5760# TAG: collapsed_forwarding (on|off)
5761# This option controls whether Squid is allowed to merge multiple
5762# potentially cachable requests for the same URI before Squid knows
5763# whether the response is going to be cachable.
5764#
5765# When enabled, instead of forwarding each concurrent request for
5766# the same URL, Squid just sends the first of them. The other, so
5767# called "collapsed" requests, wait for the response to the first
5768# request and, if it happens to be cachable, use that response.
5769# Here, "concurrent requests" means "received after the first
5770# request headers were parsed and before the corresponding response
5771# headers were parsed".
5772#
5773# This feature is disabled by default: enabling collapsed
5774# forwarding needlessly delays forwarding requests that look
5775# cachable (when they are collapsed) but then need to be forwarded
5776# individually anyway because they end up being for uncachable
5777# content. However, in some cases, such as acceleration of highly
5778# cachable content with periodic or grouped expiration times, the
5779# gains from collapsing [large volumes of simultaneous refresh
5780# requests] outweigh losses from such delays.
5781#
5782# Squid collapses two kinds of requests: regular client requests
5783# received on one of the listening ports and internal "cache
5784# revalidation" requests which are triggered by those regular
5785# requests hitting a stale cached object. Revalidation collapsing
5786# is currently disabled for Squid instances containing SMP-aware
5787# disk or memory caches and for Vary-controlled cached objects.
5788#Default:
5789# collapsed_forwarding off
5790
5791# TAG: collapsed_forwarding_shared_entries_limit (number of entries)
5792# This limits the size of a table used for sharing information
5793# about collapsible entries among SMP workers. Limiting sharing
5794# too much results in cache content duplication and missed
5795# collapsing opportunities. Using excessively large values
5796# wastes shared memory.
5797#
5798# The limit should be significantly larger then the number of
5799# concurrent collapsible entries one wants to share. For a cache
5800# that handles less than 5000 concurrent requests, the default
5801# setting of 16384 should be plenty.
5802#
5803# If the limit is set to zero, it disables sharing of collapsed
5804# forwarding between SMP workers.
5805#Default:
5806# collapsed_forwarding_shared_entries_limit 16384
5807
5808# TIMEOUTS
5809# -----------------------------------------------------------------------------
5810
5811# TAG: forward_timeout time-units
5812# This parameter specifies how long Squid should at most attempt in
5813# finding a forwarding path for the request before giving up.
5814#Default:
5815# forward_timeout 4 minutes
5816
5817# TAG: connect_timeout time-units
5818# This parameter specifies how long to wait for the TCP connect to
5819# the requested server or peer to complete before Squid should
5820# attempt to find another path where to forward the request.
5821#Default:
5822# connect_timeout 1 minute
5823
5824# TAG: peer_connect_timeout time-units
5825# This parameter specifies how long to wait for a pending TCP
5826# connection to a peer cache. The default is 30 seconds. You
5827# may also set different timeout values for individual neighbors
5828# with the 'connect-timeout' option on a 'cache_peer' line.
5829#Default:
5830# peer_connect_timeout 30 seconds
5831
5832# TAG: read_timeout time-units
5833# Applied on peer server connections.
5834#
5835# After each successful read(), the timeout will be extended by this
5836# amount. If no data is read again after this amount of time,
5837# the request is aborted and logged with ERR_READ_TIMEOUT.
5838#
5839# The default is 15 minutes.
5840#Default:
5841# read_timeout 15 minutes
5842
5843# TAG: write_timeout time-units
5844# This timeout is tracked for all connections that have data
5845# available for writing and are waiting for the socket to become
5846# ready. After each successful write, the timeout is extended by
5847# the configured amount. If Squid has data to write but the
5848# connection is not ready for the configured duration, the
5849# transaction associated with the connection is terminated. The
5850# default is 15 minutes.
5851#Default:
5852# write_timeout 15 minutes
5853
5854# TAG: request_timeout
5855# How long to wait for complete HTTP request headers after initial
5856# connection establishment.
5857#Default:
5858# request_timeout 5 minutes
5859
5860# TAG: request_start_timeout
5861# How long to wait for the first request byte after initial
5862# connection establishment.
5863#Default:
5864# request_start_timeout 5 minutes
5865
5866# TAG: client_idle_pconn_timeout
5867# How long to wait for the next HTTP request on a persistent
5868# client connection after the previous request completes.
5869#Default:
5870# client_idle_pconn_timeout 2 minutes
5871
5872# TAG: ftp_client_idle_timeout
5873# How long to wait for an FTP request on a connection to Squid ftp_port.
5874# Many FTP clients do not deal with idle connection closures well,
5875# necessitating a longer default timeout than client_idle_pconn_timeout
5876# used for incoming HTTP requests.
5877#Default:
5878# ftp_client_idle_timeout 30 minutes
5879
5880# TAG: client_lifetime time-units
5881# The maximum amount of time a client (browser) is allowed to
5882# remain connected to the cache process. This protects the Cache
5883# from having a lot of sockets (and hence file descriptors) tied up
5884# in a CLOSE_WAIT state from remote clients that go away without
5885# properly shutting down (either because of a network failure or
5886# because of a poor client implementation). The default is one
5887# day, 1440 minutes.
5888#
5889# NOTE: The default value is intended to be much larger than any
5890# client would ever need to be connected to your cache. You
5891# should probably change client_lifetime only as a last resort.
5892# If you seem to have many client connections tying up
5893# filedescriptors, we recommend first tuning the read_timeout,
5894# request_timeout, persistent_request_timeout and quick_abort values.
5895#Default:
5896# client_lifetime 1 day
5897
5898# TAG: pconn_lifetime time-units
5899# Desired maximum lifetime of a persistent connection.
5900# When set, Squid will close a now-idle persistent connection that
5901# exceeded configured lifetime instead of moving the connection into
5902# the idle connection pool (or equivalent). No effect on ongoing/active
5903# transactions. Connection lifetime is the time period from the
5904# connection acceptance or opening time until "now".
5905#
5906# This limit is useful in environments with long-lived connections
5907# where Squid configuration or environmental factors change during a
5908# single connection lifetime. If unrestricted, some connections may
5909# last for hours and even days, ignoring those changes that should
5910# have affected their behavior or their existence.
5911#
5912# Currently, a new lifetime value supplied via Squid reconfiguration
5913# has no effect on already idle connections unless they become busy.
5914#
5915# When set to '0' this limit is not used.
5916#Default:
5917# pconn_lifetime 0 seconds
5918
5919# TAG: half_closed_clients
5920# Some clients may shutdown the sending side of their TCP
5921# connections, while leaving their receiving sides open. Sometimes,
5922# Squid can not tell the difference between a half-closed and a
5923# fully-closed TCP connection.
5924#
5925# By default, Squid will immediately close client connections when
5926# read(2) returns "no more data to read."
5927#
5928# Change this option to 'on' and Squid will keep open connections
5929# until a read(2) or write(2) on the socket returns an error.
5930# This may show some benefits for reverse proxies. But if not
5931# it is recommended to leave OFF.
5932#Default:
5933# half_closed_clients off
5934
5935# TAG: server_idle_pconn_timeout
5936# Timeout for idle persistent connections to servers and other
5937# proxies.
5938#Default:
5939# server_idle_pconn_timeout 1 minute
5940
5941# TAG: ident_timeout
5942# Maximum time to wait for IDENT lookups to complete.
5943#
5944# If this is too high, and you enabled IDENT lookups from untrusted
5945# users, you might be susceptible to denial-of-service by having
5946# many ident requests going at once.
5947#Default:
5948# ident_timeout 10 seconds
5949
5950# TAG: shutdown_lifetime time-units
5951# When SIGTERM or SIGHUP is received, the cache is put into
5952# "shutdown pending" mode until all active sockets are closed.
5953# This value is the lifetime to set for all open descriptors
5954# during shutdown mode. Any active clients after this many
5955# seconds will receive a 'timeout' message.
5956#Default:
5957# shutdown_lifetime 30 seconds
5958
5959# ADMINISTRATIVE PARAMETERS
5960# -----------------------------------------------------------------------------
5961
5962# TAG: cache_mgr
5963# Email-address of local cache manager who will receive
5964# mail if the cache dies. The default is "webmaster".
5965#Default:
5966# cache_mgr webmaster
5967
5968# TAG: mail_from
5969# From: email-address for mail sent when the cache dies.
5970# The default is to use 'squid@unique_hostname'.
5971#
5972# See also: unique_hostname directive.
5973#Default:
5974# none
5975
5976# TAG: mail_program
5977# Email program used to send mail if the cache dies.
5978# The default is "mail". The specified program must comply
5979# with the standard Unix mail syntax:
5980# mail-program recipient < mailfile
5981#
5982# Optional command line options can be specified.
5983#Default:
5984# mail_program mail
5985
5986# TAG: cache_effective_user
5987# If you start Squid as root, it will change its effective/real
5988# UID/GID to the user specified below. The default is to change
5989# to UID of proxy.
5990# see also; cache_effective_group
5991#Default:
5992# cache_effective_user proxy
5993cache_effective_user proxy
5994
5995# TAG: cache_effective_group
5996# Squid sets the GID to the effective user's default group ID
5997# (taken from the password file) and supplementary group list
5998# from the groups membership.
5999#
6000# If you want Squid to run with a specific GID regardless of
6001# the group memberships of the effective user then set this
6002# to the group (or GID) you want Squid to run as. When set
6003# all other group privileges of the effective user are ignored
6004# and only this GID is effective. If Squid is not started as
6005# root the user starting Squid MUST be member of the specified
6006# group.
6007#
6008# This option is not recommended by the Squid Team.
6009# Our preference is for administrators to configure a secure
6010# user account for squid with UID/GID matching system policies.
6011#Default:
6012# Use system group memberships of the cache_effective_user account
6013
6014# TAG: httpd_suppress_version_string on|off
6015# Suppress Squid version string info in HTTP headers and HTML error pages.
6016#Default:
6017# httpd_suppress_version_string off
6018
6019# TAG: visible_hostname
6020# If you want to present a special hostname in error messages, etc,
6021# define this. Otherwise, the return value of gethostname()
6022# will be used. If you have multiple caches in a cluster and
6023# get errors about IP-forwarding you must set them to have individual
6024# names with this setting.
6025#Default:
6026# Automatically detect the system host name
6027
6028# TAG: unique_hostname
6029# If you want to have multiple machines with the same
6030# 'visible_hostname' you must give each machine a different
6031# 'unique_hostname' so forwarding loops can be detected.
6032#Default:
6033# Copy the value from visible_hostname
6034
6035# TAG: hostname_aliases
6036# A list of other DNS names your cache has.
6037#Default:
6038# none
6039
6040# TAG: umask
6041# Minimum umask which should be enforced while the proxy
6042# is running, in addition to the umask set at startup.
6043#
6044# For a traditional octal representation of umasks, start
6045# your value with 0.
6046#Default:
6047# umask 027
6048
6049# OPTIONS FOR THE CACHE REGISTRATION SERVICE
6050# -----------------------------------------------------------------------------
6051#
6052# This section contains parameters for the (optional) cache
6053# announcement service. This service is provided to help
6054# cache administrators locate one another in order to join or
6055# create cache hierarchies.
6056#
6057# An 'announcement' message is sent (via UDP) to the registration
6058# service by Squid. By default, the announcement message is NOT
6059# SENT unless you enable it with 'announce_period' below.
6060#
6061# The announcement message includes your hostname, plus the
6062# following information from this configuration file:
6063#
6064# http_port
6065# icp_port
6066# cache_mgr
6067#
6068# All current information is processed regularly and made
6069# available on the Web at http://www.ircache.net/Cache/Tracker/.
6070
6071# TAG: announce_period
6072# This is how frequently to send cache announcements.
6073#
6074# To enable announcing your cache, just set an announce period.
6075#
6076# Example:
6077# announce_period 1 day
6078#Default:
6079# Announcement messages disabled.
6080
6081# TAG: announce_host
6082# Set the hostname where announce registration messages will be sent.
6083#
6084# See also announce_port and announce_file
6085#Default:
6086# announce_host tracker.ircache.net
6087
6088# TAG: announce_file
6089# The contents of this file will be included in the announce
6090# registration messages.
6091#Default:
6092# none
6093
6094# TAG: announce_port
6095# Set the port where announce registration messages will be sent.
6096#
6097# See also announce_host and announce_file
6098#Default:
6099# announce_port 3131
6100
6101# HTTPD-ACCELERATOR OPTIONS
6102# -----------------------------------------------------------------------------
6103
6104# TAG: httpd_accel_surrogate_id
6105# Surrogates (http://www.esi.org/architecture_spec_1.0.html)
6106# need an identification token to allow control targeting. Because
6107# a farm of surrogates may all perform the same tasks, they may share
6108# an identification token.
6109#Default:
6110# visible_hostname is used if no specific ID is set.
6111
6112# TAG: http_accel_surrogate_remote on|off
6113# Remote surrogates (such as those in a CDN) honour the header
6114# "Surrogate-Control: no-store-remote".
6115#
6116# Set this to on to have squid behave as a remote surrogate.
6117#Default:
6118# http_accel_surrogate_remote off
6119
6120# TAG: esi_parser libxml2|expat
6121# Selects the XML parsing library to use when interpreting responses with
6122# Edge Side Includes.
6123#
6124# To disable ESI handling completely, ./configure Squid with --disable-esi.
6125#Default:
6126# Selects libxml2 if available at ./configure time or libexpat otherwise.
6127
6128# DELAY POOL PARAMETERS
6129# -----------------------------------------------------------------------------
6130
6131# TAG: delay_pools
6132# This represents the number of delay pools to be used. For example,
6133# if you have one class 2 delay pool and one class 3 delays pool, you
6134# have a total of 2 delay pools.
6135#
6136# See also delay_parameters, delay_class, delay_access for pool
6137# configuration details.
6138#Default:
6139# delay_pools 0
6140
6141# TAG: delay_class
6142# This defines the class of each delay pool. There must be exactly one
6143# delay_class line for each delay pool. For example, to define two
6144# delay pools, one of class 2 and one of class 3, the settings above
6145# and here would be:
6146#
6147# Example:
6148# delay_pools 4 # 4 delay pools
6149# delay_class 1 2 # pool 1 is a class 2 pool
6150# delay_class 2 3 # pool 2 is a class 3 pool
6151# delay_class 3 4 # pool 3 is a class 4 pool
6152# delay_class 4 5 # pool 4 is a class 5 pool
6153#
6154# The delay pool classes are:
6155#
6156# class 1 Everything is limited by a single aggregate
6157# bucket.
6158#
6159# class 2 Everything is limited by a single aggregate
6160# bucket as well as an "individual" bucket chosen
6161# from bits 25 through 32 of the IPv4 address.
6162#
6163# class 3 Everything is limited by a single aggregate
6164# bucket as well as a "network" bucket chosen
6165# from bits 17 through 24 of the IP address and a
6166# "individual" bucket chosen from bits 17 through
6167# 32 of the IPv4 address.
6168#
6169# class 4 Everything in a class 3 delay pool, with an
6170# additional limit on a per user basis. This
6171# only takes effect if the username is established
6172# in advance - by forcing authentication in your
6173# http_access rules.
6174#
6175# class 5 Requests are grouped according their tag (see
6176# external_acl's tag= reply).
6177#
6178#
6179# Each pool also requires a delay_parameters directive to configure the pool size
6180# and speed limits used whenever the pool is applied to a request. Along with
6181# a set of delay_access directives to determine when it is used.
6182#
6183# NOTE: If an IP address is a.b.c.d
6184# -> bits 25 through 32 are "d"
6185# -> bits 17 through 24 are "c"
6186# -> bits 17 through 32 are "c * 256 + d"
6187#
6188# NOTE-2: Due to the use of bitmasks in class 2,3,4 pools they only apply to
6189# IPv4 traffic. Class 1 and 5 pools may be used with IPv6 traffic.
6190#
6191# This clause only supports fast acl types.
6192# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6193#
6194# See also delay_parameters and delay_access.
6195#Default:
6196# none
6197
6198# TAG: delay_access
6199# This is used to determine which delay pool a request falls into.
6200#
6201# delay_access is sorted per pool and the matching starts with pool 1,
6202# then pool 2, ..., and finally pool N. The first delay pool where the
6203# request is allowed is selected for the request. If it does not allow
6204# the request to any pool then the request is not delayed (default).
6205#
6206# For example, if you want some_big_clients in delay
6207# pool 1 and lotsa_little_clients in delay pool 2:
6208#
6209# delay_access 1 allow some_big_clients
6210# delay_access 1 deny all
6211# delay_access 2 allow lotsa_little_clients
6212# delay_access 2 deny all
6213# delay_access 3 allow authenticated_clients
6214#
6215# See also delay_parameters and delay_class.
6216#
6217#Default:
6218# Deny using the pool, unless allow rules exist in squid.conf for the pool.
6219
6220# TAG: delay_parameters
6221# This defines the parameters for a delay pool. Each delay pool has
6222# a number of "buckets" associated with it, as explained in the
6223# description of delay_class.
6224#
6225# For a class 1 delay pool, the syntax is:
6226# delay_class pool 1
6227# delay_parameters pool aggregate
6228#
6229# For a class 2 delay pool:
6230# delay_class pool 2
6231# delay_parameters pool aggregate individual
6232#
6233# For a class 3 delay pool:
6234# delay_class pool 3
6235# delay_parameters pool aggregate network individual
6236#
6237# For a class 4 delay pool:
6238# delay_class pool 4
6239# delay_parameters pool aggregate network individual user
6240#
6241# For a class 5 delay pool:
6242# delay_class pool 5
6243# delay_parameters pool tagrate
6244#
6245# The option variables are:
6246#
6247# pool a pool number - ie, a number between 1 and the
6248# number specified in delay_pools as used in
6249# delay_class lines.
6250#
6251# aggregate the speed limit parameters for the aggregate bucket
6252# (class 1, 2, 3).
6253#
6254# individual the speed limit parameters for the individual
6255# buckets (class 2, 3).
6256#
6257# network the speed limit parameters for the network buckets
6258# (class 3).
6259#
6260# user the speed limit parameters for the user buckets
6261# (class 4).
6262#
6263# tagrate the speed limit parameters for the tag buckets
6264# (class 5).
6265#
6266# A pair of delay parameters is written restore/maximum, where restore is
6267# the number of bytes (not bits - modem and network speeds are usually
6268# quoted in bits) per second placed into the bucket, and maximum is the
6269# maximum number of bytes which can be in the bucket at any time.
6270#
6271# There must be one delay_parameters line for each delay pool.
6272#
6273#
6274# For example, if delay pool number 1 is a class 2 delay pool as in the
6275# above example, and is being used to strictly limit each host to 64Kbit/sec
6276# (plus overheads), with no overall limit, the line is:
6277#
6278# delay_parameters 1 none 8000/8000
6279#
6280# Note that 8 x 8K Byte/sec -> 64K bit/sec.
6281#
6282# Note that the word 'none' is used to represent no limit.
6283#
6284#
6285# And, if delay pool number 2 is a class 3 delay pool as in the above
6286# example, and you want to limit it to a total of 256Kbit/sec (strict limit)
6287# with each 8-bit network permitted 64Kbit/sec (strict limit) and each
6288# individual host permitted 4800bit/sec with a bucket maximum size of 64Kbits
6289# to permit a decent web page to be downloaded at a decent speed
6290# (if the network is not being limited due to overuse) but slow down
6291# large downloads more significantly:
6292#
6293# delay_parameters 2 32000/32000 8000/8000 600/8000
6294#
6295# Note that 8 x 32K Byte/sec -> 256K bit/sec.
6296# 8 x 8K Byte/sec -> 64K bit/sec.
6297# 8 x 600 Byte/sec -> 4800 bit/sec.
6298#
6299#
6300# Finally, for a class 4 delay pool as in the example - each user will
6301# be limited to 128Kbits/sec no matter how many workstations they are logged into.:
6302#
6303# delay_parameters 4 32000/32000 8000/8000 600/64000 16000/16000
6304#
6305#
6306# See also delay_class and delay_access.
6307#
6308#Default:
6309# none
6310
6311# TAG: delay_initial_bucket_level (percent, 0-100)
6312# The initial bucket percentage is used to determine how much is put
6313# in each bucket when squid starts, is reconfigured, or first notices
6314# a host accessing it (in class 2 and class 3, individual hosts and
6315# networks only have buckets associated with them once they have been
6316# "seen" by squid).
6317#Default:
6318# delay_initial_bucket_level 50
6319
6320# CLIENT DELAY POOL PARAMETERS
6321# -----------------------------------------------------------------------------
6322
6323# TAG: client_delay_pools
6324# This option specifies the number of client delay pools used. It must
6325# preceed other client_delay_* options.
6326#
6327# Example:
6328# client_delay_pools 2
6329#
6330# See also client_delay_parameters and client_delay_access.
6331#Default:
6332# client_delay_pools 0
6333
6334# TAG: client_delay_initial_bucket_level (percent, 0-no_limit)
6335# This option determines the initial bucket size as a percentage of
6336# max_bucket_size from client_delay_parameters. Buckets are created
6337# at the time of the "first" connection from the matching IP. Idle
6338# buckets are periodically deleted up.
6339#
6340# You can specify more than 100 percent but note that such "oversized"
6341# buckets are not refilled until their size goes down to max_bucket_size
6342# from client_delay_parameters.
6343#
6344# Example:
6345# client_delay_initial_bucket_level 50
6346#Default:
6347# client_delay_initial_bucket_level 50
6348
6349# TAG: client_delay_parameters
6350#
6351# This option configures client-side bandwidth limits using the
6352# following format:
6353#
6354# client_delay_parameters pool speed_limit max_bucket_size
6355#
6356# pool is an integer ID used for client_delay_access matching.
6357#
6358# speed_limit is bytes added to the bucket per second.
6359#
6360# max_bucket_size is the maximum size of a bucket, enforced after any
6361# speed_limit additions.
6362#
6363# Please see the delay_parameters option for more information and
6364# examples.
6365#
6366# Example:
6367# client_delay_parameters 1 1024 2048
6368# client_delay_parameters 2 51200 16384
6369#
6370# See also client_delay_access.
6371#
6372#Default:
6373# none
6374
6375# TAG: client_delay_access
6376# This option determines the client-side delay pool for the
6377# request:
6378#
6379# client_delay_access pool_ID allow|deny acl_name
6380#
6381# All client_delay_access options are checked in their pool ID
6382# order, starting with pool 1. The first checked pool with allowed
6383# request is selected for the request. If no ACL matches or there
6384# are no client_delay_access options, the request bandwidth is not
6385# limited.
6386#
6387# The ACL-selected pool is then used to find the
6388# client_delay_parameters for the request. Client-side pools are
6389# not used to aggregate clients. Clients are always aggregated
6390# based on their source IP addresses (one bucket per source IP).
6391#
6392# This clause only supports fast acl types.
6393# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6394# Additionally, only the client TCP connection details are available.
6395# ACLs testing HTTP properties will not work.
6396#
6397# Please see delay_access for more examples.
6398#
6399# Example:
6400# client_delay_access 1 allow low_rate_network
6401# client_delay_access 2 allow vips_network
6402#
6403#
6404# See also client_delay_parameters and client_delay_pools.
6405#Default:
6406# Deny use of the pool, unless allow rules exist in squid.conf for the pool.
6407
6408# WCCPv1 AND WCCPv2 CONFIGURATION OPTIONS
6409# -----------------------------------------------------------------------------
6410
6411# TAG: wccp_router
6412# Use this option to define your WCCP ``home'' router for
6413# Squid.
6414#
6415# wccp_router supports a single WCCP(v1) router
6416#
6417# wccp2_router supports multiple WCCPv2 routers
6418#
6419# only one of the two may be used at the same time and defines
6420# which version of WCCP to use.
6421#Default:
6422# WCCP disabled.
6423
6424# TAG: wccp2_router
6425# Use this option to define your WCCP ``home'' router for
6426# Squid.
6427#
6428# wccp_router supports a single WCCP(v1) router
6429#
6430# wccp2_router supports multiple WCCPv2 routers
6431#
6432# only one of the two may be used at the same time and defines
6433# which version of WCCP to use.
6434#Default:
6435# WCCPv2 disabled.
6436
6437# TAG: wccp_version
6438# This directive is only relevant if you need to set up WCCP(v1)
6439# to some very old and end-of-life Cisco routers. In all other
6440# setups it must be left unset or at the default setting.
6441# It defines an internal version in the WCCP(v1) protocol,
6442# with version 4 being the officially documented protocol.
6443#
6444# According to some users, Cisco IOS 11.2 and earlier only
6445# support WCCP version 3. If you're using that or an earlier
6446# version of IOS, you may need to change this value to 3, otherwise
6447# do not specify this parameter.
6448#Default:
6449# wccp_version 4
6450
6451# TAG: wccp2_rebuild_wait
6452# If this is enabled Squid will wait for the cache dir rebuild to finish
6453# before sending the first wccp2 HereIAm packet
6454#Default:
6455# wccp2_rebuild_wait on
6456
6457# TAG: wccp2_forwarding_method
6458# WCCP2 allows the setting of forwarding methods between the
6459# router/switch and the cache. Valid values are as follows:
6460#
6461# gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
6462# l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
6463#
6464# Currently (as of IOS 12.4) cisco routers only support GRE.
6465# Cisco switches only support the L2 redirect assignment method.
6466#Default:
6467# wccp2_forwarding_method gre
6468
6469# TAG: wccp2_return_method
6470# WCCP2 allows the setting of return methods between the
6471# router/switch and the cache for packets that the cache
6472# decides not to handle. Valid values are as follows:
6473#
6474# gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
6475# l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
6476#
6477# Currently (as of IOS 12.4) cisco routers only support GRE.
6478# Cisco switches only support the L2 redirect assignment.
6479#
6480# If the "ip wccp redirect exclude in" command has been
6481# enabled on the cache interface, then it is still safe for
6482# the proxy server to use a l2 redirect method even if this
6483# option is set to GRE.
6484#Default:
6485# wccp2_return_method gre
6486
6487# TAG: wccp2_assignment_method
6488# WCCP2 allows the setting of methods to assign the WCCP hash
6489# Valid values are as follows:
6490#
6491# hash - Hash assignment
6492# mask - Mask assignment
6493#
6494# As a general rule, cisco routers support the hash assignment method
6495# and cisco switches support the mask assignment method.
6496#Default:
6497# wccp2_assignment_method hash
6498
6499# TAG: wccp2_service
6500# WCCP2 allows for multiple traffic services. There are two
6501# types: "standard" and "dynamic". The standard type defines
6502# one service id - http (id 0). The dynamic service ids can be from
6503# 51 to 255 inclusive. In order to use a dynamic service id
6504# one must define the type of traffic to be redirected; this is done
6505# using the wccp2_service_info option.
6506#
6507# The "standard" type does not require a wccp2_service_info option,
6508# just specifying the service id will suffice.
6509#
6510# MD5 service authentication can be enabled by adding
6511# "password=<password>" to the end of this service declaration.
6512#
6513# Examples:
6514#
6515# wccp2_service standard 0 # for the 'web-cache' standard service
6516# wccp2_service dynamic 80 # a dynamic service type which will be
6517# # fleshed out with subsequent options.
6518# wccp2_service standard 0 password=foo
6519#Default:
6520# Use the 'web-cache' standard service.
6521
6522# TAG: wccp2_service_info
6523# Dynamic WCCPv2 services require further information to define the
6524# traffic you wish to have diverted.
6525#
6526# The format is:
6527#
6528# wccp2_service_info <id> protocol=<protocol> flags=<flag>,<flag>..
6529# priority=<priority> ports=<port>,<port>..
6530#
6531# The relevant WCCPv2 flags:
6532# + src_ip_hash, dst_ip_hash
6533# + source_port_hash, dst_port_hash
6534# + src_ip_alt_hash, dst_ip_alt_hash
6535# + src_port_alt_hash, dst_port_alt_hash
6536# + ports_source
6537#
6538# The port list can be one to eight entries.
6539#
6540# Example:
6541#
6542# wccp2_service_info 80 protocol=tcp flags=src_ip_hash,ports_source
6543# priority=240 ports=80
6544#
6545# Note: the service id must have been defined by a previous
6546# 'wccp2_service dynamic <id>' entry.
6547#Default:
6548# none
6549
6550# TAG: wccp2_weight
6551# Each cache server gets assigned a set of the destination
6552# hash proportional to their weight.
6553#Default:
6554# wccp2_weight 10000
6555
6556# TAG: wccp_address
6557# Use this option if you require WCCPv2 to use a specific
6558# interface address.
6559#
6560# The default behavior is to not bind to any specific address.
6561#Default:
6562# Address selected by the operating system.
6563
6564# TAG: wccp2_address
6565# Use this option if you require WCCP to use a specific
6566# interface address.
6567#
6568# The default behavior is to not bind to any specific address.
6569#Default:
6570# Address selected by the operating system.
6571
6572# PERSISTENT CONNECTION HANDLING
6573# -----------------------------------------------------------------------------
6574#
6575# Also see "pconn_timeout" in the TIMEOUTS section
6576
6577# TAG: client_persistent_connections
6578# Persistent connection support for clients.
6579# Squid uses persistent connections (when allowed). You can use
6580# this option to disable persistent connections with clients.
6581#Default:
6582# client_persistent_connections on
6583
6584# TAG: server_persistent_connections
6585# Persistent connection support for servers.
6586# Squid uses persistent connections (when allowed). You can use
6587# this option to disable persistent connections with servers.
6588#Default:
6589# server_persistent_connections on
6590
6591# TAG: persistent_connection_after_error
6592# With this directive the use of persistent connections after
6593# HTTP errors can be disabled. Useful if you have clients
6594# who fail to handle errors on persistent connections proper.
6595#Default:
6596# persistent_connection_after_error on
6597
6598# TAG: detect_broken_pconn
6599# Some servers have been found to incorrectly signal the use
6600# of HTTP/1.0 persistent connections even on replies not
6601# compatible, causing significant delays. This server problem
6602# has mostly been seen on redirects.
6603#
6604# By enabling this directive Squid attempts to detect such
6605# broken replies and automatically assume the reply is finished
6606# after 10 seconds timeout.
6607#Default:
6608# detect_broken_pconn off
6609
6610# CACHE DIGEST OPTIONS
6611# -----------------------------------------------------------------------------
6612
6613# TAG: digest_generation
6614# This controls whether the server will generate a Cache Digest
6615# of its contents. By default, Cache Digest generation is
6616# enabled if Squid is compiled with --enable-cache-digests defined.
6617#Default:
6618# digest_generation on
6619
6620# TAG: digest_bits_per_entry
6621# This is the number of bits of the server's Cache Digest which
6622# will be associated with the Digest entry for a given HTTP
6623# Method and URL (public key) combination. The default is 5.
6624#Default:
6625# digest_bits_per_entry 5
6626
6627# TAG: digest_rebuild_period (seconds)
6628# This is the wait time between Cache Digest rebuilds.
6629#Default:
6630# digest_rebuild_period 1 hour
6631
6632# TAG: digest_rewrite_period (seconds)
6633# This is the wait time between Cache Digest writes to
6634# disk.
6635#Default:
6636# digest_rewrite_period 1 hour
6637
6638# TAG: digest_swapout_chunk_size (bytes)
6639# This is the number of bytes of the Cache Digest to write to
6640# disk at a time. It defaults to 4096 bytes (4KB), the Squid
6641# default swap page.
6642#Default:
6643# digest_swapout_chunk_size 4096 bytes
6644
6645# TAG: digest_rebuild_chunk_percentage (percent, 0-100)
6646# This is the percentage of the Cache Digest to be scanned at a
6647# time. By default it is set to 10% of the Cache Digest.
6648#Default:
6649# digest_rebuild_chunk_percentage 10
6650
6651# SNMP OPTIONS
6652# -----------------------------------------------------------------------------
6653
6654# TAG: snmp_port
6655# The port number where Squid listens for SNMP requests. To enable
6656# SNMP support set this to a suitable port number. Port number
6657# 3401 is often used for the Squid SNMP agent. By default it's
6658# set to "0" (disabled)
6659#
6660# Example:
6661# snmp_port 3401
6662#Default:
6663# SNMP disabled.
6664
6665# TAG: snmp_access
6666# Allowing or denying access to the SNMP port.
6667#
6668# All access to the agent is denied by default.
6669# usage:
6670#
6671# snmp_access allow|deny [!]aclname ...
6672#
6673# This clause only supports fast acl types.
6674# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6675#
6676#Example:
6677# snmp_access allow snmppublic localhost
6678# snmp_access deny all
6679#Default:
6680# Deny, unless rules exist in squid.conf.
6681
6682# TAG: snmp_incoming_address
6683# Just like 'udp_incoming_address', but for the SNMP port.
6684#
6685# snmp_incoming_address is used for the SNMP socket receiving
6686# messages from SNMP agents.
6687#
6688# The default snmp_incoming_address is to listen on all
6689# available network interfaces.
6690#Default:
6691# Accept SNMP packets from all machine interfaces.
6692
6693# TAG: snmp_outgoing_address
6694# Just like 'udp_outgoing_address', but for the SNMP port.
6695#
6696# snmp_outgoing_address is used for SNMP packets returned to SNMP
6697# agents.
6698#
6699# If snmp_outgoing_address is not set it will use the same socket
6700# as snmp_incoming_address. Only change this if you want to have
6701# SNMP replies sent using another address than where this Squid
6702# listens for SNMP queries.
6703#
6704# NOTE, snmp_incoming_address and snmp_outgoing_address can not have
6705# the same value since they both use the same port.
6706#Default:
6707# Use snmp_incoming_address or an address selected by the operating system.
6708
6709# ICP OPTIONS
6710# -----------------------------------------------------------------------------
6711
6712# TAG: icp_port
6713# The port number where Squid sends and receives ICP queries to
6714# and from neighbor caches. The standard UDP port for ICP is 3130.
6715#
6716# Example:
6717# icp_port 3130
6718#Default:
6719# ICP disabled.
6720
6721# TAG: htcp_port
6722# The port number where Squid sends and receives HTCP queries to
6723# and from neighbor caches. To turn it on you want to set it to
6724# 4827.
6725#
6726# Example:
6727# htcp_port 4827
6728#Default:
6729# HTCP disabled.
6730
6731# TAG: log_icp_queries on|off
6732# If set, ICP queries are logged to access.log. You may wish
6733# do disable this if your ICP load is VERY high to speed things
6734# up or to simplify log analysis.
6735#Default:
6736# log_icp_queries on
6737
6738# TAG: udp_incoming_address
6739# udp_incoming_address is used for UDP packets received from other
6740# caches.
6741#
6742# The default behavior is to not bind to any specific address.
6743#
6744# Only change this if you want to have all UDP queries received on
6745# a specific interface/address.
6746#
6747# NOTE: udp_incoming_address is used by the ICP, HTCP, and DNS
6748# modules. Altering it will affect all of them in the same manner.
6749#
6750# see also; udp_outgoing_address
6751#
6752# NOTE, udp_incoming_address and udp_outgoing_address can not
6753# have the same value since they both use the same port.
6754#Default:
6755# Accept packets from all machine interfaces.
6756
6757# TAG: udp_outgoing_address
6758# udp_outgoing_address is used for UDP packets sent out to other
6759# caches.
6760#
6761# The default behavior is to not bind to any specific address.
6762#
6763# Instead it will use the same socket as udp_incoming_address.
6764# Only change this if you want to have UDP queries sent using another
6765# address than where this Squid listens for UDP queries from other
6766# caches.
6767#
6768# NOTE: udp_outgoing_address is used by the ICP, HTCP, and DNS
6769# modules. Altering it will affect all of them in the same manner.
6770#
6771# see also; udp_incoming_address
6772#
6773# NOTE, udp_incoming_address and udp_outgoing_address can not
6774# have the same value since they both use the same port.
6775#Default:
6776# Use udp_incoming_address or an address selected by the operating system.
6777
6778# TAG: icp_hit_stale on|off
6779# If you want to return ICP_HIT for stale cache objects, set this
6780# option to 'on'. If you have sibling relationships with caches
6781# in other administrative domains, this should be 'off'. If you only
6782# have sibling relationships with caches under your control,
6783# it is probably okay to set this to 'on'.
6784# If set to 'on', your siblings should use the option "allow-miss"
6785# on their cache_peer lines for connecting to you.
6786#Default:
6787# icp_hit_stale off
6788
6789# TAG: minimum_direct_hops
6790# If using the ICMP pinging stuff, do direct fetches for sites
6791# which are no more than this many hops away.
6792#Default:
6793# minimum_direct_hops 4
6794
6795# TAG: minimum_direct_rtt (msec)
6796# If using the ICMP pinging stuff, do direct fetches for sites
6797# which are no more than this many rtt milliseconds away.
6798#Default:
6799# minimum_direct_rtt 400
6800
6801# TAG: netdb_low
6802# The low water mark for the ICMP measurement database.
6803#
6804# Note: high watermark controlled by netdb_high directive.
6805#
6806# These watermarks are counts, not percents. The defaults are
6807# (low) 900 and (high) 1000. When the high water mark is
6808# reached, database entries will be deleted until the low
6809# mark is reached.
6810#Default:
6811# netdb_low 900
6812
6813# TAG: netdb_high
6814# The high water mark for the ICMP measurement database.
6815#
6816# Note: low watermark controlled by netdb_low directive.
6817#
6818# These watermarks are counts, not percents. The defaults are
6819# (low) 900 and (high) 1000. When the high water mark is
6820# reached, database entries will be deleted until the low
6821# mark is reached.
6822#Default:
6823# netdb_high 1000
6824
6825# TAG: netdb_ping_period
6826# The minimum period for measuring a site. There will be at
6827# least this much delay between successive pings to the same
6828# network. The default is five minutes.
6829#Default:
6830# netdb_ping_period 5 minutes
6831
6832# TAG: query_icmp on|off
6833# If you want to ask your peers to include ICMP data in their ICP
6834# replies, enable this option.
6835#
6836# If your peer has configured Squid (during compilation) with
6837# '--enable-icmp' that peer will send ICMP pings to origin server
6838# sites of the URLs it receives. If you enable this option the
6839# ICP replies from that peer will include the ICMP data (if available).
6840# Then, when choosing a parent cache, Squid will choose the parent with
6841# the minimal RTT to the origin server. When this happens, the
6842# hierarchy field of the access.log will be
6843# "CLOSEST_PARENT_MISS". This option is off by default.
6844#Default:
6845# query_icmp off
6846
6847# TAG: test_reachability on|off
6848# When this is 'on', ICP MISS replies will be ICP_MISS_NOFETCH
6849# instead of ICP_MISS if the target host is NOT in the ICMP
6850# database, or has a zero RTT.
6851#Default:
6852# test_reachability off
6853
6854# TAG: icp_query_timeout (msec)
6855# Normally Squid will automatically determine an optimal ICP
6856# query timeout value based on the round-trip-time of recent ICP
6857# queries. If you want to override the value determined by
6858# Squid, set this 'icp_query_timeout' to a non-zero value. This
6859# value is specified in MILLISECONDS, so, to use a 2-second
6860# timeout (the old default), you would write:
6861#
6862# icp_query_timeout 2000
6863#Default:
6864# Dynamic detection.
6865
6866# TAG: maximum_icp_query_timeout (msec)
6867# Normally the ICP query timeout is determined dynamically. But
6868# sometimes it can lead to very large values (say 5 seconds).
6869# Use this option to put an upper limit on the dynamic timeout
6870# value. Do NOT use this option to always use a fixed (instead
6871# of a dynamic) timeout value. To set a fixed timeout see the
6872# 'icp_query_timeout' directive.
6873#Default:
6874# maximum_icp_query_timeout 2000
6875
6876# TAG: minimum_icp_query_timeout (msec)
6877# Normally the ICP query timeout is determined dynamically. But
6878# sometimes it can lead to very small timeouts, even lower than
6879# the normal latency variance on your link due to traffic.
6880# Use this option to put an lower limit on the dynamic timeout
6881# value. Do NOT use this option to always use a fixed (instead
6882# of a dynamic) timeout value. To set a fixed timeout see the
6883# 'icp_query_timeout' directive.
6884#Default:
6885# minimum_icp_query_timeout 5
6886
6887# TAG: background_ping_rate time-units
6888# Controls how often the ICP pings are sent to siblings that
6889# have background-ping set.
6890#Default:
6891# background_ping_rate 10 seconds
6892
6893# MULTICAST ICP OPTIONS
6894# -----------------------------------------------------------------------------
6895
6896# TAG: mcast_groups
6897# This tag specifies a list of multicast groups which your server
6898# should join to receive multicasted ICP queries.
6899#
6900# NOTE! Be very careful what you put here! Be sure you
6901# understand the difference between an ICP _query_ and an ICP
6902# _reply_. This option is to be set only if you want to RECEIVE
6903# multicast queries. Do NOT set this option to SEND multicast
6904# ICP (use cache_peer for that). ICP replies are always sent via
6905# unicast, so this option does not affect whether or not you will
6906# receive replies from multicast group members.
6907#
6908# You must be very careful to NOT use a multicast address which
6909# is already in use by another group of caches.
6910#
6911# If you are unsure about multicast, please read the Multicast
6912# chapter in the Squid FAQ (http://www.squid-cache.org/FAQ/).
6913#
6914# Usage: mcast_groups 239.128.16.128 224.0.1.20
6915#
6916# By default, Squid doesn't listen on any multicast groups.
6917#Default:
6918# none
6919
6920# TAG: mcast_miss_addr
6921# Note: This option is only available if Squid is rebuilt with the
6922# -DMULTICAST_MISS_STREAM define
6923#
6924# If you enable this option, every "cache miss" URL will
6925# be sent out on the specified multicast address.
6926#
6927# Do not enable this option unless you are are absolutely
6928# certain you understand what you are doing.
6929#Default:
6930# disabled.
6931
6932# TAG: mcast_miss_ttl
6933# Note: This option is only available if Squid is rebuilt with the
6934# -DMULTICAST_MISS_STREAM define
6935#
6936# This is the time-to-live value for packets multicasted
6937# when multicasting off cache miss URLs is enabled. By
6938# default this is set to 'site scope', i.e. 16.
6939#Default:
6940# mcast_miss_ttl 16
6941
6942# TAG: mcast_miss_port
6943# Note: This option is only available if Squid is rebuilt with the
6944# -DMULTICAST_MISS_STREAM define
6945#
6946# This is the port number to be used in conjunction with
6947# 'mcast_miss_addr'.
6948#Default:
6949# mcast_miss_port 3135
6950
6951# TAG: mcast_miss_encode_key
6952# Note: This option is only available if Squid is rebuilt with the
6953# -DMULTICAST_MISS_STREAM define
6954#
6955# The URLs that are sent in the multicast miss stream are
6956# encrypted. This is the encryption key.
6957#Default:
6958# mcast_miss_encode_key XXXXXXXXXXXXXXXX
6959
6960# TAG: mcast_icp_query_timeout (msec)
6961# For multicast peers, Squid regularly sends out ICP "probes" to
6962# count how many other peers are listening on the given multicast
6963# address. This value specifies how long Squid should wait to
6964# count all the replies. The default is 2000 msec, or 2
6965# seconds.
6966#Default:
6967# mcast_icp_query_timeout 2000
6968
6969# INTERNAL ICON OPTIONS
6970# -----------------------------------------------------------------------------
6971
6972# TAG: icon_directory
6973# Where the icons are stored. These are normally kept in
6974# /usr/share/squid/icons
6975#Default:
6976# icon_directory /usr/share/squid/icons
6977
6978# TAG: global_internal_static
6979# This directive controls is Squid should intercept all requests for
6980# /squid-internal-static/ no matter which host the URL is requesting
6981# (default on setting), or if nothing special should be done for
6982# such URLs (off setting). The purpose of this directive is to make
6983# icons etc work better in complex cache hierarchies where it may
6984# not always be possible for all corners in the cache mesh to reach
6985# the server generating a directory listing.
6986#Default:
6987# global_internal_static on
6988
6989# TAG: short_icon_urls
6990# If this is enabled Squid will use short URLs for icons.
6991# If disabled it will revert to the old behavior of including
6992# it's own name and port in the URL.
6993#
6994# If you run a complex cache hierarchy with a mix of Squid and
6995# other proxies you may need to disable this directive.
6996#Default:
6997# short_icon_urls on
6998
6999# ERROR PAGE OPTIONS
7000# -----------------------------------------------------------------------------
7001
7002# TAG: error_directory
7003# If you wish to create your own versions of the default
7004# error files to customize them to suit your company copy
7005# the error/template files to another directory and point
7006# this tag at them.
7007#
7008# WARNING: This option will disable multi-language support
7009# on error pages if used.
7010#
7011# The squid developers are interested in making squid available in
7012# a wide variety of languages. If you are making translations for a
7013# language that Squid does not currently provide please consider
7014# contributing your translation back to the project.
7015# http://wiki.squid-cache.org/Translations
7016#
7017# The squid developers working on translations are happy to supply drop-in
7018# translated error files in exchange for any new language contributions.
7019#Default:
7020# Send error pages in the clients preferred language
7021
7022# TAG: error_default_language
7023# Set the default language which squid will send error pages in
7024# if no existing translation matches the clients language
7025# preferences.
7026#
7027# If unset (default) generic English will be used.
7028#
7029# The squid developers are interested in making squid available in
7030# a wide variety of languages. If you are interested in making
7031# translations for any language see the squid wiki for details.
7032# http://wiki.squid-cache.org/Translations
7033#Default:
7034# Generate English language pages.
7035
7036# TAG: error_log_languages
7037# Log to cache.log what languages users are attempting to
7038# auto-negotiate for translations.
7039#
7040# Successful negotiations are not logged. Only failures
7041# have meaning to indicate that Squid may need an upgrade
7042# of its error page translations.
7043#Default:
7044# error_log_languages on
7045
7046# TAG: err_page_stylesheet
7047# CSS Stylesheet to pattern the display of Squid default error pages.
7048#
7049# For information on CSS see http://www.w3.org/Style/CSS/
7050#Default:
7051# err_page_stylesheet /etc/squid/errorpage.css
7052
7053# TAG: err_html_text
7054# HTML text to include in error messages. Make this a "mailto"
7055# URL to your admin address, or maybe just a link to your
7056# organizations Web page.
7057#
7058# To include this in your error messages, you must rewrite
7059# the error template files (found in the "errors" directory).
7060# Wherever you want the 'err_html_text' line to appear,
7061# insert a %L tag in the error template file.
7062#Default:
7063# none
7064
7065# TAG: email_err_data on|off
7066# If enabled, information about the occurred error will be
7067# included in the mailto links of the ERR pages (if %W is set)
7068# so that the email body contains the data.
7069# Syntax is <A HREF="mailto:%w%W">%w</A>
7070#Default:
7071# email_err_data on
7072
7073# TAG: deny_info
7074# Usage: deny_info err_page_name acl
7075# or deny_info http://... acl
7076# or deny_info TCP_RESET acl
7077#
7078# This can be used to return a ERR_ page for requests which
7079# do not pass the 'http_access' rules. Squid remembers the last
7080# acl it evaluated in http_access, and if a 'deny_info' line exists
7081# for that ACL Squid returns a corresponding error page.
7082#
7083# The acl is typically the last acl on the http_access deny line which
7084# denied access. The exceptions to this rule are:
7085# - When Squid needs to request authentication credentials. It's then
7086# the first authentication related acl encountered
7087# - When none of the http_access lines matches. It's then the last
7088# acl processed on the last http_access line.
7089# - When the decision to deny access was made by an adaptation service,
7090# the acl name is the corresponding eCAP or ICAP service_name.
7091#
7092# NP: If providing your own custom error pages with error_directory
7093# you may also specify them by your custom file name:
7094# Example: deny_info ERR_CUSTOM_ACCESS_DENIED bad_guys
7095#
7096# By defaut Squid will send "403 Forbidden". A different 4xx or 5xx
7097# may be specified by prefixing the file name with the code and a colon.
7098# e.g. 404:ERR_CUSTOM_ACCESS_DENIED
7099#
7100# Alternatively you can tell Squid to reset the TCP connection
7101# by specifying TCP_RESET.
7102#
7103# Or you can specify an error URL or URL pattern. The browsers will
7104# get redirected to the specified URL after formatting tags have
7105# been replaced. Redirect will be done with 302 or 307 according to
7106# HTTP/1.1 specs. A different 3xx code may be specified by prefixing
7107# the URL. e.g. 303:http://example.com/
7108#
7109# URL FORMAT TAGS:
7110# %a - username (if available. Password NOT included)
7111# %B - FTP path URL
7112# %e - Error number
7113# %E - Error description
7114# %h - Squid hostname
7115# %H - Request domain name
7116# %i - Client IP Address
7117# %M - Request Method
7118# %O - Unescaped message result from external ACL helper
7119# %o - Message result from external ACL helper
7120# %p - Request Port number
7121# %P - Request Protocol name
7122# %R - Request URL path
7123# %T - Timestamp in RFC 1123 format
7124# %U - Full canonical URL from client
7125# (HTTPS URLs terminate with *)
7126# %u - Full canonical URL from client
7127# %w - Admin email from squid.conf
7128# %x - Error name
7129# %% - Literal percent (%) code
7130#
7131#Default:
7132# none
7133
7134# OPTIONS INFLUENCING REQUEST FORWARDING
7135# -----------------------------------------------------------------------------
7136
7137# TAG: nonhierarchical_direct
7138# By default, Squid will send any non-hierarchical requests
7139# (not cacheable request type) direct to origin servers.
7140#
7141# When this is set to "off", Squid will prefer to send these
7142# requests to parents.
7143#
7144# Note that in most configurations, by turning this off you will only
7145# add latency to these request without any improvement in global hit
7146# ratio.
7147#
7148# This option only sets a preference. If the parent is unavailable a
7149# direct connection to the origin server may still be attempted. To
7150# completely prevent direct connections use never_direct.
7151#Default:
7152# nonhierarchical_direct on
7153
7154# TAG: prefer_direct
7155# Normally Squid tries to use parents for most requests. If you for some
7156# reason like it to first try going direct and only use a parent if
7157# going direct fails set this to on.
7158#
7159# By combining nonhierarchical_direct off and prefer_direct on you
7160# can set up Squid to use a parent as a backup path if going direct
7161# fails.
7162#
7163# Note: If you want Squid to use parents for all requests see
7164# the never_direct directive. prefer_direct only modifies how Squid
7165# acts on cacheable requests.
7166#Default:
7167# prefer_direct off
7168
7169# TAG: cache_miss_revalidate on|off
7170# RFC 7232 defines a conditional request mechanism to prevent
7171# response objects being unnecessarily transferred over the network.
7172# If that mechanism is used by the client and a cache MISS occurs
7173# it can prevent new cache entries being created.
7174#
7175# This option determines whether Squid on cache MISS will pass the
7176# client revalidation request to the server or tries to fetch new
7177# content for caching. It can be useful while the cache is mostly
7178# empty to more quickly have the cache populated by generating
7179# non-conditional GETs.
7180#
7181# When set to 'on' (default), Squid will pass all client If-* headers
7182# to the server. This permits server responses without a cacheable
7183# payload to be delivered and on MISS no new cache entry is created.
7184#
7185# When set to 'off' and if the request is cacheable, Squid will
7186# remove the clients If-Modified-Since and If-None-Match headers from
7187# the request sent to the server. This requests a 200 status response
7188# from the server to create a new cache entry with.
7189#Default:
7190# cache_miss_revalidate on
7191
7192# TAG: always_direct
7193# Usage: always_direct allow|deny [!]aclname ...
7194#
7195# Here you can use ACL elements to specify requests which should
7196# ALWAYS be forwarded by Squid to the origin servers without using
7197# any peers. For example, to always directly forward requests for
7198# local servers ignoring any parents or siblings you may have use
7199# something like:
7200#
7201# acl local-servers dstdomain my.domain.net
7202# always_direct allow local-servers
7203#
7204# To always forward FTP requests directly, use
7205#
7206# acl FTP proto FTP
7207# always_direct allow FTP
7208#
7209# NOTE: There is a similar, but opposite option named
7210# 'never_direct'. You need to be aware that "always_direct deny
7211# foo" is NOT the same thing as "never_direct allow foo". You
7212# may need to use a deny rule to exclude a more-specific case of
7213# some other rule. Example:
7214#
7215# acl local-external dstdomain external.foo.net
7216# acl local-servers dstdomain .foo.net
7217# always_direct deny local-external
7218# always_direct allow local-servers
7219#
7220# NOTE: If your goal is to make the client forward the request
7221# directly to the origin server bypassing Squid then this needs
7222# to be done in the client configuration. Squid configuration
7223# can only tell Squid how Squid should fetch the object.
7224#
7225# NOTE: This directive is not related to caching. The replies
7226# is cached as usual even if you use always_direct. To not cache
7227# the replies see the 'cache' directive.
7228#
7229# This clause supports both fast and slow acl types.
7230# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
7231#Default:
7232# Prevent any cache_peer being used for this request.
7233
7234# TAG: never_direct
7235# Usage: never_direct allow|deny [!]aclname ...
7236#
7237# never_direct is the opposite of always_direct. Please read
7238# the description for always_direct if you have not already.
7239#
7240# With 'never_direct' you can use ACL elements to specify
7241# requests which should NEVER be forwarded directly to origin
7242# servers. For example, to force the use of a proxy for all
7243# requests, except those in your local domain use something like:
7244#
7245# acl local-servers dstdomain .foo.net
7246# never_direct deny local-servers
7247# never_direct allow all
7248#
7249# or if Squid is inside a firewall and there are local intranet
7250# servers inside the firewall use something like:
7251#
7252# acl local-intranet dstdomain .foo.net
7253# acl local-external dstdomain external.foo.net
7254# always_direct deny local-external
7255# always_direct allow local-intranet
7256# never_direct allow all
7257#
7258# This clause supports both fast and slow acl types.
7259# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
7260#Default:
7261# Allow DNS results to be used for this request.
7262
7263# ADVANCED NETWORKING OPTIONS
7264# -----------------------------------------------------------------------------
7265
7266# TAG: incoming_udp_average
7267# Heavy voodoo here. I can't even believe you are reading this.
7268# Are you crazy? Don't even think about adjusting these unless
7269# you understand the algorithms in comm_select.c first!
7270#Default:
7271# incoming_udp_average 6
7272
7273# TAG: incoming_tcp_average
7274# Heavy voodoo here. I can't even believe you are reading this.
7275# Are you crazy? Don't even think about adjusting these unless
7276# you understand the algorithms in comm_select.c first!
7277#Default:
7278# incoming_tcp_average 4
7279
7280# TAG: incoming_dns_average
7281# Heavy voodoo here. I can't even believe you are reading this.
7282# Are you crazy? Don't even think about adjusting these unless
7283# you understand the algorithms in comm_select.c first!
7284#Default:
7285# incoming_dns_average 4
7286
7287# TAG: min_udp_poll_cnt
7288# Heavy voodoo here. I can't even believe you are reading this.
7289# Are you crazy? Don't even think about adjusting these unless
7290# you understand the algorithms in comm_select.c first!
7291#Default:
7292# min_udp_poll_cnt 8
7293
7294# TAG: min_dns_poll_cnt
7295# Heavy voodoo here. I can't even believe you are reading this.
7296# Are you crazy? Don't even think about adjusting these unless
7297# you understand the algorithms in comm_select.c first!
7298#Default:
7299# min_dns_poll_cnt 8
7300
7301# TAG: min_tcp_poll_cnt
7302# Heavy voodoo here. I can't even believe you are reading this.
7303# Are you crazy? Don't even think about adjusting these unless
7304# you understand the algorithms in comm_select.c first!
7305#Default:
7306# min_tcp_poll_cnt 8
7307
7308# TAG: accept_filter
7309# FreeBSD:
7310#
7311# The name of an accept(2) filter to install on Squid's
7312# listen socket(s). This feature is perhaps specific to
7313# FreeBSD and requires support in the kernel.
7314#
7315# The 'httpready' filter delays delivering new connections
7316# to Squid until a full HTTP request has been received.
7317# See the accf_http(9) man page for details.
7318#
7319# The 'dataready' filter delays delivering new connections
7320# to Squid until there is some data to process.
7321# See the accf_dataready(9) man page for details.
7322#
7323# Linux:
7324#
7325# The 'data' filter delays delivering of new connections
7326# to Squid until there is some data to process by TCP_ACCEPT_DEFER.
7327# You may optionally specify a number of seconds to wait by
7328# 'data=N' where N is the number of seconds. Defaults to 30
7329# if not specified. See the tcp(7) man page for details.
7330#EXAMPLE:
7331## FreeBSD
7332#accept_filter httpready
7333## Linux
7334#accept_filter data
7335#Default:
7336# none
7337
7338# TAG: client_ip_max_connections
7339# Set an absolute limit on the number of connections a single
7340# client IP can use. Any more than this and Squid will begin to drop
7341# new connections from the client until it closes some links.
7342#
7343# Note that this is a global limit. It affects all HTTP, HTCP, Gopher and FTP
7344# connections from the client. For finer control use the ACL access controls.
7345#
7346# Requires client_db to be enabled (the default).
7347#
7348# WARNING: This may noticably slow down traffic received via external proxies
7349# or NAT devices and cause them to rebound error messages back to their clients.
7350#Default:
7351# No limit.
7352
7353# TAG: tcp_recv_bufsize (bytes)
7354# Size of receive buffer to set for TCP sockets. Probably just
7355# as easy to change your kernel's default.
7356# Omit from squid.conf to use the default buffer size.
7357#Default:
7358# Use operating system TCP defaults.
7359
7360# ICAP OPTIONS
7361# -----------------------------------------------------------------------------
7362
7363# TAG: icap_enable on|off
7364# If you want to enable the ICAP module support, set this to on.
7365#Default:
7366# icap_enable off
7367
7368# TAG: icap_connect_timeout
7369# This parameter specifies how long to wait for the TCP connect to
7370# the requested ICAP server to complete before giving up and either
7371# terminating the HTTP transaction or bypassing the failure.
7372#
7373# The default for optional services is peer_connect_timeout.
7374# The default for essential services is connect_timeout.
7375# If this option is explicitly set, its value applies to all services.
7376#Default:
7377# none
7378
7379# TAG: icap_io_timeout time-units
7380# This parameter specifies how long to wait for an I/O activity on
7381# an established, active ICAP connection before giving up and
7382# either terminating the HTTP transaction or bypassing the
7383# failure.
7384#Default:
7385# Use read_timeout.
7386
7387# TAG: icap_service_failure_limit limit [in memory-depth time-units]
7388# The limit specifies the number of failures that Squid tolerates
7389# when establishing a new TCP connection with an ICAP service. If
7390# the number of failures exceeds the limit, the ICAP service is
7391# not used for new ICAP requests until it is time to refresh its
7392# OPTIONS.
7393#
7394# A negative value disables the limit. Without the limit, an ICAP
7395# service will not be considered down due to connectivity failures
7396# between ICAP OPTIONS requests.
7397#
7398# Squid forgets ICAP service failures older than the specified
7399# value of memory-depth. The memory fading algorithm
7400# is approximate because Squid does not remember individual
7401# errors but groups them instead, splitting the option
7402# value into ten time slots of equal length.
7403#
7404# When memory-depth is 0 and by default this option has no
7405# effect on service failure expiration.
7406#
7407# Squid always forgets failures when updating service settings
7408# using an ICAP OPTIONS transaction, regardless of this option
7409# setting.
7410#
7411# For example,
7412# # suspend service usage after 10 failures in 5 seconds:
7413# icap_service_failure_limit 10 in 5 seconds
7414#Default:
7415# icap_service_failure_limit 10
7416
7417# TAG: icap_service_revival_delay
7418# The delay specifies the number of seconds to wait after an ICAP
7419# OPTIONS request failure before requesting the options again. The
7420# failed ICAP service is considered "down" until fresh OPTIONS are
7421# fetched.
7422#
7423# The actual delay cannot be smaller than the hardcoded minimum
7424# delay of 30 seconds.
7425#Default:
7426# icap_service_revival_delay 180
7427
7428# TAG: icap_preview_enable on|off
7429# The ICAP Preview feature allows the ICAP server to handle the
7430# HTTP message by looking only at the beginning of the message body
7431# or even without receiving the body at all. In some environments,
7432# previews greatly speedup ICAP processing.
7433#
7434# During an ICAP OPTIONS transaction, the server may tell Squid what
7435# HTTP messages should be previewed and how big the preview should be.
7436# Squid will not use Preview if the server did not request one.
7437#
7438# To disable ICAP Preview for all ICAP services, regardless of
7439# individual ICAP server OPTIONS responses, set this option to "off".
7440#Example:
7441#icap_preview_enable off
7442#Default:
7443# icap_preview_enable on
7444
7445# TAG: icap_preview_size
7446# The default size of preview data to be sent to the ICAP server.
7447# This value might be overwritten on a per server basis by OPTIONS requests.
7448#Default:
7449# No preview sent.
7450
7451# TAG: icap_206_enable on|off
7452# 206 (Partial Content) responses is an ICAP extension that allows the
7453# ICAP agents to optionally combine adapted and original HTTP message
7454# content. The decision to combine is postponed until the end of the
7455# ICAP response. Squid supports Partial Content extension by default.
7456#
7457# Activation of the Partial Content extension is negotiated with each
7458# ICAP service during OPTIONS exchange. Most ICAP servers should handle
7459# negotation correctly even if they do not support the extension, but
7460# some might fail. To disable Partial Content support for all ICAP
7461# services and to avoid any negotiation, set this option to "off".
7462#
7463# Example:
7464# icap_206_enable off
7465#Default:
7466# icap_206_enable on
7467
7468# TAG: icap_default_options_ttl
7469# The default TTL value for ICAP OPTIONS responses that don't have
7470# an Options-TTL header.
7471#Default:
7472# icap_default_options_ttl 60
7473
7474# TAG: icap_persistent_connections on|off
7475# Whether or not Squid should use persistent connections to
7476# an ICAP server.
7477#Default:
7478# icap_persistent_connections on
7479
7480# TAG: adaptation_send_client_ip on|off
7481# If enabled, Squid shares HTTP client IP information with adaptation
7482# services. For ICAP, Squid adds the X-Client-IP header to ICAP requests.
7483# For eCAP, Squid sets the libecap::metaClientIp transaction option.
7484#
7485# See also: adaptation_uses_indirect_client
7486#Default:
7487# adaptation_send_client_ip off
7488
7489# TAG: adaptation_send_username on|off
7490# This sends authenticated HTTP client username (if available) to
7491# the adaptation service.
7492#
7493# For ICAP, the username value is encoded based on the
7494# icap_client_username_encode option and is sent using the header
7495# specified by the icap_client_username_header option.
7496#Default:
7497# adaptation_send_username off
7498
7499# TAG: icap_client_username_header
7500# ICAP request header name to use for adaptation_send_username.
7501#Default:
7502# icap_client_username_header X-Client-Username
7503
7504# TAG: icap_client_username_encode on|off
7505# Whether to base64 encode the authenticated client username.
7506#Default:
7507# icap_client_username_encode off
7508
7509# TAG: icap_service
7510# Defines a single ICAP service using the following format:
7511#
7512# icap_service id vectoring_point uri [option ...]
7513#
7514# id: ID
7515# an opaque identifier or name which is used to direct traffic to
7516# this specific service. Must be unique among all adaptation
7517# services in squid.conf.
7518#
7519# vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
7520# This specifies at which point of transaction processing the
7521# ICAP service should be activated. *_postcache vectoring points
7522# are not yet supported.
7523#
7524# uri: icap://servername:port/servicepath
7525# ICAP server and service location.
7526# icaps://servername:port/servicepath
7527# The "icap:" URI scheme is used for traditional ICAP server and
7528# service location (default port is 1344, connections are not
7529# encrypted). The "icaps:" URI scheme is for Secure ICAP
7530# services that use SSL/TLS-encrypted ICAP connections (by
7531# default, on port 11344).
7532#
7533# ICAP does not allow a single service to handle both REQMOD and RESPMOD
7534# transactions. Squid does not enforce that requirement. You can specify
7535# services with the same service_url and different vectoring_points. You
7536# can even specify multiple identical services as long as their
7537# service_names differ.
7538#
7539# To activate a service, use the adaptation_access directive. To group
7540# services, use adaptation_service_chain and adaptation_service_set.
7541#
7542# Service options are separated by white space. ICAP services support
7543# the following name=value options:
7544#
7545# bypass=on|off|1|0
7546# If set to 'on' or '1', the ICAP service is treated as
7547# optional. If the service cannot be reached or malfunctions,
7548# Squid will try to ignore any errors and process the message as
7549# if the service was not enabled. No all ICAP errors can be
7550# bypassed. If set to 0, the ICAP service is treated as
7551# essential and all ICAP errors will result in an error page
7552# returned to the HTTP client.
7553#
7554# Bypass is off by default: services are treated as essential.
7555#
7556# routing=on|off|1|0
7557# If set to 'on' or '1', the ICAP service is allowed to
7558# dynamically change the current message adaptation plan by
7559# returning a chain of services to be used next. The services
7560# are specified using the X-Next-Services ICAP response header
7561# value, formatted as a comma-separated list of service names.
7562# Each named service should be configured in squid.conf. Other
7563# services are ignored. An empty X-Next-Services value results
7564# in an empty plan which ends the current adaptation.
7565#
7566# Dynamic adaptation plan may cross or cover multiple supported
7567# vectoring points in their natural processing order.
7568#
7569# Routing is not allowed by default: the ICAP X-Next-Services
7570# response header is ignored.
7571#
7572# ipv6=on|off
7573# Only has effect on split-stack systems. The default on those systems
7574# is to use IPv4-only connections. When set to 'on' this option will
7575# make Squid use IPv6-only connections to contact this ICAP service.
7576#
7577# on-overload=block|bypass|wait|force
7578# If the service Max-Connections limit has been reached, do
7579# one of the following for each new ICAP transaction:
7580# * block: send an HTTP error response to the client
7581# * bypass: ignore the "over-connected" ICAP service
7582# * wait: wait (in a FIFO queue) for an ICAP connection slot
7583# * force: proceed, ignoring the Max-Connections limit
7584#
7585# In SMP mode with N workers, each worker assumes the service
7586# connection limit is Max-Connections/N, even though not all
7587# workers may use a given service.
7588#
7589# The default value is "bypass" if service is bypassable,
7590# otherwise it is set to "wait".
7591#
7592#
7593# max-conn=number
7594# Use the given number as the Max-Connections limit, regardless
7595# of the Max-Connections value given by the service, if any.
7596#
7597# connection-encryption=on|off
7598# Determines the ICAP service effect on the connections_encrypted
7599# ACL.
7600#
7601# The default is "on" for Secure ICAP services (i.e., those
7602# with the icaps:// service URIs scheme) and "off" for plain ICAP
7603# services.
7604#
7605# Does not affect ICAP connections (e.g., does not turn Secure
7606# ICAP on or off).
7607#
7608# ==== ICAPS / TLS OPTIONS ====
7609#
7610# These options are used for Secure ICAP (icaps://....) services only.
7611#
7612# tls-cert=/path/to/ssl/certificate
7613# A client X.509 certificate to use when connecting to
7614# this ICAP server.
7615#
7616# tls-key=/path/to/ssl/key
7617# The private key corresponding to the previous
7618# tls-cert= option.
7619#
7620# If tls-key= is not specified tls-cert= is assumed to
7621# reference a PEM file containing both the certificate
7622# and private key.
7623#
7624# tls-cipher=... The list of valid TLS/SSL ciphers to use when connecting
7625# to this icap server.
7626#
7627# tls-min-version=1.N
7628# The minimum TLS protocol version to permit. To control
7629# SSLv3 use the tls-options= parameter.
7630# Supported Values: 1.0 (default), 1.1, 1.2
7631#
7632# tls-options=... Specify various OpenSSL library options:
7633#
7634# NO_SSLv3 Disallow the use of SSLv3
7635#
7636# SINGLE_DH_USE
7637# Always create a new key when using
7638# temporary/ephemeral DH key exchanges
7639#
7640# ALL Enable various bug workarounds
7641# suggested as "harmless" by OpenSSL
7642# Be warned that this reduces SSL/TLS
7643# strength to some attacks.
7644#
7645# See the OpenSSL SSL_CTX_set_options documentation for a
7646# more complete list. Options relevant only to SSLv2 are
7647# not supported.
7648#
7649# tls-cafile= PEM file containing CA certificates to use when verifying
7650# the icap server certificate.
7651# Use to specify intermediate CA certificate(s) if not sent
7652# by the server. Or the full CA chain for the server when
7653# using the tls-default-ca=off flag.
7654# May be repeated to load multiple files.
7655#
7656# tls-capath=... A directory containing additional CA certificates to
7657# use when verifying the icap server certificate.
7658# Requires OpenSSL or LibreSSL.
7659#
7660# tls-crlfile=... A certificate revocation list file to use when
7661# verifying the icap server certificate.
7662#
7663# tls-flags=... Specify various flags modifying the Squid TLS implementation:
7664#
7665# DONT_VERIFY_PEER
7666# Accept certificates even if they fail to
7667# verify.
7668# DONT_VERIFY_DOMAIN
7669# Don't verify the icap server certificate
7670# matches the server name
7671#
7672# tls-default-ca[=off]
7673# Whether to use the system Trusted CAs. Default is ON.
7674#
7675# tls-domain= The icap server name as advertised in it's certificate.
7676# Used for verifying the correctness of the received icap
7677# server certificate. If not specified the icap server
7678# hostname extracted from ICAP URI will be used.
7679#
7680# Older icap_service format without optional named parameters is
7681# deprecated but supported for backward compatibility.
7682#
7683#Example:
7684#icap_service svcBlocker reqmod_precache icap://icap1.mydomain.net:1344/reqmod bypass=0
7685#icap_service svcLogger reqmod_precache icaps://icap2.mydomain.net:11344/reqmod routing=on
7686#Default:
7687# none
7688
7689# TAG: icap_class
7690# This deprecated option was documented to define an ICAP service
7691# chain, even though it actually defined a set of similar, redundant
7692# services, and the chains were not supported.
7693#
7694# To define a set of redundant services, please use the
7695# adaptation_service_set directive. For service chains, use
7696# adaptation_service_chain.
7697#Default:
7698# none
7699
7700# TAG: icap_access
7701# This option is deprecated. Please use adaptation_access, which
7702# has the same ICAP functionality, but comes with better
7703# documentation, and eCAP support.
7704#Default:
7705# none
7706
7707# eCAP OPTIONS
7708# -----------------------------------------------------------------------------
7709
7710# TAG: ecap_enable on|off
7711# Controls whether eCAP support is enabled.
7712#Default:
7713# ecap_enable off
7714
7715# TAG: ecap_service
7716# Defines a single eCAP service
7717#
7718# ecap_service id vectoring_point uri [option ...]
7719#
7720# id: ID
7721# an opaque identifier or name which is used to direct traffic to
7722# this specific service. Must be unique among all adaptation
7723# services in squid.conf.
7724#
7725# vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
7726# This specifies at which point of transaction processing the
7727# eCAP service should be activated. *_postcache vectoring points
7728# are not yet supported.
7729#
7730# uri: ecap://vendor/service_name?custom&cgi=style¶meters=optional
7731# Squid uses the eCAP service URI to match this configuration
7732# line with one of the dynamically loaded services. Each loaded
7733# eCAP service must have a unique URI. Obtain the right URI from
7734# the service provider.
7735#
7736# To activate a service, use the adaptation_access directive. To group
7737# services, use adaptation_service_chain and adaptation_service_set.
7738#
7739# Service options are separated by white space. eCAP services support
7740# the following name=value options:
7741#
7742# bypass=on|off|1|0
7743# If set to 'on' or '1', the eCAP service is treated as optional.
7744# If the service cannot be reached or malfunctions, Squid will try
7745# to ignore any errors and process the message as if the service
7746# was not enabled. No all eCAP errors can be bypassed.
7747# If set to 'off' or '0', the eCAP service is treated as essential
7748# and all eCAP errors will result in an error page returned to the
7749# HTTP client.
7750#
7751# Bypass is off by default: services are treated as essential.
7752#
7753# routing=on|off|1|0
7754# If set to 'on' or '1', the eCAP service is allowed to
7755# dynamically change the current message adaptation plan by
7756# returning a chain of services to be used next.
7757#
7758# Dynamic adaptation plan may cross or cover multiple supported
7759# vectoring points in their natural processing order.
7760#
7761# Routing is not allowed by default.
7762#
7763# connection-encryption=on|off
7764# Determines the eCAP service effect on the connections_encrypted
7765# ACL.
7766#
7767# Defaults to "on", which does not taint the master transaction
7768# w.r.t. that ACL.
7769#
7770# Does not affect eCAP API calls.
7771#
7772# Older ecap_service format without optional named parameters is
7773# deprecated but supported for backward compatibility.
7774#
7775#
7776#Example:
7777#ecap_service s1 reqmod_precache ecap://filters.R.us/leakDetector?on_error=block bypass=off
7778#ecap_service s2 respmod_precache ecap://filters.R.us/virusFilter config=/etc/vf.cfg bypass=on
7779#Default:
7780# none
7781
7782# TAG: loadable_modules
7783# Instructs Squid to load the specified dynamic module(s) or activate
7784# preloaded module(s).
7785#Example:
7786#loadable_modules /usr/lib/MinimalAdapter.so
7787#Default:
7788# none
7789
7790# MESSAGE ADAPTATION OPTIONS
7791# -----------------------------------------------------------------------------
7792
7793# TAG: adaptation_service_set
7794#
7795# Configures an ordered set of similar, redundant services. This is
7796# useful when hot standby or backup adaptation servers are available.
7797#
7798# adaptation_service_set set_name service_name1 service_name2 ...
7799#
7800# The named services are used in the set declaration order. The first
7801# applicable adaptation service from the set is used first. The next
7802# applicable service is tried if and only if the transaction with the
7803# previous service fails and the message waiting to be adapted is still
7804# intact.
7805#
7806# When adaptation starts, broken services are ignored as if they were
7807# not a part of the set. A broken service is a down optional service.
7808#
7809# The services in a set must be attached to the same vectoring point
7810# (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
7811#
7812# If all services in a set are optional then adaptation failures are
7813# bypassable. If all services in the set are essential, then a
7814# transaction failure with one service may still be retried using
7815# another service from the set, but when all services fail, the master
7816# transaction fails as well.
7817#
7818# A set may contain a mix of optional and essential services, but that
7819# is likely to lead to surprising results because broken services become
7820# ignored (see above), making previously bypassable failures fatal.
7821# Technically, it is the bypassability of the last failed service that
7822# matters.
7823#
7824# See also: adaptation_access adaptation_service_chain
7825#
7826#Example:
7827#adaptation_service_set svcBlocker urlFilterPrimary urlFilterBackup
7828#adaptation service_set svcLogger loggerLocal loggerRemote
7829#Default:
7830# none
7831
7832# TAG: adaptation_service_chain
7833#
7834# Configures a list of complementary services that will be applied
7835# one-by-one, forming an adaptation chain or pipeline. This is useful
7836# when Squid must perform different adaptations on the same message.
7837#
7838# adaptation_service_chain chain_name service_name1 svc_name2 ...
7839#
7840# The named services are used in the chain declaration order. The first
7841# applicable adaptation service from the chain is used first. The next
7842# applicable service is applied to the successful adaptation results of
7843# the previous service in the chain.
7844#
7845# When adaptation starts, broken services are ignored as if they were
7846# not a part of the chain. A broken service is a down optional service.
7847#
7848# Request satisfaction terminates the adaptation chain because Squid
7849# does not currently allow declaration of RESPMOD services at the
7850# "reqmod_precache" vectoring point (see icap_service or ecap_service).
7851#
7852# The services in a chain must be attached to the same vectoring point
7853# (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
7854#
7855# A chain may contain a mix of optional and essential services. If an
7856# essential adaptation fails (or the failure cannot be bypassed for
7857# other reasons), the master transaction fails. Otherwise, the failure
7858# is bypassed as if the failed adaptation service was not in the chain.
7859#
7860# See also: adaptation_access adaptation_service_set
7861#
7862#Example:
7863#adaptation_service_chain svcRequest requestLogger urlFilter leakDetector
7864#Default:
7865# none
7866
7867# TAG: adaptation_access
7868# Sends an HTTP transaction to an ICAP or eCAP adaptation service.
7869#
7870# adaptation_access service_name allow|deny [!]aclname...
7871# adaptation_access set_name allow|deny [!]aclname...
7872#
7873# At each supported vectoring point, the adaptation_access
7874# statements are processed in the order they appear in this
7875# configuration file. Statements pointing to the following services
7876# are ignored (i.e., skipped without checking their ACL):
7877#
7878# - services serving different vectoring points
7879# - "broken-but-bypassable" services
7880# - "up" services configured to ignore such transactions
7881# (e.g., based on the ICAP Transfer-Ignore header).
7882#
7883# When a set_name is used, all services in the set are checked
7884# using the same rules, to find the first applicable one. See
7885# adaptation_service_set for details.
7886#
7887# If an access list is checked and there is a match, the
7888# processing stops: For an "allow" rule, the corresponding
7889# adaptation service is used for the transaction. For a "deny"
7890# rule, no adaptation service is activated.
7891#
7892# It is currently not possible to apply more than one adaptation
7893# service at the same vectoring point to the same HTTP transaction.
7894#
7895# See also: icap_service and ecap_service
7896#
7897#Example:
7898#adaptation_access service_1 allow all
7899#Default:
7900# Allow, unless rules exist in squid.conf.
7901
7902# TAG: adaptation_service_iteration_limit
7903# Limits the number of iterations allowed when applying adaptation
7904# services to a message. If your longest adaptation set or chain
7905# may have more than 16 services, increase the limit beyond its
7906# default value of 16. If detecting infinite iteration loops sooner
7907# is critical, make the iteration limit match the actual number
7908# of services in your longest adaptation set or chain.
7909#
7910# Infinite adaptation loops are most likely with routing services.
7911#
7912# See also: icap_service routing=1
7913#Default:
7914# adaptation_service_iteration_limit 16
7915
7916# TAG: adaptation_masterx_shared_names
7917# For each master transaction (i.e., the HTTP request and response
7918# sequence, including all related ICAP and eCAP exchanges), Squid
7919# maintains a table of metadata. The table entries are (name, value)
7920# pairs shared among eCAP and ICAP exchanges. The table is destroyed
7921# with the master transaction.
7922#
7923# This option specifies the table entry names that Squid must accept
7924# from and forward to the adaptation transactions.
7925#
7926# An ICAP REQMOD or RESPMOD transaction may set an entry in the
7927# shared table by returning an ICAP header field with a name
7928# specified in adaptation_masterx_shared_names.
7929#
7930# An eCAP REQMOD or RESPMOD transaction may set an entry in the
7931# shared table by implementing the libecap::visitEachOption() API
7932# to provide an option with a name specified in
7933# adaptation_masterx_shared_names.
7934#
7935# Squid will store and forward the set entry to subsequent adaptation
7936# transactions within the same master transaction scope.
7937#
7938# Only one shared entry name is supported at this time.
7939#
7940#Example:
7941## share authentication information among ICAP services
7942#adaptation_masterx_shared_names X-Subscriber-ID
7943#Default:
7944# none
7945
7946# TAG: adaptation_meta
7947# This option allows Squid administrator to add custom ICAP request
7948# headers or eCAP options to Squid ICAP requests or eCAP transactions.
7949# Use it to pass custom authentication tokens and other
7950# transaction-state related meta information to an ICAP/eCAP service.
7951#
7952# The addition of a meta header is ACL-driven:
7953# adaptation_meta name value [!]aclname ...
7954#
7955# Processing for a given header name stops after the first ACL list match.
7956# Thus, it is impossible to add two headers with the same name. If no ACL
7957# lists match for a given header name, no such header is added. For
7958# example:
7959#
7960# # do not debug transactions except for those that need debugging
7961# adaptation_meta X-Debug 1 needs_debugging
7962#
7963# # log all transactions except for those that must remain secret
7964# adaptation_meta X-Log 1 !keep_secret
7965#
7966# # mark transactions from users in the "G 1" group
7967# adaptation_meta X-Authenticated-Groups "G 1" authed_as_G1
7968#
7969# The "value" parameter may be a regular squid.conf token or a "double
7970# quoted string". Within the quoted string, use backslash (\) to escape
7971# any character, which is currently only useful for escaping backslashes
7972# and double quotes. For example,
7973# "this string has one backslash (\\) and two \"quotes\""
7974#
7975# Used adaptation_meta header values may be logged via %note
7976# logformat code. If multiple adaptation_meta headers with the same name
7977# are used during master transaction lifetime, the header values are
7978# logged in the order they were used and duplicate values are ignored
7979# (only the first repeated value will be logged).
7980#Default:
7981# none
7982
7983# TAG: icap_retry
7984# This ACL determines which retriable ICAP transactions are
7985# retried. Transactions that received a complete ICAP response
7986# and did not have to consume or produce HTTP bodies to receive
7987# that response are usually retriable.
7988#
7989# icap_retry allow|deny [!]aclname ...
7990#
7991# Squid automatically retries some ICAP I/O timeouts and errors
7992# due to persistent connection race conditions.
7993#
7994# See also: icap_retry_limit
7995#Default:
7996# icap_retry deny all
7997
7998# TAG: icap_retry_limit
7999# Limits the number of retries allowed.
8000#
8001# Communication errors due to persistent connection race
8002# conditions are unavoidable, automatically retried, and do not
8003# count against this limit.
8004#
8005# See also: icap_retry
8006#Default:
8007# No retries are allowed.
8008
8009# DNS OPTIONS
8010# -----------------------------------------------------------------------------
8011
8012# TAG: check_hostnames
8013# For security and stability reasons Squid can check
8014# hostnames for Internet standard RFC compliance. If you want
8015# Squid to perform these checks turn this directive on.
8016#Default:
8017# check_hostnames off
8018
8019# TAG: allow_underscore
8020# Underscore characters is not strictly allowed in Internet hostnames
8021# but nevertheless used by many sites. Set this to off if you want
8022# Squid to be strict about the standard.
8023# This check is performed only when check_hostnames is set to on.
8024#Default:
8025# allow_underscore on
8026
8027# TAG: dns_retransmit_interval
8028# Initial retransmit interval for DNS queries. The interval is
8029# doubled each time all configured DNS servers have been tried.
8030#Default:
8031# dns_retransmit_interval 5 seconds
8032
8033# TAG: dns_timeout
8034# DNS Query timeout. If no response is received to a DNS query
8035# within this time all DNS servers for the queried domain
8036# are assumed to be unavailable.
8037#Default:
8038# dns_timeout 30 seconds
8039
8040# TAG: dns_packet_max
8041# Maximum number of bytes packet size to advertise via EDNS.
8042# Set to "none" to disable EDNS large packet support.
8043#
8044# For legacy reasons DNS UDP replies will default to 512 bytes which
8045# is too small for many responses. EDNS provides a means for Squid to
8046# negotiate receiving larger responses back immediately without having
8047# to failover with repeat requests. Responses larger than this limit
8048# will retain the old behaviour of failover to TCP DNS.
8049#
8050# Squid has no real fixed limit internally, but allowing packet sizes
8051# over 1500 bytes requires network jumbogram support and is usually not
8052# necessary.
8053#
8054# WARNING: The RFC also indicates that some older resolvers will reply
8055# with failure of the whole request if the extension is added. Some
8056# resolvers have already been identified which will reply with mangled
8057# EDNS response on occasion. Usually in response to many-KB jumbogram
8058# sizes being advertised by Squid.
8059# Squid will currently treat these both as an unable-to-resolve domain
8060# even if it would be resolvable without EDNS.
8061#Default:
8062# EDNS disabled
8063
8064# TAG: dns_defnames on|off
8065# Normally the RES_DEFNAMES resolver option is disabled
8066# (see res_init(3)). This prevents caches in a hierarchy
8067# from interpreting single-component hostnames locally. To allow
8068# Squid to handle single-component names, enable this option.
8069#Default:
8070# Search for single-label domain names is disabled.
8071
8072# TAG: dns_multicast_local on|off
8073# When set to on, Squid sends multicast DNS lookups on the local
8074# network for domains ending in .local and .arpa.
8075# This enables local servers and devices to be contacted in an
8076# ad-hoc or zero-configuration network environment.
8077#Default:
8078# Search for .local and .arpa names is disabled.
8079
8080# TAG: dns_nameservers
8081# Use this if you want to specify a list of DNS name servers
8082# (IP addresses) to use instead of those given in your
8083# /etc/resolv.conf file.
8084#
8085# On Windows platforms, if no value is specified here or in
8086# the /etc/resolv.conf file, the list of DNS name servers are
8087# taken from the Windows registry, both static and dynamic DHCP
8088# configurations are supported.
8089#
8090# Example: dns_nameservers 10.0.0.1 192.172.0.4
8091#Default:
8092# Use operating system definitions
8093
8094# TAG: hosts_file
8095# Location of the host-local IP name-address associations
8096# database. Most Operating Systems have such a file on different
8097# default locations:
8098# - Un*X & Linux: /etc/hosts
8099# - Windows NT/2000: %SystemRoot%\system32\drivers\etc\hosts
8100# (%SystemRoot% value install default is c:\winnt)
8101# - Windows XP/2003: %SystemRoot%\system32\drivers\etc\hosts
8102# (%SystemRoot% value install default is c:\windows)
8103# - Windows 9x/Me: %windir%\hosts
8104# (%windir% value is usually c:\windows)
8105# - Cygwin: /etc/hosts
8106#
8107# The file contains newline-separated definitions, in the
8108# form ip_address_in_dotted_form name [name ...] names are
8109# whitespace-separated. Lines beginning with an hash (#)
8110# character are comments.
8111#
8112# The file is checked at startup and upon configuration.
8113# If set to 'none', it won't be checked.
8114# If append_domain is used, that domain will be added to
8115# domain-local (i.e. not containing any dot character) host
8116# definitions.
8117#Default:
8118# hosts_file /etc/hosts
8119
8120# TAG: append_domain
8121# Appends local domain name to hostnames without any dots in
8122# them. append_domain must begin with a period.
8123#
8124# Be warned there are now Internet names with no dots in
8125# them using only top-domain names, so setting this may
8126# cause some Internet sites to become unavailable.
8127#
8128#Example:
8129# append_domain .yourdomain.com
8130#Default:
8131# Use operating system definitions
8132
8133# TAG: ignore_unknown_nameservers
8134# By default Squid checks that DNS responses are received
8135# from the same IP addresses they are sent to. If they
8136# don't match, Squid ignores the response and writes a warning
8137# message to cache.log. You can allow responses from unknown
8138# nameservers by setting this option to 'off'.
8139#Default:
8140# ignore_unknown_nameservers on
8141
8142# TAG: dns_v4_first
8143# With the IPv6 Internet being as fast or faster than IPv4 Internet
8144# for most networks Squid prefers to contact websites over IPv6.
8145#
8146# This option reverses the order of preference to make Squid contact
8147# dual-stack websites over IPv4 first. Squid will still perform both
8148# IPv6 and IPv4 DNS lookups before connecting.
8149#
8150# WARNING:
8151# This option will restrict the situations under which IPv6
8152# connectivity is used (and tested), potentially hiding network
8153# problems which would otherwise be detected and warned about.
8154#Default:
8155# dns_v4_first off
8156
8157# TAG: ipcache_size (number of entries)
8158# Maximum number of DNS IP cache entries.
8159#Default:
8160# ipcache_size 1024
8161
8162# TAG: ipcache_low (percent)
8163#Default:
8164# ipcache_low 90
8165
8166# TAG: ipcache_high (percent)
8167# The size, low-, and high-water marks for the IP cache.
8168#Default:
8169# ipcache_high 95
8170
8171# TAG: fqdncache_size (number of entries)
8172# Maximum number of FQDN cache entries.
8173#Default:
8174# fqdncache_size 1024
8175
8176# MISCELLANEOUS
8177# -----------------------------------------------------------------------------
8178
8179# TAG: configuration_includes_quoted_values on|off
8180# If set, Squid will recognize each "quoted string" after a configuration
8181# directive as a single parameter. The quotes are stripped before the
8182# parameter value is interpreted or used.
8183# See "Values with spaces, quotes, and other special characters"
8184# section for more details.
8185#Default:
8186# configuration_includes_quoted_values off
8187
8188# TAG: memory_pools on|off
8189# If set, Squid will keep pools of allocated (but unused) memory
8190# available for future use. If memory is a premium on your
8191# system and you believe your malloc library outperforms Squid
8192# routines, disable this.
8193#Default:
8194# memory_pools on
8195
8196# TAG: memory_pools_limit (bytes)
8197# Used only with memory_pools on:
8198# memory_pools_limit 50 MB
8199#
8200# If set to a non-zero value, Squid will keep at most the specified
8201# limit of allocated (but unused) memory in memory pools. All free()
8202# requests that exceed this limit will be handled by your malloc
8203# library. Squid does not pre-allocate any memory, just safe-keeps
8204# objects that otherwise would be free()d. Thus, it is safe to set
8205# memory_pools_limit to a reasonably high value even if your
8206# configuration will use less memory.
8207#
8208# If set to none, Squid will keep all memory it can. That is, there
8209# will be no limit on the total amount of memory used for safe-keeping.
8210#
8211# To disable memory allocation optimization, do not set
8212# memory_pools_limit to 0 or none. Set memory_pools to "off" instead.
8213#
8214# An overhead for maintaining memory pools is not taken into account
8215# when the limit is checked. This overhead is close to four bytes per
8216# object kept. However, pools may actually _save_ memory because of
8217# reduced memory thrashing in your malloc library.
8218#Default:
8219# memory_pools_limit 5 MB
8220
8221# TAG: forwarded_for on|off|transparent|truncate|delete
8222# If set to "on", Squid will append your client's IP address
8223# in the HTTP requests it forwards. By default it looks like:
8224#
8225# X-Forwarded-For: 192.1.2.3
8226#
8227# If set to "off", it will appear as
8228#
8229# X-Forwarded-For: unknown
8230#
8231# If set to "transparent", Squid will not alter the
8232# X-Forwarded-For header in any way.
8233#
8234# If set to "delete", Squid will delete the entire
8235# X-Forwarded-For header.
8236#
8237# If set to "truncate", Squid will remove all existing
8238# X-Forwarded-For entries, and place the client IP as the sole entry.
8239#Default:
8240# forwarded_for on
8241
8242# TAG: cachemgr_passwd
8243# Specify passwords for cachemgr operations.
8244#
8245# Usage: cachemgr_passwd password action action ...
8246#
8247# Some valid actions are (see cache manager menu for a full list):
8248# 5min
8249# 60min
8250# asndb
8251# authenticator
8252# cbdata
8253# client_list
8254# comm_incoming
8255# config *
8256# counters
8257# delay
8258# digest_stats
8259# dns
8260# events
8261# filedescriptors
8262# fqdncache
8263# histograms
8264# http_headers
8265# info
8266# io
8267# ipcache
8268# mem
8269# menu
8270# netdb
8271# non_peers
8272# objects
8273# offline_toggle *
8274# pconn
8275# peer_select
8276# reconfigure *
8277# redirector
8278# refresh
8279# server_list
8280# shutdown *
8281# store_digest
8282# storedir
8283# utilization
8284# via_headers
8285# vm_objects
8286#
8287# * Indicates actions which will not be performed without a
8288# valid password, others can be performed if not listed here.
8289#
8290# To disable an action, set the password to "disable".
8291# To allow performing an action without a password, set the
8292# password to "none".
8293#
8294# Use the keyword "all" to set the same password for all actions.
8295#
8296#Example:
8297# cachemgr_passwd secret shutdown
8298# cachemgr_passwd lesssssssecret info stats/objects
8299# cachemgr_passwd disable all
8300#Default:
8301# No password. Actions which require password are denied.
8302
8303# TAG: client_db on|off
8304# If you want to disable collecting per-client statistics,
8305# turn off client_db here.
8306#Default:
8307# client_db on
8308
8309# TAG: refresh_all_ims on|off
8310# When you enable this option, squid will always check
8311# the origin server for an update when a client sends an
8312# If-Modified-Since request. Many browsers use IMS
8313# requests when the user requests a reload, and this
8314# ensures those clients receive the latest version.
8315#
8316# By default (off), squid may return a Not Modified response
8317# based on the age of the cached version.
8318#Default:
8319# refresh_all_ims off
8320
8321# TAG: reload_into_ims on|off
8322# When you enable this option, client no-cache or ``reload''
8323# requests will be changed to If-Modified-Since requests.
8324# Doing this VIOLATES the HTTP standard. Enabling this
8325# feature could make you liable for problems which it
8326# causes.
8327#
8328# see also refresh_pattern for a more selective approach.
8329#Default:
8330# reload_into_ims off
8331
8332# TAG: connect_retries
8333# Limits the number of reopening attempts when establishing a single
8334# TCP connection. All these attempts must still complete before the
8335# applicable connection opening timeout expires.
8336#
8337# By default and when connect_retries is set to zero, Squid does not
8338# retry failed connection opening attempts.
8339#
8340# The (not recommended) maximum is 10 tries. An attempt to configure a
8341# higher value results in the value of 10 being used (with a warning).
8342#
8343# Squid may open connections to retry various high-level forwarding
8344# failures. For an outside observer, that activity may look like a
8345# low-level connection reopening attempt, but those high-level retries
8346# are governed by forward_max_tries instead.
8347#
8348# See also: connect_timeout, forward_timeout, icap_connect_timeout,
8349# ident_timeout, and forward_max_tries.
8350#Default:
8351# Do not retry failed connections.
8352
8353# TAG: retry_on_error
8354# If set to ON Squid will automatically retry requests when
8355# receiving an error response with status 403 (Forbidden),
8356# 500 (Internal Error), 501 or 503 (Service not available).
8357# Status 502 and 504 (Gateway errors) are always retried.
8358#
8359# This is mainly useful if you are in a complex cache hierarchy to
8360# work around access control errors.
8361#
8362# NOTE: This retry will attempt to find another working destination.
8363# Which is different from the server which just failed.
8364#Default:
8365# retry_on_error off
8366
8367# TAG: as_whois_server
8368# WHOIS server to query for AS numbers. NOTE: AS numbers are
8369# queried only when Squid starts up, not for every request.
8370#Default:
8371# as_whois_server whois.ra.net
8372
8373# TAG: offline_mode
8374# Enable this option and Squid will never try to validate cached
8375# objects.
8376#Default:
8377# offline_mode off
8378
8379# TAG: uri_whitespace
8380# What to do with requests that have whitespace characters in the
8381# URI. Options:
8382#
8383# strip: The whitespace characters are stripped out of the URL.
8384# This is the behavior recommended by RFC2396 and RFC3986
8385# for tolerant handling of generic URI.
8386# NOTE: This is one difference between generic URI and HTTP URLs.
8387#
8388# deny: The request is denied. The user receives an "Invalid
8389# Request" message.
8390# This is the behaviour recommended by RFC2616 for safe
8391# handling of HTTP request URL.
8392#
8393# allow: The request is allowed and the URI is not changed. The
8394# whitespace characters remain in the URI. Note the
8395# whitespace is passed to redirector processes if they
8396# are in use.
8397# Note this may be considered a violation of RFC2616
8398# request parsing where whitespace is prohibited in the
8399# URL field.
8400#
8401# encode: The request is allowed and the whitespace characters are
8402# encoded according to RFC1738.
8403#
8404# chop: The request is allowed and the URI is chopped at the
8405# first whitespace.
8406#
8407#
8408# NOTE the current Squid implementation of encode and chop violates
8409# RFC2616 by not using a 301 redirect after altering the URL.
8410#Default:
8411# uri_whitespace strip
8412
8413# TAG: chroot
8414# Specifies a directory where Squid should do a chroot() while
8415# initializing. This also causes Squid to fully drop root
8416# privileges after initializing. This means, for example, if you
8417# use a HTTP port less than 1024 and try to reconfigure, you may
8418# get an error saying that Squid can not open the port.
8419#Default:
8420# none
8421
8422# TAG: pipeline_prefetch
8423# HTTP clients may send a pipeline of 1+N requests to Squid using a
8424# single connection, without waiting for Squid to respond to the first
8425# of those requests. This option limits the number of concurrent
8426# requests Squid will try to handle in parallel. If set to N, Squid
8427# will try to receive and process up to 1+N requests on the same
8428# connection concurrently.
8429#
8430# Defaults to 0 (off) for bandwidth management and access logging
8431# reasons.
8432#
8433# NOTE: pipelining requires persistent connections to clients.
8434#
8435# WARNING: pipelining breaks NTLM and Negotiate/Kerberos authentication.
8436#Default:
8437# Do not pre-parse pipelined requests.
8438
8439# TAG: high_response_time_warning (msec)
8440# If the one-minute median response time exceeds this value,
8441# Squid prints a WARNING with debug level 0 to get the
8442# administrators attention. The value is in milliseconds.
8443#Default:
8444# disabled.
8445
8446# TAG: high_page_fault_warning
8447# If the one-minute average page fault rate exceeds this
8448# value, Squid prints a WARNING with debug level 0 to get
8449# the administrators attention. The value is in page faults
8450# per second.
8451#Default:
8452# disabled.
8453
8454# TAG: high_memory_warning
8455# Note: This option is only available if Squid is rebuilt with the
8456# GNU Malloc with mstats()
8457#
8458# If the memory usage (as determined by gnumalloc, if available and used)
8459# exceeds this amount, Squid prints a WARNING with debug level 0 to get
8460# the administrators attention.
8461#Default:
8462# disabled.
8463
8464# TAG: sleep_after_fork (microseconds)
8465# When this is set to a non-zero value, the main Squid process
8466# sleeps the specified number of microseconds after a fork()
8467# system call. This sleep may help the situation where your
8468# system reports fork() failures due to lack of (virtual)
8469# memory. Note, however, if you have a lot of child
8470# processes, these sleep delays will add up and your
8471# Squid will not service requests for some amount of time
8472# until all the child processes have been started.
8473# On Windows value less then 1000 (1 milliseconds) are
8474# rounded to 1000.
8475#Default:
8476# sleep_after_fork 0
8477
8478# TAG: windows_ipaddrchangemonitor on|off
8479# Note: This option is only available if Squid is rebuilt with the
8480# MS Windows
8481#
8482# On Windows Squid by default will monitor IP address changes and will
8483# reconfigure itself after any detected event. This is very useful for
8484# proxies connected to internet with dial-up interfaces.
8485# In some cases (a Proxy server acting as VPN gateway is one) it could be
8486# desiderable to disable this behaviour setting this to 'off'.
8487# Note: after changing this, Squid service must be restarted.
8488#Default:
8489# windows_ipaddrchangemonitor on
8490
8491# TAG: eui_lookup
8492# Whether to lookup the EUI or MAC address of a connected client.
8493#Default:
8494# eui_lookup on
8495
8496# TAG: max_filedescriptors
8497# Reduce the maximum number of filedescriptors supported below
8498# the usual operating system defaults.
8499#
8500# Remove from squid.conf to inherit the current ulimit setting.
8501#
8502# Note: Changing this requires a restart of Squid. Also
8503# not all I/O types supports large values (eg on Windows).
8504#Default:
8505# Use operating system limits set by ulimit.
8506
8507# TAG: force_request_body_continuation
8508# This option controls how Squid handles data upload requests from HTTP
8509# and FTP agents that require a "Please Continue" control message response
8510# to actually send the request body to Squid. It is mostly useful in
8511# adaptation environments.
8512#
8513# When Squid receives an HTTP request with an "Expect: 100-continue"
8514# header or an FTP upload command (e.g., STOR), Squid normally sends the
8515# request headers or FTP command information to an adaptation service (or
8516# peer) and waits for a response. Most adaptation services (and some
8517# broken peers) may not respond to Squid at that stage because they may
8518# decide to wait for the HTTP request body or FTP data transfer. However,
8519# that request body or data transfer may never come because Squid has not
8520# responded with the HTTP 100 or FTP 150 (Please Continue) control message
8521# to the request sender yet!
8522#
8523# An allow match tells Squid to respond with the HTTP 100 or FTP 150
8524# (Please Continue) control message on its own, before forwarding the
8525# request to an adaptation service or peer. Such a response usually forces
8526# the request sender to proceed with sending the body. A deny match tells
8527# Squid to delay that control response until the origin server confirms
8528# that the request body is needed. Delaying is the default behavior.
8529#Default:
8530# Deny, unless rules exist in squid.conf.
8531
8532# TAG: server_pconn_for_nonretriable
8533# This option provides fine-grained control over persistent connection
8534# reuse when forwarding HTTP requests that Squid cannot retry. It is useful
8535# in environments where opening new connections is very expensive
8536# (e.g., all connections are secured with TLS with complex client and server
8537# certificate validation) and race conditions associated with persistent
8538# connections are very rare and/or only cause minor problems.
8539#
8540# HTTP prohibits retrying unsafe and non-idempotent requests (e.g., POST).
8541# Squid limitations also prohibit retrying all requests with bodies (e.g., PUT).
8542# By default, when forwarding such "risky" requests, Squid opens a new
8543# connection to the server or cache_peer, even if there is an idle persistent
8544# connection available. When Squid is configured to risk sending a non-retriable
8545# request on a previously used persistent connection, and the server closes
8546# the connection before seeing that risky request, the user gets an error response
8547# from Squid. In most cases, that error response will be HTTP 502 (Bad Gateway)
8548# with ERR_ZERO_SIZE_OBJECT or ERR_WRITE_ERROR (peer connection reset) error detail.
8549#
8550# If an allow rule matches, Squid reuses an available idle persistent connection
8551# (if any) for the request that Squid cannot retry. If a deny rule matches, then
8552# Squid opens a new connection for the request that Squid cannot retry.
8553#
8554# This option does not affect requests that Squid can retry. They will reuse idle
8555# persistent connections (if any).
8556#
8557# This clause only supports fast acl types.
8558# See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
8559#
8560# Example:
8561# acl SpeedIsWorthTheRisk method POST
8562# server_pconn_for_nonretriable allow SpeedIsWorthTheRisk
8563#Default:
8564# Open new connections for forwarding requests Squid cannot retry safely.
8565
8566cache_effective_group proxy