1 ## Copyright (C) 1996-2016 The Squid Software Foundation and contributors
3 ## Squid software is distributed under GPLv2+ license and includes
4 ## contributions from numerous individuals and organizations.
5 ## Please see the COPYING and CONTRIBUTORS files for details.
10 ----------------------------
12 This is the documentation for the Squid configuration file.
13 This documentation can also be found online at:
14 http://www.squid-cache.org/Doc/config/
16 You may wish to look at the Squid home page and wiki for the
17 FAQ and other documentation:
18 http://www.squid-cache.org/
19 http://wiki.squid-cache.org/SquidFaq
20 http://wiki.squid-cache.org/ConfigExamples
22 This documentation shows what the defaults for various directives
23 happen to be. If you don't need to change the default, you should
24 leave the line out of your squid.conf in most cases.
26 In some cases "none" refers to no default setting at all,
27 while in other cases it refers to the value of the option
28 - the comments for that keyword indicate if this is the case.
33 Configuration options can be included using the "include" directive.
34 Include takes a list of files to include. Quoting and wildcards are
39 include /path/to/included/file/squid.acl.config
41 Includes can be nested up to a hard-coded depth of 16 levels.
42 This arbitrary restriction is to prevent recursive include references
43 from causing Squid entering an infinite loop whilst trying to load
46 Values with byte units
48 Squid accepts size units on some size related directives. All
49 such directives are documented with a default value displaying
52 Units accepted by Squid are:
54 KB - Kilobyte (1024 bytes)
58 Values with spaces, quotes, and other special characters
60 Squid supports directive parameters with spaces, quotes, and other
61 special characters. Surround such parameters with "double quotes". Use
62 the configuration_includes_quoted_values directive to enable or
65 Squid supports reading configuration option parameters from external
66 files using the syntax:
67 parameters("/path/filename")
69 acl whitelist dstdomain parameters("/etc/squid/whitelist.txt")
71 Conditional configuration
73 If-statements can be used to make configuration directives
77 ... regular configuration directives ...
79 ... regular configuration directives ...]
82 The else part is optional. The keywords "if", "else", and "endif"
83 must be typed on their own lines, as if they were regular
84 configuration directives.
86 NOTE: An else-if condition is not supported.
88 These individual conditions types are supported:
91 Always evaluates to true.
93 Always evaluates to false.
95 Equality comparison of two integer numbers.
100 The following SMP-related preprocessor macros can be used.
102 ${process_name} expands to the current Squid process "name"
103 (e.g., squid1, squid2, or cache1).
105 ${process_number} expands to the current Squid process
106 identifier, which is an integer number (e.g., 1, 2, 3) unique
107 across all Squid processes of the current service instance.
109 ${service_name} expands into the current Squid service instance
110 name identifier which is provided by -n on the command line.
114 # options still not yet ported from 2.7 to 3.x
115 NAME: broken_vary_encoding
118 This option is not yet supported by Squid-3.
124 This option is not yet supported by Squid-3.
130 This option is not yet supported by Squid-3.
133 NAME: external_refresh_check
136 This option is not yet supported by Squid-3.
139 NAME: location_rewrite_program location_rewrite_access location_rewrite_children location_rewrite_concurrency
142 This option is not yet supported by Squid-3.
145 NAME: refresh_stale_hit
148 This option is not yet supported by Squid-3.
151 # Options removed in 3.6
152 NAME: cache_peer_domain cache_host_domain
155 Replace with dstdomain ACLs and cache_peer_access.
158 NAME: sslproxy_cafile
161 Remove this line. Use tls_outgoing_options cafile= instead.
164 NAME: sslproxy_capath
167 Remove this line. Use tls_outgoing_options capath= instead.
170 NAME: sslproxy_cipher
173 Remove this line. Use tls_outgoing_options cipher= instead.
176 NAME: sslproxy_client_certificate
179 Remove this line. Use tls_outgoing_options cert= instead.
182 NAME: sslproxy_client_key
185 Remove this line. Use tls_outgoing_options key= instead.
191 Remove this line. Use tls_outgoing_options flags= instead.
194 NAME: sslproxy_options
197 Remove this line. Use tls_outgoing_options options= instead.
200 NAME: sslproxy_version
203 Remove this line. Use tls_outgoing_options options= instead.
206 # Options removed in 3.5
207 NAME: hierarchy_stoplist
210 Remove this line. Use always_direct or cache_peer_access ACLs instead if you need to prevent cache_peer use.
213 # Options removed in 3.4
217 Remove this line. Use acls with access_log directives to control access logging
223 Remove this line. Use acls with icap_log directives to control icap logging
226 # Options Removed in 3.3
227 NAME: ignore_ims_on_miss
230 Remove this line. The HTTP/1.1 feature is now configured by 'cache_miss_revalidate'.
233 # Options Removed in 3.2
234 NAME: chunked_request_body_max_size
237 Remove this line. Squid is now HTTP/1.1 compliant.
240 NAME: dns_v4_fallback
243 Remove this line. Squid performs a 'Happy Eyeballs' algorithm, the 'fallback' algorithm is no longer relevant.
246 NAME: emulate_httpd_log
249 Replace this with an access_log directive using the format 'common' or 'combined'.
255 Use a regular access.log with ACL limiting it to MISS events.
261 Remove this line. Configure FTP page display using the CSS controls in errorpages.css instead.
264 NAME: ignore_expect_100
267 Remove this line. The HTTP/1.1 feature is now fully supported by default.
273 Remove this option from your config. To log FQDN use %>A in the log format.
276 NAME: log_ip_on_direct
279 Remove this option from your config. To log server or peer names use %<A in the log format.
282 NAME: maximum_single_addr_tries
285 Replaced by connect_retries. The behaviour has changed, please read the documentation before altering.
288 NAME: referer_log referrer_log
291 Replace this with an access_log directive using the format 'referrer'.
297 Remove this line. The feature is supported by default in storage types where update is implemented.
300 NAME: url_rewrite_concurrency
303 Remove this line. Set the 'concurrency=' option of url_rewrite_children instead.
309 Replace this with an access_log directive using the format 'useragent'.
312 # Options Removed in 3.1
316 Remove this line. DNS is no longer tested on startup.
319 NAME: extension_methods
322 Remove this line. All valid methods for HTTP are accepted by default.
325 # 2.7 Options Removed/Replaced in 3.2
330 # 2.7 Options Removed/Replaced in 3.1
338 Remove this line. HTTP/1.1 is supported by default.
341 NAME: upgrade_http0.9
344 Remove this line. ICY/1.0 streaming protocol is supported by default.
347 NAME: zph_local zph_mode zph_option zph_parent zph_sibling
350 Alter these entries. Use the qos_flows directive instead.
353 # Options Removed in 3.0
357 Since squid-3.0 replace with request_header_access or reply_header_access
358 depending on whether you wish to match client requests or server replies.
361 NAME: httpd_accel_no_pmtu_disc
364 Since squid-3.0 use the 'disable-pmtu-discovery' flag on http_port instead.
367 NAME: wais_relay_host
370 Replace this line with 'cache_peer' configuration.
373 NAME: wais_relay_port
376 Replace this line with 'cache_peer' configuration.
381 -----------------------------------------------------------------------------
388 DEFAULT_DOC: SMP support disabled.
390 Number of main Squid processes or "workers" to fork and maintain.
391 0: "no daemon" mode, like running "squid -N ..."
392 1: "no SMP" mode, start one main Squid process daemon (default)
393 N: start N main Squid process daemons (i.e., SMP mode)
395 In SMP mode, each worker does nearly all what a single Squid daemon
396 does (e.g., listen on http_port and forward HTTP requests).
399 NAME: cpu_affinity_map
401 LOC: Config.cpuAffinityMap
403 DEFAULT_DOC: Let operating system decide.
405 Usage: cpu_affinity_map process_numbers=P1,P2,... cores=C1,C2,...
407 Sets 1:1 mapping between Squid processes and CPU cores. For example,
409 cpu_affinity_map process_numbers=1,2,3,4 cores=1,3,5,7
411 affects processes 1 through 4 only and places them on the first
412 four even cores, starting with core #1.
414 CPU cores are numbered starting from 1. Requires support for
415 sched_getaffinity(2) and sched_setaffinity(2) system calls.
417 Multiple cpu_affinity_map options are merged.
423 OPTIONS FOR AUTHENTICATION
424 -----------------------------------------------------------------------------
433 This is used to define parameters for the various authentication
434 schemes supported by Squid.
436 format: auth_param scheme parameter [setting]
438 The order in which authentication schemes are presented to the client is
439 dependent on the order the scheme first appears in config file. IE
440 has a bug (it's not RFC 2617 compliant) in that it will use the basic
441 scheme if basic is the first entry presented, even if more secure
442 schemes are presented. For now use the order in the recommended
443 settings section below. If other browsers have difficulties (don't
444 recognize the schemes offered even if you are using basic) either
445 put basic first, or disable the other schemes (by commenting out their
448 Once an authentication scheme is fully configured, it can only be
449 shutdown by shutting squid down and restarting. Changes can be made on
450 the fly and activated with a reconfigure. I.E. You can change to a
451 different helper, but not unconfigure the helper completely.
453 Please note that while this directive defines how Squid processes
454 authentication it does not automatically activate authentication.
455 To use authentication you must in addition make use of ACLs based
456 on login name in http_access (proxy_auth, proxy_auth_regex or
457 external with %LOGIN used in the format tag). The browser will be
458 challenged for authentication on the first such acl encountered
459 in http_access processing and will also be re-challenged for new
460 login credentials if the request is being denied by a proxy_auth
463 WARNING: authentication can't be used in a transparently intercepting
464 proxy as the client then thinks it is talking to an origin server and
465 not the proxy. This is a limitation of bending the TCP/IP protocol to
466 transparently intercepting port 80, not a limitation in Squid.
467 Ports flagged 'transparent', 'intercept', or 'tproxy' have
468 authentication disabled.
470 === Parameters common to all schemes. ===
473 Specifies the command for the external authenticator.
475 By default, each authentication scheme is not used unless a
476 program is specified.
478 See http://wiki.squid-cache.org/Features/AddonHelpers for
479 more details on helper operations and creating your own.
482 Specifies a string to be append to request line format for
483 the authentication helper. "Quoted" format values may contain
484 spaces and logformat %macros. In theory, any logformat %macro
485 can be used. In practice, a %macro expands as a dash (-) if
486 the helper request is sent before the required macro
487 information is available to Squid.
489 By default, Squid uses request formats provided in
490 scheme-specific examples below (search for %credentials).
492 The expanded key_extras value is added to the Squid credentials
493 cache and, hence, will affect authentication. It can be used to
494 autenticate different users with identical user names (e.g.,
495 when user authentication depends on http_port).
497 Avoid adding frequently changing information to key_extras. For
498 example, if you add user source IP, and it changes frequently
499 in your environment, then max_user_ip ACL is going to treat
500 every user+IP combination as a unique "user", breaking the ACL
501 and wasting a lot of memory on those user records. It will also
502 force users to authenticate from scratch whenever their IP
506 Specifies the protection scope (aka realm name) which is to be
507 reported to the client for the authentication scheme. It is
508 commonly part of the text the user will see when prompted for
509 their username and password.
511 For Basic the default is "Squid proxy-caching web server".
512 For Digest there is no default, this parameter is mandatory.
513 For NTLM and Negotiate this parameter is ignored.
515 "children" numberofchildren [startup=N] [idle=N] [concurrency=N] [queue-size=N]
517 The maximum number of authenticator processes to spawn. If
518 you start too few Squid will have to wait for them to process
519 a backlog of credential verifications, slowing it down. When
520 password verifications are done via a (slow) network you are
521 likely to need lots of authenticator processes.
523 The startup= and idle= options permit some skew in the exact
524 amount run. A minimum of startup=N will begin during startup
525 and reconfigure. Squid will start more in groups of up to
526 idle=N in an attempt to meet traffic needs and to keep idle=N
527 free above those traffic needs up to the maximum.
529 The concurrency= option sets the number of concurrent requests
530 the helper can process. The default of 0 is used for helpers
531 who only supports one request at a time. Setting this to a
532 number greater than 0 changes the protocol used to include a
533 channel ID field first on the request/response line, allowing
534 multiple requests to be sent to the same helper in parallel
535 without waiting for the response.
537 Concurrency must not be set unless it's known the helper
538 supports the input format with channel-ID fields.
540 The queue-size= option sets the maximum number of queued
541 requests. If the queued requests exceed queue size for more
542 than 3 minutes then squid aborts its operation.
543 The default value is set to 2*numberofchildren/
545 NOTE: NTLM and Negotiate schemes do not support concurrency
546 in the Squid code module even though some helpers can.
549 IF HAVE_AUTH_MODULE_BASIC
550 === Basic authentication parameters ===
553 HTTP uses iso-latin-1 as character set, while some
554 authentication backends such as LDAP expects UTF-8. If this is
555 set to on Squid will translate the HTTP iso-latin-1 charset to
556 UTF-8 before sending the username and password to the helper.
558 "credentialsttl" timetolive
559 Specifies how long squid assumes an externally validated
560 username:password pair is valid for - in other words how
561 often the helper program is called for that user. Set this
562 low to force revalidation with short lived passwords.
564 NOTE: setting this high does not impact your susceptibility
565 to replay attacks unless you are using an one-time password
566 system (such as SecureID). If you are using such a system,
567 you will be vulnerable to replay attacks unless you also
568 use the max_user_ip ACL in an http_access rule.
570 "casesensitive" on|off
571 Specifies if usernames are case sensitive. Most user databases
572 are case insensitive allowing the same username to be spelled
573 using both lower and upper case letters, but some are case
574 sensitive. This makes a big difference for user_max_ip ACL
575 processing and similar.
578 IF HAVE_AUTH_MODULE_DIGEST
579 === Digest authentication parameters ===
582 HTTP uses iso-latin-1 as character set, while some
583 authentication backends such as LDAP expects UTF-8. If this is
584 set to on Squid will translate the HTTP iso-latin-1 charset to
585 UTF-8 before sending the username and password to the helper.
587 "nonce_garbage_interval" timeinterval
588 Specifies the interval that nonces that have been issued
589 to client_agent's are checked for validity.
591 "nonce_max_duration" timeinterval
592 Specifies the maximum length of time a given nonce will be
595 "nonce_max_count" number
596 Specifies the maximum number of times a given nonce can be
599 "nonce_strictness" on|off
600 Determines if squid requires strict increment-by-1 behavior
601 for nonce counts, or just incrementing (off - for use when
602 user agents generate nonce counts that occasionally miss 1
603 (ie, 1,2,4,6)). Default off.
605 "check_nonce_count" on|off
606 This directive if set to off can disable the nonce count check
607 completely to work around buggy digest qop implementations in
608 certain mainstream browser versions. Default on to check the
609 nonce count to protect from authentication replay attacks.
611 "post_workaround" on|off
612 This is a workaround to certain buggy browsers who send an
613 incorrect request digest in POST requests when reusing the
614 same nonce as acquired earlier on a GET request.
617 IF HAVE_AUTH_MODULE_NEGOTIATE
618 === Negotiate authentication parameters ===
621 If you experience problems with PUT/POST requests when using
622 the this authentication scheme then you can try setting this
623 to off. This will cause Squid to forcibly close the connection
624 on the initial request where the browser asks which schemes
625 are supported by the proxy.
628 IF HAVE_AUTH_MODULE_NTLM
629 === NTLM authentication parameters ===
632 If you experience problems with PUT/POST requests when using
633 the this authentication scheme then you can try setting this
634 to off. This will cause Squid to forcibly close the connection
635 on the initial request where the browser asks which schemes
636 are supported by the proxy.
639 === Example Configuration ===
641 This configuration displays the recommended authentication scheme
642 order from most to least secure with recommended minimum configuration
643 settings for each scheme:
645 #auth_param negotiate program <uncomment and complete this line to activate>
646 #auth_param negotiate children 20 startup=0 idle=1
647 #auth_param negotiate keep_alive on
649 #auth_param digest program <uncomment and complete this line to activate>
650 #auth_param digest children 20 startup=0 idle=1
651 #auth_param digest realm Squid proxy-caching web server
652 #auth_param digest nonce_garbage_interval 5 minutes
653 #auth_param digest nonce_max_duration 30 minutes
654 #auth_param digest nonce_max_count 50
656 #auth_param ntlm program <uncomment and complete this line to activate>
657 #auth_param ntlm children 20 startup=0 idle=1
658 #auth_param ntlm keep_alive on
660 #auth_param basic program <uncomment and complete this line>
661 #auth_param basic children 5 startup=5 idle=1
662 #auth_param basic realm Squid proxy-caching web server
663 #auth_param basic credentialsttl 2 hours
666 NAME: authenticate_cache_garbage_interval
669 LOC: Config.authenticateGCInterval
671 The time period between garbage collection across the username cache.
672 This is a trade-off between memory utilization (long intervals - say
673 2 days) and CPU (short intervals - say 1 minute). Only change if you
677 NAME: authenticate_ttl
680 LOC: Config.authenticateTTL
682 The time a user & their credentials stay in the logged in
683 user cache since their last request. When the garbage
684 interval passes, all user credentials that have passed their
685 TTL are removed from memory.
688 NAME: authenticate_ip_ttl
690 LOC: Config.authenticateIpTTL
693 If you use proxy authentication and the 'max_user_ip' ACL,
694 this directive controls how long Squid remembers the IP
695 addresses associated with each user. Use a small value
696 (e.g., 60 seconds) if your users might change addresses
697 quickly, as is the case with dialup. You might be safe
698 using a larger value (e.g., 2 hours) in a corporate LAN
699 environment with relatively static address assignments.
704 -----------------------------------------------------------------------------
707 NAME: external_acl_type
708 TYPE: externalAclHelper
709 LOC: Config.externalAclHelperList
712 This option defines external acl classes using a helper program
713 to look up the status
715 external_acl_type name [options] FORMAT /path/to/helper [helper arguments]
719 ttl=n TTL in seconds for cached results (defaults to 3600
723 TTL for cached negative lookups (default same
726 grace=n Percentage remaining of TTL where a refresh of a
727 cached entry should be initiated without needing to
728 wait for a new reply. (default is for no grace period)
730 cache=n The maximum number of entries in the result cache. The
731 default limit is 262144 entries. Each cache entry usually
732 consumes at least 256 bytes. Squid currently does not remove
733 expired cache entries until the limit is reached, so a proxy
734 will sooner or later reach the limit. The expanded FORMAT
735 value is used as the cache key, so if the details in FORMAT
736 are highly variable, a larger cache may be needed to produce
737 reduction in helper load.
740 Maximum number of acl helper processes spawned to service
741 external acl lookups of this type. (default 20)
744 Minimum number of acl helper processes to spawn during
745 startup and reconfigure to service external acl lookups
746 of this type. (default 0)
749 Number of acl helper processes to keep ahead of traffic
750 loads. Squid will spawn this many at once whenever load
751 rises above the capabilities of existing processes.
752 Up to the value of children-max. (default 1)
754 concurrency=n concurrency level per process. Only used with helpers
755 capable of processing more than one query at a time.
757 queue-size=N The queue-size= option sets the maximum number of queued
758 requests. If the queued requests exceed queue size
760 The default value is set to 2*children-max.
762 protocol=2.5 Compatibility mode for Squid-2.5 external acl helpers.
764 ipv4 / ipv6 IP protocol used to communicate with this helper.
765 The default is to auto-detect IPv6 and use it when available.
768 FORMAT is a series of %macro codes. See logformat directive for a full list
769 of the accepted codes. Although note that at the time of any external ACL
770 being tested data may not be available and thus some %macro expand to '-'.
772 In addition to the logformat codes; when processing external ACLs these
773 additional macros are made available:
775 %ACL The name of the ACL being tested.
777 %DATA The ACL arguments separated by spaces (an
780 By default, Squid applies URL-encoding to each ACL
781 argument inside the argument string. If an explicit
782 encoding modifier is used (e.g., %#DATA), then Squid
783 encodes the whole argument string as a single token
784 (e.g., with %#DATA, spaces between arguments become
787 If you do not specify a DATA macro inside FORMAT,
788 Squid automatically appends %DATA to your FORMAT.
790 If SSL is enabled, the following formating codes become available:
792 %USER_CERT SSL User certificate in PEM format
793 %USER_CERTCHAIN SSL User certificate chain in PEM format
794 %USER_CERT_xx SSL User certificate subject attribute xx
795 %USER_CA_CERT_xx SSL User certificate issuer attribute xx
798 NOTE: all other format codes accepted by older Squid versions
802 General request syntax:
804 [channel-ID] FORMAT-values [acl-values ...]
807 FORMAT-values consists of transaction details expanded with
808 whitespace separation per the config file FORMAT specification
809 using the FORMAT macros listed above.
811 acl-values consists of any string specified in the referencing
812 config 'acl ... external' line. see the "acl external" directive.
814 Request values sent to the helper are URL escaped to protect
815 each value in requests against whitespaces.
817 If using protocol=2.5 then the request sent to the helper is not
818 URL escaped to protect against whitespace.
820 NOTE: protocol=3.0 is deprecated as no longer necessary.
822 When using the concurrency= option the protocol is changed by
823 introducing a query channel tag in front of the request/response.
824 The query channel tag is a number between 0 and concurrency-1.
825 This value must be echoed back unchanged to Squid as the first part
826 of the response relating to its request.
829 The helper receives lines expanded per the above format specification
830 and for each input line returns 1 line starting with OK/ERR/BH result
831 code and optionally followed by additional keywords with more details.
834 General result syntax:
836 [channel-ID] result keyword=value ...
838 Result consists of one of the codes:
841 the ACL test produced a match.
844 the ACL test does not produce a match.
847 An internal error occurred in the helper, preventing
848 a result being identified.
850 The meaning of 'a match' is determined by your squid.conf
851 access control configuration. See the Squid wiki for details.
855 user= The users name (login)
857 password= The users password (for login= cache_peer option)
859 message= Message describing the reason for this response.
860 Available as %o in error pages.
861 Useful on (ERR and BH results).
863 tag= Apply a tag to a request. Only sets a tag once,
864 does not alter existing tags.
866 log= String to be logged in access.log. Available as
867 %ea in logformat specifications.
869 clt_conn_tag= Associates a TAG with the client TCP connection.
870 Please see url_rewrite_program related documentation
873 Any keywords may be sent on any response whether OK, ERR or BH.
875 All response keyword values need to be a single token with URL
876 escaping, or enclosed in double quotes (") and escaped using \ on
877 any double quotes or \ characters within the value. The wrapping
878 double quotes are removed before the value is interpreted by Squid.
879 \r and \n are also replace by CR and LF.
881 Some example key values:
885 user="J. \"Bob\" Smith"
892 DEFAULT: ssl::certHasExpired ssl_error X509_V_ERR_CERT_HAS_EXPIRED
893 DEFAULT: ssl::certNotYetValid ssl_error X509_V_ERR_CERT_NOT_YET_VALID
894 DEFAULT: ssl::certDomainMismatch ssl_error SQUID_X509_V_ERR_DOMAIN_MISMATCH
895 DEFAULT: ssl::certUntrusted ssl_error X509_V_ERR_INVALID_CA X509_V_ERR_SELF_SIGNED_CERT_IN_CHAIN X509_V_ERR_UNABLE_TO_VERIFY_LEAF_SIGNATURE X509_V_ERR_UNABLE_TO_GET_ISSUER_CERT X509_V_ERR_UNABLE_TO_GET_ISSUER_CERT_LOCALLY X509_V_ERR_CERT_UNTRUSTED
896 DEFAULT: ssl::certSelfSigned ssl_error X509_V_ERR_DEPTH_ZERO_SELF_SIGNED_CERT
899 DEFAULT: manager url_regex -i ^cache_object:// +i ^https?://[^/]+/squid-internal-mgr/
900 DEFAULT: localhost src 127.0.0.1/32 ::1
901 DEFAULT: to_localhost dst 127.0.0.0/8 0.0.0.0/32 ::1
902 DEFAULT_DOC: ACLs all, manager, localhost, and to_localhost are predefined.
904 Defining an Access List
906 Every access list definition must begin with an aclname and acltype,
907 followed by either type-specific arguments or a quoted filename that
910 acl aclname acltype argument ...
911 acl aclname acltype "file" ...
913 When using "file", the file should contain one item per line.
918 Some acl types supports options which changes their default behaviour:
920 -i,+i By default, regular expressions are CASE-SENSITIVE. To make them
921 case-insensitive, use the -i option. To return case-sensitive
922 use the +i option between patterns, or make a new ACL line
925 -n Disable lookups and address type conversions. If lookup or
926 conversion is required because the parameter type (IP or
927 domain name) does not match the message address type (domain
928 name or IP), then the ACL would immediately declare a mismatch
929 without any warnings or lookups.
932 Perform a list membership test, interpreting values as
933 comma-separated token lists and matching against individual
934 tokens instead of whole values.
935 The optional "delimiters" parameter specifies one or more
936 alternative non-alphanumeric delimiter characters.
937 non-alphanumeric delimiter characters.
939 -- Used to stop processing all options, in the case the first acl
940 value has '-' character as first character (for example the '-'
941 is a valid domain name)
943 Some acl types require suspending the current request in order
944 to access some external data source.
945 Those which do are marked with the tag [slow], those which
946 don't are marked as [fast].
947 See http://wiki.squid-cache.org/SquidFaq/SquidAcl
948 for further information
950 ***** ACL TYPES AVAILABLE *****
952 acl aclname src ip-address/mask ... # clients IP address [fast]
953 acl aclname src addr1-addr2/mask ... # range of addresses [fast]
954 acl aclname dst [-n] ip-address/mask ... # URL host's IP address [slow]
955 acl aclname localip ip-address/mask ... # IP address the client connected to [fast]
957 acl aclname arp mac-address ... (xx:xx:xx:xx:xx:xx notation)
959 # The 'arp' ACL code is not portable to all operating systems.
960 # It works on Linux, Solaris, Windows, FreeBSD, and some other
963 # NOTE: Squid can only determine the MAC/EUI address for IPv4
964 # clients that are on the same subnet. If the client is on a
965 # different subnet, then Squid cannot find out its address.
967 # NOTE 2: IPv6 protocol does not contain ARP. MAC/EUI is either
968 # encoded directly in the IPv6 address or not available.
970 acl aclname srcdomain .foo.com ...
971 # reverse lookup, from client IP [slow]
972 acl aclname dstdomain [-n] .foo.com ...
973 # Destination server from URL [fast]
974 acl aclname srcdom_regex [-i] \.foo\.com ...
975 # regex matching client name [slow]
976 acl aclname dstdom_regex [-n] [-i] \.foo\.com ...
977 # regex matching server [fast]
979 # For dstdomain and dstdom_regex a reverse lookup is tried if a IP
980 # based URL is used and no match is found. The name "none" is used
981 # if the reverse lookup fails.
983 acl aclname src_as number ...
984 acl aclname dst_as number ...
986 # Except for access control, AS numbers can be used for
987 # routing of requests to specific caches. Here's an
988 # example for routing all requests for AS#1241 and only
989 # those to mycache.mydomain.net:
990 # acl asexample dst_as 1241
991 # cache_peer_access mycache.mydomain.net allow asexample
992 # cache_peer_access mycache_mydomain.net deny all
994 acl aclname peername myPeer ...
996 # match against a named cache_peer entry
997 # set unique name= on cache_peer lines for reliable use.
999 acl aclname time [day-abbrevs] [h1:m1-h2:m2]
1009 # h1:m1 must be less than h2:m2
1011 acl aclname url_regex [-i] ^http:// ...
1012 # regex matching on whole URL [fast]
1013 acl aclname urllogin [-i] [^a-zA-Z0-9] ...
1014 # regex matching on URL login field
1015 acl aclname urlpath_regex [-i] \.gif$ ...
1016 # regex matching on URL path [fast]
1018 acl aclname port 80 70 21 0-1024... # destination TCP port [fast]
1020 acl aclname localport 3128 ... # TCP port the client connected to [fast]
1021 # NP: for interception mode this is usually '80'
1023 acl aclname myportname 3128 ... # *_port name [fast]
1025 acl aclname proto HTTP FTP ... # request protocol [fast]
1027 acl aclname method GET POST ... # HTTP request method [fast]
1029 acl aclname http_status 200 301 500- 400-403 ...
1030 # status code in reply [fast]
1032 acl aclname browser [-i] regexp ...
1033 # pattern match on User-Agent header (see also req_header below) [fast]
1035 acl aclname referer_regex [-i] regexp ...
1036 # pattern match on Referer header [fast]
1037 # Referer is highly unreliable, so use with care
1039 acl aclname ident username ...
1040 acl aclname ident_regex [-i] pattern ...
1041 # string match on ident output [slow]
1042 # use REQUIRED to accept any non-null ident.
1044 acl aclname proxy_auth [-i] username ...
1045 acl aclname proxy_auth_regex [-i] pattern ...
1046 # perform http authentication challenge to the client and match against
1047 # supplied credentials [slow]
1049 # takes a list of allowed usernames.
1050 # use REQUIRED to accept any valid username.
1052 # Will use proxy authentication in forward-proxy scenarios, and plain
1053 # http authenticaiton in reverse-proxy scenarios
1055 # NOTE: when a Proxy-Authentication header is sent but it is not
1056 # needed during ACL checking the username is NOT logged
1059 # NOTE: proxy_auth requires a EXTERNAL authentication program
1060 # to check username/password combinations (see
1061 # auth_param directive).
1063 # NOTE: proxy_auth can't be used in a transparent/intercepting proxy
1064 # as the browser needs to be configured for using a proxy in order
1065 # to respond to proxy authentication.
1067 acl aclname snmp_community string ...
1068 # A community string to limit access to your SNMP Agent [fast]
1071 # acl snmppublic snmp_community public
1073 acl aclname maxconn number
1074 # This will be matched when the client's IP address has
1075 # more than <number> TCP connections established. [fast]
1076 # NOTE: This only measures direct TCP links so X-Forwarded-For
1077 # indirect clients are not counted.
1079 acl aclname max_user_ip [-s] number
1080 # This will be matched when the user attempts to log in from more
1081 # than <number> different ip addresses. The authenticate_ip_ttl
1082 # parameter controls the timeout on the ip entries. [fast]
1083 # If -s is specified the limit is strict, denying browsing
1084 # from any further IP addresses until the ttl has expired. Without
1085 # -s Squid will just annoy the user by "randomly" denying requests.
1086 # (the counter is reset each time the limit is reached and a
1087 # request is denied)
1088 # NOTE: in acceleration mode or where there is mesh of child proxies,
1089 # clients may appear to come from multiple addresses if they are
1090 # going through proxy farms, so a limit of 1 may cause user problems.
1092 acl aclname random probability
1093 # Pseudo-randomly match requests. Based on the probability given.
1094 # Probability may be written as a decimal (0.333), fraction (1/3)
1095 # or ratio of matches:non-matches (3:5).
1097 acl aclname req_mime_type [-i] mime-type ...
1098 # regex match against the mime type of the request generated
1099 # by the client. Can be used to detect file upload or some
1100 # types HTTP tunneling requests [fast]
1101 # NOTE: This does NOT match the reply. You cannot use this
1102 # to match the returned file type.
1104 acl aclname req_header header-name [-i] any\.regex\.here
1105 # regex match against any of the known request headers. May be
1106 # thought of as a superset of "browser", "referer" and "mime-type"
1109 acl aclname rep_mime_type [-i] mime-type ...
1110 # regex match against the mime type of the reply received by
1111 # squid. Can be used to detect file download or some
1112 # types HTTP tunneling requests. [fast]
1113 # NOTE: This has no effect in http_access rules. It only has
1114 # effect in rules that affect the reply data stream such as
1115 # http_reply_access.
1117 acl aclname rep_header header-name [-i] any\.regex\.here
1118 # regex match against any of the known reply headers. May be
1119 # thought of as a superset of "browser", "referer" and "mime-type"
1122 acl aclname external class_name [arguments...]
1123 # external ACL lookup via a helper class defined by the
1124 # external_acl_type directive [slow]
1126 acl aclname user_cert attribute values...
1127 # match against attributes in a user SSL certificate
1128 # attribute is one of DN/C/O/CN/L/ST or a numerical OID [fast]
1130 acl aclname ca_cert attribute values...
1131 # match against attributes a users issuing CA SSL certificate
1132 # attribute is one of DN/C/O/CN/L/ST or a numerical OID [fast]
1134 acl aclname ext_user username ...
1135 acl aclname ext_user_regex [-i] pattern ...
1136 # string match on username returned by external acl helper [slow]
1137 # use REQUIRED to accept any non-null user name.
1139 acl aclname tag tagvalue ...
1140 # string match on tag returned by external acl helper [fast]
1141 # DEPRECATED. Only the first tag will match with this ACL.
1142 # Use the 'note' ACL instead for handling multiple tag values.
1144 acl aclname hier_code codename ...
1145 # string match against squid hierarchy code(s); [fast]
1146 # e.g., DIRECT, PARENT_HIT, NONE, etc.
1148 # NOTE: This has no effect in http_access rules. It only has
1149 # effect in rules that affect the reply data stream such as
1150 # http_reply_access.
1152 acl aclname note [-m[=delimiters]] name [value ...]
1153 # match transaction annotation [fast]
1154 # Without values, matches any annotation with a given name.
1155 # With value(s), matches any annotation with a given name that
1156 # also has one of the given values.
1157 # If the -m flag is used, then the value of the named
1158 # annotation is interpreted as a list of tokens, and the ACL
1159 # matches individual name=token pairs rather than whole
1160 # name=value pairs. See "ACL Options" above for more info.
1161 # Annotation sources include note and adaptation_meta directives
1162 # as well as helper and eCAP responses.
1164 acl aclname adaptation_service service ...
1165 # Matches the name of any icap_service, ecap_service,
1166 # adaptation_service_set, or adaptation_service_chain that Squid
1167 # has used (or attempted to use) for the master transaction.
1168 # This ACL must be defined after the corresponding adaptation
1169 # service is named in squid.conf. This ACL is usable with
1170 # adaptation_meta because it starts matching immediately after
1171 # the service has been selected for adaptation.
1174 acl aclname ssl_error errorname
1175 # match against SSL certificate validation error [fast]
1177 # For valid error names see in @DEFAULT_ERROR_DIR@/templates/error-details.txt
1180 # The following can be used as shortcuts for certificate properties:
1181 # [ssl::]certHasExpired: the "not after" field is in the past
1182 # [ssl::]certNotYetValid: the "not before" field is in the future
1183 # [ssl::]certUntrusted: The certificate issuer is not to be trusted.
1184 # [ssl::]certSelfSigned: The certificate is self signed.
1185 # [ssl::]certDomainMismatch: The certificate CN domain does not
1186 # match the name the name of the host we are connecting to.
1188 # The ssl::certHasExpired, ssl::certNotYetValid, ssl::certDomainMismatch,
1189 # ssl::certUntrusted, and ssl::certSelfSigned can also be used as
1190 # predefined ACLs, just like the 'all' ACL.
1192 # NOTE: The ssl_error ACL is only supported with sslproxy_cert_error,
1193 # sslproxy_cert_sign, and sslproxy_cert_adapt options.
1195 acl aclname server_cert_fingerprint [-sha1] fingerprint
1196 # match against server SSL certificate fingerprint [fast]
1198 # The fingerprint is the digest of the DER encoded version
1199 # of the whole certificate. The user should use the form: XX:XX:...
1200 # Optional argument specifies the digest algorithm to use.
1201 # The SHA1 digest algorithm is the default and is currently
1202 # the only algorithm supported (-sha1).
1204 acl aclname at_step step
1205 # match against the current step during ssl_bump evaluation [fast]
1206 # Never matches and should not be used outside the ssl_bump context.
1208 # At each SslBump step, Squid evaluates ssl_bump directives to find
1209 # the next bumping action (e.g., peek or splice). Valid SslBump step
1210 # values and the corresponding ssl_bump evaluation moments are:
1211 # SslBump1: After getting TCP-level and HTTP CONNECT info.
1212 # SslBump2: After getting SSL Client Hello info.
1213 # SslBump3: After getting SSL Server Hello info.
1215 acl aclname ssl::server_name .foo.com ...
1216 # matches server name obtained from various sources [fast]
1218 # The server name is obtained during Ssl-Bump steps from such sources
1219 # as CONNECT request URI, client SNI, and SSL server certificate CN.
1220 # During each Ssl-Bump step, Squid may improve its understanding of a
1221 # "true server name". Unlike dstdomain, this ACL does not perform
1224 acl aclname ssl::server_name_regex [-i] \.foo\.com ...
1225 # regex matches server name obtained from various sources [fast]
1227 acl aclname connections_encrypted
1228 # matches transactions with all HTTP messages received over TLS
1229 # transport connections. [fast]
1231 # The master transaction deals with HTTP messages received from
1232 # various sources. All sources used by the master transaction in the
1233 # past are considered by the ACL. The following rules define whether
1234 # a given message source taints the entire master transaction,
1235 # resulting in ACL mismatches:
1237 # * The HTTP client transport connection is not TLS.
1238 # * An adaptation service connection-encryption flag is off.
1239 # * The peer or origin server transport connection is not TLS.
1241 # Caching currently does not affect these rules. This cache ignorance
1242 # implies that only the current HTTP client transport and REQMOD
1243 # services status determine whether this ACL matches a from-cache
1244 # transaction. The source of the cached response does not have any
1245 # effect on future transaction that use the cached response without
1246 # revalidation. This may change.
1248 # DNS, ICP, and HTCP exchanges during the master transaction do not
1249 # affect these rules.
1251 acl aclname any-of acl1 acl2 ...
1252 # match any one of the acls [fast or slow]
1253 # The first matching ACL stops further ACL evaluation.
1255 # ACLs from multiple any-of lines with the same name are ORed.
1256 # For example, A = (a1 or a2) or (a3 or a4) can be written as
1257 # acl A any-of a1 a2
1258 # acl A any-of a3 a4
1260 # This group ACL is fast if all evaluated ACLs in the group are fast
1261 # and slow otherwise.
1263 acl aclname all-of acl1 acl2 ...
1264 # match all of the acls [fast or slow]
1265 # The first mismatching ACL stops further ACL evaluation.
1267 # ACLs from multiple all-of lines with the same name are ORed.
1268 # For example, B = (b1 and b2) or (b3 and b4) can be written as
1269 # acl B all-of b1 b2
1270 # acl B all-of b3 b4
1272 # This group ACL is fast if all evaluated ACLs in the group are fast
1273 # and slow otherwise.
1276 acl macaddress arp 09:00:2b:23:45:67
1277 acl myexample dst_as 1241
1278 acl password proxy_auth REQUIRED
1279 acl fileupload req_mime_type -i ^multipart/form-data$
1280 acl javascript rep_mime_type -i ^application/x-javascript$
1284 # Recommended minimum configuration:
1287 # Example rule allowing access from your local networks.
1288 # Adapt to list your (internal) IP networks from where browsing
1290 acl localnet src 0.0.0.1-0.255.255.255 # RFC 1122 "this" network (LAN)
1291 acl localnet src 10.0.0.0/8 # RFC 1918 local private network (LAN)
1292 acl localnet src 100.64.0.0/10 # RFC 6598 shared address space (CGN)
1293 acl localhet src 169.254.0.0/16 # RFC 3927 link-local (directly plugged) machines
1294 acl localnet src 172.16.0.0/12 # RFC 1918 local private network (LAN)
1295 acl localnet src 192.168.0.0/16 # RFC 1918 local private network (LAN)
1296 acl localnet src fc00::/7 # RFC 4193 local private network range
1297 acl localnet src fe80::/10 # RFC 4291 link-local (directly plugged) machines
1299 acl SSL_ports port 443
1300 acl Safe_ports port 80 # http
1301 acl Safe_ports port 21 # ftp
1302 acl Safe_ports port 443 # https
1303 acl Safe_ports port 70 # gopher
1304 acl Safe_ports port 210 # wais
1305 acl Safe_ports port 1025-65535 # unregistered ports
1306 acl Safe_ports port 280 # http-mgmt
1307 acl Safe_ports port 488 # gss-http
1308 acl Safe_ports port 591 # filemaker
1309 acl Safe_ports port 777 # multiling http
1310 acl CONNECT method CONNECT
1314 NAME: proxy_protocol_access
1316 LOC: Config.accessList.proxyProtocol
1318 DEFAULT_DOC: all TCP connections to ports with require-proxy-header will be denied
1320 Determine which client proxies can be trusted to provide correct
1321 information regarding real client IP address using PROXY protocol.
1323 Requests may pass through a chain of several other proxies
1324 before reaching us. The original source details may by sent in:
1325 * HTTP message Forwarded header, or
1326 * HTTP message X-Forwarded-For header, or
1327 * PROXY protocol connection header.
1329 This directive is solely for validating new PROXY protocol
1330 connections received from a port flagged with require-proxy-header.
1331 It is checked only once after TCP connection setup.
1333 A deny match results in TCP connection closure.
1335 An allow match is required for Squid to permit the corresponding
1336 TCP connection, before Squid even looks for HTTP request headers.
1337 If there is an allow match, Squid starts using PROXY header information
1338 to determine the source address of the connection for all future ACL
1339 checks, logging, etc.
1341 SECURITY CONSIDERATIONS:
1343 Any host from which we accept client IP details can place
1344 incorrect information in the relevant header, and Squid
1345 will use the incorrect information as if it were the
1346 source address of the request. This may enable remote
1347 hosts to bypass any access control restrictions that are
1348 based on the client's source addresses.
1350 This clause only supports fast acl types.
1351 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1354 NAME: follow_x_forwarded_for
1356 IFDEF: FOLLOW_X_FORWARDED_FOR
1357 LOC: Config.accessList.followXFF
1358 DEFAULT_IF_NONE: deny all
1359 DEFAULT_DOC: X-Forwarded-For header will be ignored.
1361 Determine which client proxies can be trusted to provide correct
1362 information regarding real client IP address.
1364 Requests may pass through a chain of several other proxies
1365 before reaching us. The original source details may by sent in:
1366 * HTTP message Forwarded header, or
1367 * HTTP message X-Forwarded-For header, or
1368 * PROXY protocol connection header.
1370 PROXY protocol connections are controlled by the proxy_protocol_access
1371 directive which is checked before this.
1373 If a request reaches us from a source that is allowed by this
1374 directive, then we trust the information it provides regarding
1375 the IP of the client it received from (if any).
1377 For the purpose of ACLs used in this directive the src ACL type always
1378 matches the address we are testing and srcdomain matches its rDNS.
1380 On each HTTP request Squid checks for X-Forwarded-For header fields.
1381 If found the header values are iterated in reverse order and an allow
1382 match is required for Squid to continue on to the next value.
1383 The verification ends when a value receives a deny match, cannot be
1384 tested, or there are no more values to test.
1385 NOTE: Squid does not yet follow the Forwarded HTTP header.
1387 The end result of this process is an IP address that we will
1388 refer to as the indirect client address. This address may
1389 be treated as the client address for access control, ICAP, delay
1390 pools and logging, depending on the acl_uses_indirect_client,
1391 icap_uses_indirect_client, delay_pool_uses_indirect_client,
1392 log_uses_indirect_client and tproxy_uses_indirect_client options.
1394 This clause only supports fast acl types.
1395 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1397 SECURITY CONSIDERATIONS:
1399 Any host from which we accept client IP details can place
1400 incorrect information in the relevant header, and Squid
1401 will use the incorrect information as if it were the
1402 source address of the request. This may enable remote
1403 hosts to bypass any access control restrictions that are
1404 based on the client's source addresses.
1408 acl localhost src 127.0.0.1
1409 acl my_other_proxy srcdomain .proxy.example.com
1410 follow_x_forwarded_for allow localhost
1411 follow_x_forwarded_for allow my_other_proxy
1414 NAME: acl_uses_indirect_client
1417 IFDEF: FOLLOW_X_FORWARDED_FOR
1419 LOC: Config.onoff.acl_uses_indirect_client
1421 Controls whether the indirect client address
1422 (see follow_x_forwarded_for) is used instead of the
1423 direct client address in acl matching.
1425 NOTE: maxconn ACL considers direct TCP links and indirect
1426 clients will always have zero. So no match.
1429 NAME: delay_pool_uses_indirect_client
1432 IFDEF: FOLLOW_X_FORWARDED_FOR&&USE_DELAY_POOLS
1434 LOC: Config.onoff.delay_pool_uses_indirect_client
1436 Controls whether the indirect client address
1437 (see follow_x_forwarded_for) is used instead of the
1438 direct client address in delay pools.
1441 NAME: log_uses_indirect_client
1444 IFDEF: FOLLOW_X_FORWARDED_FOR
1446 LOC: Config.onoff.log_uses_indirect_client
1448 Controls whether the indirect client address
1449 (see follow_x_forwarded_for) is used instead of the
1450 direct client address in the access log.
1453 NAME: tproxy_uses_indirect_client
1456 IFDEF: FOLLOW_X_FORWARDED_FOR&&LINUX_NETFILTER
1458 LOC: Config.onoff.tproxy_uses_indirect_client
1460 Controls whether the indirect client address
1461 (see follow_x_forwarded_for) is used instead of the
1462 direct client address when spoofing the outgoing client.
1464 This has no effect on requests arriving in non-tproxy
1467 SECURITY WARNING: Usage of this option is dangerous
1468 and should not be used trivially. Correct configuration
1469 of follow_x_forwarded_for with a limited set of trusted
1470 sources is required to prevent abuse of your proxy.
1473 NAME: spoof_client_ip
1475 LOC: Config.accessList.spoof_client_ip
1477 DEFAULT_DOC: Allow spoofing on all TPROXY traffic.
1479 Control client IP address spoofing of TPROXY traffic based on
1480 defined access lists.
1482 spoof_client_ip allow|deny [!]aclname ...
1484 If there are no "spoof_client_ip" lines present, the default
1485 is to "allow" spoofing of any suitable request.
1487 Note that the cache_peer "no-tproxy" option overrides this ACL.
1489 This clause supports fast acl types.
1490 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1495 LOC: Config.accessList.http
1496 DEFAULT_IF_NONE: deny all
1497 DEFAULT_DOC: Deny, unless rules exist in squid.conf.
1499 Allowing or Denying access based on defined access lists
1501 To allow or deny a message received on an HTTP, HTTPS, or FTP port:
1502 http_access allow|deny [!]aclname ...
1504 NOTE on default values:
1506 If there are no "access" lines present, the default is to deny
1509 If none of the "access" lines cause a match, the default is the
1510 opposite of the last line in the list. If the last line was
1511 deny, the default is allow. Conversely, if the last line
1512 is allow, the default will be deny. For these reasons, it is a
1513 good idea to have an "deny all" entry at the end of your access
1514 lists to avoid potential confusion.
1516 This clause supports both fast and slow acl types.
1517 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1522 # Recommended minimum Access Permission configuration:
1524 # Deny requests to certain unsafe ports
1525 http_access deny !Safe_ports
1527 # Deny CONNECT to other than secure SSL ports
1528 http_access deny CONNECT !SSL_ports
1530 # Only allow cachemgr access from localhost
1531 http_access allow localhost manager
1532 http_access deny manager
1534 # We strongly recommend the following be uncommented to protect innocent
1535 # web applications running on the proxy server who think the only
1536 # one who can access services on "localhost" is a local user
1537 #http_access deny to_localhost
1540 # INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS
1543 # Example rule allowing access from your local networks.
1544 # Adapt localnet in the ACL section to list your (internal) IP networks
1545 # from where browsing should be allowed
1546 http_access allow localnet
1547 http_access allow localhost
1549 # And finally deny all other access to this proxy
1550 http_access deny all
1554 NAME: adapted_http_access http_access2
1556 LOC: Config.accessList.adapted_http
1558 DEFAULT_DOC: Allow, unless rules exist in squid.conf.
1560 Allowing or Denying access based on defined access lists
1562 Essentially identical to http_access, but runs after redirectors
1563 and ICAP/eCAP adaptation. Allowing access control based on their
1566 If not set then only http_access is used.
1569 NAME: http_reply_access
1571 LOC: Config.accessList.reply
1573 DEFAULT_DOC: Allow, unless rules exist in squid.conf.
1575 Allow replies to client requests. This is complementary to http_access.
1577 http_reply_access allow|deny [!] aclname ...
1579 NOTE: if there are no access lines present, the default is to allow
1582 If none of the access lines cause a match the opposite of the
1583 last line will apply. Thus it is good practice to end the rules
1584 with an "allow all" or "deny all" entry.
1586 This clause supports both fast and slow acl types.
1587 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1592 LOC: Config.accessList.icp
1594 DEFAULT_DOC: Deny, unless rules exist in squid.conf.
1596 Allowing or Denying access to the ICP port based on defined
1599 icp_access allow|deny [!]aclname ...
1601 NOTE: The default if no icp_access lines are present is to
1602 deny all traffic. This default may cause problems with peers
1605 This clause only supports fast acl types.
1606 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1608 # Allow ICP queries from local networks only
1609 #icp_access allow localnet
1610 #icp_access deny all
1616 LOC: Config.accessList.htcp
1618 DEFAULT_DOC: Deny, unless rules exist in squid.conf.
1620 Allowing or Denying access to the HTCP port based on defined
1623 htcp_access allow|deny [!]aclname ...
1625 See also htcp_clr_access for details on access control for
1626 cache purge (CLR) HTCP messages.
1628 NOTE: The default if no htcp_access lines are present is to
1629 deny all traffic. This default may cause problems with peers
1630 using the htcp option.
1632 This clause only supports fast acl types.
1633 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1635 # Allow HTCP queries from local networks only
1636 #htcp_access allow localnet
1637 #htcp_access deny all
1640 NAME: htcp_clr_access
1643 LOC: Config.accessList.htcp_clr
1645 DEFAULT_DOC: Deny, unless rules exist in squid.conf.
1647 Allowing or Denying access to purge content using HTCP based
1648 on defined access lists.
1649 See htcp_access for details on general HTCP access control.
1651 htcp_clr_access allow|deny [!]aclname ...
1653 This clause only supports fast acl types.
1654 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1656 # Allow HTCP CLR requests from trusted peers
1657 acl htcp_clr_peer src 192.0.2.2 2001:DB8::2
1658 htcp_clr_access allow htcp_clr_peer
1659 htcp_clr_access deny all
1664 LOC: Config.accessList.miss
1666 DEFAULT_DOC: Allow, unless rules exist in squid.conf.
1668 Determines whether network access is permitted when satisfying a request.
1671 to force your neighbors to use you as a sibling instead of
1674 acl localclients src 192.0.2.0/24 2001:DB8::a:0/64
1675 miss_access deny !localclients
1676 miss_access allow all
1678 This means only your local clients are allowed to fetch relayed/MISS
1679 replies from the network and all other clients can only fetch cached
1682 The default for this setting allows all clients who passed the
1683 http_access rules to relay via this proxy.
1685 This clause only supports fast acl types.
1686 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1689 NAME: ident_lookup_access
1693 DEFAULT_DOC: Unless rules exist in squid.conf, IDENT is not fetched.
1694 LOC: Ident::TheConfig.identLookup
1696 A list of ACL elements which, if matched, cause an ident
1697 (RFC 931) lookup to be performed for this request. For
1698 example, you might choose to always perform ident lookups
1699 for your main multi-user Unix boxes, but not for your Macs
1700 and PCs. By default, ident lookups are not performed for
1703 To enable ident lookups for specific client addresses, you
1704 can follow this example:
1706 acl ident_aware_hosts src 198.168.1.0/24
1707 ident_lookup_access allow ident_aware_hosts
1708 ident_lookup_access deny all
1710 Only src type ACL checks are fully supported. A srcdomain
1711 ACL might work at times, but it will not always provide
1714 This clause only supports fast acl types.
1715 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1718 NAME: reply_body_max_size
1719 COMMENT: size [acl acl...]
1722 DEFAULT_DOC: No limit is applied.
1723 LOC: Config.ReplyBodySize
1725 This option specifies the maximum size of a reply body. It can be
1726 used to prevent users from downloading very large files, such as
1727 MP3's and movies. When the reply headers are received, the
1728 reply_body_max_size lines are processed, and the first line where
1729 all (if any) listed ACLs are true is used as the maximum body size
1732 This size is checked twice. First when we get the reply headers,
1733 we check the content-length value. If the content length value exists
1734 and is larger than the allowed size, the request is denied and the
1735 user receives an error message that says "the request or reply
1736 is too large." If there is no content-length, and the reply
1737 size exceeds this limit, the client's connection is just closed
1738 and they will receive a partial reply.
1740 WARNING: downstream caches probably can not detect a partial reply
1741 if there is no content-length header, so they will cache
1742 partial responses and give them out as hits. You should NOT
1743 use this option if you have downstream caches.
1745 WARNING: A maximum size smaller than the size of squid's error messages
1746 will cause an infinite loop and crash squid. Ensure that the smallest
1747 non-zero value you use is greater that the maximum header size plus
1748 the size of your largest error page.
1750 If you set this parameter none (the default), there will be
1753 Configuration Format is:
1754 reply_body_max_size SIZE UNITS [acl ...]
1756 reply_body_max_size 10 MB
1760 NAME: on_unsupported_protocol
1761 TYPE: on_unsupported_protocol
1762 LOC: Config.accessList.on_unsupported_protocol
1764 DEFAULT_DOC: Respond with an error message to unidentifiable traffic
1766 Determines Squid behavior when encountering strange requests at the
1767 beginning of an accepted TCP connection or the beginning of a bumped
1768 CONNECT tunnel. Controlling Squid reaction to unexpected traffic is
1769 especially useful in interception environments where Squid is likely
1770 to see connections for unsupported protocols that Squid should either
1771 terminate or tunnel at TCP level.
1773 on_unsupported_protocol <action> [!]acl ...
1775 The first matching action wins. Only fast ACLs are supported.
1777 Supported actions are:
1779 tunnel: Establish a TCP connection with the intended server and
1780 blindly shovel TCP packets between the client and server.
1782 respond: Respond with an error message, using the transfer protocol
1783 for the Squid port that received the request (e.g., HTTP
1784 for connections intercepted at the http_port). This is the
1787 Squid expects the following traffic patterns:
1789 http_port: a plain HTTP request
1790 https_port: SSL/TLS handshake followed by an [encrypted] HTTP request
1791 ftp_port: a plain FTP command (no on_unsupported_protocol support yet!)
1792 CONNECT tunnel on http_port: same as https_port
1793 CONNECT tunnel on https_port: same as https_port
1795 Currently, this directive has effect on intercepted connections and
1796 bumped tunnels only. Other cases are not supported because Squid
1797 cannot know the intended destination of other traffic.
1800 # define what Squid errors indicate receiving non-HTTP traffic:
1801 acl foreignProtocol squid_error ERR_PROTOCOL_UNKNOWN ERR_TOO_BIG
1802 # define what Squid errors indicate receiving nothing:
1803 acl serverTalksFirstProtocol squid_error ERR_REQUEST_START_TIMEOUT
1804 # tunnel everything that does not look like HTTP:
1805 on_unsupported_protocol tunnel foreignProtocol
1806 # tunnel if we think the client waits for the server to talk first:
1807 on_unsupported_protocol tunnel serverTalksFirstProtocol
1808 # in all other error cases, just send an HTTP "error page" response:
1809 on_unsupported_protocol respond all
1811 See also: squid_error ACL
1816 -----------------------------------------------------------------------------
1819 NAME: http_port ascii_port
1824 Usage: port [mode] [options]
1825 hostname:port [mode] [options]
1826 1.2.3.4:port [mode] [options]
1828 The socket addresses where Squid will listen for HTTP client
1829 requests. You may specify multiple socket addresses.
1830 There are three forms: port alone, hostname with port, and
1831 IP address with port. If you specify a hostname or IP
1832 address, Squid binds the socket to that specific
1833 address. Most likely, you do not need to bind to a specific
1834 address, so you can use the port number alone.
1836 If you are running Squid in accelerator mode, you
1837 probably want to listen on port 80 also, or instead.
1839 The -a command line option may be used to specify additional
1840 port(s) where Squid listens for proxy request. Such ports will
1841 be plain proxy ports with no options.
1843 You may specify multiple socket addresses on multiple lines.
1847 intercept Support for IP-Layer NAT interception delivering
1848 traffic to this Squid port.
1849 NP: disables authentication on the port.
1851 tproxy Support Linux TPROXY (or BSD divert-to) with spoofing
1852 of outgoing connections using the client IP address.
1853 NP: disables authentication on the port.
1855 accel Accelerator / reverse proxy mode
1857 ssl-bump For each CONNECT request allowed by ssl_bump ACLs,
1858 establish secure connection with the client and with
1859 the server, decrypt HTTPS messages as they pass through
1860 Squid, and treat them as unencrypted HTTP messages,
1861 becoming the man-in-the-middle.
1863 The ssl_bump option is required to fully enable
1864 bumping of CONNECT requests.
1866 Omitting the mode flag causes default forward proxy mode to be used.
1869 Accelerator Mode Options:
1871 defaultsite=domainname
1872 What to use for the Host: header if it is not present
1873 in a request. Determines what site (not origin server)
1874 accelerators should consider the default.
1876 no-vhost Disable using HTTP/1.1 Host header for virtual domain support.
1878 protocol= Protocol to reconstruct accelerated and intercepted
1879 requests with. Defaults to HTTP/1.1 for http_port and
1880 HTTPS/1.1 for https_port.
1881 When an unsupported value is configured Squid will
1882 produce a FATAL error.
1883 Values: HTTP or HTTP/1.1, HTTPS or HTTPS/1.1
1885 vport Virtual host port support. Using the http_port number
1886 instead of the port passed on Host: headers.
1888 vport=NN Virtual host port support. Using the specified port
1889 number instead of the port passed on Host: headers.
1892 Act as if this Squid is the origin server.
1893 This currently means generate new Date: and Expires:
1894 headers on HIT instead of adding Age:.
1896 ignore-cc Ignore request Cache-Control headers.
1898 WARNING: This option violates HTTP specifications if
1899 used in non-accelerator setups.
1901 allow-direct Allow direct forwarding in accelerator mode. Normally
1902 accelerated requests are denied direct forwarding as if
1903 never_direct was used.
1905 WARNING: this option opens accelerator mode to security
1906 vulnerabilities usually only affecting in interception
1907 mode. Make sure to protect forwarding with suitable
1908 http_access rules when using this.
1911 SSL Bump Mode Options:
1912 In addition to these options ssl-bump requires TLS/SSL options.
1914 generate-host-certificates[=<on|off>]
1915 Dynamically create SSL server certificates for the
1916 destination hosts of bumped CONNECT requests.When
1917 enabled, the cert and key options are used to sign
1918 generated certificates. Otherwise generated
1919 certificate will be selfsigned.
1920 If there is a CA certificate lifetime of the generated
1921 certificate equals lifetime of the CA certificate. If
1922 generated certificate is selfsigned lifetime is three
1924 This option is enabled by default when ssl-bump is used.
1925 See the ssl-bump option above for more information.
1927 dynamic_cert_mem_cache_size=SIZE
1928 Approximate total RAM size spent on cached generated
1929 certificates. If set to zero, caching is disabled. The
1930 default value is 4MB.
1934 cert= Path to SSL certificate (PEM format).
1936 key= Path to SSL private key file (PEM format)
1937 if not specified, the certificate file is
1938 assumed to be a combined certificate and
1941 cipher= Colon separated list of supported ciphers.
1942 NOTE: some ciphers such as EDH ciphers depend on
1943 additional settings. If those settings are
1944 omitted the ciphers may be silently ignored
1945 by the OpenSSL library.
1947 options= Various SSL implementation options. The most important
1950 NO_SSLv3 Disallow the use of SSLv3
1952 NO_TLSv1 Disallow the use of TLSv1.0
1954 NO_TLSv1_1 Disallow the use of TLSv1.1
1956 NO_TLSv1_2 Disallow the use of TLSv1.2
1959 Always create a new key when using
1960 temporary/ephemeral DH key exchanges
1963 Enable ephemeral ECDH key exchange.
1964 The adopted curve should be specified
1965 using the tls-dh option.
1968 Disable use of RFC5077 session tickets.
1969 Some servers may have problems
1970 understanding the TLS extension due
1971 to ambiguous specification in RFC4507.
1973 ALL Enable various bug workarounds
1974 suggested as "harmless" by OpenSSL
1975 Be warned that this reduces SSL/TLS
1976 strength to some attacks.
1978 See the OpenSSL SSL_CTX_set_options documentation for a
1981 clientca= File containing the list of CAs to use when
1982 requesting a client certificate.
1984 tls-cafile= PEM file containing CA certificates to use when verifying
1985 client certificates. If not configured clientca will be
1986 used. May be repeated to load multiple files.
1988 capath= Directory containing additional CA certificates
1989 and CRL lists to use when verifying client certificates.
1990 Requires OpenSSL or LibreSSL.
1992 crlfile= File of additional CRL lists to use when verifying
1993 the client certificate, in addition to CRLs stored in
1994 the capath. Implies VERIFY_CRL flag below.
1997 File containing DH parameters for temporary/ephemeral DH key
1998 exchanges, optionally prefixed by a curve for ephemeral ECDH
2000 See OpenSSL documentation for details on how to create the
2001 DH parameter file. Supported curves for ECDH can be listed
2002 using the "openssl ecparam -list_curves" command.
2003 WARNING: EDH and EECDH ciphers will be silently disabled if
2004 this option is not set.
2006 sslflags= Various flags modifying the use of SSL:
2008 Don't request client certificates
2009 immediately, but wait until acl processing
2010 requires a certificate (not yet implemented).
2012 Don't allow for session reuse. Each connection
2013 will result in a new SSL session.
2015 Verify CRL lists when accepting client
2018 Verify CRL lists for all certificates in the
2019 client certificate chain.
2021 tls-default-ca[=off]
2022 Whether to use the system Trusted CAs. Default is OFF.
2024 tls-no-npn Do not use the TLS NPN extension to advertise HTTP/1.1.
2026 sslcontext= SSL session ID context identifier.
2030 connection-auth[=on|off]
2031 use connection-auth=off to tell Squid to prevent
2032 forwarding Microsoft connection oriented authentication
2033 (NTLM, Negotiate and Kerberos)
2035 disable-pmtu-discovery=
2036 Control Path-MTU discovery usage:
2037 off lets OS decide on what to do (default).
2038 transparent disable PMTU discovery when transparent
2040 always disable always PMTU discovery.
2042 In many setups of transparently intercepting proxies
2043 Path-MTU discovery can not work on traffic towards the
2044 clients. This is the case when the intercepting device
2045 does not fully track connections and fails to forward
2046 ICMP must fragment messages to the cache server. If you
2047 have such setup and experience that certain clients
2048 sporadically hang or never complete requests set
2049 disable-pmtu-discovery option to 'transparent'.
2051 name= Specifies a internal name for the port. Defaults to
2052 the port specification (port or addr:port)
2054 tcpkeepalive[=idle,interval,timeout]
2055 Enable TCP keepalive probes of idle connections.
2056 In seconds; idle is the initial time before TCP starts
2057 probing the connection, interval how often to probe, and
2058 timeout the time before giving up.
2060 require-proxy-header
2061 Require PROXY protocol version 1 or 2 connections.
2062 The proxy_protocol_access is required to whitelist
2063 downstream proxies which can be trusted.
2065 If you run Squid on a dual-homed machine with an internal
2066 and an external interface we recommend you to specify the
2067 internal address:port in http_port. This way Squid will only be
2068 visible on the internal address.
2072 # Squid normally listens to port 3128
2073 http_port @DEFAULT_HTTP_PORT@
2078 IFDEF: USE_GNUTLS||USE_OPENSSL
2083 Usage: [ip:]port [mode] cert=certificate.pem [options]
2085 The socket address where Squid will listen for client requests made
2086 over TLS or SSL connections. Commonly referred to as HTTPS.
2088 This is most useful for situations where you are running squid in
2089 accelerator mode and you want to do the TLS work at the accelerator level.
2091 You may specify multiple socket addresses on multiple lines,
2092 each with their own certificate and/or options.
2094 The TLS cert= option is mandatory on HTTPS ports.
2096 See http_port for a list of modes and options.
2104 Enables Native FTP proxy by specifying the socket address where Squid
2105 listens for FTP client requests. See http_port directive for various
2106 ways to specify the listening address and mode.
2108 Usage: ftp_port address [mode] [options]
2110 WARNING: This is a new, experimental, complex feature that has seen
2111 limited production exposure. Some Squid modules (e.g., caching) do not
2112 currently work with native FTP proxying, and many features have not
2113 even been tested for compatibility. Test well before deploying!
2115 Native FTP proxying differs substantially from proxying HTTP requests
2116 with ftp:// URIs because Squid works as an FTP server and receives
2117 actual FTP commands (rather than HTTP requests with FTP URLs).
2119 Native FTP commands accepted at ftp_port are internally converted or
2120 wrapped into HTTP-like messages. The same happens to Native FTP
2121 responses received from FTP origin servers. Those HTTP-like messages
2122 are shoveled through regular access control and adaptation layers
2123 between the FTP client and the FTP origin server. This allows Squid to
2124 examine, adapt, block, and log FTP exchanges. Squid reuses most HTTP
2125 mechanisms when shoveling wrapped FTP messages. For example,
2126 http_access and adaptation_access directives are used.
2130 intercept Same as http_port intercept. The FTP origin address is
2131 determined based on the intended destination of the
2132 intercepted connection.
2134 tproxy Support Linux TPROXY for spoofing outgoing
2135 connections using the client IP address.
2136 NP: disables authentication and maybe IPv6 on the port.
2138 By default (i.e., without an explicit mode option), Squid extracts the
2139 FTP origin address from the login@origin parameter of the FTP USER
2140 command. Many popular FTP clients support such native FTP proxying.
2144 name=token Specifies an internal name for the port. Defaults to
2145 the port address. Usable with myportname ACL.
2148 Enables tracking of FTP directories by injecting extra
2149 PWD commands and adjusting Request-URI (in wrapping
2150 HTTP requests) to reflect the current FTP server
2151 directory. Tracking is disabled by default.
2153 protocol=FTP Protocol to reconstruct accelerated and intercepted
2154 requests with. Defaults to FTP. No other accepted
2155 values have been tested with. An unsupported value
2156 results in a FATAL error. Accepted values are FTP,
2157 HTTP (or HTTP/1.1), and HTTPS (or HTTPS/1.1).
2159 Other http_port modes and options that are not specific to HTTP and
2160 HTTPS may also work.
2163 NAME: tcp_outgoing_tos tcp_outgoing_ds tcp_outgoing_dscp
2166 LOC: Ip::Qos::TheConfig.tosToServer
2168 Allows you to select a TOS/Diffserv value for packets outgoing
2169 on the server side, based on an ACL.
2171 tcp_outgoing_tos ds-field [!]aclname ...
2173 Example where normal_service_net uses the TOS value 0x00
2174 and good_service_net uses 0x20
2176 acl normal_service_net src 10.0.0.0/24
2177 acl good_service_net src 10.0.1.0/24
2178 tcp_outgoing_tos 0x00 normal_service_net
2179 tcp_outgoing_tos 0x20 good_service_net
2181 TOS/DSCP values really only have local significance - so you should
2182 know what you're specifying. For more information, see RFC2474,
2183 RFC2475, and RFC3260.
2185 The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
2186 "default" to use whatever default your host has.
2187 Note that only multiples of 4 are usable as the two rightmost bits have
2188 been redefined for use by ECN (RFC 3168 section 23.1).
2189 The squid parser will enforce this by masking away the ECN bits.
2191 Processing proceeds in the order specified, and stops at first fully
2194 Only fast ACLs are supported.
2197 NAME: clientside_tos
2200 LOC: Ip::Qos::TheConfig.tosToClient
2202 Allows you to select a TOS/DSCP value for packets being transmitted
2203 on the client-side, based on an ACL.
2205 clientside_tos ds-field [!]aclname ...
2207 Example where normal_service_net uses the TOS value 0x00
2208 and good_service_net uses 0x20
2210 acl normal_service_net src 10.0.0.0/24
2211 acl good_service_net src 10.0.1.0/24
2212 clientside_tos 0x00 normal_service_net
2213 clientside_tos 0x20 good_service_net
2215 Note: This feature is incompatible with qos_flows. Any TOS values set here
2216 will be overwritten by TOS values in qos_flows.
2218 The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
2219 "default" to use whatever default your host has.
2220 Note that only multiples of 4 are usable as the two rightmost bits have
2221 been redefined for use by ECN (RFC 3168 section 23.1).
2222 The squid parser will enforce this by masking away the ECN bits.
2226 NAME: tcp_outgoing_mark
2228 IFDEF: SO_MARK&&USE_LIBCAP
2230 LOC: Ip::Qos::TheConfig.nfmarkToServer
2232 Allows you to apply a Netfilter mark value to outgoing packets
2233 on the server side, based on an ACL.
2235 tcp_outgoing_mark mark-value [!]aclname ...
2237 Example where normal_service_net uses the mark value 0x00
2238 and good_service_net uses 0x20
2240 acl normal_service_net src 10.0.0.0/24
2241 acl good_service_net src 10.0.1.0/24
2242 tcp_outgoing_mark 0x00 normal_service_net
2243 tcp_outgoing_mark 0x20 good_service_net
2245 Only fast ACLs are supported.
2248 NAME: clientside_mark
2250 IFDEF: SO_MARK&&USE_LIBCAP
2252 LOC: Ip::Qos::TheConfig.nfmarkToClient
2254 Allows you to apply a Netfilter mark value to packets being transmitted
2255 on the client-side, based on an ACL.
2257 clientside_mark mark-value [!]aclname ...
2259 Example where normal_service_net uses the mark value 0x00
2260 and good_service_net uses 0x20
2262 acl normal_service_net src 10.0.0.0/24
2263 acl good_service_net src 10.0.1.0/24
2264 clientside_mark 0x00 normal_service_net
2265 clientside_mark 0x20 good_service_net
2267 Note: This feature is incompatible with qos_flows. Any mark values set here
2268 will be overwritten by mark values in qos_flows.
2275 LOC: Ip::Qos::TheConfig
2277 Allows you to select a TOS/DSCP value to mark outgoing
2278 connections to the client, based on where the reply was sourced.
2279 For platforms using netfilter, allows you to set a netfilter mark
2280 value instead of, or in addition to, a TOS value.
2282 By default this functionality is disabled. To enable it with the default
2283 settings simply use "qos_flows mark" or "qos_flows tos". Default
2284 settings will result in the netfilter mark or TOS value being copied
2285 from the upstream connection to the client. Note that it is the connection
2286 CONNMARK value not the packet MARK value that is copied.
2288 It is not currently possible to copy the mark or TOS value from the
2289 client to the upstream connection request.
2291 TOS values really only have local significance - so you should
2292 know what you're specifying. For more information, see RFC2474,
2293 RFC2475, and RFC3260.
2295 The TOS/DSCP byte must be exactly that - a octet value 0 - 255.
2296 Note that only multiples of 4 are usable as the two rightmost bits have
2297 been redefined for use by ECN (RFC 3168 section 23.1).
2298 The squid parser will enforce this by masking away the ECN bits.
2300 Mark values can be any unsigned 32-bit integer value.
2302 This setting is configured by setting the following values:
2304 tos|mark Whether to set TOS or netfilter mark values
2306 local-hit=0xFF Value to mark local cache hits.
2308 sibling-hit=0xFF Value to mark hits from sibling peers.
2310 parent-hit=0xFF Value to mark hits from parent peers.
2312 miss=0xFF[/mask] Value to mark cache misses. Takes precedence
2313 over the preserve-miss feature (see below), unless
2314 mask is specified, in which case only the bits
2315 specified in the mask are written.
2317 The TOS variant of the following features are only possible on Linux
2318 and require your kernel to be patched with the TOS preserving ZPH
2319 patch, available from http://zph.bratcheda.org
2320 No patch is needed to preserve the netfilter mark, which will work
2321 with all variants of netfilter.
2323 disable-preserve-miss
2324 This option disables the preservation of the TOS or netfilter
2325 mark. By default, the existing TOS or netfilter mark value of
2326 the response coming from the remote server will be retained
2327 and masked with miss-mark.
2328 NOTE: in the case of a netfilter mark, the mark must be set on
2329 the connection (using the CONNMARK target) not on the packet
2333 Allows you to mask certain bits in the TOS or mark value
2334 received from the remote server, before copying the value to
2335 the TOS sent towards clients.
2336 Default for tos: 0xFF (TOS from server is not changed).
2337 Default for mark: 0xFFFFFFFF (mark from server is not changed).
2339 All of these features require the --enable-zph-qos compilation flag
2340 (enabled by default). Netfilter marking also requires the
2341 libnetfilter_conntrack libraries (--with-netfilter-conntrack) and
2342 libcap 2.09+ (--with-libcap).
2346 NAME: tcp_outgoing_address
2349 DEFAULT_DOC: Address selection is performed by the operating system.
2350 LOC: Config.accessList.outgoing_address
2352 Allows you to map requests to different outgoing IP addresses
2353 based on the username or source address of the user making
2356 tcp_outgoing_address ipaddr [[!]aclname] ...
2359 Forwarding clients with dedicated IPs for certain subnets.
2361 acl normal_service_net src 10.0.0.0/24
2362 acl good_service_net src 10.0.2.0/24
2364 tcp_outgoing_address 2001:db8::c001 good_service_net
2365 tcp_outgoing_address 10.1.0.2 good_service_net
2367 tcp_outgoing_address 2001:db8::beef normal_service_net
2368 tcp_outgoing_address 10.1.0.1 normal_service_net
2370 tcp_outgoing_address 2001:db8::1
2371 tcp_outgoing_address 10.1.0.3
2373 Processing proceeds in the order specified, and stops at first fully
2376 Squid will add an implicit IP version test to each line.
2377 Requests going to IPv4 websites will use the outgoing 10.1.0.* addresses.
2378 Requests going to IPv6 websites will use the outgoing 2001:db8:* addresses.
2381 NOTE: The use of this directive using client dependent ACLs is
2382 incompatible with the use of server side persistent connections. To
2383 ensure correct results it is best to set server_persistent_connections
2384 to off when using this directive in such configurations.
2386 NOTE: The use of this directive to set a local IP on outgoing TCP links
2387 is incompatible with using TPROXY to set client IP out outbound TCP links.
2388 When needing to contact peers use the no-tproxy cache_peer option and the
2389 client_dst_passthru directive re-enable normal forwarding such as this.
2393 NAME: host_verify_strict
2396 LOC: Config.onoff.hostStrictVerify
2398 Regardless of this option setting, when dealing with intercepted
2399 traffic, Squid always verifies that the destination IP address matches
2400 the Host header domain or IP (called 'authority form URL').
2402 This enforcement is performed to satisfy a MUST-level requirement in
2403 RFC 2616 section 14.23: "The Host field value MUST represent the naming
2404 authority of the origin server or gateway given by the original URL".
2407 Squid always responds with an HTTP 409 (Conflict) error
2408 page and logs a security warning if there is no match.
2410 Squid verifies that the destination IP address matches
2411 the Host header for forward-proxy and reverse-proxy traffic
2412 as well. For those traffic types, Squid also enables the
2413 following checks, comparing the corresponding Host header
2414 and Request-URI components:
2416 * The host names (domain or IP) must be identical,
2417 but valueless or missing Host header disables all checks.
2418 For the two host names to match, both must be either IP
2421 * Port numbers must be identical, but if a port is missing
2422 the scheme-default port is assumed.
2425 When set to OFF (the default):
2426 Squid allows suspicious requests to continue but logs a
2427 security warning and blocks caching of the response.
2429 * Forward-proxy traffic is not checked at all.
2431 * Reverse-proxy traffic is not checked at all.
2433 * Intercepted traffic which passes verification is handled
2434 according to client_dst_passthru.
2436 * Intercepted requests which fail verification are sent
2437 to the client original destination instead of DIRECT.
2438 This overrides 'client_dst_passthru off'.
2440 For now suspicious intercepted CONNECT requests are always
2441 responded to with an HTTP 409 (Conflict) error page.
2446 As described in CVE-2009-0801 when the Host: header alone is used
2447 to determine the destination of a request it becomes trivial for
2448 malicious scripts on remote websites to bypass browser same-origin
2449 security policy and sandboxing protections.
2451 The cause of this is that such applets are allowed to perform their
2452 own HTTP stack, in which case the same-origin policy of the browser
2453 sandbox only verifies that the applet tries to contact the same IP
2454 as from where it was loaded at the IP level. The Host: header may
2455 be different from the connected IP and approved origin.
2459 NAME: client_dst_passthru
2462 LOC: Config.onoff.client_dst_passthru
2464 With NAT or TPROXY intercepted traffic Squid may pass the request
2465 directly to the original client destination IP or seek a faster
2466 source using the HTTP Host header.
2468 Using Host to locate alternative servers can provide faster
2469 connectivity with a range of failure recovery options.
2470 But can also lead to connectivity trouble when the client and
2471 server are attempting stateful interactions unaware of the proxy.
2473 This option (on by default) prevents alternative DNS entries being
2474 located to send intercepted traffic DIRECT to an origin server.
2475 The clients original destination IP and port will be used instead.
2477 Regardless of this option setting, when dealing with intercepted
2478 traffic Squid will verify the Host: header and any traffic which
2479 fails Host verification will be treated as if this option were ON.
2481 see host_verify_strict for details on the verification process.
2486 -----------------------------------------------------------------------------
2489 NAME: tls_outgoing_options
2490 IFDEF: USE_GNUTLS||USE_OPENSSL
2491 TYPE: securePeerOptions
2492 DEFAULT: min-version=1.0
2493 LOC: Security::ProxyOutgoingConfig
2495 disable Do not support https:// URLs.
2497 cert=/path/to/client/certificate
2498 A client TLS certificate to use when connecting.
2500 key=/path/to/client/private_key
2501 The private TLS key corresponding to the cert= above.
2502 If key= is not specified cert= is assumed to reference
2503 a PEM file containing both the certificate and the key.
2505 cipher=... The list of valid TLS ciphers to use.
2508 The minimum TLS protocol version to permit.
2509 To control SSLv3 use the options= parameter.
2510 Supported Values: 1.0 (default), 1.1, 1.2
2512 options=... Specify various TLS/SSL implementation options:
2514 NO_SSLv3 Disallow the use of SSLv3
2516 NO_TLSv1 Disallow the use of TLSv1.0
2518 NO_TLSv1_1 Disallow the use of TLSv1.1
2520 NO_TLSv1_2 Disallow the use of TLSv1.2
2523 Always create a new key when using
2524 temporary/ephemeral DH key exchanges
2527 Disable use of RFC5077 session tickets.
2528 Some servers may have problems
2529 understanding the TLS extension due
2530 to ambiguous specification in RFC4507.
2532 ALL Enable various bug workarounds
2533 suggested as "harmless" by OpenSSL
2534 Be warned that this reduces SSL/TLS
2535 strength to some attacks.
2537 See the OpenSSL SSL_CTX_set_options documentation for a
2540 cafile= PEM file containing CA certificates to use when verifying
2541 the peer certificate. May be repeated to load multiple files.
2543 capath= A directory containing additional CA certificates to
2544 use when verifying the peer certificate.
2545 Requires OpenSSL or LibreSSL.
2547 crlfile=... A certificate revocation list file to use when
2548 verifying the peer certificate.
2550 flags=... Specify various flags modifying the TLS implementation:
2553 Accept certificates even if they fail to
2556 Don't verify the peer certificate
2557 matches the server name
2560 Whether to use the system Trusted CAs. Default is ON.
2562 domain= The peer name as advertised in its certificate.
2563 Used for verifying the correctness of the received peer
2564 certificate. If not specified the peer hostname will be
2570 -----------------------------------------------------------------------------
2573 NAME: ssl_unclean_shutdown
2577 LOC: Config.SSL.unclean_shutdown
2579 Some browsers (especially MSIE) bugs out on SSL shutdown
2586 LOC: Config.SSL.ssl_engine
2589 The OpenSSL engine to use. You will need to set this if you
2590 would like to use hardware SSL acceleration for example.
2593 NAME: sslproxy_session_ttl
2596 LOC: Config.SSL.session_ttl
2599 Sets the timeout value for SSL sessions
2602 NAME: sslproxy_session_cache_size
2605 LOC: Config.SSL.sessionCacheSize
2608 Sets the cache size to use for ssl session
2611 NAME: sslproxy_foreign_intermediate_certs
2614 LOC: Config.ssl_client.foreignIntermediateCertsPath
2617 Many origin servers fail to send their full server certificate
2618 chain for verification, assuming the client already has or can
2619 easily locate any missing intermediate certificates.
2621 Squid uses the certificates from the specified file to fill in
2622 these missing chains when trying to validate origin server
2625 The file is expected to contain zero or more PEM-encoded
2626 intermediate certificates. These certificates are not treated
2627 as trusted root certificates, and any self-signed certificate in
2628 this file will be ignored.
2631 NAME: sslproxy_cert_sign_hash
2634 LOC: Config.SSL.certSignHash
2637 Sets the hashing algorithm to use when signing generated certificates.
2638 Valid algorithm names depend on the OpenSSL library used. The following
2639 names are usually available: sha1, sha256, sha512, and md5. Please see
2640 your OpenSSL library manual for the available hashes. By default, Squids
2641 that support this option use sha256 hashes.
2643 Squid does not forcefully purge cached certificates that were generated
2644 with an algorithm other than the currently configured one. They remain
2645 in the cache, subject to the regular cache eviction policy, and become
2646 useful if the algorithm changes again.
2651 TYPE: sslproxy_ssl_bump
2652 LOC: Config.accessList.ssl_bump
2653 DEFAULT_DOC: Become a TCP tunnel without decrypting proxied traffic.
2656 This option is consulted when a CONNECT request is received on
2657 an http_port (or a new connection is intercepted at an
2658 https_port), provided that port was configured with an ssl-bump
2659 flag. The subsequent data on the connection is either treated as
2660 HTTPS and decrypted OR tunneled at TCP level without decryption,
2661 depending on the first matching bumping "action".
2663 ssl_bump <action> [!]acl ...
2665 The following bumping actions are currently supported:
2668 Become a TCP tunnel without decrypting proxied traffic.
2669 This is the default action.
2672 Establish a secure connection with the server and, using a
2673 mimicked server certificate, with the client.
2676 Receive client (step SslBump1) or server (step SslBump2)
2677 certificate while preserving the possibility of splicing the
2678 connection. Peeking at the server certificate (during step 2)
2679 usually precludes bumping of the connection at step 3.
2682 Receive client (step SslBump1) or server (step SslBump2)
2683 certificate while preserving the possibility of bumping the
2684 connection. Staring at the server certificate (during step 2)
2685 usually precludes splicing of the connection at step 3.
2688 Close client and server connections.
2690 Backward compatibility actions available at step SslBump1:
2693 Bump the connection. Establish a secure connection with the
2694 client first, then connect to the server. This old mode does
2695 not allow Squid to mimic server SSL certificate and does not
2696 work with intercepted SSL connections.
2699 Bump the connection. Establish a secure connection with the
2700 server first, then establish a secure connection with the
2701 client, using a mimicked server certificate. Works with both
2702 CONNECT requests and intercepted SSL connections, but does
2703 not allow to make decisions based on SSL handshake info.
2706 Decide whether to bump or splice the connection based on
2707 client-to-squid and server-to-squid SSL hello messages.
2711 Same as the "splice" action.
2713 All ssl_bump rules are evaluated at each of the supported bumping
2714 steps. Rules with actions that are impossible at the current step are
2715 ignored. The first matching ssl_bump action wins and is applied at the
2716 end of the current step. If no rules match, the splice action is used.
2717 See the at_step ACL for a list of the supported SslBump steps.
2719 This clause supports both fast and slow acl types.
2720 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2722 See also: http_port ssl-bump, https_port ssl-bump, and acl at_step.
2725 # Example: Bump all TLS connections except those originating from
2726 # localhost or those going to example.com.
2728 acl broken_sites ssl::server_name .example.com
2729 ssl_bump splice localhost
2730 ssl_bump splice broken_sites
2734 NAME: sslproxy_cert_error
2737 DEFAULT_DOC: Server certificate errors terminate the transaction.
2738 LOC: Config.ssl_client.cert_error
2741 Use this ACL to bypass server certificate validation errors.
2743 For example, the following lines will bypass all validation errors
2744 when talking to servers for example.com. All other
2745 validation errors will result in ERR_SECURE_CONNECT_FAIL error.
2747 acl BrokenButTrustedServers dstdomain example.com
2748 sslproxy_cert_error allow BrokenButTrustedServers
2749 sslproxy_cert_error deny all
2751 This clause only supports fast acl types.
2752 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
2753 Using slow acl types may result in server crashes
2755 Without this option, all server certificate validation errors
2756 terminate the transaction to protect Squid and the client.
2758 SQUID_X509_V_ERR_INFINITE_VALIDATION error cannot be bypassed
2759 but should not happen unless your OpenSSL library is buggy.
2762 Bypassing validation errors is dangerous because an
2763 error usually implies that the server cannot be trusted
2764 and the connection may be insecure.
2766 See also: sslproxy_flags and DONT_VERIFY_PEER.
2769 NAME: sslproxy_cert_sign
2772 POSTSCRIPTUM: signUntrusted ssl::certUntrusted
2773 POSTSCRIPTUM: signSelf ssl::certSelfSigned
2774 POSTSCRIPTUM: signTrusted all
2775 TYPE: sslproxy_cert_sign
2776 LOC: Config.ssl_client.cert_sign
2779 sslproxy_cert_sign <signing algorithm> acl ...
2781 The following certificate signing algorithms are supported:
2784 Sign using the configured CA certificate which is usually
2785 placed in and trusted by end-user browsers. This is the
2786 default for trusted origin server certificates.
2789 Sign to guarantee an X509_V_ERR_CERT_UNTRUSTED browser error.
2790 This is the default for untrusted origin server certificates
2791 that are not self-signed (see ssl::certUntrusted).
2794 Sign using a self-signed certificate with the right CN to
2795 generate a X509_V_ERR_DEPTH_ZERO_SELF_SIGNED_CERT error in the
2796 browser. This is the default for self-signed origin server
2797 certificates (see ssl::certSelfSigned).
2799 This clause only supports fast acl types.
2801 When sslproxy_cert_sign acl(s) match, Squid uses the corresponding
2802 signing algorithm to generate the certificate and ignores all
2803 subsequent sslproxy_cert_sign options (the first match wins). If no
2804 acl(s) match, the default signing algorithm is determined by errors
2805 detected when obtaining and validating the origin server certificate.
2807 WARNING: SQUID_X509_V_ERR_DOMAIN_MISMATCH and ssl:certDomainMismatch can
2808 be used with sslproxy_cert_adapt, but if and only if Squid is bumping a
2809 CONNECT request that carries a domain name. In all other cases (CONNECT
2810 to an IP address or an intercepted SSL connection), Squid cannot detect
2811 the domain mismatch at certificate generation time when
2812 bump-server-first is used.
2815 NAME: sslproxy_cert_adapt
2818 TYPE: sslproxy_cert_adapt
2819 LOC: Config.ssl_client.cert_adapt
2822 sslproxy_cert_adapt <adaptation algorithm> acl ...
2824 The following certificate adaptation algorithms are supported:
2827 Sets the "Not After" property to the "Not After" property of
2828 the CA certificate used to sign generated certificates.
2831 Sets the "Not Before" property to the "Not Before" property of
2832 the CA certificate used to sign generated certificates.
2834 setCommonName or setCommonName{CN}
2835 Sets Subject.CN property to the host name specified as a
2836 CN parameter or, if no explicit CN parameter was specified,
2837 extracted from the CONNECT request. It is a misconfiguration
2838 to use setCommonName without an explicit parameter for
2839 intercepted or tproxied SSL connections.
2841 This clause only supports fast acl types.
2843 Squid first groups sslproxy_cert_adapt options by adaptation algorithm.
2844 Within a group, when sslproxy_cert_adapt acl(s) match, Squid uses the
2845 corresponding adaptation algorithm to generate the certificate and
2846 ignores all subsequent sslproxy_cert_adapt options in that algorithm's
2847 group (i.e., the first match wins within each algorithm group). If no
2848 acl(s) match, the default mimicking action takes place.
2850 WARNING: SQUID_X509_V_ERR_DOMAIN_MISMATCH and ssl:certDomainMismatch can
2851 be used with sslproxy_cert_adapt, but if and only if Squid is bumping a
2852 CONNECT request that carries a domain name. In all other cases (CONNECT
2853 to an IP address or an intercepted SSL connection), Squid cannot detect
2854 the domain mismatch at certificate generation time when
2855 bump-server-first is used.
2858 NAME: sslpassword_program
2861 LOC: Config.Program.ssl_password
2864 Specify a program used for entering SSL key passphrases
2865 when using encrypted SSL certificate keys. If not specified
2866 keys must either be unencrypted, or Squid started with the -N
2867 option to allow it to query interactively for the passphrase.
2869 The key file name is given as argument to the program allowing
2870 selection of the right password if you have multiple encrypted
2875 OPTIONS RELATING TO EXTERNAL SSL_CRTD
2876 -----------------------------------------------------------------------------
2879 NAME: sslcrtd_program
2882 DEFAULT: @DEFAULT_SSL_CRTD@ -s @DEFAULT_SSL_DB_DIR@ -M 4MB
2883 LOC: Ssl::TheConfig.ssl_crtd
2885 Specify the location and options of the executable for certificate
2887 @DEFAULT_SSL_CRTD@ program requires -s and -M parameters
2888 For more information use:
2889 @DEFAULT_SSL_CRTD@ -h
2892 NAME: sslcrtd_children
2893 TYPE: HelperChildConfig
2895 DEFAULT: 32 startup=5 idle=1
2896 LOC: Ssl::TheConfig.ssl_crtdChildren
2898 The maximum number of processes spawn to service ssl server.
2899 The maximum this may be safely set to is 32.
2901 The startup= and idle= options allow some measure of skew in your
2906 Sets the minimum number of processes to spawn when Squid
2907 starts or reconfigures. When set to zero the first request will
2908 cause spawning of the first child process to handle it.
2910 Starting too few children temporary slows Squid under load while it
2911 tries to spawn enough additional processes to cope with traffic.
2915 Sets a minimum of how many processes Squid is to try and keep available
2916 at all times. When traffic begins to rise above what the existing
2917 processes can handle this many more will be spawned up to the maximum
2918 configured. A minimum setting of 1 is required.
2922 Sets the maximum number of queued requests.
2923 If the queued requests exceed queue size for more than 3 minutes
2924 squid aborts its operation.
2925 The default value is set to 2*numberofchildren.
2927 You must have at least one ssl_crtd process.
2930 NAME: sslcrtvalidator_program
2934 LOC: Ssl::TheConfig.ssl_crt_validator
2936 Specify the location and options of the executable for ssl_crt_validator
2939 Usage: sslcrtvalidator_program [ttl=n] [cache=n] path ...
2942 ttl=n TTL in seconds for cached results. The default is 60 secs
2943 cache=n limit the result cache size. The default value is 2048
2946 NAME: sslcrtvalidator_children
2947 TYPE: HelperChildConfig
2949 DEFAULT: 32 startup=5 idle=1 concurrency=1
2950 LOC: Ssl::TheConfig.ssl_crt_validator_Children
2952 The maximum number of processes spawn to service SSL server.
2953 The maximum this may be safely set to is 32.
2955 The startup= and idle= options allow some measure of skew in your
2960 Sets the minimum number of processes to spawn when Squid
2961 starts or reconfigures. When set to zero the first request will
2962 cause spawning of the first child process to handle it.
2964 Starting too few children temporary slows Squid under load while it
2965 tries to spawn enough additional processes to cope with traffic.
2969 Sets a minimum of how many processes Squid is to try and keep available
2970 at all times. When traffic begins to rise above what the existing
2971 processes can handle this many more will be spawned up to the maximum
2972 configured. A minimum setting of 1 is required.
2976 The number of requests each certificate validator helper can handle in
2977 parallel. A value of 0 indicates the certficate validator does not
2978 support concurrency. Defaults to 1.
2980 When this directive is set to a value >= 1 then the protocol
2981 used to communicate with the helper is modified to include
2982 a request ID in front of the request/response. The request
2983 ID from the request must be echoed back with the response
2988 Sets the maximum number of queued requests.
2989 If the queued requests exceed queue size for more than 3 minutes
2990 squid aborts its operation.
2991 The default value is set to 2*numberofchildren.
2993 You must have at least one ssl_crt_validator process.
2997 OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
2998 -----------------------------------------------------------------------------
3006 To specify other caches in a hierarchy, use the format:
3008 cache_peer hostname type http-port icp-port [options]
3013 # hostname type port port options
3014 # -------------------- -------- ----- ----- -----------
3015 cache_peer parent.foo.net parent 3128 3130 default
3016 cache_peer sib1.foo.net sibling 3128 3130 proxy-only
3017 cache_peer sib2.foo.net sibling 3128 3130 proxy-only
3018 cache_peer example.com parent 80 0 default
3019 cache_peer cdn.example.com sibling 3128 0
3021 type: either 'parent', 'sibling', or 'multicast'.
3023 proxy-port: The port number where the peer accept HTTP requests.
3024 For other Squid proxies this is usually 3128
3025 For web servers this is usually 80
3027 icp-port: Used for querying neighbor caches about objects.
3028 Set to 0 if the peer does not support ICP or HTCP.
3029 See ICP and HTCP options below for additional details.
3032 ==== ICP OPTIONS ====
3034 You MUST also set icp_port and icp_access explicitly when using these options.
3035 The defaults will prevent peer traffic using ICP.
3038 no-query Disable ICP queries to this neighbor.
3041 Indicates the named peer is a member of a multicast group.
3042 ICP queries will not be sent directly to the peer, but ICP
3043 replies will be accepted from it.
3045 closest-only Indicates that, for ICP_OP_MISS replies, we'll only forward
3046 CLOSEST_PARENT_MISSes and never FIRST_PARENT_MISSes.
3049 To only send ICP queries to this neighbor infrequently.
3050 This is used to keep the neighbor round trip time updated
3051 and is usually used in conjunction with weighted-round-robin.
3054 ==== HTCP OPTIONS ====
3056 You MUST also set htcp_port and htcp_access explicitly when using these options.
3057 The defaults will prevent peer traffic using HTCP.
3060 htcp Send HTCP, instead of ICP, queries to the neighbor.
3061 You probably also want to set the "icp-port" to 4827
3062 instead of 3130. This directive accepts a comma separated
3063 list of options described below.
3065 htcp=oldsquid Send HTCP to old Squid versions (2.5 or earlier).
3067 htcp=no-clr Send HTCP to the neighbor but without
3068 sending any CLR requests. This cannot be used with
3071 htcp=only-clr Send HTCP to the neighbor but ONLY CLR requests.
3072 This cannot be used with no-clr.
3075 Send HTCP to the neighbor including CLRs but only when
3076 they do not result from PURGE requests.
3079 Forward any HTCP CLR requests this proxy receives to the peer.
3082 ==== PEER SELECTION METHODS ====
3084 The default peer selection method is ICP, with the first responding peer
3085 being used as source. These options can be used for better load balancing.
3088 default This is a parent cache which can be used as a "last-resort"
3089 if a peer cannot be located by any of the peer-selection methods.
3090 If specified more than once, only the first is used.
3092 round-robin Load-Balance parents which should be used in a round-robin
3093 fashion in the absence of any ICP queries.
3094 weight=N can be used to add bias.
3096 weighted-round-robin
3097 Load-Balance parents which should be used in a round-robin
3098 fashion with the frequency of each parent being based on the
3099 round trip time. Closer parents are used more often.
3100 Usually used for background-ping parents.
3101 weight=N can be used to add bias.
3103 carp Load-Balance parents which should be used as a CARP array.
3104 The requests will be distributed among the parents based on the
3105 CARP load balancing hash function based on their weight.
3107 userhash Load-balance parents based on the client proxy_auth or ident username.
3109 sourcehash Load-balance parents based on the client source IP.
3112 To be used only for cache peers of type "multicast".
3113 ALL members of this multicast group have "sibling"
3114 relationship with it, not "parent". This is to a multicast
3115 group when the requested object would be fetched only from
3116 a "parent" cache, anyway. It's useful, e.g., when
3117 configuring a pool of redundant Squid proxies, being
3118 members of the same multicast group.
3121 ==== PEER SELECTION OPTIONS ====
3123 weight=N use to affect the selection of a peer during any weighted
3124 peer-selection mechanisms.
3125 The weight must be an integer; default is 1,
3126 larger weights are favored more.
3127 This option does not affect parent selection if a peering
3128 protocol is not in use.
3130 basetime=N Specify a base amount to be subtracted from round trip
3132 It is subtracted before division by weight in calculating
3133 which parent to fectch from. If the rtt is less than the
3134 base time the rtt is set to a minimal value.
3136 ttl=N Specify a TTL to use when sending multicast ICP queries
3138 Only useful when sending to a multicast group.
3139 Because we don't accept ICP replies from random
3140 hosts, you must configure other group members as
3141 peers with the 'multicast-responder' option.
3143 no-delay To prevent access to this neighbor from influencing the
3146 digest-url=URL Tell Squid to fetch the cache digest (if digests are
3147 enabled) for this host from the specified URL rather
3148 than the Squid default location.
3151 ==== CARP OPTIONS ====
3153 carp-key=key-specification
3154 use a different key than the full URL to hash against the peer.
3155 the key-specification is a comma-separated list of the keywords
3156 scheme, host, port, path, params
3157 Order is not important.
3159 ==== ACCELERATOR / REVERSE-PROXY OPTIONS ====
3161 originserver Causes this parent to be contacted as an origin server.
3162 Meant to be used in accelerator setups when the peer
3166 Set the Host header of requests forwarded to this peer.
3167 Useful in accelerator setups where the server (peer)
3168 expects a certain domain name but clients may request
3169 others. ie example.com or www.example.com
3171 no-digest Disable request of cache digests.
3174 Disables requesting ICMP RTT database (NetDB).
3177 ==== AUTHENTICATION OPTIONS ====
3180 If this is a personal/workgroup proxy and your parent
3181 requires proxy authentication.
3183 Note: The string can include URL escapes (i.e. %20 for
3184 spaces). This also means % must be written as %%.
3187 Send login details received from client to this peer.
3188 Both Proxy- and WWW-Authorization headers are passed
3189 without alteration to the peer.
3190 Authentication is not required by Squid for this to work.
3192 Note: This will pass any form of authentication but
3193 only Basic auth will work through a proxy unless the
3194 connection-auth options are also used.
3196 login=PASS Send login details received from client to this peer.
3197 Authentication is not required by this option.
3199 If there are no client-provided authentication headers
3200 to pass on, but username and password are available
3201 from an external ACL user= and password= result tags
3202 they may be sent instead.
3204 Note: To combine this with proxy_auth both proxies must
3205 share the same user database as HTTP only allows for
3206 a single login (one for proxy, one for origin server).
3207 Also be warned this will expose your users proxy
3208 password to the peer. USE WITH CAUTION
3211 Send the username to the upstream cache, but with a
3212 fixed password. This is meant to be used when the peer
3213 is in another administrative domain, but it is still
3214 needed to identify each user.
3215 The star can optionally be followed by some extra
3216 information which is added to the username. This can
3217 be used to identify this proxy to the peer, similar to
3218 the login=username:password option above.
3221 If this is a personal/workgroup proxy and your parent
3222 requires a secure proxy authentication.
3223 The first principal from the default keytab or defined by
3224 the environment variable KRB5_KTNAME will be used.
3226 WARNING: The connection may transmit requests from multiple
3227 clients. Negotiate often assumes end-to-end authentication
3228 and a single-client. Which is not strictly true here.
3230 login=NEGOTIATE:principal_name
3231 If this is a personal/workgroup proxy and your parent
3232 requires a secure proxy authentication.
3233 The principal principal_name from the default keytab or
3234 defined by the environment variable KRB5_KTNAME will be
3237 WARNING: The connection may transmit requests from multiple
3238 clients. Negotiate often assumes end-to-end authentication
3239 and a single-client. Which is not strictly true here.
3241 connection-auth=on|off
3242 Tell Squid that this peer does or not support Microsoft
3243 connection oriented authentication, and any such
3244 challenges received from there should be ignored.
3245 Default is auto to automatically determine the status
3249 Do not use a keytab to authenticate to a peer when
3250 login=NEGOTIATE is specified. Let the GSSAPI
3251 implementation determine which already existing
3252 credentials cache to use instead.
3255 ==== SSL / HTTPS / TLS OPTIONS ====
3257 ssl Encrypt connections to this peer with SSL/TLS.
3259 sslcert=/path/to/ssl/certificate
3260 A client SSL certificate to use when connecting to
3263 sslkey=/path/to/ssl/key
3264 The private SSL key corresponding to sslcert above.
3265 If 'sslkey' is not specified 'sslcert' is assumed to
3266 reference a combined file containing both the
3267 certificate and the key.
3269 sslcipher=... The list of valid SSL ciphers to use when connecting
3273 The minimum TLS protocol version to permit. To control
3274 SSLv3 use the ssloptions= parameter.
3275 Supported Values: 1.0 (default), 1.1, 1.2
3277 ssloptions=... Specify various SSL implementation options:
3279 NO_SSLv3 Disallow the use of SSLv3
3281 NO_TLSv1 Disallow the use of TLSv1.0
3283 NO_TLSv1_1 Disallow the use of TLSv1.1
3285 NO_TLSv1_2 Disallow the use of TLSv1.2
3288 Always create a new key when using
3289 temporary/ephemeral DH key exchanges
3292 Disable use of RFC5077 session tickets.
3293 Some servers may have problems
3294 understanding the TLS extension due
3295 to ambiguous specification in RFC4507.
3297 ALL Enable various bug workarounds
3298 suggested as "harmless" by OpenSSL
3299 Be warned that this reduces SSL/TLS
3300 strength to some attacks.
3302 See the OpenSSL SSL_CTX_set_options documentation for a
3305 tls-cafile= PEM file containing CA certificates to use when verifying
3306 the peer certificate. May be repeated to load multiple files.
3308 sslcapath=... A directory containing additional CA certificates to
3309 use when verifying the peer certificate.
3310 Requires OpenSSL or LibreSSL.
3312 sslcrlfile=... A certificate revocation list file to use when
3313 verifying the peer certificate.
3315 sslflags=... Specify various flags modifying the SSL implementation:
3318 Accept certificates even if they fail to
3322 Don't verify the peer certificate
3323 matches the server name
3325 ssldomain= The peer name as advertised in it's certificate.
3326 Used for verifying the correctness of the received peer
3327 certificate. If not specified the peer hostname will be
3331 Enable the "Front-End-Https: On" header needed when
3332 using Squid as a SSL frontend in front of Microsoft OWA.
3333 See MS KB document Q307347 for details on this header.
3334 If set to auto the header will only be added if the
3335 request is forwarded as a https:// URL.
3337 tls-default-ca[=off]
3338 Whether to use the system Trusted CAs. Default is ON.
3340 tls-no-npn Do not use the TLS NPN extension to advertise HTTP/1.1.
3342 ==== GENERAL OPTIONS ====
3345 A peer-specific connect timeout.
3346 Also see the peer_connect_timeout directive.
3348 connect-fail-limit=N
3349 How many times connecting to a peer must fail before
3350 it is marked as down. Standby connection failures
3351 count towards this limit. Default is 10.
3353 allow-miss Disable Squid's use of only-if-cached when forwarding
3354 requests to siblings. This is primarily useful when
3355 icp_hit_stale is used by the sibling. Excessive use
3356 of this option may result in forwarding loops. One way
3357 to prevent peering loops when using this option, is to
3358 deny cache peer usage on requests from a peer:
3360 cache_peer_access peerName deny fromPeer
3362 max-conn=N Limit the number of concurrent connections the Squid
3363 may open to this peer, including already opened idle
3364 and standby connections. There is no peer-specific
3365 connection limit by default.
3367 A peer exceeding the limit is not used for new
3368 requests unless a standby connection is available.
3370 max-conn currently works poorly with idle persistent
3371 connections: When a peer reaches its max-conn limit,
3372 and there are idle persistent connections to the peer,
3373 the peer may not be selected because the limiting code
3374 does not know whether Squid can reuse those idle
3377 standby=N Maintain a pool of N "hot standby" connections to an
3378 UP peer, available for requests when no idle
3379 persistent connection is available (or safe) to use.
3380 By default and with zero N, no such pool is maintained.
3381 N must not exceed the max-conn limit (if any).
3383 At start or after reconfiguration, Squid opens new TCP
3384 standby connections until there are N connections
3385 available and then replenishes the standby pool as
3386 opened connections are used up for requests. A used
3387 connection never goes back to the standby pool, but
3388 may go to the regular idle persistent connection pool
3389 shared by all peers and origin servers.
3391 Squid never opens multiple new standby connections
3392 concurrently. This one-at-a-time approach minimizes
3393 flooding-like effect on peers. Furthermore, just a few
3394 standby connections should be sufficient in most cases
3395 to supply most new requests with a ready-to-use
3398 Standby connections obey server_idle_pconn_timeout.
3399 For the feature to work as intended, the peer must be
3400 configured to accept and keep them open longer than
3401 the idle timeout at the connecting Squid, to minimize
3402 race conditions typical to idle used persistent
3403 connections. Default request_timeout and
3404 server_idle_pconn_timeout values ensure such a
3407 name=xxx Unique name for the peer.
3408 Required if you have multiple peers on the same host
3409 but different ports.
3410 This name can be used in cache_peer_access and similar
3411 directives to identify the peer.
3412 Can be used by outgoing access controls through the
3415 no-tproxy Do not use the client-spoof TPROXY support when forwarding
3416 requests to this peer. Use normal address selection instead.
3417 This overrides the spoof_client_ip ACL.
3419 proxy-only objects fetched from the peer will not be stored locally.
3423 NAME: cache_peer_access
3426 DEFAULT_DOC: No peer usage restrictions.
3429 Restricts usage of cache_peer proxies.
3432 cache_peer_access peer-name allow|deny [!]aclname ...
3434 For the required peer-name parameter, use either the value of the
3435 cache_peer name=value parameter or, if name=value is missing, the
3436 cache_peer hostname parameter.
3438 This directive narrows down the selection of peering candidates, but
3439 does not determine the order in which the selected candidates are
3440 contacted. That order is determined by the peer selection algorithms
3441 (see PEER SELECTION sections in the cache_peer documentation).
3443 If a deny rule matches, the corresponding peer will not be contacted
3444 for the current transaction -- Squid will not send ICP queries and
3445 will not forward HTTP requests to that peer. An allow match leaves
3446 the corresponding peer in the selection. The first match for a given
3447 peer wins for that peer.
3449 The relative order of cache_peer_access directives for the same peer
3450 matters. The relative order of any two cache_peer_access directives
3451 for different peers does not matter. To ease interpretation, it is a
3452 good idea to group cache_peer_access directives for the same peer
3455 A single cache_peer_access directive may be evaluated multiple times
3456 for a given transaction because individual peer selection algorithms
3457 may check it independently from each other. These redundant checks
3458 may be optimized away in future Squid versions.
3460 This clause only supports fast acl types.
3461 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
3465 NAME: neighbor_type_domain
3466 TYPE: hostdomaintype
3468 DEFAULT_DOC: The peer type from cache_peer directive is used for all requests to that peer.
3471 Modify the cache_peer neighbor type when passing requests
3472 about specific domains to the peer.
3475 neighbor_type_domain neighbor parent|sibling domain domain ...
3478 cache_peer foo.example.com parent 3128 3130
3479 neighbor_type_domain foo.example.com sibling .au .de
3481 The above configuration treats all requests to foo.example.com as a
3482 parent proxy unless the request is for a .au or .de ccTLD domain name.
3485 NAME: dead_peer_timeout
3489 LOC: Config.Timeout.deadPeer
3491 This controls how long Squid waits to declare a peer cache
3492 as "dead." If there are no ICP replies received in this
3493 amount of time, Squid will declare the peer dead and not
3494 expect to receive any further ICP replies. However, it
3495 continues to send ICP queries, and will mark the peer as
3496 alive upon receipt of the first subsequent ICP reply.
3498 This timeout also affects when Squid expects to receive ICP
3499 replies from peers. If more than 'dead_peer' seconds have
3500 passed since the last ICP reply was received, Squid will not
3501 expect to receive an ICP reply on the next query. Thus, if
3502 your time between requests is greater than this timeout, you
3503 will see a lot of requests sent DIRECT to origin servers
3504 instead of to your parents.
3507 NAME: forward_max_tries
3510 LOC: Config.forward_max_tries
3512 Controls how many different forward paths Squid will try
3513 before giving up. See also forward_timeout.
3515 NOTE: connect_retries (default: none) can make each of these
3516 possible forwarding paths be tried multiple times.
3520 MEMORY CACHE OPTIONS
3521 -----------------------------------------------------------------------------
3528 LOC: Config.memMaxSize
3530 NOTE: THIS PARAMETER DOES NOT SPECIFY THE MAXIMUM PROCESS SIZE.
3531 IT ONLY PLACES A LIMIT ON HOW MUCH ADDITIONAL MEMORY SQUID WILL
3532 USE AS A MEMORY CACHE OF OBJECTS. SQUID USES MEMORY FOR OTHER
3533 THINGS AS WELL. SEE THE SQUID FAQ SECTION 8 FOR DETAILS.
3535 'cache_mem' specifies the ideal amount of memory to be used
3537 * In-Transit objects
3539 * Negative-Cached objects
3541 Data for these objects are stored in 4 KB blocks. This
3542 parameter specifies the ideal upper limit on the total size of
3543 4 KB blocks allocated. In-Transit objects take the highest
3546 In-transit objects have priority over the others. When
3547 additional space is needed for incoming data, negative-cached
3548 and hot objects will be released. In other words, the
3549 negative-cached and hot objects will fill up any unused space
3550 not needed for in-transit objects.
3552 If circumstances require, this limit will be exceeded.
3553 Specifically, if your incoming request rate requires more than
3554 'cache_mem' of memory to hold in-transit objects, Squid will
3555 exceed this limit to satisfy the new requests. When the load
3556 decreases, blocks will be freed until the high-water mark is
3557 reached. Thereafter, blocks will be used to store hot
3560 If shared memory caching is enabled, Squid does not use the shared
3561 cache space for in-transit objects, but they still consume as much
3562 local memory as they need. For more details about the shared memory
3563 cache, see memory_cache_shared.
3566 NAME: maximum_object_size_in_memory
3570 LOC: Config.Store.maxInMemObjSize
3572 Objects greater than this size will not be attempted to kept in
3573 the memory cache. This should be set high enough to keep objects
3574 accessed frequently in memory to improve performance whilst low
3575 enough to keep larger objects from hoarding cache_mem.
3578 NAME: memory_cache_shared
3581 LOC: Config.memShared
3583 DEFAULT_DOC: "on" where supported if doing memory caching with multiple SMP workers.
3585 Controls whether the memory cache is shared among SMP workers.
3587 The shared memory cache is meant to occupy cache_mem bytes and replace
3588 the non-shared memory cache, although some entities may still be
3589 cached locally by workers for now (e.g., internal and in-transit
3590 objects may be served from a local memory cache even if shared memory
3591 caching is enabled).
3593 By default, the memory cache is shared if and only if all of the
3594 following conditions are satisfied: Squid runs in SMP mode with
3595 multiple workers, cache_mem is positive, and Squid environment
3596 supports required IPC primitives (e.g., POSIX shared memory segments
3597 and GCC-style atomic operations).
3599 To avoid blocking locks, shared memory uses opportunistic algorithms
3600 that do not guarantee that every cachable entity that could have been
3601 shared among SMP workers will actually be shared.
3603 Currently, entities exceeding 32KB in size cannot be shared.
3606 NAME: memory_cache_mode
3610 DEFAULT_DOC: Keep the most recently fetched objects in memory
3612 Controls which objects to keep in the memory cache (cache_mem)
3614 always Keep most recently fetched objects in memory (default)
3616 disk Only disk cache hits are kept in memory, which means
3617 an object must first be cached on disk and then hit
3618 a second time before cached in memory.
3620 network Only objects fetched from network is kept in memory
3623 NAME: memory_replacement_policy
3625 LOC: Config.memPolicy
3628 The memory replacement policy parameter determines which
3629 objects are purged from memory when memory space is needed.
3631 See cache_replacement_policy for details on algorithms.
3636 -----------------------------------------------------------------------------
3639 NAME: cache_replacement_policy
3641 LOC: Config.replPolicy
3644 The cache replacement policy parameter determines which
3645 objects are evicted (replaced) when disk space is needed.
3647 lru : Squid's original list based LRU policy
3648 heap GDSF : Greedy-Dual Size Frequency
3649 heap LFUDA: Least Frequently Used with Dynamic Aging
3650 heap LRU : LRU policy implemented using a heap
3652 Applies to any cache_dir lines listed below this directive.
3654 The LRU policies keeps recently referenced objects.
3656 The heap GDSF policy optimizes object hit rate by keeping smaller
3657 popular objects in cache so it has a better chance of getting a
3658 hit. It achieves a lower byte hit rate than LFUDA though since
3659 it evicts larger (possibly popular) objects.
3661 The heap LFUDA policy keeps popular objects in cache regardless of
3662 their size and thus optimizes byte hit rate at the expense of
3663 hit rate since one large, popular object will prevent many
3664 smaller, slightly less popular objects from being cached.
3666 Both policies utilize a dynamic aging mechanism that prevents
3667 cache pollution that can otherwise occur with frequency-based
3668 replacement policies.
3670 NOTE: if using the LFUDA replacement policy you should increase
3671 the value of maximum_object_size above its default of 4 MB to
3672 to maximize the potential byte hit rate improvement of LFUDA.
3674 For more information about the GDSF and LFUDA cache replacement
3675 policies see http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html
3676 and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html.
3679 NAME: minimum_object_size
3683 DEFAULT_DOC: no limit
3684 LOC: Config.Store.minObjectSize
3686 Objects smaller than this size will NOT be saved on disk. The
3687 value is specified in bytes, and the default is 0 KB, which
3688 means all responses can be stored.
3691 NAME: maximum_object_size
3695 LOC: Config.Store.maxObjectSize
3697 Set the default value for max-size parameter on any cache_dir.
3698 The value is specified in bytes, and the default is 4 MB.
3700 If you wish to get a high BYTES hit ratio, you should probably
3701 increase this (one 32 MB object hit counts for 3200 10KB
3704 If you wish to increase hit ratio more than you want to
3705 save bandwidth you should leave this low.
3707 NOTE: if using the LFUDA replacement policy you should increase
3708 this value to maximize the byte hit rate improvement of LFUDA!
3709 See cache_replacement_policy for a discussion of this policy.
3715 DEFAULT_DOC: No disk cache. Store cache ojects only in memory.
3716 LOC: Config.cacheSwap
3719 cache_dir Type Directory-Name Fs-specific-data [options]
3721 You can specify multiple cache_dir lines to spread the
3722 cache among different disk partitions.
3724 Type specifies the kind of storage system to use. Only "ufs"
3725 is built by default. To enable any of the other storage systems
3726 see the --enable-storeio configure option.
3728 'Directory' is a top-level directory where cache swap
3729 files will be stored. If you want to use an entire disk
3730 for caching, this can be the mount-point directory.
3731 The directory must exist and be writable by the Squid
3732 process. Squid will NOT create this directory for you.
3734 In SMP configurations, cache_dir must not precede the workers option
3735 and should use configuration macros or conditionals to give each
3736 worker interested in disk caching a dedicated cache directory.
3739 ==== The ufs store type ====
3741 "ufs" is the old well-known Squid storage format that has always
3745 cache_dir ufs Directory-Name Mbytes L1 L2 [options]
3747 'Mbytes' is the amount of disk space (MB) to use under this
3748 directory. The default is 100 MB. Change this to suit your
3749 configuration. Do NOT put the size of your disk drive here.
3750 Instead, if you want Squid to use the entire disk drive,
3751 subtract 20% and use that value.
3753 'L1' is the number of first-level subdirectories which
3754 will be created under the 'Directory'. The default is 16.
3756 'L2' is the number of second-level subdirectories which
3757 will be created under each first-level directory. The default
3761 ==== The aufs store type ====
3763 "aufs" uses the same storage format as "ufs", utilizing
3764 POSIX-threads to avoid blocking the main Squid process on
3765 disk-I/O. This was formerly known in Squid as async-io.
3768 cache_dir aufs Directory-Name Mbytes L1 L2 [options]
3770 see argument descriptions under ufs above
3773 ==== The diskd store type ====
3775 "diskd" uses the same storage format as "ufs", utilizing a
3776 separate process to avoid blocking the main Squid process on
3780 cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]
3782 see argument descriptions under ufs above
3784 Q1 specifies the number of unacknowledged I/O requests when Squid
3785 stops opening new files. If this many messages are in the queues,
3786 Squid won't open new files. Default is 64
3788 Q2 specifies the number of unacknowledged messages when Squid
3789 starts blocking. If this many messages are in the queues,
3790 Squid blocks until it receives some replies. Default is 72
3792 When Q1 < Q2 (the default), the cache directory is optimized
3793 for lower response time at the expense of a decrease in hit
3794 ratio. If Q1 > Q2, the cache directory is optimized for
3795 higher hit ratio at the expense of an increase in response
3799 ==== The rock store type ====
3802 cache_dir rock Directory-Name Mbytes [options]
3804 The Rock Store type is a database-style storage. All cached
3805 entries are stored in a "database" file, using fixed-size slots.
3806 A single entry occupies one or more slots.
3808 If possible, Squid using Rock Store creates a dedicated kid
3809 process called "disker" to avoid blocking Squid worker(s) on disk
3810 I/O. One disker kid is created for each rock cache_dir. Diskers
3811 are created only when Squid, running in daemon mode, has support
3812 for the IpcIo disk I/O module.
3814 swap-timeout=msec: Squid will not start writing a miss to or
3815 reading a hit from disk if it estimates that the swap operation
3816 will take more than the specified number of milliseconds. By
3817 default and when set to zero, disables the disk I/O time limit
3818 enforcement. Ignored when using blocking I/O module because
3819 blocking synchronous I/O does not allow Squid to estimate the
3820 expected swap wait time.
3822 max-swap-rate=swaps/sec: Artificially limits disk access using
3823 the specified I/O rate limit. Swap out requests that
3824 would cause the average I/O rate to exceed the limit are
3825 delayed. Individual swap in requests (i.e., hits or reads) are
3826 not delayed, but they do contribute to measured swap rate and
3827 since they are placed in the same FIFO queue as swap out
3828 requests, they may wait longer if max-swap-rate is smaller.
3829 This is necessary on file systems that buffer "too
3830 many" writes and then start blocking Squid and other processes
3831 while committing those writes to disk. Usually used together
3832 with swap-timeout to avoid excessive delays and queue overflows
3833 when disk demand exceeds available disk "bandwidth". By default
3834 and when set to zero, disables the disk I/O rate limit
3835 enforcement. Currently supported by IpcIo module only.
3837 slot-size=bytes: The size of a database "record" used for
3838 storing cached responses. A cached response occupies at least
3839 one slot and all database I/O is done using individual slots so
3840 increasing this parameter leads to more disk space waste while
3841 decreasing it leads to more disk I/O overheads. Should be a
3842 multiple of your operating system I/O page size. Defaults to
3843 16KBytes. A housekeeping header is stored with each slot and
3844 smaller slot-sizes will be rejected. The header is smaller than
3848 ==== COMMON OPTIONS ====
3850 no-store no new objects should be stored to this cache_dir.
3852 min-size=n the minimum object size in bytes this cache_dir
3853 will accept. It's used to restrict a cache_dir
3854 to only store large objects (e.g. AUFS) while
3855 other stores are optimized for smaller objects
3859 max-size=n the maximum object size in bytes this cache_dir
3861 The value in maximum_object_size directive sets
3862 the default unless more specific details are
3863 available (ie a small store capacity).
3865 Note: To make optimal use of the max-size limits you should order
3866 the cache_dir lines with the smallest max-size value first.
3870 # Uncomment and adjust the following to add a disk cache directory.
3871 #cache_dir ufs @DEFAULT_SWAP_DIR@ 100 16 256
3875 NAME: store_dir_select_algorithm
3877 LOC: Config.store_dir_select_algorithm
3880 How Squid selects which cache_dir to use when the response
3881 object will fit into more than one.
3883 Regardless of which algorithm is used the cache_dir min-size
3884 and max-size parameters are obeyed. As such they can affect
3885 the selection algorithm by limiting the set of considered
3892 This algorithm is suited to caches with similar cache_dir
3893 sizes and disk speeds.
3895 The disk with the least I/O pending is selected.
3896 When there are multiple disks with the same I/O load ranking
3897 the cache_dir with most available capacity is selected.
3899 When a mix of cache_dir sizes are configured the faster disks
3900 have a naturally lower I/O loading and larger disks have more
3901 capacity. So space used to store objects and data throughput
3902 may be very unbalanced towards larger disks.
3907 This algorithm is suited to caches with unequal cache_dir
3910 Each cache_dir is selected in a rotation. The next suitable
3913 Available cache_dir capacity is only considered in relation
3914 to whether the object will fit and meets the min-size and
3915 max-size parameters.
3917 Disk I/O loading is only considered to prevent overload on slow
3918 disks. This algorithm does not spread objects by size, so any
3919 I/O loading per-disk may appear very unbalanced and volatile.
3921 If several cache_dirs use similar min-size, max-size, or other
3922 limits to to reject certain responses, then do not group such
3923 cache_dir lines together, to avoid round-robin selection bias
3924 towards the first cache_dir after the group. Instead, interleave
3925 cache_dir lines from different groups. For example:
3927 store_dir_select_algorithm round-robin
3928 cache_dir rock /hdd1 ... min-size=100000
3929 cache_dir rock /ssd1 ... max-size=99999
3930 cache_dir rock /hdd2 ... min-size=100000
3931 cache_dir rock /ssd2 ... max-size=99999
3932 cache_dir rock /hdd3 ... min-size=100000
3933 cache_dir rock /ssd3 ... max-size=99999
3936 NAME: max_open_disk_fds
3938 LOC: Config.max_open_disk_fds
3940 DEFAULT_DOC: no limit
3942 To avoid having disk as the I/O bottleneck Squid can optionally
3943 bypass the on-disk cache if more than this amount of disk file
3944 descriptors are open.
3946 A value of 0 indicates no limit.
3949 NAME: cache_swap_low
3950 COMMENT: (percent, 0-100)
3953 LOC: Config.Swap.lowWaterMark
3955 The low-water mark for AUFS/UFS/diskd cache object eviction by
3956 the cache_replacement_policy algorithm.
3958 Removal begins when the swap (disk) usage of a cache_dir is
3959 above this low-water mark and attempts to maintain utilization
3960 near the low-water mark.
3962 As swap utilization increases towards the high-water mark set
3963 by cache_swap_high object eviction becomes more agressive.
3965 The value difference in percentages between low- and high-water
3966 marks represent an eviction rate of 300 objects per second and
3967 the rate continues to scale in agressiveness by multiples of
3968 this above the high-water mark.
3970 Defaults are 90% and 95%. If you have a large cache, 5% could be
3971 hundreds of MB. If this is the case you may wish to set these
3972 numbers closer together.
3974 See also cache_swap_high and cache_replacement_policy
3977 NAME: cache_swap_high
3978 COMMENT: (percent, 0-100)
3981 LOC: Config.Swap.highWaterMark
3983 The high-water mark for AUFS/UFS/diskd cache object eviction by
3984 the cache_replacement_policy algorithm.
3986 Removal begins when the swap (disk) usage of a cache_dir is
3987 above the low-water mark set by cache_swap_low and attempts to
3988 maintain utilization near the low-water mark.
3990 As swap utilization increases towards this high-water mark object
3991 eviction becomes more agressive.
3993 The value difference in percentages between low- and high-water
3994 marks represent an eviction rate of 300 objects per second and
3995 the rate continues to scale in agressiveness by multiples of
3996 this above the high-water mark.
3998 Defaults are 90% and 95%. If you have a large cache, 5% could be
3999 hundreds of MB. If this is the case you may wish to set these
4000 numbers closer together.
4002 See also cache_swap_low and cache_replacement_policy
4007 -----------------------------------------------------------------------------
4014 DEFAULT_DOC: The format definitions squid, common, combined, referrer, useragent are built in.
4018 logformat <name> <format specification>
4020 Defines an access log format.
4022 The <format specification> is a string with embedded % format codes
4024 % format codes all follow the same basic structure where all but
4025 the formatcode is optional. Output strings are automatically escaped
4026 as required according to their context and the output format
4027 modifiers are usually not needed, but can be specified if an explicit
4028 output format is desired.
4030 % ["|[|'|#|/] [-] [[0]width] [{arg}] formatcode [{arg}]
4032 " output in quoted string format
4033 [ output in squid text log format as used by log_mime_hdrs
4034 # output in URL quoted format
4035 / output in shell \-escaped format
4040 width minimum and/or maximum field width:
4041 [width_min][.width_max]
4042 When minimum starts with 0, the field is zero-padded.
4043 String values exceeding maximum width are truncated.
4045 {arg} argument such as header name etc. This field may be
4046 placed before or after the token, but not both at once.
4050 % a literal % character
4051 sn Unique sequence number per log line entry
4052 err_code The ID of an error response served by Squid or
4053 a similar internal error identifier.
4054 err_detail Additional err_code-dependent error information.
4055 note The annotation specified by the argument. Also
4056 logs the adaptation meta headers set by the
4057 adaptation_meta configuration parameter.
4058 If no argument given all annotations logged.
4059 The argument may include a separator to use with
4062 By default, multiple note values are separated with ","
4063 and multiple notes are separated with "\r\n".
4064 When logging named notes with %{name}note, the
4065 explicitly configured separator is used between note
4066 values. When logging all notes with %note, the
4067 explicitly configured separator is used between
4068 individual notes. There is currently no way to
4069 specify both value and notes separators when logging
4070 all notes with %note.
4072 Connection related format codes:
4074 >a Client source IP address
4076 >p Client source port
4077 >eui Client source EUI (MAC address, EUI-48 or EUI-64 identifier)
4078 >la Local IP address the client connected to
4079 >lp Local port number the client connected to
4080 >qos Client connection TOS/DSCP value set by Squid
4081 >nfmark Client connection netfilter mark set by Squid
4083 la Local listening IP address the client connection was connected to.
4084 lp Local listening port number the client connection was connected to.
4086 <a Server IP address of the last server or peer connection
4087 <A Server FQDN or peer name
4088 <p Server port number of the last server or peer connection
4089 <la Local IP address of the last server or peer connection
4090 <lp Local port number of the last server or peer connection
4091 <qos Server connection TOS/DSCP value set by Squid
4092 <nfmark Server connection netfilter mark set by Squid
4094 Time related format codes:
4096 ts Seconds since epoch
4097 tu subsecond time (milliseconds)
4098 tl Local time. Optional strftime format argument
4099 default %d/%b/%Y:%H:%M:%S %z
4100 tg GMT time. Optional strftime format argument
4101 default %d/%b/%Y:%H:%M:%S %z
4102 tr Response time (milliseconds)
4103 dt Total time spent making DNS lookups (milliseconds)
4104 tS Approximate master transaction start time in
4105 <full seconds since epoch>.<fractional seconds> format.
4106 Currently, Squid considers the master transaction
4107 started when a complete HTTP request header initiating
4108 the transaction is received from the client. This is
4109 the same value that Squid uses to calculate transaction
4110 response time when logging %tr to access.log. Currently,
4111 Squid uses millisecond resolution for %tS values,
4112 similar to the default access.log "current time" field
4115 Access Control related format codes:
4117 et Tag returned by external acl
4118 ea Log string returned by external acl
4119 un User name (any available)
4120 ul User name from authentication
4121 ue User name from external acl helper
4122 ui User name from ident
4123 un A user name. Expands to the first available name
4124 from the following list of information sources:
4125 - authenticated user name, like %ul
4126 - user name supplied by an external ACL, like %ue
4127 - SSL client name, like %us
4128 - ident user name, like %ui
4129 credentials Client credentials. The exact meaning depends on
4130 the authentication scheme: For Basic authentication,
4131 it is the password; for Digest, the realm sent by the
4132 client; for NTLM and Negotiate, the client challenge
4133 or client credentials prefixed with "YR " or "KK ".
4135 HTTP related format codes:
4139 [http::]rm Request method (GET/POST etc)
4140 [http::]>rm Request method from client
4141 [http::]<rm Request method sent to server or peer
4142 [http::]ru Request URL from client (historic, filtered for logging)
4143 [http::]>ru Request URL from client
4144 [http::]<ru Request URL sent to server or peer
4145 [http::]>rs Request URL scheme from client
4146 [http::]<rs Request URL scheme sent to server or peer
4147 [http::]>rd Request URL domain from client
4148 [http::]<rd Request URL domain sent to server or peer
4149 [http::]>rP Request URL port from client
4150 [http::]<rP Request URL port sent to server or peer
4151 [http::]rp Request URL path excluding hostname
4152 [http::]>rp Request URL path excluding hostname from client
4153 [http::]<rp Request URL path excluding hostname sent to server or peer
4154 [http::]rv Request protocol version
4155 [http::]>rv Request protocol version from client
4156 [http::]<rv Request protocol version sent to server or peer
4158 [http::]>h Original received request header.
4159 Usually differs from the request header sent by
4160 Squid, although most fields are often preserved.
4161 Accepts optional header field name/value filter
4162 argument using name[:[separator]element] format.
4163 [http::]>ha Received request header after adaptation and
4164 redirection (pre-cache REQMOD vectoring point).
4165 Usually differs from the request header sent by
4166 Squid, although most fields are often preserved.
4167 Optional header name argument as for >h
4171 [http::]<Hs HTTP status code received from the next hop
4172 [http::]>Hs HTTP status code sent to the client
4174 [http::]<h Reply header. Optional header name argument
4177 [http::]mt MIME content type
4182 [http::]st Total size of request + reply traffic with client
4183 [http::]>st Total size of request received from client.
4184 Excluding chunked encoding bytes.
4185 [http::]<st Total size of reply sent to client (after adaptation)
4187 [http::]>sh Size of request headers received from client
4188 [http::]<sh Size of reply headers sent to client (after adaptation)
4190 [http::]<sH Reply high offset sent
4191 [http::]<sS Upstream object size
4193 [http::]<bs Number of HTTP-equivalent message body bytes
4194 received from the next hop, excluding chunked
4195 transfer encoding and control messages.
4196 Generated FTP/Gopher listings are treated as
4201 [http::]<pt Peer response time in milliseconds. The timer starts
4202 when the last request byte is sent to the next hop
4203 and stops when the last response byte is received.
4204 [http::]<tt Total time in milliseconds. The timer
4205 starts with the first connect request (or write I/O)
4206 sent to the first selected peer. The timer stops
4207 with the last I/O with the last peer.
4209 Squid handling related format codes:
4211 Ss Squid request status (TCP_MISS etc)
4212 Sh Squid hierarchy status (DEFAULT_PARENT etc)
4214 SSL-related format codes:
4216 ssl::bump_mode SslBump decision for the transaction:
4218 For CONNECT requests that initiated bumping of
4219 a connection and for any request received on
4220 an already bumped connection, Squid logs the
4221 corresponding SslBump mode ("server-first" or
4222 "client-first"). See the ssl_bump option for
4223 more information about these modes.
4225 A "none" token is logged for requests that
4226 triggered "ssl_bump" ACL evaluation matching
4227 either a "none" rule or no rules at all.
4229 In all other cases, a single dash ("-") is
4232 ssl::>sni SSL client SNI sent to Squid. Available only
4233 after the peek, stare, or splice SSL bumping
4237 The Subject field of the received client
4238 SSL certificate or a dash ('-') if Squid has
4239 received an invalid/malformed certificate or
4240 no certificate at all. Consider encoding the
4241 logged value because Subject often has spaces.
4244 The Issuer field of the received client
4245 SSL certificate or a dash ('-') if Squid has
4246 received an invalid/malformed certificate or
4247 no certificate at all. Consider encoding the
4248 logged value because Issuer often has spaces.
4251 The list of certificate validation errors
4252 detected by Squid (including OpenSSL and
4253 certificate validation helper components). The
4254 errors are listed in the discovery order. By
4255 default, the error codes are separated by ':'.
4256 Accepts an optional separator argument.
4258 %ssl::>negotiated_version The negotiated TLS version of the
4261 %ssl::<negotiated_version The negotiated TLS version of the
4262 last server or peer connection.
4264 %ssl::>received_hello_version The TLS version of the Hello
4265 message received from TLS client.
4267 %ssl::<received_hello_version The TLS version of the Hello
4268 message received from TLS server.
4270 %ssl::>received_supported_version The maximum TLS version
4271 supported by the TLS client.
4273 %ssl::<received_supported_version The maximum TLS version
4274 supported by the TLS server.
4276 %ssl::>negotiated_cipher The negotiated cipher of the
4279 %ssl::<negotiated_cipher The negotiated cipher of the
4280 last server or peer connection.
4282 If ICAP is enabled, the following code becomes available (as
4283 well as ICAP log codes documented with the icap_log option):
4285 icap::tt Total ICAP processing time for the HTTP
4286 transaction. The timer ticks when ICAP
4287 ACLs are checked and when ICAP
4288 transaction is in progress.
4290 If adaptation is enabled the following codes become available:
4292 adapt::<last_h The header of the last ICAP response or
4293 meta-information from the last eCAP
4294 transaction related to the HTTP transaction.
4295 Like <h, accepts an optional header name
4298 adapt::sum_trs Summed adaptation transaction response
4299 times recorded as a comma-separated list in
4300 the order of transaction start time. Each time
4301 value is recorded as an integer number,
4302 representing response time of one or more
4303 adaptation (ICAP or eCAP) transaction in
4304 milliseconds. When a failed transaction is
4305 being retried or repeated, its time is not
4306 logged individually but added to the
4307 replacement (next) transaction. See also:
4310 adapt::all_trs All adaptation transaction response times.
4311 Same as adaptation_strs but response times of
4312 individual transactions are never added
4313 together. Instead, all transaction response
4314 times are recorded individually.
4316 You can prefix adapt::*_trs format codes with adaptation
4317 service name in curly braces to record response time(s) specific
4318 to that service. For example: %{my_service}adapt::sum_trs
4320 The default formats available (which do not need re-defining) are:
4322 logformat squid %ts.%03tu %6tr %>a %Ss/%03>Hs %<st %rm %ru %[un %Sh/%<a %mt
4323 logformat common %>a %[ui %[un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st %Ss:%Sh
4324 logformat combined %>a %[ui %[un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st "%{Referer}>h" "%{User-Agent}>h" %Ss:%Sh
4325 logformat referrer %ts.%03tu %>a %{Referer}>h %ru
4326 logformat useragent %>a [%tl] "%{User-Agent}>h"
4328 NOTE: When the log_mime_hdrs directive is set to ON.
4329 The squid, common and combined formats have a safely encoded copy
4330 of the mime headers appended to each line within a pair of brackets.
4332 NOTE: The common and combined formats are not quite true to the Apache definition.
4333 The logs from Squid contain an extra status and hierarchy code appended.
4337 NAME: access_log cache_access_log
4339 LOC: Config.Log.accesslogs
4340 DEFAULT_IF_NONE: daemon:@DEFAULT_ACCESS_LOG@ squid
4342 Configures whether and how Squid logs HTTP and ICP transactions.
4343 If access logging is enabled, a single line is logged for every
4344 matching HTTP or ICP request. The recommended directive formats are:
4346 access_log <module>:<place> [option ...] [acl acl ...]
4347 access_log none [acl acl ...]
4349 The following directive format is accepted but may be deprecated:
4350 access_log <module>:<place> [<logformat name> [acl acl ...]]
4352 In most cases, the first ACL name must not contain the '=' character
4353 and should not be equal to an existing logformat name. You can always
4354 start with an 'all' ACL to work around those restrictions.
4356 Will log to the specified module:place using the specified format (which
4357 must be defined in a logformat directive) those entries which match
4358 ALL the acl's specified (which must be defined in acl clauses).
4359 If no acl is specified, all requests will be logged to this destination.
4361 ===== Available options for the recommended directive format =====
4363 logformat=name Names log line format (either built-in or
4364 defined by a logformat directive). Defaults
4367 buffer-size=64KB Defines approximate buffering limit for log
4368 records (see buffered_logs). Squid should not
4369 keep more than the specified size and, hence,
4370 should flush records before the buffer becomes
4371 full to avoid overflows under normal
4372 conditions (the exact flushing algorithm is
4373 module-dependent though). The on-error option
4374 controls overflow handling.
4376 on-error=die|drop Defines action on unrecoverable errors. The
4377 'drop' action ignores (i.e., does not log)
4378 affected log records. The default 'die' action
4379 kills the affected worker. The drop action
4380 support has not been tested for modules other
4383 rotate=N Specifies the number of log file rotations to
4384 make when you run 'squid -k rotate'. The default
4385 is to obey the logfile_rotate directive. Setting
4386 rotate=0 will disable the file name rotation,
4387 but the log files are still closed and re-opened.
4388 This will enable you to rename the logfiles
4389 yourself just before sending the rotate signal.
4390 Only supported by the stdio module.
4392 ===== Modules Currently available =====
4394 none Do not log any requests matching these ACL.
4395 Do not specify Place or logformat name.
4397 stdio Write each log line to disk immediately at the completion of
4399 Place: the filename and path to be written.
4401 daemon Very similar to stdio. But instead of writing to disk the log
4402 line is passed to a daemon helper for asychronous handling instead.
4403 Place: varies depending on the daemon.
4405 log_file_daemon Place: the file name and path to be written.
4407 syslog To log each request via syslog facility.
4408 Place: The syslog facility and priority level for these entries.
4409 Place Format: facility.priority
4411 where facility could be any of:
4412 authpriv, daemon, local0 ... local7 or user.
4414 And priority could be any of:
4415 err, warning, notice, info, debug.
4417 udp To send each log line as text data to a UDP receiver.
4418 Place: The destination host name or IP and port.
4419 Place Format: //host:port
4421 tcp To send each log line as text data to a TCP receiver.
4422 Lines may be accumulated before sending (see buffered_logs).
4423 Place: The destination host name or IP and port.
4424 Place Format: //host:port
4427 access_log daemon:@DEFAULT_ACCESS_LOG@ squid
4433 LOC: Config.Log.icaplogs
4436 ICAP log files record ICAP transaction summaries, one line per
4439 The icap_log option format is:
4440 icap_log <filepath> [<logformat name> [acl acl ...]]
4441 icap_log none [acl acl ...]]
4443 Please see access_log option documentation for details. The two
4444 kinds of logs share the overall configuration approach and many
4447 ICAP processing of a single HTTP message or transaction may
4448 require multiple ICAP transactions. In such cases, multiple
4449 ICAP transaction log lines will correspond to a single access
4452 ICAP log uses logformat codes that make sense for an ICAP
4453 transaction. Header-related codes are applied to the HTTP header
4454 embedded in an ICAP server response, with the following caveats:
4455 For REQMOD, there is no HTTP response header unless the ICAP
4456 server performed request satisfaction. For RESPMOD, the HTTP
4457 request header is the header sent to the ICAP server. For
4458 OPTIONS, there are no HTTP headers.
4460 The following format codes are also available for ICAP logs:
4462 icap::<A ICAP server IP address. Similar to <A.
4464 icap::<service_name ICAP service name from the icap_service
4465 option in Squid configuration file.
4467 icap::ru ICAP Request-URI. Similar to ru.
4469 icap::rm ICAP request method (REQMOD, RESPMOD, or
4470 OPTIONS). Similar to existing rm.
4472 icap::>st Bytes sent to the ICAP server (TCP payload
4473 only; i.e., what Squid writes to the socket).
4475 icap::<st Bytes received from the ICAP server (TCP
4476 payload only; i.e., what Squid reads from
4479 icap::<bs Number of message body bytes received from the
4480 ICAP server. ICAP message body, if any, usually
4481 includes encapsulated HTTP message headers and
4482 possibly encapsulated HTTP message body. The
4483 HTTP body part is dechunked before its size is
4486 icap::tr Transaction response time (in
4487 milliseconds). The timer starts when
4488 the ICAP transaction is created and
4489 stops when the transaction is completed.
4492 icap::tio Transaction I/O time (in milliseconds). The
4493 timer starts when the first ICAP request
4494 byte is scheduled for sending. The timers
4495 stops when the last byte of the ICAP response
4498 icap::to Transaction outcome: ICAP_ERR* for all
4499 transaction errors, ICAP_OPT for OPTION
4500 transactions, ICAP_ECHO for 204
4501 responses, ICAP_MOD for message
4502 modification, and ICAP_SAT for request
4503 satisfaction. Similar to Ss.
4505 icap::Hs ICAP response status code. Similar to Hs.
4507 icap::>h ICAP request header(s). Similar to >h.
4509 icap::<h ICAP response header(s). Similar to <h.
4511 The default ICAP log format, which can be used without an explicit
4512 definition, is called icap_squid:
4514 logformat icap_squid %ts.%03tu %6icap::tr %>a %icap::to/%03icap::Hs %icap::<size %icap::rm %icap::ru% %un -/%icap::<A -
4516 See also: logformat, log_icap, and %adapt::<last_h
4519 NAME: logfile_daemon
4521 DEFAULT: @DEFAULT_LOGFILED@
4522 LOC: Log::TheConfig.logfile_daemon
4524 Specify the path to the logfile-writing daemon. This daemon is
4525 used to write the access and store logs, if configured.
4527 Squid sends a number of commands to the log daemon:
4528 L<data>\n - logfile data
4533 r<n>\n - set rotate count to <n>
4534 b<n>\n - 1 = buffer output, 0 = don't buffer output
4536 No responses is expected.
4539 NAME: stats_collection
4541 LOC: Config.accessList.stats_collection
4543 DEFAULT_DOC: Allow logging for all transactions.
4544 COMMENT: allow|deny acl acl...
4546 This options allows you to control which requests gets accounted
4547 in performance counters.
4549 This clause only supports fast acl types.
4550 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4553 NAME: cache_store_log
4556 LOC: Config.Log.store
4558 Logs the activities of the storage manager. Shows which
4559 objects are ejected from the cache, and which objects are
4560 saved and for how long.
4561 There are not really utilities to analyze this data, so you can safely
4562 disable it (the default).
4564 Store log uses modular logging outputs. See access_log for the list
4565 of modules supported.
4568 cache_store_log stdio:@DEFAULT_STORE_LOG@
4569 cache_store_log daemon:@DEFAULT_STORE_LOG@
4572 NAME: cache_swap_state cache_swap_log
4574 LOC: Config.Log.swap
4576 DEFAULT_DOC: Store the journal inside its cache_dir
4578 Location for the cache "swap.state" file. This index file holds
4579 the metadata of objects saved on disk. It is used to rebuild
4580 the cache during startup. Normally this file resides in each
4581 'cache_dir' directory, but you may specify an alternate
4582 pathname here. Note you must give a full filename, not just
4583 a directory. Since this is the index for the whole object
4584 list you CANNOT periodically rotate it!
4586 If %s can be used in the file name it will be replaced with a
4587 a representation of the cache_dir name where each / is replaced
4588 with '.'. This is needed to allow adding/removing cache_dir
4589 lines when cache_swap_log is being used.
4591 If have more than one 'cache_dir', and %s is not used in the name
4592 these swap logs will have names such as:
4598 The numbered extension (which is added automatically)
4599 corresponds to the order of the 'cache_dir' lines in this
4600 configuration file. If you change the order of the 'cache_dir'
4601 lines in this file, these index files will NOT correspond to
4602 the correct 'cache_dir' entry (unless you manually rename
4603 them). We recommend you do NOT use this option. It is
4604 better to keep these index files in each 'cache_dir' directory.
4607 NAME: logfile_rotate
4610 LOC: Config.Log.rotateNumber
4612 Specifies the default number of logfile rotations to make when you
4613 type 'squid -k rotate'. The default is 10, which will rotate
4614 with extensions 0 through 9. Setting logfile_rotate to 0 will
4615 disable the file name rotation, but the logfiles are still closed
4616 and re-opened. This will enable you to rename the logfiles
4617 yourself just before sending the rotate signal.
4619 Note, from Squid-3.1 this option is only a default for cache.log,
4620 that log can be rotated separately by using debug_options.
4622 Note, from Squid-3.6 this option is only a default for access.log
4623 recorded by stdio: module. Those logs can be rotated separately by
4624 using the rotate=N option on their access_log directive.
4626 Note, the 'squid -k rotate' command normally sends a USR1
4627 signal to the running squid process. In certain situations
4628 (e.g. on Linux with Async I/O), USR1 is used for other
4629 purposes, so -k rotate uses another signal. It is best to get
4630 in the habit of using 'squid -k rotate' instead of 'kill -USR1
4637 DEFAULT: @DEFAULT_MIME_TABLE@
4638 LOC: Config.mimeTablePathname
4640 Path to Squid's icon configuration file.
4642 You shouldn't need to change this, but the default file contains
4643 examples and formatting information if you do.
4649 LOC: Config.onoff.log_mime_hdrs
4652 The Cache can record both the request and the response MIME
4653 headers for each HTTP transaction. The headers are encoded
4654 safely and will appear as two bracketed fields at the end of
4655 the access log (for either the native or httpd-emulated log
4656 formats). To enable this logging set log_mime_hdrs to 'on'.
4661 DEFAULT: @DEFAULT_PID_FILE@
4662 LOC: Config.pidFilename
4664 A filename to write the process-id to. To disable, enter "none".
4667 NAME: client_netmask
4669 LOC: Config.Addrs.client_netmask
4671 DEFAULT_DOC: Log full client IP address
4673 A netmask for client addresses in logfiles and cachemgr output.
4674 Change this to protect the privacy of your cache clients.
4675 A netmask of 255.255.255.0 will log all IP's in that range with
4676 the last digit set to '0'.
4679 NAME: strip_query_terms
4681 LOC: Config.onoff.strip_query_terms
4684 By default, Squid strips query terms from requested URLs before
4685 logging. This protects your user's privacy and reduces log size.
4687 When investigating HIT/MISS or other caching behaviour you
4688 will need to disable this to see the full URL used by Squid.
4695 LOC: Config.onoff.buffered_logs
4697 Whether to write/send access_log records ASAP or accumulate them and
4698 then write/send them in larger chunks. Buffering may improve
4699 performance because it decreases the number of I/Os. However,
4700 buffering increases the delay before log records become available to
4701 the final recipient (e.g., a disk file or logging daemon) and,
4702 hence, increases the risk of log records loss.
4704 Note that even when buffered_logs are off, Squid may have to buffer
4705 records if it cannot write/send them immediately due to pending I/Os
4706 (e.g., the I/O writing the previous log record) or connectivity loss.
4708 Currently honored by 'daemon' and 'tcp' access_log modules only.
4711 NAME: netdb_filename
4713 DEFAULT: stdio:@DEFAULT_NETDB_FILE@
4714 LOC: Config.netdbFilename
4717 Where Squid stores it's netdb journal.
4718 When enabled this journal preserves netdb state between restarts.
4720 To disable, enter "none".
4724 OPTIONS FOR TROUBLESHOOTING
4725 -----------------------------------------------------------------------------
4730 DEFAULT_IF_NONE: @DEFAULT_CACHE_LOG@
4731 LOC: Debug::cache_log
4733 Squid administrative logging file.
4735 This is where general information about Squid behavior goes. You can
4736 increase the amount of data logged to this file and how often it is
4737 rotated with "debug_options"
4743 DEFAULT_DOC: Log all critical and important messages.
4744 LOC: Debug::debugOptions
4746 Logging options are set as section,level where each source file
4747 is assigned a unique section. Lower levels result in less
4748 output, Full debugging (level 9) can result in a very large
4749 log file, so be careful.
4751 The magic word "ALL" sets debugging levels for all sections.
4752 The default is to run with "ALL,1" to record important warnings.
4754 The rotate=N option can be used to keep more or less of these logs
4755 than would otherwise be kept by logfile_rotate.
4756 For most uses a single log should be enough to monitor current
4757 events affecting Squid.
4762 LOC: Config.coredump_dir
4763 DEFAULT_IF_NONE: none
4764 DEFAULT_DOC: Use the directory from where Squid was started.
4766 By default Squid leaves core files in the directory from where
4767 it was started. If you set 'coredump_dir' to a directory
4768 that exists, Squid will chdir() to that directory at startup
4769 and coredump files will be left there.
4773 # Leave coredumps in the first cache dir
4774 coredump_dir @DEFAULT_SWAP_DIR@
4780 OPTIONS FOR FTP GATEWAYING
4781 -----------------------------------------------------------------------------
4787 LOC: Config.Ftp.anon_user
4789 If you want the anonymous login password to be more informative
4790 (and enable the use of picky FTP servers), set this to something
4791 reasonable for your domain, like wwwuser@somewhere.net
4793 The reason why this is domainless by default is the
4794 request can be made on the behalf of a user in any domain,
4795 depending on how the cache is used.
4796 Some FTP server also validate the email address is valid
4797 (for example perl.com).
4803 LOC: Config.Ftp.passive
4805 If your firewall does not allow Squid to use passive
4806 connections, turn off this option.
4808 Use of ftp_epsv_all option requires this to be ON.
4814 LOC: Config.Ftp.epsv_all
4816 FTP Protocol extensions permit the use of a special "EPSV ALL" command.
4818 NATs may be able to put the connection on a "fast path" through the
4819 translator, as the EPRT command will never be used and therefore,
4820 translation of the data portion of the segments will never be needed.
4822 When a client only expects to do two-way FTP transfers this may be
4824 If squid finds that it must do a three-way FTP transfer after issuing
4825 an EPSV ALL command, the FTP session will fail.
4827 If you have any doubts about this option do not use it.
4828 Squid will nicely attempt all other connection methods.
4830 Requires ftp_passive to be ON (default) for any effect.
4836 LOC: Config.accessList.ftp_epsv
4838 FTP Protocol extensions permit the use of a special "EPSV" command.
4840 NATs may be able to put the connection on a "fast path" through the
4841 translator using EPSV, as the EPRT command will never be used
4842 and therefore, translation of the data portion of the segments
4843 will never be needed.
4845 EPSV is often required to interoperate with FTP servers on IPv6
4846 networks. On the other hand, it may break some IPv4 servers.
4848 By default, EPSV may try EPSV with any FTP server. To fine tune
4849 that decision, you may restrict EPSV to certain clients or servers
4852 ftp_epsv allow|deny al1 acl2 ...
4854 WARNING: Disabling EPSV may cause problems with external NAT and IPv6.
4856 Only fast ACLs are supported.
4857 Requires ftp_passive to be ON (default) for any effect.
4863 LOC: Config.Ftp.eprt
4865 FTP Protocol extensions permit the use of a special "EPRT" command.
4867 This extension provides a protocol neutral alternative to the
4868 IPv4-only PORT command. When supported it enables active FTP data
4869 channels over IPv6 and efficient NAT handling.
4871 Turning this OFF will prevent EPRT being attempted and will skip
4872 straight to using PORT for IPv4 servers.
4874 Some devices are known to not handle this extension correctly and
4875 may result in crashes. Devices which suport EPRT enough to fail
4876 cleanly will result in Squid attempting PORT anyway. This directive
4877 should only be disabled when EPRT results in device failures.
4879 WARNING: Doing so will convert Squid back to the old behavior with all
4880 the related problems with external NAT devices/layers and IPv4-only FTP.
4883 NAME: ftp_sanitycheck
4886 LOC: Config.Ftp.sanitycheck
4888 For security and data integrity reasons Squid by default performs
4889 sanity checks of the addresses of FTP data connections ensure the
4890 data connection is to the requested server. If you need to allow
4891 FTP connections to servers using another IP address for the data
4892 connection turn this off.
4895 NAME: ftp_telnet_protocol
4898 LOC: Config.Ftp.telnet
4900 The FTP protocol is officially defined to use the telnet protocol
4901 as transport channel for the control connection. However, many
4902 implementations are broken and does not respect this aspect of
4905 If you have trouble accessing files with ASCII code 255 in the
4906 path or similar problems involving this ASCII code you can
4907 try setting this directive to off. If that helps, report to the
4908 operator of the FTP server in question that their FTP server
4909 is broken and does not follow the FTP standard.
4913 OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
4914 -----------------------------------------------------------------------------
4919 DEFAULT: @DEFAULT_DISKD@
4920 LOC: Config.Program.diskd
4922 Specify the location of the diskd executable.
4923 Note this is only useful if you have compiled in
4924 diskd as one of the store io modules.
4927 NAME: unlinkd_program
4930 DEFAULT: @DEFAULT_UNLINKD@
4931 LOC: Config.Program.unlinkd
4933 Specify the location of the executable for file deletion process.
4936 NAME: pinger_program
4939 DEFAULT: @DEFAULT_PINGER@
4942 Specify the location of the executable for the pinger process.
4951 Control whether the pinger is active at run-time.
4952 Enables turning ICMP pinger on and off with a simple
4953 squid -k reconfigure.
4958 OPTIONS FOR URL REWRITING
4959 -----------------------------------------------------------------------------
4962 NAME: url_rewrite_program redirect_program
4964 LOC: Config.Program.redirect
4967 Specify the location of the executable URL rewriter to use.
4968 Since they can perform almost any function there isn't one included.
4970 For each requested URL, the rewriter will receive on line with the format
4972 [channel-ID <SP>] URL [<SP> extras]<NL>
4974 See url_rewrite_extras on how to send "extras" with optional values to
4976 After processing the request the helper must reply using the following format:
4978 [channel-ID <SP>] result [<SP> kv-pairs]
4980 The result code can be:
4982 OK status=30N url="..."
4983 Redirect the URL to the one supplied in 'url='.
4984 'status=' is optional and contains the status code to send
4985 the client in Squids HTTP response. It must be one of the
4986 HTTP redirect status codes: 301, 302, 303, 307, 308.
4987 When no status is given Squid will use 302.
4989 OK rewrite-url="..."
4990 Rewrite the URL to the one supplied in 'rewrite-url='.
4991 The new URL is fetched directly by Squid and returned to
4992 the client as the response to its request.
4995 When neither of url= and rewrite-url= are sent Squid does
4999 Do not change the URL.
5002 An internal error occurred in the helper, preventing
5003 a result being identified. The 'message=' key name is
5004 reserved for delivering a log message.
5007 In addition to the above kv-pairs Squid also understands the following
5008 optional kv-pairs received from URL rewriters:
5010 Associates a TAG with the client TCP connection.
5011 The TAG is treated as a regular annotation but persists across
5012 future requests on the client connection rather than just the
5013 current request. A helper may update the TAG during subsequent
5014 requests be returning a new kv-pair.
5016 When using the concurrency= option the protocol is changed by
5017 introducing a query channel tag in front of the request/response.
5018 The query channel tag is a number between 0 and concurrency-1.
5019 This value must be echoed back unchanged to Squid as the first part
5020 of the response relating to its request.
5022 WARNING: URL re-writing ability should be avoided whenever possible.
5023 Use the URL redirect form of response instead.
5025 Re-write creates a difference in the state held by the client
5026 and server. Possibly causing confusion when the server response
5027 contains snippets of its view state. Embeded URLs, response
5028 and content Location headers, etc. are not re-written by this
5031 By default, a URL rewriter is not used.
5034 NAME: url_rewrite_children redirect_children
5035 TYPE: HelperChildConfig
5036 DEFAULT: 20 startup=0 idle=1 concurrency=0
5037 LOC: Config.redirectChildren
5039 The maximum number of redirector processes to spawn. If you limit
5040 it too few Squid will have to wait for them to process a backlog of
5041 URLs, slowing it down. If you allow too many they will use RAM
5042 and other system resources noticably.
5044 The startup= and idle= options allow some measure of skew in your
5049 Sets a minimum of how many processes are to be spawned when Squid
5050 starts or reconfigures. When set to zero the first request will
5051 cause spawning of the first child process to handle it.
5053 Starting too few will cause an initial slowdown in traffic as Squid
5054 attempts to simultaneously spawn enough processes to cope.
5058 Sets a minimum of how many processes Squid is to try and keep available
5059 at all times. When traffic begins to rise above what the existing
5060 processes can handle this many more will be spawned up to the maximum
5061 configured. A minimum setting of 1 is required.
5065 The number of requests each redirector helper can handle in
5066 parallel. Defaults to 0 which indicates the redirector
5067 is a old-style single threaded redirector.
5069 When this directive is set to a value >= 1 then the protocol
5070 used to communicate with the helper is modified to include
5071 an ID in front of the request/response. The ID from the request
5072 must be echoed back with the response to that request.
5076 Sets the maximum number of queued requests.
5077 If the queued requests exceed queue size and redirector_bypass
5078 configuration option is set, then redirector is bypassed. Otherwise, if
5079 overloading persists squid may abort its operation.
5080 The default value is set to 2*numberofchildren.
5083 NAME: url_rewrite_host_header redirect_rewrites_host_header
5086 LOC: Config.onoff.redir_rewrites_host
5088 To preserve same-origin security policies in browsers and
5089 prevent Host: header forgery by redirectors Squid rewrites
5090 any Host: header in redirected requests.
5092 If you are running an accelerator this may not be a wanted
5093 effect of a redirector. This directive enables you disable
5094 Host: alteration in reverse-proxy traffic.
5096 WARNING: Entries are cached on the result of the URL rewriting
5097 process, so be careful if you have domain-virtual hosts.
5099 WARNING: Squid and other software verifies the URL and Host
5100 are matching, so be careful not to relay through other proxies
5101 or inspecting firewalls with this disabled.
5104 NAME: url_rewrite_access redirector_access
5107 DEFAULT_DOC: Allow, unless rules exist in squid.conf.
5108 LOC: Config.accessList.redirector
5110 If defined, this access list specifies which requests are
5111 sent to the redirector processes.
5113 This clause supports both fast and slow acl types.
5114 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5117 NAME: url_rewrite_bypass redirector_bypass
5119 LOC: Config.onoff.redirector_bypass
5122 When this is 'on', a request will not go through the
5123 redirector if all the helpers are busy. If this is 'off'
5124 and the redirector queue grows too large, Squid will exit
5125 with a FATAL error and ask you to increase the number of
5126 redirectors. You should only enable this if the redirectors
5127 are not critical to your caching system. If you use
5128 redirectors for access control, and you enable this option,
5129 users may have access to pages they should not
5130 be allowed to request.
5131 This options sets default queue-size option of the url_rewrite_children
5135 NAME: url_rewrite_extras
5136 TYPE: TokenOrQuotedString
5137 LOC: Config.redirector_extras
5138 DEFAULT: "%>a/%>A %un %>rm myip=%la myport=%lp"
5140 Specifies a string to be append to request line format for the
5141 rewriter helper. "Quoted" format values may contain spaces and
5142 logformat %macros. In theory, any logformat %macro can be used.
5143 In practice, a %macro expands as a dash (-) if the helper request is
5144 sent before the required macro information is available to Squid.
5147 NAME: url_rewrite_timeout
5148 TYPE: UrlHelperTimeout
5149 LOC: Config.onUrlRewriteTimeout
5151 DEFAULT_DOC: Squid waits for the helper response forever
5153 Squid times active requests to redirector. The timeout value and Squid
5154 reaction to a timed out request are configurable using the following
5157 url_rewrite_timeout timeout time-units on_timeout=<action> [response=<quoted-response>]
5159 supported timeout actions:
5160 fail Squid return a ERR_GATEWAY_FAILURE error page
5162 bypass Do not re-write the URL
5164 retry Send the lookup to the helper again
5166 use_configured_response
5167 Use the <quoted-response> as helper response
5171 OPTIONS FOR STORE ID
5172 -----------------------------------------------------------------------------
5175 NAME: store_id_program storeurl_rewrite_program
5177 LOC: Config.Program.store_id
5180 Specify the location of the executable StoreID helper to use.
5181 Since they can perform almost any function there isn't one included.
5183 For each requested URL, the helper will receive one line with the format
5185 [channel-ID <SP>] URL [<SP> extras]<NL>
5188 After processing the request the helper must reply using the following format:
5190 [channel-ID <SP>] result [<SP> kv-pairs]
5192 The result code can be:
5195 Use the StoreID supplied in 'store-id='.
5198 The default is to use HTTP request URL as the store ID.
5201 An internal error occured in the helper, preventing
5202 a result being identified.
5204 In addition to the above kv-pairs Squid also understands the following
5205 optional kv-pairs received from URL rewriters:
5207 Associates a TAG with the client TCP connection.
5208 Please see url_rewrite_program related documentation for this
5211 Helper programs should be prepared to receive and possibly ignore
5212 additional whitespace-separated tokens on each input line.
5214 When using the concurrency= option the protocol is changed by
5215 introducing a query channel tag in front of the request/response.
5216 The query channel tag is a number between 0 and concurrency-1.
5217 This value must be echoed back unchanged to Squid as the first part
5218 of the response relating to its request.
5220 NOTE: when using StoreID refresh_pattern will apply to the StoreID
5221 returned from the helper and not the URL.
5223 WARNING: Wrong StoreID value returned by a careless helper may result
5224 in the wrong cached response returned to the user.
5226 By default, a StoreID helper is not used.
5229 NAME: store_id_extras
5230 TYPE: TokenOrQuotedString
5231 LOC: Config.storeId_extras
5232 DEFAULT: "%>a/%>A %un %>rm myip=%la myport=%lp"
5234 Specifies a string to be append to request line format for the
5235 StoreId helper. "Quoted" format values may contain spaces and
5236 logformat %macros. In theory, any logformat %macro can be used.
5237 In practice, a %macro expands as a dash (-) if the helper request is
5238 sent before the required macro information is available to Squid.
5241 NAME: store_id_children storeurl_rewrite_children
5242 TYPE: HelperChildConfig
5243 DEFAULT: 20 startup=0 idle=1 concurrency=0
5244 LOC: Config.storeIdChildren
5246 The maximum number of StoreID helper processes to spawn. If you limit
5247 it too few Squid will have to wait for them to process a backlog of
5248 requests, slowing it down. If you allow too many they will use RAM
5249 and other system resources noticably.
5251 The startup= and idle= options allow some measure of skew in your
5256 Sets a minimum of how many processes are to be spawned when Squid
5257 starts or reconfigures. When set to zero the first request will
5258 cause spawning of the first child process to handle it.
5260 Starting too few will cause an initial slowdown in traffic as Squid
5261 attempts to simultaneously spawn enough processes to cope.
5265 Sets a minimum of how many processes Squid is to try and keep available
5266 at all times. When traffic begins to rise above what the existing
5267 processes can handle this many more will be spawned up to the maximum
5268 configured. A minimum setting of 1 is required.
5272 The number of requests each storeID helper can handle in
5273 parallel. Defaults to 0 which indicates the helper
5274 is a old-style single threaded program.
5276 When this directive is set to a value >= 1 then the protocol
5277 used to communicate with the helper is modified to include
5278 an ID in front of the request/response. The ID from the request
5279 must be echoed back with the response to that request.
5283 Sets the maximum number of queued requests.
5284 If the queued requests exceed queue size and store_id_bypass
5285 configuration option is set, then storeID helper is bypassed. Otherwise,
5286 if overloading persists squid may abort its operation.
5287 The default value is set to 2*numberofchildren.
5290 NAME: store_id_access storeurl_rewrite_access
5293 DEFAULT_DOC: Allow, unless rules exist in squid.conf.
5294 LOC: Config.accessList.store_id
5296 If defined, this access list specifies which requests are
5297 sent to the StoreID processes. By default all requests
5300 This clause supports both fast and slow acl types.
5301 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5304 NAME: store_id_bypass storeurl_rewrite_bypass
5306 LOC: Config.onoff.store_id_bypass
5309 When this is 'on', a request will not go through the
5310 helper if all helpers are busy. If this is 'off'
5311 and the helper queue grows too large, Squid will exit
5312 with a FATAL error and ask you to increase the number of
5313 helpers. You should only enable this if the helperss
5314 are not critical to your caching system. If you use
5315 helpers for critical caching components, and you enable this
5316 option, users may not get objects from cache.
5317 This options sets default queue-size option of the store_id_children
5322 OPTIONS FOR TUNING THE CACHE
5323 -----------------------------------------------------------------------------
5326 NAME: cache no_cache
5329 DEFAULT_DOC: By default, this directive is unused and has no effect.
5330 LOC: Config.accessList.noCache
5332 Requests denied by this directive will not be served from the cache
5333 and their responses will not be stored in the cache. This directive
5334 has no effect on other transactions and on already cached responses.
5336 This clause supports both fast and slow acl types.
5337 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5339 This and the two other similar caching directives listed below are
5340 checked at different transaction processing stages, have different
5341 access to response information, affect different cache operations,
5342 and differ in slow ACLs support:
5344 * cache: Checked before Squid makes a hit/miss determination.
5345 No access to reply information!
5346 Denies both serving a hit and storing a miss.
5347 Supports both fast and slow ACLs.
5348 * send_hit: Checked after a hit was detected.
5349 Has access to reply (hit) information.
5350 Denies serving a hit only.
5351 Supports fast ACLs only.
5352 * store_miss: Checked before storing a cachable miss.
5353 Has access to reply (miss) information.
5354 Denies storing a miss only.
5355 Supports fast ACLs only.
5357 If you are not sure which of the three directives to use, apply the
5358 following decision logic:
5360 * If your ACL(s) are of slow type _and_ need response info, redesign.
5361 Squid does not support that particular combination at this time.
5363 * If your directive ACL(s) are of slow type, use "cache"; and/or
5364 * if your directive ACL(s) need no response info, use "cache".
5366 * If you do not want the response cached, use store_miss; and/or
5367 * if you do not want a hit on a cached response, use send_hit.
5373 DEFAULT_DOC: By default, this directive is unused and has no effect.
5374 LOC: Config.accessList.sendHit
5376 Responses denied by this directive will not be served from the cache
5377 (but may still be cached, see store_miss). This directive has no
5378 effect on the responses it allows and on the cached objects.
5380 Please see the "cache" directive for a summary of differences among
5381 store_miss, send_hit, and cache directives.
5383 Unlike the "cache" directive, send_hit only supports fast acl
5384 types. See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5388 # apply custom Store ID mapping to some URLs
5389 acl MapMe dstdomain .c.example.com
5390 store_id_program ...
5391 store_id_access allow MapMe
5393 # but prevent caching of special responses
5394 # such as 302 redirects that cause StoreID loops
5395 acl Ordinary http_status 200-299
5396 store_miss deny MapMe !Ordinary
5398 # and do not serve any previously stored special responses
5399 # from the cache (in case they were already cached before
5400 # the above store_miss rule was in effect).
5401 send_hit deny MapMe !Ordinary
5407 DEFAULT_DOC: By default, this directive is unused and has no effect.
5408 LOC: Config.accessList.storeMiss
5410 Responses denied by this directive will not be cached (but may still
5411 be served from the cache, see send_hit). This directive has no
5412 effect on the responses it allows and on the already cached responses.
5414 Please see the "cache" directive for a summary of differences among
5415 store_miss, send_hit, and cache directives. See the
5416 send_hit directive for a usage example.
5418 Unlike the "cache" directive, store_miss only supports fast acl
5419 types. See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5425 LOC: Config.maxStale
5428 This option puts an upper limit on how stale content Squid
5429 will serve from the cache if cache validation fails.
5430 Can be overriden by the refresh_pattern max-stale option.
5433 NAME: refresh_pattern
5434 TYPE: refreshpattern
5438 usage: refresh_pattern [-i] regex min percent max [options]
5440 By default, regular expressions are CASE-SENSITIVE. To make
5441 them case-insensitive, use the -i option.
5443 'Min' is the time (in minutes) an object without an explicit
5444 expiry time should be considered fresh. The recommended
5445 value is 0, any higher values may cause dynamic applications
5446 to be erroneously cached unless the application designer
5447 has taken the appropriate actions.
5449 'Percent' is a percentage of the objects age (time since last
5450 modification age) an object without explicit expiry time
5451 will be considered fresh.
5453 'Max' is an upper limit on how long objects without an explicit
5454 expiry time will be considered fresh.
5456 options: override-expire
5466 override-expire enforces min age even if the server
5467 sent an explicit expiry time (e.g., with the
5468 Expires: header or Cache-Control: max-age). Doing this
5469 VIOLATES the HTTP standard. Enabling this feature
5470 could make you liable for problems which it causes.
5472 Note: override-expire does not enforce staleness - it only extends
5473 freshness / min. If the server returns a Expires time which
5474 is longer than your max time, Squid will still consider
5475 the object fresh for that period of time.
5477 override-lastmod enforces min age even on objects
5478 that were modified recently.
5480 reload-into-ims changes a client no-cache or ``reload''
5481 request for a cached entry into a conditional request using
5482 If-Modified-Since and/or If-None-Match headers, provided the
5483 cached entry has a Last-Modified and/or a strong ETag header.
5484 Doing this VIOLATES the HTTP standard. Enabling this feature
5485 could make you liable for problems which it causes.
5487 ignore-reload ignores a client no-cache or ``reload''
5488 header. Doing this VIOLATES the HTTP standard. Enabling
5489 this feature could make you liable for problems which
5492 ignore-no-store ignores any ``Cache-control: no-store''
5493 headers received from a server. Doing this VIOLATES
5494 the HTTP standard. Enabling this feature could make you
5495 liable for problems which it causes.
5497 ignore-private ignores any ``Cache-control: private''
5498 headers received from a server. Doing this VIOLATES
5499 the HTTP standard. Enabling this feature could make you
5500 liable for problems which it causes.
5502 refresh-ims causes squid to contact the origin server
5503 when a client issues an If-Modified-Since request. This
5504 ensures that the client will receive an updated version
5505 if one is available.
5507 store-stale stores responses even if they don't have explicit
5508 freshness or a validator (i.e., Last-Modified or an ETag)
5509 present, or if they're already stale. By default, Squid will
5510 not cache such responses because they usually can't be
5511 reused. Note that such responses will be stale by default.
5513 max-stale=NN provide a maximum staleness factor. Squid won't
5514 serve objects more stale than this even if it failed to
5515 validate the object. Default: use the max_stale global limit.
5517 Basically a cached object is:
5519 FRESH if expire > now, else STALE
5521 FRESH if lm-factor < percent, else STALE
5525 The refresh_pattern lines are checked in the order listed here.
5526 The first entry which matches is used. If none of the entries
5527 match the default will be used.
5529 Note, you must uncomment all the default lines if you want
5530 to change one. The default setting is only active if none is
5536 # Add any of your own refresh_pattern entries above these.
5538 refresh_pattern ^ftp: 1440 20% 10080
5539 refresh_pattern ^gopher: 1440 0% 1440
5540 refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
5541 refresh_pattern . 0 20% 4320
5545 NAME: quick_abort_min
5549 LOC: Config.quickAbort.min
5552 NAME: quick_abort_max
5556 LOC: Config.quickAbort.max
5559 NAME: quick_abort_pct
5563 LOC: Config.quickAbort.pct
5565 The cache by default continues downloading aborted requests
5566 which are almost completed (less than 16 KB remaining). This
5567 may be undesirable on slow (e.g. SLIP) links and/or very busy
5568 caches. Impatient users may tie up file descriptors and
5569 bandwidth by repeatedly requesting and immediately aborting
5572 When the user aborts a request, Squid will check the
5573 quick_abort values to the amount of data transferred until
5576 If the transfer has less than 'quick_abort_min' KB remaining,
5577 it will finish the retrieval.
5579 If the transfer has more than 'quick_abort_max' KB remaining,
5580 it will abort the retrieval.
5582 If more than 'quick_abort_pct' of the transfer has completed,
5583 it will finish the retrieval.
5585 If you do not want any retrieval to continue after the client
5586 has aborted, set both 'quick_abort_min' and 'quick_abort_max'
5589 If you want retrievals to always continue if they are being
5590 cached set 'quick_abort_min' to '-1 KB'.
5593 NAME: read_ahead_gap
5594 COMMENT: buffer-size
5596 LOC: Config.readAheadGap
5599 The amount of data the cache will buffer ahead of what has been
5600 sent to the client when retrieving an object from another server.
5604 IFDEF: USE_HTTP_VIOLATIONS
5607 LOC: Config.negativeTtl
5610 Set the Default Time-to-Live (TTL) for failed requests.
5611 Certain types of failures (such as "connection refused" and
5612 "404 Not Found") are able to be negatively-cached for a short time.
5613 Modern web servers should provide Expires: header, however if they
5614 do not this can provide a minimum TTL.
5615 The default is not to cache errors with unknown expiry details.
5617 Note that this is different from negative caching of DNS lookups.
5619 WARNING: Doing this VIOLATES the HTTP standard. Enabling
5620 this feature could make you liable for problems which it
5624 NAME: positive_dns_ttl
5627 LOC: Config.positiveDnsTtl
5630 Upper limit on how long Squid will cache positive DNS responses.
5631 Default is 6 hours (360 minutes). This directive must be set
5632 larger than negative_dns_ttl.
5635 NAME: negative_dns_ttl
5638 LOC: Config.negativeDnsTtl
5641 Time-to-Live (TTL) for negative caching of failed DNS lookups.
5642 This also sets the lower cache limit on positive lookups.
5643 Minimum value is 1 second, and it is not recommendable to go
5644 much below 10 seconds.
5647 NAME: range_offset_limit
5648 COMMENT: size [acl acl...]
5650 LOC: Config.rangeOffsetLimit
5653 usage: (size) [units] [[!]aclname]
5655 Sets an upper limit on how far (number of bytes) into the file
5656 a Range request may be to cause Squid to prefetch the whole file.
5657 If beyond this limit, Squid forwards the Range request as it is and
5658 the result is NOT cached.
5660 This is to stop a far ahead range request (lets say start at 17MB)
5661 from making Squid fetch the whole object up to that point before
5662 sending anything to the client.
5664 Multiple range_offset_limit lines may be specified, and they will
5665 be searched from top to bottom on each request until a match is found.
5666 The first match found will be used. If no line matches a request, the
5667 default limit of 0 bytes will be used.
5669 'size' is the limit specified as a number of units.
5671 'units' specifies whether to use bytes, KB, MB, etc.
5672 If no units are specified bytes are assumed.
5674 A size of 0 causes Squid to never fetch more than the
5675 client requested. (default)
5677 A size of 'none' causes Squid to always fetch the object from the
5678 beginning so it may cache the result. (2.0 style)
5680 'aclname' is the name of a defined ACL.
5682 NP: Using 'none' as the byte value here will override any quick_abort settings
5683 that may otherwise apply to the range request. The range request will
5684 be fully fetched from start to finish regardless of the client
5685 actions. This affects bandwidth usage.
5688 NAME: minimum_expiry_time
5691 LOC: Config.minimum_expiry_time
5694 The minimum caching time according to (Expires - Date)
5695 headers Squid honors if the object can't be revalidated.
5696 The default is 60 seconds.
5698 In reverse proxy environments it might be desirable to honor
5699 shorter object lifetimes. It is most likely better to make
5700 your server return a meaningful Last-Modified header however.
5702 In ESI environments where page fragments often have short
5703 lifetimes, this will often be best set to 0.
5706 NAME: store_avg_object_size
5710 LOC: Config.Store.avgObjectSize
5712 Average object size, used to estimate number of objects your
5713 cache can hold. The default is 13 KB.
5715 This is used to pre-seed the cache index memory allocation to
5716 reduce expensive reallocate operations while handling clients
5717 traffic. Too-large values may result in memory allocation during
5718 peak traffic, too-small values will result in wasted memory.
5720 Check the cache manager 'info' report metrics for the real
5721 object sizes seen by your Squid before tuning this.
5724 NAME: store_objects_per_bucket
5727 LOC: Config.Store.objectsPerBucket
5729 Target number of objects per bucket in the store hash table.
5730 Lowering this value increases the total number of buckets and
5731 also the storage maintenance rate. The default is 20.
5736 -----------------------------------------------------------------------------
5739 NAME: request_header_max_size
5743 LOC: Config.maxRequestHeaderSize
5745 This specifies the maximum size for HTTP headers in a request.
5746 Request headers are usually relatively small (about 512 bytes).
5747 Placing a limit on the request header size will catch certain
5748 bugs (for example with persistent connections) and possibly
5749 buffer-overflow or denial-of-service attacks.
5752 NAME: reply_header_max_size
5756 LOC: Config.maxReplyHeaderSize
5758 This specifies the maximum size for HTTP headers in a reply.
5759 Reply headers are usually relatively small (about 512 bytes).
5760 Placing a limit on the reply header size will catch certain
5761 bugs (for example with persistent connections) and possibly
5762 buffer-overflow or denial-of-service attacks.
5765 NAME: request_body_max_size
5769 DEFAULT_DOC: No limit.
5770 LOC: Config.maxRequestBodySize
5772 This specifies the maximum size for an HTTP request body.
5773 In other words, the maximum size of a PUT/POST request.
5774 A user who attempts to send a request with a body larger
5775 than this limit receives an "Invalid Request" error message.
5776 If you set this parameter to a zero (the default), there will
5777 be no limit imposed.
5779 See also client_request_buffer_max_size for an alternative
5780 limitation on client uploads which can be configured.
5783 NAME: client_request_buffer_max_size
5787 LOC: Config.maxRequestBufferSize
5789 This specifies the maximum buffer size of a client request.
5790 It prevents squid eating too much memory when somebody uploads
5795 IFDEF: USE_HTTP_VIOLATIONS
5798 DEFAULT_DOC: Obey RFC 2616.
5799 LOC: Config.accessList.brokenPosts
5801 A list of ACL elements which, if matched, causes Squid to send
5802 an extra CRLF pair after the body of a PUT/POST request.
5804 Some HTTP servers has broken implementations of PUT/POST,
5805 and rely on an extra CRLF pair sent by some WWW clients.
5807 Quote from RFC2616 section 4.1 on this matter:
5809 Note: certain buggy HTTP/1.0 client implementations generate an
5810 extra CRLF's after a POST request. To restate what is explicitly
5811 forbidden by the BNF, an HTTP/1.1 client must not preface or follow
5812 a request with an extra CRLF.
5814 This clause only supports fast acl types.
5815 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5818 acl buggy_server url_regex ^http://....
5819 broken_posts allow buggy_server
5822 NAME: adaptation_uses_indirect_client icap_uses_indirect_client
5825 IFDEF: FOLLOW_X_FORWARDED_FOR&&USE_ADAPTATION
5827 LOC: Adaptation::Config::use_indirect_client
5829 Controls whether the indirect client IP address (instead of the direct
5830 client IP address) is passed to adaptation services.
5832 See also: follow_x_forwarded_for adaptation_send_client_ip
5836 IFDEF: USE_HTTP_VIOLATIONS
5840 LOC: Config.onoff.via
5842 If set (default), Squid will include a Via header in requests and
5843 replies as required by RFC2616.
5849 LOC: Config.onoff.ie_refresh
5852 Microsoft Internet Explorer up until version 5.5 Service
5853 Pack 1 has an issue with transparent proxies, wherein it
5854 is impossible to force a refresh. Turning this on provides
5855 a partial fix to the problem, by causing all IMS-REFRESH
5856 requests from older IE versions to check the origin server
5857 for fresh content. This reduces hit ratio by some amount
5858 (~10% in my experience), but allows users to actually get
5859 fresh content when they want it. Note because Squid
5860 cannot tell if the user is using 5.5 or 5.5SP1, the behavior
5861 of 5.5 is unchanged from old versions of Squid (i.e. a
5862 forced refresh is impossible). Newer versions of IE will,
5863 hopefully, continue to have the new behavior and will be
5864 handled based on that assumption. This option defaults to
5865 the old Squid behavior, which is better for hit ratios but
5866 worse for clients using IE, if they need to be able to
5867 force fresh content.
5870 NAME: vary_ignore_expire
5873 LOC: Config.onoff.vary_ignore_expire
5876 Many HTTP servers supporting Vary gives such objects
5877 immediate expiry time with no cache-control header
5878 when requested by a HTTP/1.0 client. This option
5879 enables Squid to ignore such expiry times until
5880 HTTP/1.1 is fully implemented.
5882 WARNING: If turned on this may eventually cause some
5883 varying objects not intended for caching to get cached.
5886 NAME: request_entities
5888 LOC: Config.onoff.request_entities
5891 Squid defaults to deny GET and HEAD requests with request entities,
5892 as the meaning of such requests are undefined in the HTTP standard
5893 even if not explicitly forbidden.
5895 Set this directive to on if you have clients which insists
5896 on sending request entities in GET or HEAD requests. But be warned
5897 that there is server software (both proxies and web servers) which
5898 can fail to properly process this kind of request which may make you
5899 vulnerable to cache pollution attacks if enabled.
5902 NAME: request_header_access
5903 IFDEF: USE_HTTP_VIOLATIONS
5904 TYPE: http_header_access
5905 LOC: Config.request_header_access
5907 DEFAULT_DOC: No limits.
5909 Usage: request_header_access header_name allow|deny [!]aclname ...
5911 WARNING: Doing this VIOLATES the HTTP standard. Enabling
5912 this feature could make you liable for problems which it
5915 This option replaces the old 'anonymize_headers' and the
5916 older 'http_anonymizer' option with something that is much
5917 more configurable. A list of ACLs for each header name allows
5918 removal of specific header fields under specific conditions.
5920 This option only applies to outgoing HTTP request headers (i.e.,
5921 headers sent by Squid to the next HTTP hop such as a cache peer
5922 or an origin server). The option has no effect during cache hit
5923 detection. The equivalent adaptation vectoring point in ICAP
5924 terminology is post-cache REQMOD.
5926 The option is applied to individual outgoing request header
5927 fields. For each request header field F, Squid uses the first
5928 qualifying sets of request_header_access rules:
5930 1. Rules with header_name equal to F's name.
5931 2. Rules with header_name 'Other', provided F's name is not
5932 on the hard-coded list of commonly used HTTP header names.
5933 3. Rules with header_name 'All'.
5935 Within that qualifying rule set, rule ACLs are checked as usual.
5936 If ACLs of an "allow" rule match, the header field is allowed to
5937 go through as is. If ACLs of a "deny" rule match, the header is
5938 removed and request_header_replace is then checked to identify
5939 if the removed header has a replacement. If no rules within the
5940 set have matching ACLs, the header field is left as is.
5942 For example, to achieve the same behavior as the old
5943 'http_anonymizer standard' option, you should use:
5945 request_header_access From deny all
5946 request_header_access Referer deny all
5947 request_header_access User-Agent deny all
5949 Or, to reproduce the old 'http_anonymizer paranoid' feature
5952 request_header_access Authorization allow all
5953 request_header_access Proxy-Authorization allow all
5954 request_header_access Cache-Control allow all
5955 request_header_access Content-Length allow all
5956 request_header_access Content-Type allow all
5957 request_header_access Date allow all
5958 request_header_access Host allow all
5959 request_header_access If-Modified-Since allow all
5960 request_header_access Pragma allow all
5961 request_header_access Accept allow all
5962 request_header_access Accept-Charset allow all
5963 request_header_access Accept-Encoding allow all
5964 request_header_access Accept-Language allow all
5965 request_header_access Connection allow all
5966 request_header_access All deny all
5968 HTTP reply headers are controlled with the reply_header_access directive.
5970 By default, all headers are allowed (no anonymizing is performed).
5973 NAME: reply_header_access
5974 IFDEF: USE_HTTP_VIOLATIONS
5975 TYPE: http_header_access
5976 LOC: Config.reply_header_access
5978 DEFAULT_DOC: No limits.
5980 Usage: reply_header_access header_name allow|deny [!]aclname ...
5982 WARNING: Doing this VIOLATES the HTTP standard. Enabling
5983 this feature could make you liable for problems which it
5986 This option only applies to reply headers, i.e., from the
5987 server to the client.
5989 This is the same as request_header_access, but in the other
5990 direction. Please see request_header_access for detailed
5993 For example, to achieve the same behavior as the old
5994 'http_anonymizer standard' option, you should use:
5996 reply_header_access Server deny all
5997 reply_header_access WWW-Authenticate deny all
5998 reply_header_access Link deny all
6000 Or, to reproduce the old 'http_anonymizer paranoid' feature
6003 reply_header_access Allow allow all
6004 reply_header_access WWW-Authenticate allow all
6005 reply_header_access Proxy-Authenticate allow all
6006 reply_header_access Cache-Control allow all
6007 reply_header_access Content-Encoding allow all
6008 reply_header_access Content-Length allow all
6009 reply_header_access Content-Type allow all
6010 reply_header_access Date allow all
6011 reply_header_access Expires allow all
6012 reply_header_access Last-Modified allow all
6013 reply_header_access Location allow all
6014 reply_header_access Pragma allow all
6015 reply_header_access Content-Language allow all
6016 reply_header_access Retry-After allow all
6017 reply_header_access Title allow all
6018 reply_header_access Content-Disposition allow all
6019 reply_header_access Connection allow all
6020 reply_header_access All deny all
6022 HTTP request headers are controlled with the request_header_access directive.
6024 By default, all headers are allowed (no anonymizing is
6028 NAME: request_header_replace header_replace
6029 IFDEF: USE_HTTP_VIOLATIONS
6030 TYPE: http_header_replace
6031 LOC: Config.request_header_access
6034 Usage: request_header_replace header_name message
6035 Example: request_header_replace User-Agent Nutscrape/1.0 (CP/M; 8-bit)
6037 This option allows you to change the contents of headers
6038 denied with request_header_access above, by replacing them
6039 with some fixed string.
6041 This only applies to request headers, not reply headers.
6043 By default, headers are removed if denied.
6046 NAME: reply_header_replace
6047 IFDEF: USE_HTTP_VIOLATIONS
6048 TYPE: http_header_replace
6049 LOC: Config.reply_header_access
6052 Usage: reply_header_replace header_name message
6053 Example: reply_header_replace Server Foo/1.0
6055 This option allows you to change the contents of headers
6056 denied with reply_header_access above, by replacing them
6057 with some fixed string.
6059 This only applies to reply headers, not request headers.
6061 By default, headers are removed if denied.
6064 NAME: request_header_add
6065 TYPE: HeaderWithAclList
6066 LOC: Config.request_header_add
6069 Usage: request_header_add field-name field-value acl1 [acl2] ...
6070 Example: request_header_add X-Client-CA "CA=%ssl::>cert_issuer" all
6072 This option adds header fields to outgoing HTTP requests (i.e.,
6073 request headers sent by Squid to the next HTTP hop such as a
6074 cache peer or an origin server). The option has no effect during
6075 cache hit detection. The equivalent adaptation vectoring point
6076 in ICAP terminology is post-cache REQMOD.
6078 Field-name is a token specifying an HTTP header name. If a
6079 standard HTTP header name is used, Squid does not check whether
6080 the new header conflicts with any existing headers or violates
6081 HTTP rules. If the request to be modified already contains a
6082 field with the same name, the old field is preserved but the
6083 header field values are not merged.
6085 Field-value is either a token or a quoted string. If quoted
6086 string format is used, then the surrounding quotes are removed
6087 while escape sequences and %macros are processed.
6089 In theory, all of the logformat codes can be used as %macros.
6090 However, unlike logging (which happens at the very end of
6091 transaction lifetime), the transaction may not yet have enough
6092 information to expand a macro when the new header value is needed.
6093 And some information may already be available to Squid but not yet
6094 committed where the macro expansion code can access it (report
6095 such instances!). The macro will be expanded into a single dash
6096 ('-') in such cases. Not all macros have been tested.
6098 One or more Squid ACLs may be specified to restrict header
6099 injection to matching requests. As always in squid.conf, all
6100 ACLs in an option ACL list must be satisfied for the insertion
6101 to happen. The request_header_add option supports fast ACLs
6110 This option used to log custom information about the master
6111 transaction. For example, an admin may configure Squid to log
6112 which "user group" the transaction belongs to, where "user group"
6113 will be determined based on a set of ACLs and not [just]
6114 authentication information.
6115 Values of key/value pairs can be logged using %{key}note macros:
6117 note key value acl ...
6118 logformat myFormat ... %{key}note ...
6121 NAME: relaxed_header_parser
6122 COMMENT: on|off|warn
6124 LOC: Config.onoff.relaxed_header_parser
6127 In the default "on" setting Squid accepts certain forms
6128 of non-compliant HTTP messages where it is unambiguous
6129 what the sending application intended even if the message
6130 is not correctly formatted. The messages is then normalized
6131 to the correct form when forwarded by Squid.
6133 If set to "warn" then a warning will be emitted in cache.log
6134 each time such HTTP error is encountered.
6136 If set to "off" then such HTTP errors will cause the request
6137 or response to be rejected.
6140 NAME: collapsed_forwarding
6143 LOC: Config.onoff.collapsed_forwarding
6146 This option controls whether Squid is allowed to merge multiple
6147 potentially cachable requests for the same URI before Squid knows
6148 whether the response is going to be cachable.
6150 This feature is disabled by default: Enabling collapsed forwarding
6151 needlessly delays forwarding requests that look cachable (when they are
6152 collapsed) but then need to be forwarded individually anyway because
6153 they end up being for uncachable content. However, in some cases, such
6154 as accelleration of highly cachable content with periodic or groupped
6155 expiration times, the gains from collapsing [large volumes of
6156 simultenous refresh requests] outweigh losses from such delays.
6161 -----------------------------------------------------------------------------
6164 NAME: forward_timeout
6167 LOC: Config.Timeout.forward
6170 This parameter specifies how long Squid should at most attempt in
6171 finding a forwarding path for the request before giving up.
6174 NAME: connect_timeout
6177 LOC: Config.Timeout.connect
6180 This parameter specifies how long to wait for the TCP connect to
6181 the requested server or peer to complete before Squid should
6182 attempt to find another path where to forward the request.
6185 NAME: peer_connect_timeout
6188 LOC: Config.Timeout.peer_connect
6191 This parameter specifies how long to wait for a pending TCP
6192 connection to a peer cache. The default is 30 seconds. You
6193 may also set different timeout values for individual neighbors
6194 with the 'connect-timeout' option on a 'cache_peer' line.
6200 LOC: Config.Timeout.read
6203 Applied on peer server connections.
6205 After each successful read(), the timeout will be extended by this
6206 amount. If no data is read again after this amount of time,
6207 the request is aborted and logged with ERR_READ_TIMEOUT.
6209 The default is 15 minutes.
6215 LOC: Config.Timeout.write
6218 This timeout is tracked for all connections that have data
6219 available for writing and are waiting for the socket to become
6220 ready. After each successful write, the timeout is extended by
6221 the configured amount. If Squid has data to write but the
6222 connection is not ready for the configured duration, the
6223 transaction associated with the connection is terminated. The
6224 default is 15 minutes.
6227 NAME: request_timeout
6229 LOC: Config.Timeout.request
6232 How long to wait for complete HTTP request headers after initial
6233 connection establishment.
6236 NAME: request_start_timeout
6238 LOC: Config.Timeout.request_start_timeout
6241 How long to wait for the first request byte after initial
6242 connection establishment.
6245 NAME: client_idle_pconn_timeout persistent_request_timeout
6247 LOC: Config.Timeout.clientIdlePconn
6250 How long to wait for the next HTTP request on a persistent
6251 client connection after the previous request completes.
6254 NAME: ftp_client_idle_timeout
6256 LOC: Config.Timeout.ftpClientIdle
6259 How long to wait for an FTP request on a connection to Squid ftp_port.
6260 Many FTP clients do not deal with idle connection closures well,
6261 necessitating a longer default timeout than client_idle_pconn_timeout
6262 used for incoming HTTP requests.
6265 NAME: client_lifetime
6268 LOC: Config.Timeout.lifetime
6271 The maximum amount of time a client (browser) is allowed to
6272 remain connected to the cache process. This protects the Cache
6273 from having a lot of sockets (and hence file descriptors) tied up
6274 in a CLOSE_WAIT state from remote clients that go away without
6275 properly shutting down (either because of a network failure or
6276 because of a poor client implementation). The default is one
6279 NOTE: The default value is intended to be much larger than any
6280 client would ever need to be connected to your cache. You
6281 should probably change client_lifetime only as a last resort.
6282 If you seem to have many client connections tying up
6283 filedescriptors, we recommend first tuning the read_timeout,
6284 request_timeout, persistent_request_timeout and quick_abort values.
6287 NAME: pconn_lifetime
6290 LOC: Config.Timeout.pconnLifetime
6293 Desired maximum lifetime of a persistent connection.
6294 When set, Squid will close a now-idle persistent connection that
6295 exceeded configured lifetime instead of moving the connection into
6296 the idle connection pool (or equivalent). No effect on ongoing/active
6297 transactions. Connection lifetime is the time period from the
6298 connection acceptance or opening time until "now".
6300 This limit is useful in environments with long-lived connections
6301 where Squid configuration or environmental factors change during a
6302 single connection lifetime. If unrestricted, some connections may
6303 last for hours and even days, ignoring those changes that should
6304 have affected their behavior or their existence.
6306 Currently, a new lifetime value supplied via Squid reconfiguration
6307 has no effect on already idle connections unless they become busy.
6309 When set to '0' this limit is not used.
6312 NAME: half_closed_clients
6314 LOC: Config.onoff.half_closed_clients
6317 Some clients may shutdown the sending side of their TCP
6318 connections, while leaving their receiving sides open. Sometimes,
6319 Squid can not tell the difference between a half-closed and a
6320 fully-closed TCP connection.
6322 By default, Squid will immediately close client connections when
6323 read(2) returns "no more data to read."
6325 Change this option to 'on' and Squid will keep open connections
6326 until a read(2) or write(2) on the socket returns an error.
6327 This may show some benefits for reverse proxies. But if not
6328 it is recommended to leave OFF.
6331 NAME: server_idle_pconn_timeout pconn_timeout
6333 LOC: Config.Timeout.serverIdlePconn
6336 Timeout for idle persistent connections to servers and other
6343 LOC: Ident::TheConfig.timeout
6346 Maximum time to wait for IDENT lookups to complete.
6348 If this is too high, and you enabled IDENT lookups from untrusted
6349 users, you might be susceptible to denial-of-service by having
6350 many ident requests going at once.
6353 NAME: shutdown_lifetime
6356 LOC: Config.shutdownLifetime
6359 When SIGTERM or SIGHUP is received, the cache is put into
6360 "shutdown pending" mode until all active sockets are closed.
6361 This value is the lifetime to set for all open descriptors
6362 during shutdown mode. Any active clients after this many
6363 seconds will receive a 'timeout' message.
6367 ADMINISTRATIVE PARAMETERS
6368 -----------------------------------------------------------------------------
6374 LOC: Config.adminEmail
6376 Email-address of local cache manager who will receive
6377 mail if the cache dies. The default is "webmaster".
6383 LOC: Config.EmailFrom
6385 From: email-address for mail sent when the cache dies.
6386 The default is to use 'squid@unique_hostname'.
6388 See also: unique_hostname directive.
6394 LOC: Config.EmailProgram
6396 Email program used to send mail if the cache dies.
6397 The default is "mail". The specified program must comply
6398 with the standard Unix mail syntax:
6399 mail-program recipient < mailfile
6401 Optional command line options can be specified.
6404 NAME: cache_effective_user
6406 DEFAULT: @DEFAULT_CACHE_EFFECTIVE_USER@
6407 LOC: Config.effectiveUser
6409 If you start Squid as root, it will change its effective/real
6410 UID/GID to the user specified below. The default is to change
6411 to UID of @DEFAULT_CACHE_EFFECTIVE_USER@.
6412 see also; cache_effective_group
6415 NAME: cache_effective_group
6418 DEFAULT_DOC: Use system group memberships of the cache_effective_user account
6419 LOC: Config.effectiveGroup
6421 Squid sets the GID to the effective user's default group ID
6422 (taken from the password file) and supplementary group list
6423 from the groups membership.
6425 If you want Squid to run with a specific GID regardless of
6426 the group memberships of the effective user then set this
6427 to the group (or GID) you want Squid to run as. When set
6428 all other group privileges of the effective user are ignored
6429 and only this GID is effective. If Squid is not started as
6430 root the user starting Squid MUST be member of the specified
6433 This option is not recommended by the Squid Team.
6434 Our preference is for administrators to configure a secure
6435 user account for squid with UID/GID matching system policies.
6438 NAME: httpd_suppress_version_string
6442 LOC: Config.onoff.httpd_suppress_version_string
6444 Suppress Squid version string info in HTTP headers and HTML error pages.
6447 NAME: visible_hostname
6449 LOC: Config.visibleHostname
6451 DEFAULT_DOC: Automatically detect the system host name
6453 If you want to present a special hostname in error messages, etc,
6454 define this. Otherwise, the return value of gethostname()
6455 will be used. If you have multiple caches in a cluster and
6456 get errors about IP-forwarding you must set them to have individual
6457 names with this setting.
6460 NAME: unique_hostname
6462 LOC: Config.uniqueHostname
6464 DEFAULT_DOC: Copy the value from visible_hostname
6466 If you want to have multiple machines with the same
6467 'visible_hostname' you must give each machine a different
6468 'unique_hostname' so forwarding loops can be detected.
6471 NAME: hostname_aliases
6473 LOC: Config.hostnameAliases
6476 A list of other DNS names your cache has.
6484 Minimum umask which should be enforced while the proxy
6485 is running, in addition to the umask set at startup.
6487 For a traditional octal representation of umasks, start
6492 OPTIONS FOR THE CACHE REGISTRATION SERVICE
6493 -----------------------------------------------------------------------------
6495 This section contains parameters for the (optional) cache
6496 announcement service. This service is provided to help
6497 cache administrators locate one another in order to join or
6498 create cache hierarchies.
6500 An 'announcement' message is sent (via UDP) to the registration
6501 service by Squid. By default, the announcement message is NOT
6502 SENT unless you enable it with 'announce_period' below.
6504 The announcement message includes your hostname, plus the
6505 following information from this configuration file:
6511 All current information is processed regularly and made
6512 available on the Web at http://www.ircache.net/Cache/Tracker/.
6515 NAME: announce_period
6517 LOC: Config.Announce.period
6519 DEFAULT_DOC: Announcement messages disabled.
6521 This is how frequently to send cache announcements.
6523 To enable announcing your cache, just set an announce period.
6526 announce_period 1 day
6531 DEFAULT: tracker.ircache.net
6532 LOC: Config.Announce.host
6534 Set the hostname where announce registration messages will be sent.
6536 See also announce_port and announce_file
6542 LOC: Config.Announce.file
6544 The contents of this file will be included in the announce
6545 registration messages.
6551 LOC: Config.Announce.port
6553 Set the port where announce registration messages will be sent.
6555 See also announce_host and announce_file
6559 HTTPD-ACCELERATOR OPTIONS
6560 -----------------------------------------------------------------------------
6563 NAME: httpd_accel_surrogate_id
6566 DEFAULT_DOC: visible_hostname is used if no specific ID is set.
6567 LOC: Config.Accel.surrogate_id
6569 Surrogates (http://www.esi.org/architecture_spec_1.0.html)
6570 need an identification token to allow control targeting. Because
6571 a farm of surrogates may all perform the same tasks, they may share
6572 an identification token.
6575 NAME: http_accel_surrogate_remote
6579 LOC: Config.onoff.surrogate_is_remote
6581 Remote surrogates (such as those in a CDN) honour the header
6582 "Surrogate-Control: no-store-remote".
6584 Set this to on to have squid behave as a remote surrogate.
6588 IFDEF: USE_SQUID_ESI
6589 COMMENT: libxml2|expat|custom
6591 LOC: ESIParser::Type
6594 ESI markup is not strictly XML compatible. The custom ESI parser
6595 will give higher performance, but cannot handle non ASCII character
6600 DELAY POOL PARAMETERS
6601 -----------------------------------------------------------------------------
6605 TYPE: delay_pool_count
6607 IFDEF: USE_DELAY_POOLS
6610 This represents the number of delay pools to be used. For example,
6611 if you have one class 2 delay pool and one class 3 delays pool, you
6612 have a total of 2 delay pools.
6614 See also delay_parameters, delay_class, delay_access for pool
6615 configuration details.
6619 TYPE: delay_pool_class
6621 IFDEF: USE_DELAY_POOLS
6624 This defines the class of each delay pool. There must be exactly one
6625 delay_class line for each delay pool. For example, to define two
6626 delay pools, one of class 2 and one of class 3, the settings above
6630 delay_pools 4 # 4 delay pools
6631 delay_class 1 2 # pool 1 is a class 2 pool
6632 delay_class 2 3 # pool 2 is a class 3 pool
6633 delay_class 3 4 # pool 3 is a class 4 pool
6634 delay_class 4 5 # pool 4 is a class 5 pool
6636 The delay pool classes are:
6638 class 1 Everything is limited by a single aggregate
6641 class 2 Everything is limited by a single aggregate
6642 bucket as well as an "individual" bucket chosen
6643 from bits 25 through 32 of the IPv4 address.
6645 class 3 Everything is limited by a single aggregate
6646 bucket as well as a "network" bucket chosen
6647 from bits 17 through 24 of the IP address and a
6648 "individual" bucket chosen from bits 17 through
6649 32 of the IPv4 address.
6651 class 4 Everything in a class 3 delay pool, with an
6652 additional limit on a per user basis. This
6653 only takes effect if the username is established
6654 in advance - by forcing authentication in your
6657 class 5 Requests are grouped according their tag (see
6658 external_acl's tag= reply).
6661 Each pool also requires a delay_parameters directive to configure the pool size
6662 and speed limits used whenever the pool is applied to a request. Along with
6663 a set of delay_access directives to determine when it is used.
6665 NOTE: If an IP address is a.b.c.d
6666 -> bits 25 through 32 are "d"
6667 -> bits 17 through 24 are "c"
6668 -> bits 17 through 32 are "c * 256 + d"
6670 NOTE-2: Due to the use of bitmasks in class 2,3,4 pools they only apply to
6671 IPv4 traffic. Class 1 and 5 pools may be used with IPv6 traffic.
6673 This clause only supports fast acl types.
6674 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6676 See also delay_parameters and delay_access.
6680 TYPE: delay_pool_access
6682 DEFAULT_DOC: Deny using the pool, unless allow rules exist in squid.conf for the pool.
6683 IFDEF: USE_DELAY_POOLS
6686 This is used to determine which delay pool a request falls into.
6688 delay_access is sorted per pool and the matching starts with pool 1,
6689 then pool 2, ..., and finally pool N. The first delay pool where the
6690 request is allowed is selected for the request. If it does not allow
6691 the request to any pool then the request is not delayed (default).
6693 For example, if you want some_big_clients in delay
6694 pool 1 and lotsa_little_clients in delay pool 2:
6696 delay_access 1 allow some_big_clients
6697 delay_access 1 deny all
6698 delay_access 2 allow lotsa_little_clients
6699 delay_access 2 deny all
6700 delay_access 3 allow authenticated_clients
6702 See also delay_parameters and delay_class.
6706 NAME: delay_parameters
6707 TYPE: delay_pool_rates
6709 IFDEF: USE_DELAY_POOLS
6712 This defines the parameters for a delay pool. Each delay pool has
6713 a number of "buckets" associated with it, as explained in the
6714 description of delay_class.
6716 For a class 1 delay pool, the syntax is:
6718 delay_parameters pool aggregate
6720 For a class 2 delay pool:
6722 delay_parameters pool aggregate individual
6724 For a class 3 delay pool:
6726 delay_parameters pool aggregate network individual
6728 For a class 4 delay pool:
6730 delay_parameters pool aggregate network individual user
6732 For a class 5 delay pool:
6734 delay_parameters pool tagrate
6736 The option variables are:
6738 pool a pool number - ie, a number between 1 and the
6739 number specified in delay_pools as used in
6742 aggregate the speed limit parameters for the aggregate bucket
6745 individual the speed limit parameters for the individual
6746 buckets (class 2, 3).
6748 network the speed limit parameters for the network buckets
6751 user the speed limit parameters for the user buckets
6754 tagrate the speed limit parameters for the tag buckets
6757 A pair of delay parameters is written restore/maximum, where restore is
6758 the number of bytes (not bits - modem and network speeds are usually
6759 quoted in bits) per second placed into the bucket, and maximum is the
6760 maximum number of bytes which can be in the bucket at any time.
6762 There must be one delay_parameters line for each delay pool.
6765 For example, if delay pool number 1 is a class 2 delay pool as in the
6766 above example, and is being used to strictly limit each host to 64Kbit/sec
6767 (plus overheads), with no overall limit, the line is:
6769 delay_parameters 1 none 8000/8000
6771 Note that 8 x 8K Byte/sec -> 64K bit/sec.
6773 Note that the word 'none' is used to represent no limit.
6776 And, if delay pool number 2 is a class 3 delay pool as in the above
6777 example, and you want to limit it to a total of 256Kbit/sec (strict limit)
6778 with each 8-bit network permitted 64Kbit/sec (strict limit) and each
6779 individual host permitted 4800bit/sec with a bucket maximum size of 64Kbits
6780 to permit a decent web page to be downloaded at a decent speed
6781 (if the network is not being limited due to overuse) but slow down
6782 large downloads more significantly:
6784 delay_parameters 2 32000/32000 8000/8000 600/8000
6786 Note that 8 x 32K Byte/sec -> 256K bit/sec.
6787 8 x 8K Byte/sec -> 64K bit/sec.
6788 8 x 600 Byte/sec -> 4800 bit/sec.
6791 Finally, for a class 4 delay pool as in the example - each user will
6792 be limited to 128Kbits/sec no matter how many workstations they are logged into.:
6794 delay_parameters 4 32000/32000 8000/8000 600/64000 16000/16000
6797 See also delay_class and delay_access.
6801 NAME: delay_initial_bucket_level
6802 COMMENT: (percent, 0-100)
6805 IFDEF: USE_DELAY_POOLS
6806 LOC: Config.Delay.initial
6808 The initial bucket percentage is used to determine how much is put
6809 in each bucket when squid starts, is reconfigured, or first notices
6810 a host accessing it (in class 2 and class 3, individual hosts and
6811 networks only have buckets associated with them once they have been
6816 CLIENT DELAY POOL PARAMETERS
6817 -----------------------------------------------------------------------------
6820 NAME: client_delay_pools
6821 TYPE: client_delay_pool_count
6823 IFDEF: USE_DELAY_POOLS
6824 LOC: Config.ClientDelay
6826 This option specifies the number of client delay pools used. It must
6827 preceed other client_delay_* options.
6830 client_delay_pools 2
6832 See also client_delay_parameters and client_delay_access.
6835 NAME: client_delay_initial_bucket_level
6836 COMMENT: (percent, 0-no_limit)
6839 IFDEF: USE_DELAY_POOLS
6840 LOC: Config.ClientDelay.initial
6842 This option determines the initial bucket size as a percentage of
6843 max_bucket_size from client_delay_parameters. Buckets are created
6844 at the time of the "first" connection from the matching IP. Idle
6845 buckets are periodically deleted up.
6847 You can specify more than 100 percent but note that such "oversized"
6848 buckets are not refilled until their size goes down to max_bucket_size
6849 from client_delay_parameters.
6852 client_delay_initial_bucket_level 50
6855 NAME: client_delay_parameters
6856 TYPE: client_delay_pool_rates
6858 IFDEF: USE_DELAY_POOLS
6859 LOC: Config.ClientDelay
6862 This option configures client-side bandwidth limits using the
6865 client_delay_parameters pool speed_limit max_bucket_size
6867 pool is an integer ID used for client_delay_access matching.
6869 speed_limit is bytes added to the bucket per second.
6871 max_bucket_size is the maximum size of a bucket, enforced after any
6872 speed_limit additions.
6874 Please see the delay_parameters option for more information and
6878 client_delay_parameters 1 1024 2048
6879 client_delay_parameters 2 51200 16384
6881 See also client_delay_access.
6885 NAME: client_delay_access
6886 TYPE: client_delay_pool_access
6888 DEFAULT_DOC: Deny use of the pool, unless allow rules exist in squid.conf for the pool.
6889 IFDEF: USE_DELAY_POOLS
6890 LOC: Config.ClientDelay
6892 This option determines the client-side delay pool for the
6895 client_delay_access pool_ID allow|deny acl_name
6897 All client_delay_access options are checked in their pool ID
6898 order, starting with pool 1. The first checked pool with allowed
6899 request is selected for the request. If no ACL matches or there
6900 are no client_delay_access options, the request bandwidth is not
6903 The ACL-selected pool is then used to find the
6904 client_delay_parameters for the request. Client-side pools are
6905 not used to aggregate clients. Clients are always aggregated
6906 based on their source IP addresses (one bucket per source IP).
6908 This clause only supports fast acl types.
6909 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6910 Additionally, only the client TCP connection details are available.
6911 ACLs testing HTTP properties will not work.
6913 Please see delay_access for more examples.
6916 client_delay_access 1 allow low_rate_network
6917 client_delay_access 2 allow vips_network
6920 See also client_delay_parameters and client_delay_pools.
6924 WCCPv1 AND WCCPv2 CONFIGURATION OPTIONS
6925 -----------------------------------------------------------------------------
6930 LOC: Config.Wccp.router
6932 DEFAULT_DOC: WCCP disabled.
6935 Use this option to define your WCCP ``home'' router for
6938 wccp_router supports a single WCCP(v1) router
6940 wccp2_router supports multiple WCCPv2 routers
6942 only one of the two may be used at the same time and defines
6943 which version of WCCP to use.
6947 TYPE: IpAddress_list
6948 LOC: Config.Wccp2.router
6950 DEFAULT_DOC: WCCPv2 disabled.
6953 Use this option to define your WCCP ``home'' router for
6956 wccp_router supports a single WCCP(v1) router
6958 wccp2_router supports multiple WCCPv2 routers
6960 only one of the two may be used at the same time and defines
6961 which version of WCCP to use.
6966 LOC: Config.Wccp.version
6970 This directive is only relevant if you need to set up WCCP(v1)
6971 to some very old and end-of-life Cisco routers. In all other
6972 setups it must be left unset or at the default setting.
6973 It defines an internal version in the WCCP(v1) protocol,
6974 with version 4 being the officially documented protocol.
6976 According to some users, Cisco IOS 11.2 and earlier only
6977 support WCCP version 3. If you're using that or an earlier
6978 version of IOS, you may need to change this value to 3, otherwise
6979 do not specify this parameter.
6982 NAME: wccp2_rebuild_wait
6984 LOC: Config.Wccp2.rebuildwait
6988 If this is enabled Squid will wait for the cache dir rebuild to finish
6989 before sending the first wccp2 HereIAm packet
6992 NAME: wccp2_forwarding_method
6994 LOC: Config.Wccp2.forwarding_method
6998 WCCP2 allows the setting of forwarding methods between the
6999 router/switch and the cache. Valid values are as follows:
7001 gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
7002 l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
7004 Currently (as of IOS 12.4) cisco routers only support GRE.
7005 Cisco switches only support the L2 redirect assignment method.
7008 NAME: wccp2_return_method
7010 LOC: Config.Wccp2.return_method
7014 WCCP2 allows the setting of return methods between the
7015 router/switch and the cache for packets that the cache
7016 decides not to handle. Valid values are as follows:
7018 gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
7019 l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
7021 Currently (as of IOS 12.4) cisco routers only support GRE.
7022 Cisco switches only support the L2 redirect assignment.
7024 If the "ip wccp redirect exclude in" command has been
7025 enabled on the cache interface, then it is still safe for
7026 the proxy server to use a l2 redirect method even if this
7027 option is set to GRE.
7030 NAME: wccp2_assignment_method
7032 LOC: Config.Wccp2.assignment_method
7036 WCCP2 allows the setting of methods to assign the WCCP hash
7037 Valid values are as follows:
7039 hash - Hash assignment
7040 mask - Mask assignment
7042 As a general rule, cisco routers support the hash assignment method
7043 and cisco switches support the mask assignment method.
7048 LOC: Config.Wccp2.info
7049 DEFAULT_IF_NONE: standard 0
7050 DEFAULT_DOC: Use the 'web-cache' standard service.
7053 WCCP2 allows for multiple traffic services. There are two
7054 types: "standard" and "dynamic". The standard type defines
7055 one service id - http (id 0). The dynamic service ids can be from
7056 51 to 255 inclusive. In order to use a dynamic service id
7057 one must define the type of traffic to be redirected; this is done
7058 using the wccp2_service_info option.
7060 The "standard" type does not require a wccp2_service_info option,
7061 just specifying the service id will suffice.
7063 MD5 service authentication can be enabled by adding
7064 "password=<password>" to the end of this service declaration.
7068 wccp2_service standard 0 # for the 'web-cache' standard service
7069 wccp2_service dynamic 80 # a dynamic service type which will be
7070 # fleshed out with subsequent options.
7071 wccp2_service standard 0 password=foo
7074 NAME: wccp2_service_info
7075 TYPE: wccp2_service_info
7076 LOC: Config.Wccp2.info
7080 Dynamic WCCPv2 services require further information to define the
7081 traffic you wish to have diverted.
7085 wccp2_service_info <id> protocol=<protocol> flags=<flag>,<flag>..
7086 priority=<priority> ports=<port>,<port>..
7088 The relevant WCCPv2 flags:
7089 + src_ip_hash, dst_ip_hash
7090 + source_port_hash, dst_port_hash
7091 + src_ip_alt_hash, dst_ip_alt_hash
7092 + src_port_alt_hash, dst_port_alt_hash
7095 The port list can be one to eight entries.
7099 wccp2_service_info 80 protocol=tcp flags=src_ip_hash,ports_source
7100 priority=240 ports=80
7102 Note: the service id must have been defined by a previous
7103 'wccp2_service dynamic <id>' entry.
7108 LOC: Config.Wccp2.weight
7112 Each cache server gets assigned a set of the destination
7113 hash proportional to their weight.
7118 LOC: Config.Wccp.address
7120 DEFAULT_DOC: Address selected by the operating system.
7123 Use this option if you require WCCPv2 to use a specific
7126 The default behavior is to not bind to any specific address.
7131 LOC: Config.Wccp2.address
7133 DEFAULT_DOC: Address selected by the operating system.
7136 Use this option if you require WCCP to use a specific
7139 The default behavior is to not bind to any specific address.
7143 PERSISTENT CONNECTION HANDLING
7144 -----------------------------------------------------------------------------
7146 Also see "pconn_timeout" in the TIMEOUTS section
7149 NAME: client_persistent_connections
7151 LOC: Config.onoff.client_pconns
7154 Persistent connection support for clients.
7155 Squid uses persistent connections (when allowed). You can use
7156 this option to disable persistent connections with clients.
7159 NAME: server_persistent_connections
7161 LOC: Config.onoff.server_pconns
7164 Persistent connection support for servers.
7165 Squid uses persistent connections (when allowed). You can use
7166 this option to disable persistent connections with servers.
7169 NAME: persistent_connection_after_error
7171 LOC: Config.onoff.error_pconns
7174 With this directive the use of persistent connections after
7175 HTTP errors can be disabled. Useful if you have clients
7176 who fail to handle errors on persistent connections proper.
7179 NAME: detect_broken_pconn
7181 LOC: Config.onoff.detect_broken_server_pconns
7184 Some servers have been found to incorrectly signal the use
7185 of HTTP/1.0 persistent connections even on replies not
7186 compatible, causing significant delays. This server problem
7187 has mostly been seen on redirects.
7189 By enabling this directive Squid attempts to detect such
7190 broken replies and automatically assume the reply is finished
7191 after 10 seconds timeout.
7195 CACHE DIGEST OPTIONS
7196 -----------------------------------------------------------------------------
7199 NAME: digest_generation
7200 IFDEF: USE_CACHE_DIGESTS
7202 LOC: Config.onoff.digest_generation
7205 This controls whether the server will generate a Cache Digest
7206 of its contents. By default, Cache Digest generation is
7207 enabled if Squid is compiled with --enable-cache-digests defined.
7210 NAME: digest_bits_per_entry
7211 IFDEF: USE_CACHE_DIGESTS
7213 LOC: Config.digest.bits_per_entry
7216 This is the number of bits of the server's Cache Digest which
7217 will be associated with the Digest entry for a given HTTP
7218 Method and URL (public key) combination. The default is 5.
7221 NAME: digest_rebuild_period
7222 IFDEF: USE_CACHE_DIGESTS
7225 LOC: Config.digest.rebuild_period
7228 This is the wait time between Cache Digest rebuilds.
7231 NAME: digest_rewrite_period
7233 IFDEF: USE_CACHE_DIGESTS
7235 LOC: Config.digest.rewrite_period
7238 This is the wait time between Cache Digest writes to
7242 NAME: digest_swapout_chunk_size
7245 IFDEF: USE_CACHE_DIGESTS
7246 LOC: Config.digest.swapout_chunk_size
7249 This is the number of bytes of the Cache Digest to write to
7250 disk at a time. It defaults to 4096 bytes (4KB), the Squid
7254 NAME: digest_rebuild_chunk_percentage
7255 COMMENT: (percent, 0-100)
7256 IFDEF: USE_CACHE_DIGESTS
7258 LOC: Config.digest.rebuild_chunk_percentage
7261 This is the percentage of the Cache Digest to be scanned at a
7262 time. By default it is set to 10% of the Cache Digest.
7267 -----------------------------------------------------------------------------
7272 LOC: Config.Port.snmp
7274 DEFAULT_DOC: SNMP disabled.
7277 The port number where Squid listens for SNMP requests. To enable
7278 SNMP support set this to a suitable port number. Port number
7279 3401 is often used for the Squid SNMP agent. By default it's
7280 set to "0" (disabled)
7288 LOC: Config.accessList.snmp
7290 DEFAULT_DOC: Deny, unless rules exist in squid.conf.
7293 Allowing or denying access to the SNMP port.
7295 All access to the agent is denied by default.
7298 snmp_access allow|deny [!]aclname ...
7300 This clause only supports fast acl types.
7301 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
7304 snmp_access allow snmppublic localhost
7305 snmp_access deny all
7308 NAME: snmp_incoming_address
7310 LOC: Config.Addrs.snmp_incoming
7312 DEFAULT_DOC: Accept SNMP packets from all machine interfaces.
7315 Just like 'udp_incoming_address', but for the SNMP port.
7317 snmp_incoming_address is used for the SNMP socket receiving
7318 messages from SNMP agents.
7320 The default snmp_incoming_address is to listen on all
7321 available network interfaces.
7324 NAME: snmp_outgoing_address
7326 LOC: Config.Addrs.snmp_outgoing
7328 DEFAULT_DOC: Use snmp_incoming_address or an address selected by the operating system.
7331 Just like 'udp_outgoing_address', but for the SNMP port.
7333 snmp_outgoing_address is used for SNMP packets returned to SNMP
7336 If snmp_outgoing_address is not set it will use the same socket
7337 as snmp_incoming_address. Only change this if you want to have
7338 SNMP replies sent using another address than where this Squid
7339 listens for SNMP queries.
7341 NOTE, snmp_incoming_address and snmp_outgoing_address can not have
7342 the same value since they both use the same port.
7347 -----------------------------------------------------------------------------
7350 NAME: icp_port udp_port
7353 DEFAULT_DOC: ICP disabled.
7354 LOC: Config.Port.icp
7356 The port number where Squid sends and receives ICP queries to
7357 and from neighbor caches. The standard UDP port for ICP is 3130.
7360 icp_port @DEFAULT_ICP_PORT@
7367 DEFAULT_DOC: HTCP disabled.
7368 LOC: Config.Port.htcp
7370 The port number where Squid sends and receives HTCP queries to
7371 and from neighbor caches. To turn it on you want to set it to
7378 NAME: log_icp_queries
7382 LOC: Config.onoff.log_udp
7384 If set, ICP queries are logged to access.log. You may wish
7385 do disable this if your ICP load is VERY high to speed things
7386 up or to simplify log analysis.
7389 NAME: udp_incoming_address
7391 LOC:Config.Addrs.udp_incoming
7393 DEFAULT_DOC: Accept packets from all machine interfaces.
7395 udp_incoming_address is used for UDP packets received from other
7398 The default behavior is to not bind to any specific address.
7400 Only change this if you want to have all UDP queries received on
7401 a specific interface/address.
7403 NOTE: udp_incoming_address is used by the ICP, HTCP, and DNS
7404 modules. Altering it will affect all of them in the same manner.
7406 see also; udp_outgoing_address
7408 NOTE, udp_incoming_address and udp_outgoing_address can not
7409 have the same value since they both use the same port.
7412 NAME: udp_outgoing_address
7414 LOC: Config.Addrs.udp_outgoing
7416 DEFAULT_DOC: Use udp_incoming_address or an address selected by the operating system.
7418 udp_outgoing_address is used for UDP packets sent out to other
7421 The default behavior is to not bind to any specific address.
7423 Instead it will use the same socket as udp_incoming_address.
7424 Only change this if you want to have UDP queries sent using another
7425 address than where this Squid listens for UDP queries from other
7428 NOTE: udp_outgoing_address is used by the ICP, HTCP, and DNS
7429 modules. Altering it will affect all of them in the same manner.
7431 see also; udp_incoming_address
7433 NOTE, udp_incoming_address and udp_outgoing_address can not
7434 have the same value since they both use the same port.
7441 LOC: Config.onoff.icp_hit_stale
7443 If you want to return ICP_HIT for stale cache objects, set this
7444 option to 'on'. If you have sibling relationships with caches
7445 in other administrative domains, this should be 'off'. If you only
7446 have sibling relationships with caches under your control,
7447 it is probably okay to set this to 'on'.
7448 If set to 'on', your siblings should use the option "allow-miss"
7449 on their cache_peer lines for connecting to you.
7452 NAME: minimum_direct_hops
7455 LOC: Config.minDirectHops
7457 If using the ICMP pinging stuff, do direct fetches for sites
7458 which are no more than this many hops away.
7461 NAME: minimum_direct_rtt
7465 LOC: Config.minDirectRtt
7467 If using the ICMP pinging stuff, do direct fetches for sites
7468 which are no more than this many rtt milliseconds away.
7474 LOC: Config.Netdb.low
7476 The low water mark for the ICMP measurement database.
7478 Note: high watermark controlled by netdb_high directive.
7480 These watermarks are counts, not percents. The defaults are
7481 (low) 900 and (high) 1000. When the high water mark is
7482 reached, database entries will be deleted until the low
7489 LOC: Config.Netdb.high
7491 The high water mark for the ICMP measurement database.
7493 Note: low watermark controlled by netdb_low directive.
7495 These watermarks are counts, not percents. The defaults are
7496 (low) 900 and (high) 1000. When the high water mark is
7497 reached, database entries will be deleted until the low
7501 NAME: netdb_ping_period
7503 LOC: Config.Netdb.period
7506 The minimum period for measuring a site. There will be at
7507 least this much delay between successive pings to the same
7508 network. The default is five minutes.
7515 LOC: Config.onoff.query_icmp
7517 If you want to ask your peers to include ICMP data in their ICP
7518 replies, enable this option.
7520 If your peer has configured Squid (during compilation) with
7521 '--enable-icmp' that peer will send ICMP pings to origin server
7522 sites of the URLs it receives. If you enable this option the
7523 ICP replies from that peer will include the ICMP data (if available).
7524 Then, when choosing a parent cache, Squid will choose the parent with
7525 the minimal RTT to the origin server. When this happens, the
7526 hierarchy field of the access.log will be
7527 "CLOSEST_PARENT_MISS". This option is off by default.
7530 NAME: test_reachability
7534 LOC: Config.onoff.test_reachability
7536 When this is 'on', ICP MISS replies will be ICP_MISS_NOFETCH
7537 instead of ICP_MISS if the target host is NOT in the ICMP
7538 database, or has a zero RTT.
7541 NAME: icp_query_timeout
7544 DEFAULT_DOC: Dynamic detection.
7546 LOC: Config.Timeout.icp_query
7548 Normally Squid will automatically determine an optimal ICP
7549 query timeout value based on the round-trip-time of recent ICP
7550 queries. If you want to override the value determined by
7551 Squid, set this 'icp_query_timeout' to a non-zero value. This
7552 value is specified in MILLISECONDS, so, to use a 2-second
7553 timeout (the old default), you would write:
7555 icp_query_timeout 2000
7558 NAME: maximum_icp_query_timeout
7562 LOC: Config.Timeout.icp_query_max
7564 Normally the ICP query timeout is determined dynamically. But
7565 sometimes it can lead to very large values (say 5 seconds).
7566 Use this option to put an upper limit on the dynamic timeout
7567 value. Do NOT use this option to always use a fixed (instead
7568 of a dynamic) timeout value. To set a fixed timeout see the
7569 'icp_query_timeout' directive.
7572 NAME: minimum_icp_query_timeout
7576 LOC: Config.Timeout.icp_query_min
7578 Normally the ICP query timeout is determined dynamically. But
7579 sometimes it can lead to very small timeouts, even lower than
7580 the normal latency variance on your link due to traffic.
7581 Use this option to put an lower limit on the dynamic timeout
7582 value. Do NOT use this option to always use a fixed (instead
7583 of a dynamic) timeout value. To set a fixed timeout see the
7584 'icp_query_timeout' directive.
7587 NAME: background_ping_rate
7591 LOC: Config.backgroundPingRate
7593 Controls how often the ICP pings are sent to siblings that
7594 have background-ping set.
7598 MULTICAST ICP OPTIONS
7599 -----------------------------------------------------------------------------
7604 LOC: Config.mcast_group_list
7607 This tag specifies a list of multicast groups which your server
7608 should join to receive multicasted ICP queries.
7610 NOTE! Be very careful what you put here! Be sure you
7611 understand the difference between an ICP _query_ and an ICP
7612 _reply_. This option is to be set only if you want to RECEIVE
7613 multicast queries. Do NOT set this option to SEND multicast
7614 ICP (use cache_peer for that). ICP replies are always sent via
7615 unicast, so this option does not affect whether or not you will
7616 receive replies from multicast group members.
7618 You must be very careful to NOT use a multicast address which
7619 is already in use by another group of caches.
7621 If you are unsure about multicast, please read the Multicast
7622 chapter in the Squid FAQ (http://www.squid-cache.org/FAQ/).
7624 Usage: mcast_groups 239.128.16.128 224.0.1.20
7626 By default, Squid doesn't listen on any multicast groups.
7629 NAME: mcast_miss_addr
7630 IFDEF: MULTICAST_MISS_STREAM
7632 LOC: Config.mcast_miss.addr
7634 DEFAULT_DOC: disabled.
7636 If you enable this option, every "cache miss" URL will
7637 be sent out on the specified multicast address.
7639 Do not enable this option unless you are are absolutely
7640 certain you understand what you are doing.
7643 NAME: mcast_miss_ttl
7644 IFDEF: MULTICAST_MISS_STREAM
7646 LOC: Config.mcast_miss.ttl
7649 This is the time-to-live value for packets multicasted
7650 when multicasting off cache miss URLs is enabled. By
7651 default this is set to 'site scope', i.e. 16.
7654 NAME: mcast_miss_port
7655 IFDEF: MULTICAST_MISS_STREAM
7657 LOC: Config.mcast_miss.port
7660 This is the port number to be used in conjunction with
7664 NAME: mcast_miss_encode_key
7665 IFDEF: MULTICAST_MISS_STREAM
7667 LOC: Config.mcast_miss.encode_key
7668 DEFAULT: XXXXXXXXXXXXXXXX
7670 The URLs that are sent in the multicast miss stream are
7671 encrypted. This is the encryption key.
7674 NAME: mcast_icp_query_timeout
7678 LOC: Config.Timeout.mcast_icp_query
7680 For multicast peers, Squid regularly sends out ICP "probes" to
7681 count how many other peers are listening on the given multicast
7682 address. This value specifies how long Squid should wait to
7683 count all the replies. The default is 2000 msec, or 2
7688 INTERNAL ICON OPTIONS
7689 -----------------------------------------------------------------------------
7692 NAME: icon_directory
7694 LOC: Config.icons.directory
7695 DEFAULT: @DEFAULT_ICON_DIR@
7697 Where the icons are stored. These are normally kept in
7701 NAME: global_internal_static
7703 LOC: Config.onoff.global_internal_static
7706 This directive controls is Squid should intercept all requests for
7707 /squid-internal-static/ no matter which host the URL is requesting
7708 (default on setting), or if nothing special should be done for
7709 such URLs (off setting). The purpose of this directive is to make
7710 icons etc work better in complex cache hierarchies where it may
7711 not always be possible for all corners in the cache mesh to reach
7712 the server generating a directory listing.
7715 NAME: short_icon_urls
7717 LOC: Config.icons.use_short_names
7720 If this is enabled Squid will use short URLs for icons.
7721 If disabled it will revert to the old behavior of including
7722 it's own name and port in the URL.
7724 If you run a complex cache hierarchy with a mix of Squid and
7725 other proxies you may need to disable this directive.
7730 -----------------------------------------------------------------------------
7733 NAME: error_directory
7735 LOC: Config.errorDirectory
7737 DEFAULT_DOC: Send error pages in the clients preferred language
7739 If you wish to create your own versions of the default
7740 error files to customize them to suit your company copy
7741 the error/template files to another directory and point
7744 WARNING: This option will disable multi-language support
7745 on error pages if used.
7747 The squid developers are interested in making squid available in
7748 a wide variety of languages. If you are making translations for a
7749 language that Squid does not currently provide please consider
7750 contributing your translation back to the project.
7751 http://wiki.squid-cache.org/Translations
7753 The squid developers working on translations are happy to supply drop-in
7754 translated error files in exchange for any new language contributions.
7757 NAME: error_default_language
7758 IFDEF: USE_ERR_LOCALES
7760 LOC: Config.errorDefaultLanguage
7762 DEFAULT_DOC: Generate English language pages.
7764 Set the default language which squid will send error pages in
7765 if no existing translation matches the clients language
7768 If unset (default) generic English will be used.
7770 The squid developers are interested in making squid available in
7771 a wide variety of languages. If you are interested in making
7772 translations for any language see the squid wiki for details.
7773 http://wiki.squid-cache.org/Translations
7776 NAME: error_log_languages
7777 IFDEF: USE_ERR_LOCALES
7779 LOC: Config.errorLogMissingLanguages
7782 Log to cache.log what languages users are attempting to
7783 auto-negotiate for translations.
7785 Successful negotiations are not logged. Only failures
7786 have meaning to indicate that Squid may need an upgrade
7787 of its error page translations.
7790 NAME: err_page_stylesheet
7792 LOC: Config.errorStylesheet
7793 DEFAULT: @DEFAULT_CONFIG_DIR@/errorpage.css
7795 CSS Stylesheet to pattern the display of Squid default error pages.
7797 For information on CSS see http://www.w3.org/Style/CSS/
7802 LOC: Config.errHtmlText
7805 HTML text to include in error messages. Make this a "mailto"
7806 URL to your admin address, or maybe just a link to your
7807 organizations Web page.
7809 To include this in your error messages, you must rewrite
7810 the error template files (found in the "errors" directory).
7811 Wherever you want the 'err_html_text' line to appear,
7812 insert a %L tag in the error template file.
7815 NAME: email_err_data
7818 LOC: Config.onoff.emailErrData
7821 If enabled, information about the occurred error will be
7822 included in the mailto links of the ERR pages (if %W is set)
7823 so that the email body contains the data.
7824 Syntax is <A HREF="mailto:%w%W">%w</A>
7829 LOC: Config.denyInfoList
7832 Usage: deny_info err_page_name acl
7833 or deny_info http://... acl
7834 or deny_info TCP_RESET acl
7836 This can be used to return a ERR_ page for requests which
7837 do not pass the 'http_access' rules. Squid remembers the last
7838 acl it evaluated in http_access, and if a 'deny_info' line exists
7839 for that ACL Squid returns a corresponding error page.
7841 The acl is typically the last acl on the http_access deny line which
7842 denied access. The exceptions to this rule are:
7843 - When Squid needs to request authentication credentials. It's then
7844 the first authentication related acl encountered
7845 - When none of the http_access lines matches. It's then the last
7846 acl processed on the last http_access line.
7847 - When the decision to deny access was made by an adaptation service,
7848 the acl name is the corresponding eCAP or ICAP service_name.
7850 NP: If providing your own custom error pages with error_directory
7851 you may also specify them by your custom file name:
7852 Example: deny_info ERR_CUSTOM_ACCESS_DENIED bad_guys
7854 By defaut Squid will send "403 Forbidden". A different 4xx or 5xx
7855 may be specified by prefixing the file name with the code and a colon.
7856 e.g. 404:ERR_CUSTOM_ACCESS_DENIED
7858 Alternatively you can tell Squid to reset the TCP connection
7859 by specifying TCP_RESET.
7861 Or you can specify an error URL or URL pattern. The browsers will
7862 get redirected to the specified URL after formatting tags have
7863 been replaced. Redirect will be done with 302 or 307 according to
7864 HTTP/1.1 specs. A different 3xx code may be specified by prefixing
7865 the URL. e.g. 303:http://example.com/
7868 %a - username (if available. Password NOT included)
7871 %E - Error description
7873 %H - Request domain name
7874 %i - Client IP Address
7876 %o - Message result from external ACL helper
7877 %p - Request Port number
7878 %P - Request Protocol name
7879 %R - Request URL path
7880 %T - Timestamp in RFC 1123 format
7881 %U - Full canonical URL from client
7882 (HTTPS URLs terminate with *)
7883 %u - Full canonical URL from client
7884 %w - Admin email from squid.conf
7886 %% - Literal percent (%) code
7891 OPTIONS INFLUENCING REQUEST FORWARDING
7892 -----------------------------------------------------------------------------
7895 NAME: nonhierarchical_direct
7897 LOC: Config.onoff.nonhierarchical_direct
7900 By default, Squid will send any non-hierarchical requests
7901 (not cacheable request type) direct to origin servers.
7903 When this is set to "off", Squid will prefer to send these
7904 requests to parents.
7906 Note that in most configurations, by turning this off you will only
7907 add latency to these request without any improvement in global hit
7910 This option only sets a preference. If the parent is unavailable a
7911 direct connection to the origin server may still be attempted. To
7912 completely prevent direct connections use never_direct.
7917 LOC: Config.onoff.prefer_direct
7920 Normally Squid tries to use parents for most requests. If you for some
7921 reason like it to first try going direct and only use a parent if
7922 going direct fails set this to on.
7924 By combining nonhierarchical_direct off and prefer_direct on you
7925 can set up Squid to use a parent as a backup path if going direct
7928 Note: If you want Squid to use parents for all requests see
7929 the never_direct directive. prefer_direct only modifies how Squid
7930 acts on cacheable requests.
7933 NAME: cache_miss_revalidate
7937 LOC: Config.onoff.cache_miss_revalidate
7939 RFC 7232 defines a conditional request mechanism to prevent
7940 response objects being unnecessarily transferred over the network.
7941 If that mechanism is used by the client and a cache MISS occurs
7942 it can prevent new cache entries being created.
7944 This option determines whether Squid on cache MISS will pass the
7945 client revalidation request to the server or tries to fetch new
7946 content for caching. It can be useful while the cache is mostly
7947 empty to more quickly have the cache populated by generating
7948 non-conditional GETs.
7950 When set to 'on' (default), Squid will pass all client If-* headers
7951 to the server. This permits server responses without a cacheable
7952 payload to be delivered and on MISS no new cache entry is created.
7954 When set to 'off' and if the request is cacheable, Squid will
7955 remove the clients If-Modified-Since and If-None-Match headers from
7956 the request sent to the server. This requests a 200 status response
7957 from the server to create a new cache entry with.
7962 LOC: Config.accessList.AlwaysDirect
7964 DEFAULT_DOC: Prevent any cache_peer being used for this request.
7966 Usage: always_direct allow|deny [!]aclname ...
7968 Here you can use ACL elements to specify requests which should
7969 ALWAYS be forwarded by Squid to the origin servers without using
7970 any peers. For example, to always directly forward requests for
7971 local servers ignoring any parents or siblings you may have use
7974 acl local-servers dstdomain my.domain.net
7975 always_direct allow local-servers
7977 To always forward FTP requests directly, use
7980 always_direct allow FTP
7982 NOTE: There is a similar, but opposite option named
7983 'never_direct'. You need to be aware that "always_direct deny
7984 foo" is NOT the same thing as "never_direct allow foo". You
7985 may need to use a deny rule to exclude a more-specific case of
7986 some other rule. Example:
7988 acl local-external dstdomain external.foo.net
7989 acl local-servers dstdomain .foo.net
7990 always_direct deny local-external
7991 always_direct allow local-servers
7993 NOTE: If your goal is to make the client forward the request
7994 directly to the origin server bypassing Squid then this needs
7995 to be done in the client configuration. Squid configuration
7996 can only tell Squid how Squid should fetch the object.
7998 NOTE: This directive is not related to caching. The replies
7999 is cached as usual even if you use always_direct. To not cache
8000 the replies see the 'cache' directive.
8002 This clause supports both fast and slow acl types.
8003 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
8008 LOC: Config.accessList.NeverDirect
8010 DEFAULT_DOC: Allow DNS results to be used for this request.
8012 Usage: never_direct allow|deny [!]aclname ...
8014 never_direct is the opposite of always_direct. Please read
8015 the description for always_direct if you have not already.
8017 With 'never_direct' you can use ACL elements to specify
8018 requests which should NEVER be forwarded directly to origin
8019 servers. For example, to force the use of a proxy for all
8020 requests, except those in your local domain use something like:
8022 acl local-servers dstdomain .foo.net
8023 never_direct deny local-servers
8024 never_direct allow all
8026 or if Squid is inside a firewall and there are local intranet
8027 servers inside the firewall use something like:
8029 acl local-intranet dstdomain .foo.net
8030 acl local-external dstdomain external.foo.net
8031 always_direct deny local-external
8032 always_direct allow local-intranet
8033 never_direct allow all
8035 This clause supports both fast and slow acl types.
8036 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
8040 ADVANCED NETWORKING OPTIONS
8041 -----------------------------------------------------------------------------
8044 NAME: incoming_udp_average incoming_icp_average
8047 LOC: Config.comm_incoming.udp.average
8049 Heavy voodoo here. I can't even believe you are reading this.
8050 Are you crazy? Don't even think about adjusting these unless
8051 you understand the algorithms in comm_select.c first!
8054 NAME: incoming_tcp_average incoming_http_average
8057 LOC: Config.comm_incoming.tcp.average
8059 Heavy voodoo here. I can't even believe you are reading this.
8060 Are you crazy? Don't even think about adjusting these unless
8061 you understand the algorithms in comm_select.c first!
8064 NAME: incoming_dns_average
8067 LOC: Config.comm_incoming.dns.average
8069 Heavy voodoo here. I can't even believe you are reading this.
8070 Are you crazy? Don't even think about adjusting these unless
8071 you understand the algorithms in comm_select.c first!
8074 NAME: min_udp_poll_cnt min_icp_poll_cnt
8077 LOC: Config.comm_incoming.udp.min_poll
8079 Heavy voodoo here. I can't even believe you are reading this.
8080 Are you crazy? Don't even think about adjusting these unless
8081 you understand the algorithms in comm_select.c first!
8084 NAME: min_dns_poll_cnt
8087 LOC: Config.comm_incoming.dns.min_poll
8089 Heavy voodoo here. I can't even believe you are reading this.
8090 Are you crazy? Don't even think about adjusting these unless
8091 you understand the algorithms in comm_select.c first!
8094 NAME: min_tcp_poll_cnt min_http_poll_cnt
8097 LOC: Config.comm_incoming.tcp.min_poll
8099 Heavy voodoo here. I can't even believe you are reading this.
8100 Are you crazy? Don't even think about adjusting these unless
8101 you understand the algorithms in comm_select.c first!
8107 LOC: Config.accept_filter
8111 The name of an accept(2) filter to install on Squid's
8112 listen socket(s). This feature is perhaps specific to
8113 FreeBSD and requires support in the kernel.
8115 The 'httpready' filter delays delivering new connections
8116 to Squid until a full HTTP request has been received.
8117 See the accf_http(9) man page for details.
8119 The 'dataready' filter delays delivering new connections
8120 to Squid until there is some data to process.
8121 See the accf_dataready(9) man page for details.
8125 The 'data' filter delays delivering of new connections
8126 to Squid until there is some data to process by TCP_ACCEPT_DEFER.
8127 You may optionally specify a number of seconds to wait by
8128 'data=N' where N is the number of seconds. Defaults to 30
8129 if not specified. See the tcp(7) man page for details.
8132 accept_filter httpready
8137 NAME: client_ip_max_connections
8139 LOC: Config.client_ip_max_connections
8141 DEFAULT_DOC: No limit.
8143 Set an absolute limit on the number of connections a single
8144 client IP can use. Any more than this and Squid will begin to drop
8145 new connections from the client until it closes some links.
8147 Note that this is a global limit. It affects all HTTP, HTCP, Gopher and FTP
8148 connections from the client. For finer control use the ACL access controls.
8150 Requires client_db to be enabled (the default).
8152 WARNING: This may noticably slow down traffic received via external proxies
8153 or NAT devices and cause them to rebound error messages back to their clients.
8156 NAME: tcp_recv_bufsize
8160 DEFAULT_DOC: Use operating system TCP defaults.
8161 LOC: Config.tcpRcvBufsz
8163 Size of receive buffer to set for TCP sockets. Probably just
8164 as easy to change your kernel's default.
8165 Omit from squid.conf to use the default buffer size.
8170 -----------------------------------------------------------------------------
8177 LOC: Adaptation::Icap::TheConfig.onoff
8180 If you want to enable the ICAP module support, set this to on.
8183 NAME: icap_connect_timeout
8186 LOC: Adaptation::Icap::TheConfig.connect_timeout_raw
8189 This parameter specifies how long to wait for the TCP connect to
8190 the requested ICAP server to complete before giving up and either
8191 terminating the HTTP transaction or bypassing the failure.
8193 The default for optional services is peer_connect_timeout.
8194 The default for essential services is connect_timeout.
8195 If this option is explicitly set, its value applies to all services.
8198 NAME: icap_io_timeout
8202 DEFAULT_DOC: Use read_timeout.
8203 LOC: Adaptation::Icap::TheConfig.io_timeout_raw
8206 This parameter specifies how long to wait for an I/O activity on
8207 an established, active ICAP connection before giving up and
8208 either terminating the HTTP transaction or bypassing the
8212 NAME: icap_service_failure_limit
8213 COMMENT: limit [in memory-depth time-units]
8214 TYPE: icap_service_failure_limit
8216 LOC: Adaptation::Icap::TheConfig
8219 The limit specifies the number of failures that Squid tolerates
8220 when establishing a new TCP connection with an ICAP service. If
8221 the number of failures exceeds the limit, the ICAP service is
8222 not used for new ICAP requests until it is time to refresh its
8225 A negative value disables the limit. Without the limit, an ICAP
8226 service will not be considered down due to connectivity failures
8227 between ICAP OPTIONS requests.
8229 Squid forgets ICAP service failures older than the specified
8230 value of memory-depth. The memory fading algorithm
8231 is approximate because Squid does not remember individual
8232 errors but groups them instead, splitting the option
8233 value into ten time slots of equal length.
8235 When memory-depth is 0 and by default this option has no
8236 effect on service failure expiration.
8238 Squid always forgets failures when updating service settings
8239 using an ICAP OPTIONS transaction, regardless of this option
8243 # suspend service usage after 10 failures in 5 seconds:
8244 icap_service_failure_limit 10 in 5 seconds
8247 NAME: icap_service_revival_delay
8250 LOC: Adaptation::Icap::TheConfig.service_revival_delay
8253 The delay specifies the number of seconds to wait after an ICAP
8254 OPTIONS request failure before requesting the options again. The
8255 failed ICAP service is considered "down" until fresh OPTIONS are
8258 The actual delay cannot be smaller than the hardcoded minimum
8259 delay of 30 seconds.
8262 NAME: icap_preview_enable
8266 LOC: Adaptation::Icap::TheConfig.preview_enable
8269 The ICAP Preview feature allows the ICAP server to handle the
8270 HTTP message by looking only at the beginning of the message body
8271 or even without receiving the body at all. In some environments,
8272 previews greatly speedup ICAP processing.
8274 During an ICAP OPTIONS transaction, the server may tell Squid what
8275 HTTP messages should be previewed and how big the preview should be.
8276 Squid will not use Preview if the server did not request one.
8278 To disable ICAP Preview for all ICAP services, regardless of
8279 individual ICAP server OPTIONS responses, set this option to "off".
8281 icap_preview_enable off
8284 NAME: icap_preview_size
8287 LOC: Adaptation::Icap::TheConfig.preview_size
8289 DEFAULT_DOC: No preview sent.
8291 The default size of preview data to be sent to the ICAP server.
8292 This value might be overwritten on a per server basis by OPTIONS requests.
8295 NAME: icap_206_enable
8299 LOC: Adaptation::Icap::TheConfig.allow206_enable
8302 206 (Partial Content) responses is an ICAP extension that allows the
8303 ICAP agents to optionally combine adapted and original HTTP message
8304 content. The decision to combine is postponed until the end of the
8305 ICAP response. Squid supports Partial Content extension by default.
8307 Activation of the Partial Content extension is negotiated with each
8308 ICAP service during OPTIONS exchange. Most ICAP servers should handle
8309 negotation correctly even if they do not support the extension, but
8310 some might fail. To disable Partial Content support for all ICAP
8311 services and to avoid any negotiation, set this option to "off".
8317 NAME: icap_default_options_ttl
8320 LOC: Adaptation::Icap::TheConfig.default_options_ttl
8323 The default TTL value for ICAP OPTIONS responses that don't have
8324 an Options-TTL header.
8327 NAME: icap_persistent_connections
8331 LOC: Adaptation::Icap::TheConfig.reuse_connections
8334 Whether or not Squid should use persistent connections to
8338 NAME: adaptation_send_client_ip icap_send_client_ip
8340 IFDEF: USE_ADAPTATION
8342 LOC: Adaptation::Config::send_client_ip
8345 If enabled, Squid shares HTTP client IP information with adaptation
8346 services. For ICAP, Squid adds the X-Client-IP header to ICAP requests.
8347 For eCAP, Squid sets the libecap::metaClientIp transaction option.
8349 See also: adaptation_uses_indirect_client
8352 NAME: adaptation_send_username icap_send_client_username
8354 IFDEF: USE_ADAPTATION
8356 LOC: Adaptation::Config::send_username
8359 This sends authenticated HTTP client username (if available) to
8360 the adaptation service.
8362 For ICAP, the username value is encoded based on the
8363 icap_client_username_encode option and is sent using the header
8364 specified by the icap_client_username_header option.
8367 NAME: icap_client_username_header
8370 LOC: Adaptation::Icap::TheConfig.client_username_header
8371 DEFAULT: X-Client-Username
8373 ICAP request header name to use for adaptation_send_username.
8376 NAME: icap_client_username_encode
8380 LOC: Adaptation::Icap::TheConfig.client_username_encode
8383 Whether to base64 encode the authenticated client username.
8387 TYPE: icap_service_type
8389 LOC: Adaptation::Icap::TheConfig
8392 Defines a single ICAP service using the following format:
8394 icap_service id vectoring_point uri [option ...]
8397 an opaque identifier or name which is used to direct traffic to
8398 this specific service. Must be unique among all adaptation
8399 services in squid.conf.
8401 vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
8402 This specifies at which point of transaction processing the
8403 ICAP service should be activated. *_postcache vectoring points
8404 are not yet supported.
8406 uri: icap://servername:port/servicepath
8407 ICAP server and service location.
8408 icaps://servername:port/servicepath
8409 The "icap:" URI scheme is used for traditional ICAP server and
8410 service location (default port is 1344, connections are not
8411 encrypted). The "icaps:" URI scheme is for Secure ICAP
8412 services that use SSL/TLS-encrypted ICAP connections (by
8413 default, on port 11344).
8415 ICAP does not allow a single service to handle both REQMOD and RESPMOD
8416 transactions. Squid does not enforce that requirement. You can specify
8417 services with the same service_url and different vectoring_points. You
8418 can even specify multiple identical services as long as their
8419 service_names differ.
8421 To activate a service, use the adaptation_access directive. To group
8422 services, use adaptation_service_chain and adaptation_service_set.
8424 Service options are separated by white space. ICAP services support
8425 the following name=value options:
8428 If set to 'on' or '1', the ICAP service is treated as
8429 optional. If the service cannot be reached or malfunctions,
8430 Squid will try to ignore any errors and process the message as
8431 if the service was not enabled. No all ICAP errors can be
8432 bypassed. If set to 0, the ICAP service is treated as
8433 essential and all ICAP errors will result in an error page
8434 returned to the HTTP client.
8436 Bypass is off by default: services are treated as essential.
8439 If set to 'on' or '1', the ICAP service is allowed to
8440 dynamically change the current message adaptation plan by
8441 returning a chain of services to be used next. The services
8442 are specified using the X-Next-Services ICAP response header
8443 value, formatted as a comma-separated list of service names.
8444 Each named service should be configured in squid.conf. Other
8445 services are ignored. An empty X-Next-Services value results
8446 in an empty plan which ends the current adaptation.
8448 Dynamic adaptation plan may cross or cover multiple supported
8449 vectoring points in their natural processing order.
8451 Routing is not allowed by default: the ICAP X-Next-Services
8452 response header is ignored.
8455 Only has effect on split-stack systems. The default on those systems
8456 is to use IPv4-only connections. When set to 'on' this option will
8457 make Squid use IPv6-only connections to contact this ICAP service.
8459 on-overload=block|bypass|wait|force
8460 If the service Max-Connections limit has been reached, do
8461 one of the following for each new ICAP transaction:
8462 * block: send an HTTP error response to the client
8463 * bypass: ignore the "over-connected" ICAP service
8464 * wait: wait (in a FIFO queue) for an ICAP connection slot
8465 * force: proceed, ignoring the Max-Connections limit
8467 In SMP mode with N workers, each worker assumes the service
8468 connection limit is Max-Connections/N, even though not all
8469 workers may use a given service.
8471 The default value is "bypass" if service is bypassable,
8472 otherwise it is set to "wait".
8476 Use the given number as the Max-Connections limit, regardless
8477 of the Max-Connections value given by the service, if any.
8479 connection-encryption=on|off
8480 Determines the ICAP service effect on the connections_encrypted
8483 The default is "on" for Secure ICAP services (i.e., those
8484 with the icaps:// service URIs scheme) and "off" for plain ICAP
8487 Does not affect ICAP connections (e.g., does not turn Secure
8490 ==== ICAPS / TLS OPTIONS ====
8492 These options are used for Secure ICAP (icaps://....) services only.
8494 tls-cert=/path/to/ssl/certificate
8495 A client SSL certificate to use when connecting to
8498 tls-key=/path/to/ssl/key
8499 The private TLS/SSL key corresponding to sslcert above.
8500 If 'tls-key' is not specified 'tls-cert' is assumed to
8501 reference a combined PEM format file containing both the
8502 certificate and the key.
8504 tls-cipher=... The list of valid TLS/SSL ciphers to use when connecting
8505 to this icap server.
8508 The minimum TLS protocol version to permit. To control
8509 SSLv3 use the ssloptions= parameter.
8510 Supported Values: 1.0 (default), 1.1, 1.2
8512 tls-options=... Specify various OpenSSL library options:
8514 NO_SSLv3 Disallow the use of SSLv3
8516 NO_TLSv1 Disallow the use of TLSv1.0
8517 NO_TLSv1_1 Disallow the use of TLSv1.1
8518 NO_TLSv1_2 Disallow the use of TLSv1.2
8521 Always create a new key when using
8522 temporary/ephemeral DH key exchanges
8524 ALL Enable various bug workarounds
8525 suggested as "harmless" by OpenSSL
8526 Be warned that this reduces SSL/TLS
8527 strength to some attacks.
8529 See the OpenSSL SSL_CTX_set_options documentation for a
8530 more complete list. Options relevant only to SSLv2 are
8533 tls-cafile= PEM file containing CA certificates to use when verifying
8534 the icap server certificate.
8535 Use to specify intermediate CA certificate(s) if not sent
8536 by the server. Or the full CA chain for the server when
8537 using the tls-default-ca=off flag.
8538 May be repeated to load multiple files.
8540 tls-capath=... A directory containing additional CA certificates to
8541 use when verifying the icap server certificate.
8542 Requires OpenSSL or LibreSSL.
8544 tls-crlfile=... A certificate revocation list file to use when
8545 verifying the icap server certificate.
8547 tls-flags=... Specify various flags modifying the Squid TLS implementation:
8550 Accept certificates even if they fail to
8553 Don't verify the icap server certificate
8554 matches the server name
8556 tls-default-ca[=off]
8557 Whether to use the system Trusted CAs. Default is ON.
8559 tls-domain= The icap server name as advertised in it's certificate.
8560 Used for verifying the correctness of the received icap
8561 server certificate. If not specified the icap server
8562 hostname extracted from ICAP URI will be used.
8564 Older icap_service format without optional named parameters is
8565 deprecated but supported for backward compatibility.
8568 icap_service svcBlocker reqmod_precache icap://icap1.mydomain.net:1344/reqmod bypass=0
8569 icap_service svcLogger reqmod_precache icaps://icap2.mydomain.net:11344/reqmod routing=on
8573 TYPE: icap_class_type
8578 This deprecated option was documented to define an ICAP service
8579 chain, even though it actually defined a set of similar, redundant
8580 services, and the chains were not supported.
8582 To define a set of redundant services, please use the
8583 adaptation_service_set directive. For service chains, use
8584 adaptation_service_chain.
8588 TYPE: icap_access_type
8593 This option is deprecated. Please use adaptation_access, which
8594 has the same ICAP functionality, but comes with better
8595 documentation, and eCAP support.
8600 -----------------------------------------------------------------------------
8607 LOC: Adaptation::Ecap::TheConfig.onoff
8610 Controls whether eCAP support is enabled.
8614 TYPE: ecap_service_type
8616 LOC: Adaptation::Ecap::TheConfig
8619 Defines a single eCAP service
8621 ecap_service id vectoring_point uri [option ...]
8624 an opaque identifier or name which is used to direct traffic to
8625 this specific service. Must be unique among all adaptation
8626 services in squid.conf.
8628 vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
8629 This specifies at which point of transaction processing the
8630 eCAP service should be activated. *_postcache vectoring points
8631 are not yet supported.
8633 uri: ecap://vendor/service_name?custom&cgi=style¶meters=optional
8634 Squid uses the eCAP service URI to match this configuration
8635 line with one of the dynamically loaded services. Each loaded
8636 eCAP service must have a unique URI. Obtain the right URI from
8637 the service provider.
8639 To activate a service, use the adaptation_access directive. To group
8640 services, use adaptation_service_chain and adaptation_service_set.
8642 Service options are separated by white space. eCAP services support
8643 the following name=value options:
8646 If set to 'on' or '1', the eCAP service is treated as optional.
8647 If the service cannot be reached or malfunctions, Squid will try
8648 to ignore any errors and process the message as if the service
8649 was not enabled. No all eCAP errors can be bypassed.
8650 If set to 'off' or '0', the eCAP service is treated as essential
8651 and all eCAP errors will result in an error page returned to the
8654 Bypass is off by default: services are treated as essential.
8657 If set to 'on' or '1', the eCAP service is allowed to
8658 dynamically change the current message adaptation plan by
8659 returning a chain of services to be used next.
8661 Dynamic adaptation plan may cross or cover multiple supported
8662 vectoring points in their natural processing order.
8664 Routing is not allowed by default.
8666 connection-encryption=on|off
8667 Determines the eCAP service effect on the connections_encrypted
8670 Defaults to "on", which does not taint the master transaction
8673 Does not affect eCAP API calls.
8675 Older ecap_service format without optional named parameters is
8676 deprecated but supported for backward compatibility.
8680 ecap_service s1 reqmod_precache ecap://filters.R.us/leakDetector?on_error=block bypass=off
8681 ecap_service s2 respmod_precache ecap://filters.R.us/virusFilter config=/etc/vf.cfg bypass=on
8684 NAME: loadable_modules
8686 IFDEF: USE_LOADABLE_MODULES
8687 LOC: Config.loadable_module_names
8690 Instructs Squid to load the specified dynamic module(s) or activate
8691 preloaded module(s).
8693 loadable_modules @DEFAULT_PREFIX@/lib/MinimalAdapter.so
8697 MESSAGE ADAPTATION OPTIONS
8698 -----------------------------------------------------------------------------
8701 NAME: adaptation_service_set
8702 TYPE: adaptation_service_set_type
8703 IFDEF: USE_ADAPTATION
8708 Configures an ordered set of similar, redundant services. This is
8709 useful when hot standby or backup adaptation servers are available.
8711 adaptation_service_set set_name service_name1 service_name2 ...
8713 The named services are used in the set declaration order. The first
8714 applicable adaptation service from the set is used first. The next
8715 applicable service is tried if and only if the transaction with the
8716 previous service fails and the message waiting to be adapted is still
8719 When adaptation starts, broken services are ignored as if they were
8720 not a part of the set. A broken service is a down optional service.
8722 The services in a set must be attached to the same vectoring point
8723 (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
8725 If all services in a set are optional then adaptation failures are
8726 bypassable. If all services in the set are essential, then a
8727 transaction failure with one service may still be retried using
8728 another service from the set, but when all services fail, the master
8729 transaction fails as well.
8731 A set may contain a mix of optional and essential services, but that
8732 is likely to lead to surprising results because broken services become
8733 ignored (see above), making previously bypassable failures fatal.
8734 Technically, it is the bypassability of the last failed service that
8737 See also: adaptation_access adaptation_service_chain
8740 adaptation_service_set svcBlocker urlFilterPrimary urlFilterBackup
8741 adaptation service_set svcLogger loggerLocal loggerRemote
8744 NAME: adaptation_service_chain
8745 TYPE: adaptation_service_chain_type
8746 IFDEF: USE_ADAPTATION
8751 Configures a list of complementary services that will be applied
8752 one-by-one, forming an adaptation chain or pipeline. This is useful
8753 when Squid must perform different adaptations on the same message.
8755 adaptation_service_chain chain_name service_name1 svc_name2 ...
8757 The named services are used in the chain declaration order. The first
8758 applicable adaptation service from the chain is used first. The next
8759 applicable service is applied to the successful adaptation results of
8760 the previous service in the chain.
8762 When adaptation starts, broken services are ignored as if they were
8763 not a part of the chain. A broken service is a down optional service.
8765 Request satisfaction terminates the adaptation chain because Squid
8766 does not currently allow declaration of RESPMOD services at the
8767 "reqmod_precache" vectoring point (see icap_service or ecap_service).
8769 The services in a chain must be attached to the same vectoring point
8770 (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
8772 A chain may contain a mix of optional and essential services. If an
8773 essential adaptation fails (or the failure cannot be bypassed for
8774 other reasons), the master transaction fails. Otherwise, the failure
8775 is bypassed as if the failed adaptation service was not in the chain.
8777 See also: adaptation_access adaptation_service_set
8780 adaptation_service_chain svcRequest requestLogger urlFilter leakDetector
8783 NAME: adaptation_access
8784 TYPE: adaptation_access_type
8785 IFDEF: USE_ADAPTATION
8788 DEFAULT_DOC: Allow, unless rules exist in squid.conf.
8790 Sends an HTTP transaction to an ICAP or eCAP adaptation service.
8792 adaptation_access service_name allow|deny [!]aclname...
8793 adaptation_access set_name allow|deny [!]aclname...
8795 At each supported vectoring point, the adaptation_access
8796 statements are processed in the order they appear in this
8797 configuration file. Statements pointing to the following services
8798 are ignored (i.e., skipped without checking their ACL):
8800 - services serving different vectoring points
8801 - "broken-but-bypassable" services
8802 - "up" services configured to ignore such transactions
8803 (e.g., based on the ICAP Transfer-Ignore header).
8805 When a set_name is used, all services in the set are checked
8806 using the same rules, to find the first applicable one. See
8807 adaptation_service_set for details.
8809 If an access list is checked and there is a match, the
8810 processing stops: For an "allow" rule, the corresponding
8811 adaptation service is used for the transaction. For a "deny"
8812 rule, no adaptation service is activated.
8814 It is currently not possible to apply more than one adaptation
8815 service at the same vectoring point to the same HTTP transaction.
8817 See also: icap_service and ecap_service
8820 adaptation_access service_1 allow all
8823 NAME: adaptation_service_iteration_limit
8825 IFDEF: USE_ADAPTATION
8826 LOC: Adaptation::Config::service_iteration_limit
8829 Limits the number of iterations allowed when applying adaptation
8830 services to a message. If your longest adaptation set or chain
8831 may have more than 16 services, increase the limit beyond its
8832 default value of 16. If detecting infinite iteration loops sooner
8833 is critical, make the iteration limit match the actual number
8834 of services in your longest adaptation set or chain.
8836 Infinite adaptation loops are most likely with routing services.
8838 See also: icap_service routing=1
8841 NAME: adaptation_masterx_shared_names
8843 IFDEF: USE_ADAPTATION
8844 LOC: Adaptation::Config::masterx_shared_name
8847 For each master transaction (i.e., the HTTP request and response
8848 sequence, including all related ICAP and eCAP exchanges), Squid
8849 maintains a table of metadata. The table entries are (name, value)
8850 pairs shared among eCAP and ICAP exchanges. The table is destroyed
8851 with the master transaction.
8853 This option specifies the table entry names that Squid must accept
8854 from and forward to the adaptation transactions.
8856 An ICAP REQMOD or RESPMOD transaction may set an entry in the
8857 shared table by returning an ICAP header field with a name
8858 specified in adaptation_masterx_shared_names.
8860 An eCAP REQMOD or RESPMOD transaction may set an entry in the
8861 shared table by implementing the libecap::visitEachOption() API
8862 to provide an option with a name specified in
8863 adaptation_masterx_shared_names.
8865 Squid will store and forward the set entry to subsequent adaptation
8866 transactions within the same master transaction scope.
8868 Only one shared entry name is supported at this time.
8871 # share authentication information among ICAP services
8872 adaptation_masterx_shared_names X-Subscriber-ID
8875 NAME: adaptation_meta
8877 IFDEF: USE_ADAPTATION
8878 LOC: Adaptation::Config::metaHeaders
8881 This option allows Squid administrator to add custom ICAP request
8882 headers or eCAP options to Squid ICAP requests or eCAP transactions.
8883 Use it to pass custom authentication tokens and other
8884 transaction-state related meta information to an ICAP/eCAP service.
8886 The addition of a meta header is ACL-driven:
8887 adaptation_meta name value [!]aclname ...
8889 Processing for a given header name stops after the first ACL list match.
8890 Thus, it is impossible to add two headers with the same name. If no ACL
8891 lists match for a given header name, no such header is added. For
8894 # do not debug transactions except for those that need debugging
8895 adaptation_meta X-Debug 1 needs_debugging
8897 # log all transactions except for those that must remain secret
8898 adaptation_meta X-Log 1 !keep_secret
8900 # mark transactions from users in the "G 1" group
8901 adaptation_meta X-Authenticated-Groups "G 1" authed_as_G1
8903 The "value" parameter may be a regular squid.conf token or a "double
8904 quoted string". Within the quoted string, use backslash (\) to escape
8905 any character, which is currently only useful for escaping backslashes
8906 and double quotes. For example,
8907 "this string has one backslash (\\) and two \"quotes\""
8909 Used adaptation_meta header values may be logged via %note
8910 logformat code. If multiple adaptation_meta headers with the same name
8911 are used during master transaction lifetime, the header values are
8912 logged in the order they were used and duplicate values are ignored
8913 (only the first repeated value will be logged).
8919 LOC: Adaptation::Icap::TheConfig.repeat
8920 DEFAULT_IF_NONE: deny all
8922 This ACL determines which retriable ICAP transactions are
8923 retried. Transactions that received a complete ICAP response
8924 and did not have to consume or produce HTTP bodies to receive
8925 that response are usually retriable.
8927 icap_retry allow|deny [!]aclname ...
8929 Squid automatically retries some ICAP I/O timeouts and errors
8930 due to persistent connection race conditions.
8932 See also: icap_retry_limit
8935 NAME: icap_retry_limit
8938 LOC: Adaptation::Icap::TheConfig.repeat_limit
8940 DEFAULT_DOC: No retries are allowed.
8942 Limits the number of retries allowed.
8944 Communication errors due to persistent connection race
8945 conditions are unavoidable, automatically retried, and do not
8946 count against this limit.
8948 See also: icap_retry
8954 -----------------------------------------------------------------------------
8957 NAME: check_hostnames
8960 LOC: Config.onoff.check_hostnames
8962 For security and stability reasons Squid can check
8963 hostnames for Internet standard RFC compliance. If you want
8964 Squid to perform these checks turn this directive on.
8967 NAME: allow_underscore
8970 LOC: Config.onoff.allow_underscore
8972 Underscore characters is not strictly allowed in Internet hostnames
8973 but nevertheless used by many sites. Set this to off if you want
8974 Squid to be strict about the standard.
8975 This check is performed only when check_hostnames is set to on.
8978 NAME: dns_retransmit_interval
8981 LOC: Config.Timeout.idns_retransmit
8983 Initial retransmit interval for DNS queries. The interval is
8984 doubled each time all configured DNS servers have been tried.
8990 LOC: Config.Timeout.idns_query
8992 DNS Query timeout. If no response is received to a DNS query
8993 within this time all DNS servers for the queried domain
8994 are assumed to be unavailable.
8997 NAME: dns_packet_max
8999 DEFAULT_DOC: EDNS disabled
9001 LOC: Config.dns.packet_max
9003 Maximum number of bytes packet size to advertise via EDNS.
9004 Set to "none" to disable EDNS large packet support.
9006 For legacy reasons DNS UDP replies will default to 512 bytes which
9007 is too small for many responses. EDNS provides a means for Squid to
9008 negotiate receiving larger responses back immediately without having
9009 to failover with repeat requests. Responses larger than this limit
9010 will retain the old behaviour of failover to TCP DNS.
9012 Squid has no real fixed limit internally, but allowing packet sizes
9013 over 1500 bytes requires network jumbogram support and is usually not
9016 WARNING: The RFC also indicates that some older resolvers will reply
9017 with failure of the whole request if the extension is added. Some
9018 resolvers have already been identified which will reply with mangled
9019 EDNS response on occasion. Usually in response to many-KB jumbogram
9020 sizes being advertised by Squid.
9021 Squid will currently treat these both as an unable-to-resolve domain
9022 even if it would be resolvable without EDNS.
9029 DEFAULT_DOC: Search for single-label domain names is disabled.
9030 LOC: Config.onoff.res_defnames
9032 Normally the RES_DEFNAMES resolver option is disabled
9033 (see res_init(3)). This prevents caches in a hierarchy
9034 from interpreting single-component hostnames locally. To allow
9035 Squid to handle single-component names, enable this option.
9038 NAME: dns_multicast_local
9042 DEFAULT_DOC: Search for .local and .arpa names is disabled.
9043 LOC: Config.onoff.dns_mdns
9045 When set to on, Squid sends multicast DNS lookups on the local
9046 network for domains ending in .local and .arpa.
9047 This enables local servers and devices to be contacted in an
9048 ad-hoc or zero-configuration network environment.
9051 NAME: dns_nameservers
9054 DEFAULT_DOC: Use operating system definitions
9055 LOC: Config.dns_nameservers
9057 Use this if you want to specify a list of DNS name servers
9058 (IP addresses) to use instead of those given in your
9059 /etc/resolv.conf file.
9061 On Windows platforms, if no value is specified here or in
9062 the /etc/resolv.conf file, the list of DNS name servers are
9063 taken from the Windows registry, both static and dynamic DHCP
9064 configurations are supported.
9066 Example: dns_nameservers 10.0.0.1 192.172.0.4
9071 DEFAULT: @DEFAULT_HOSTS@
9072 LOC: Config.etcHostsPath
9074 Location of the host-local IP name-address associations
9075 database. Most Operating Systems have such a file on different
9077 - Un*X & Linux: /etc/hosts
9078 - Windows NT/2000: %SystemRoot%\system32\drivers\etc\hosts
9079 (%SystemRoot% value install default is c:\winnt)
9080 - Windows XP/2003: %SystemRoot%\system32\drivers\etc\hosts
9081 (%SystemRoot% value install default is c:\windows)
9082 - Windows 9x/Me: %windir%\hosts
9083 (%windir% value is usually c:\windows)
9084 - Cygwin: /etc/hosts
9086 The file contains newline-separated definitions, in the
9087 form ip_address_in_dotted_form name [name ...] names are
9088 whitespace-separated. Lines beginning with an hash (#)
9089 character are comments.
9091 The file is checked at startup and upon configuration.
9092 If set to 'none', it won't be checked.
9093 If append_domain is used, that domain will be added to
9094 domain-local (i.e. not containing any dot character) host
9100 LOC: Config.appendDomain
9102 DEFAULT_DOC: Use operating system definitions
9104 Appends local domain name to hostnames without any dots in
9105 them. append_domain must begin with a period.
9107 Be warned there are now Internet names with no dots in
9108 them using only top-domain names, so setting this may
9109 cause some Internet sites to become unavailable.
9112 append_domain .yourdomain.com
9115 NAME: ignore_unknown_nameservers
9117 LOC: Config.onoff.ignore_unknown_nameservers
9120 By default Squid checks that DNS responses are received
9121 from the same IP addresses they are sent to. If they
9122 don't match, Squid ignores the response and writes a warning
9123 message to cache.log. You can allow responses from unknown
9124 nameservers by setting this option to 'off'.
9130 LOC: Config.dns.v4_first
9132 With the IPv6 Internet being as fast or faster than IPv4 Internet
9133 for most networks Squid prefers to contact websites over IPv6.
9135 This option reverses the order of preference to make Squid contact
9136 dual-stack websites over IPv4 first. Squid will still perform both
9137 IPv6 and IPv4 DNS lookups before connecting.
9140 This option will restrict the situations under which IPv6
9141 connectivity is used (and tested). Hiding network problems
9142 which would otherwise be detected and warned about.
9146 COMMENT: (number of entries)
9149 LOC: Config.ipcache.size
9151 Maximum number of DNS IP cache entries.
9158 LOC: Config.ipcache.low
9165 LOC: Config.ipcache.high
9167 The size, low-, and high-water marks for the IP cache.
9170 NAME: fqdncache_size
9171 COMMENT: (number of entries)
9174 LOC: Config.fqdncache.size
9176 Maximum number of FQDN cache entries.
9181 -----------------------------------------------------------------------------
9184 NAME: configuration_includes_quoted_values
9186 TYPE: configuration_includes_quoted_values
9188 LOC: ConfigParser::RecognizeQuotedValues
9190 If set, Squid will recognize each "quoted string" after a configuration
9191 directive as a single parameter. The quotes are stripped before the
9192 parameter value is interpreted or used.
9193 See "Values with spaces, quotes, and other special characters"
9194 section for more details.
9201 LOC: Config.onoff.mem_pools
9203 If set, Squid will keep pools of allocated (but unused) memory
9204 available for future use. If memory is a premium on your
9205 system and you believe your malloc library outperforms Squid
9206 routines, disable this.
9209 NAME: memory_pools_limit
9213 LOC: Config.MemPools.limit
9215 Used only with memory_pools on:
9216 memory_pools_limit 50 MB
9218 If set to a non-zero value, Squid will keep at most the specified
9219 limit of allocated (but unused) memory in memory pools. All free()
9220 requests that exceed this limit will be handled by your malloc
9221 library. Squid does not pre-allocate any memory, just safe-keeps
9222 objects that otherwise would be free()d. Thus, it is safe to set
9223 memory_pools_limit to a reasonably high value even if your
9224 configuration will use less memory.
9226 If set to none, Squid will keep all memory it can. That is, there
9227 will be no limit on the total amount of memory used for safe-keeping.
9229 To disable memory allocation optimization, do not set
9230 memory_pools_limit to 0 or none. Set memory_pools to "off" instead.
9232 An overhead for maintaining memory pools is not taken into account
9233 when the limit is checked. This overhead is close to four bytes per
9234 object kept. However, pools may actually _save_ memory because of
9235 reduced memory thrashing in your malloc library.
9239 COMMENT: on|off|transparent|truncate|delete
9242 LOC: opt_forwarded_for
9244 If set to "on", Squid will append your client's IP address
9245 in the HTTP requests it forwards. By default it looks like:
9247 X-Forwarded-For: 192.1.2.3
9249 If set to "off", it will appear as
9251 X-Forwarded-For: unknown
9253 If set to "transparent", Squid will not alter the
9254 X-Forwarded-For header in any way.
9256 If set to "delete", Squid will delete the entire
9257 X-Forwarded-For header.
9259 If set to "truncate", Squid will remove all existing
9260 X-Forwarded-For entries, and place the client IP as the sole entry.
9263 NAME: cachemgr_passwd
9264 TYPE: cachemgrpasswd
9266 DEFAULT_DOC: No password. Actions which require password are denied.
9267 LOC: Config.passwd_list
9269 Specify passwords for cachemgr operations.
9271 Usage: cachemgr_passwd password action action ...
9273 Some valid actions are (see cache manager menu for a full list):
9313 * Indicates actions which will not be performed without a
9314 valid password, others can be performed if not listed here.
9316 To disable an action, set the password to "disable".
9317 To allow performing an action without a password, set the
9320 Use the keyword "all" to set the same password for all actions.
9323 cachemgr_passwd secret shutdown
9324 cachemgr_passwd lesssssssecret info stats/objects
9325 cachemgr_passwd disable all
9332 LOC: Config.onoff.client_db
9334 If you want to disable collecting per-client statistics,
9335 turn off client_db here.
9338 NAME: refresh_all_ims
9342 LOC: Config.onoff.refresh_all_ims
9344 When you enable this option, squid will always check
9345 the origin server for an update when a client sends an
9346 If-Modified-Since request. Many browsers use IMS
9347 requests when the user requests a reload, and this
9348 ensures those clients receive the latest version.
9350 By default (off), squid may return a Not Modified response
9351 based on the age of the cached version.
9354 NAME: reload_into_ims
9355 IFDEF: USE_HTTP_VIOLATIONS
9359 LOC: Config.onoff.reload_into_ims
9361 When you enable this option, client no-cache or ``reload''
9362 requests will be changed to If-Modified-Since requests.
9363 Doing this VIOLATES the HTTP standard. Enabling this
9364 feature could make you liable for problems which it
9367 see also refresh_pattern for a more selective approach.
9370 NAME: connect_retries
9372 LOC: Config.connect_retries
9374 DEFAULT_DOC: Do not retry failed connections.
9376 This sets the maximum number of connection attempts made for each
9377 TCP connection. The connect_retries attempts must all still
9378 complete within the connection timeout period.
9380 The default is not to re-try if the first connection attempt fails.
9381 The (not recommended) maximum is 10 tries.
9383 A warning message will be generated if it is set to a too-high
9384 value and the configured value will be over-ridden.
9386 Note: These re-tries are in addition to forward_max_tries
9387 which limit how many different addresses may be tried to find
9391 NAME: retry_on_error
9393 LOC: Config.retry.onerror
9396 If set to ON Squid will automatically retry requests when
9397 receiving an error response with status 403 (Forbidden),
9398 500 (Internal Error), 501 or 503 (Service not available).
9399 Status 502 and 504 (Gateway errors) are always retried.
9401 This is mainly useful if you are in a complex cache hierarchy to
9402 work around access control errors.
9404 NOTE: This retry will attempt to find another working destination.
9405 Which is different from the server which just failed.
9408 NAME: as_whois_server
9410 LOC: Config.as_whois_server
9411 DEFAULT: whois.ra.net
9413 WHOIS server to query for AS numbers. NOTE: AS numbers are
9414 queried only when Squid starts up, not for every request.
9419 LOC: Config.onoff.offline
9422 Enable this option and Squid will never try to validate cached
9426 NAME: uri_whitespace
9427 TYPE: uri_whitespace
9428 LOC: Config.uri_whitespace
9431 What to do with requests that have whitespace characters in the
9434 strip: The whitespace characters are stripped out of the URL.
9435 This is the behavior recommended by RFC2396 and RFC3986
9436 for tolerant handling of generic URI.
9437 NOTE: This is one difference between generic URI and HTTP URLs.
9439 deny: The request is denied. The user receives an "Invalid
9441 This is the behaviour recommended by RFC2616 for safe
9442 handling of HTTP request URL.
9444 allow: The request is allowed and the URI is not changed. The
9445 whitespace characters remain in the URI. Note the
9446 whitespace is passed to redirector processes if they
9448 Note this may be considered a violation of RFC2616
9449 request parsing where whitespace is prohibited in the
9452 encode: The request is allowed and the whitespace characters are
9453 encoded according to RFC1738.
9455 chop: The request is allowed and the URI is chopped at the
9459 NOTE the current Squid implementation of encode and chop violates
9460 RFC2616 by not using a 301 redirect after altering the URL.
9465 LOC: Config.chroot_dir
9468 Specifies a directory where Squid should do a chroot() while
9469 initializing. This also causes Squid to fully drop root
9470 privileges after initializing. This means, for example, if you
9471 use a HTTP port less than 1024 and try to reconfigure, you may
9472 get an error saying that Squid can not open the port.
9475 NAME: balance_on_multiple_ip
9477 LOC: Config.onoff.balance_on_multiple_ip
9480 Modern IP resolvers in squid sort lookup results by preferred access.
9481 By default squid will use these IP in order and only rotates to
9482 the next listed when the most preffered fails.
9484 Some load balancing servers based on round robin DNS have been
9485 found not to preserve user session state across requests
9486 to different IP addresses.
9488 Enabling this directive Squid rotates IP's per request.
9491 NAME: pipeline_prefetch
9492 TYPE: pipelinePrefetch
9493 LOC: Config.pipeline_max_prefetch
9495 DEFAULT_DOC: Do not pre-parse pipelined requests.
9497 HTTP clients may send a pipeline of 1+N requests to Squid using a
9498 single connection, without waiting for Squid to respond to the first
9499 of those requests. This option limits the number of concurrent
9500 requests Squid will try to handle in parallel. If set to N, Squid
9501 will try to receive and process up to 1+N requests on the same
9502 connection concurrently.
9504 Defaults to 0 (off) for bandwidth management and access logging
9507 NOTE: pipelining requires persistent connections to clients.
9509 WARNING: pipelining breaks NTLM and Negotiate/Kerberos authentication.
9512 NAME: high_response_time_warning
9515 LOC: Config.warnings.high_rptm
9517 DEFAULT_DOC: disabled.
9519 If the one-minute median response time exceeds this value,
9520 Squid prints a WARNING with debug level 0 to get the
9521 administrators attention. The value is in milliseconds.
9524 NAME: high_page_fault_warning
9526 LOC: Config.warnings.high_pf
9528 DEFAULT_DOC: disabled.
9530 If the one-minute average page fault rate exceeds this
9531 value, Squid prints a WARNING with debug level 0 to get
9532 the administrators attention. The value is in page faults
9536 NAME: high_memory_warning
9538 LOC: Config.warnings.high_memory
9539 IFDEF: HAVE_MSTATS&&HAVE_GNUMALLOC_H
9541 DEFAULT_DOC: disabled.
9543 If the memory usage (as determined by gnumalloc, if available and used)
9544 exceeds this amount, Squid prints a WARNING with debug level 0 to get
9545 the administrators attention.
9547 # TODO: link high_memory_warning to mempools?
9549 NAME: sleep_after_fork
9550 COMMENT: (microseconds)
9552 LOC: Config.sleep_after_fork
9555 When this is set to a non-zero value, the main Squid process
9556 sleeps the specified number of microseconds after a fork()
9557 system call. This sleep may help the situation where your
9558 system reports fork() failures due to lack of (virtual)
9559 memory. Note, however, if you have a lot of child
9560 processes, these sleep delays will add up and your
9561 Squid will not service requests for some amount of time
9562 until all the child processes have been started.
9563 On Windows value less then 1000 (1 milliseconds) are
9567 NAME: windows_ipaddrchangemonitor
9568 IFDEF: _SQUID_WINDOWS_
9572 LOC: Config.onoff.WIN32_IpAddrChangeMonitor
9574 On Windows Squid by default will monitor IP address changes and will
9575 reconfigure itself after any detected event. This is very useful for
9576 proxies connected to internet with dial-up interfaces.
9577 In some cases (a Proxy server acting as VPN gateway is one) it could be
9578 desiderable to disable this behaviour setting this to 'off'.
9579 Note: after changing this, Squid service must be restarted.
9584 IFDEF: USE_SQUID_EUI
9586 LOC: Eui::TheConfig.euiLookup
9588 Whether to lookup the EUI or MAC address of a connected client.
9591 NAME: max_filedescriptors max_filedesc
9594 DEFAULT_DOC: Use operating system limits set by ulimit.
9595 LOC: Config.max_filedescriptors
9597 Reduce the maximum number of filedescriptors supported below
9598 the usual operating system defaults.
9600 Remove from squid.conf to inherit the current ulimit setting.
9602 Note: Changing this requires a restart of Squid. Also
9603 not all I/O types supports large values (eg on Windows).
9606 NAME: force_request_body_continuation
9608 LOC: Config.accessList.forceRequestBodyContinuation
9610 DEFAULT_DOC: Deny, unless rules exist in squid.conf.
9612 This option controls how Squid handles data upload requests from HTTP
9613 and FTP agents that require a "Please Continue" control message response
9614 to actually send the request body to Squid. It is mostly useful in
9615 adaptation environments.
9617 When Squid receives an HTTP request with an "Expect: 100-continue"
9618 header or an FTP upload command (e.g., STOR), Squid normally sends the
9619 request headers or FTP command information to an adaptation service (or
9620 peer) and waits for a response. Most adaptation services (and some
9621 broken peers) may not respond to Squid at that stage because they may
9622 decide to wait for the HTTP request body or FTP data transfer. However,
9623 that request body or data transfer may never come because Squid has not
9624 responded with the HTTP 100 or FTP 150 (Please Continue) control message
9625 to the request sender yet!
9627 An allow match tells Squid to respond with the HTTP 100 or FTP 150
9628 (Please Continue) control message on its own, before forwarding the
9629 request to an adaptation service or peer. Such a response usually forces
9630 the request sender to proceed with sending the body. A deny match tells
9631 Squid to delay that control response until the origin server confirms
9632 that the request body is needed. Delaying is the default behavior.
9635 NAME: server_pconn_for_nonretriable
9638 DEFAULT_DOC: Open new connections for forwarding requests Squid cannot retry safely.
9639 LOC: Config.accessList.serverPconnForNonretriable
9641 This option provides fine-grained control over persistent connection
9642 reuse when forwarding HTTP requests that Squid cannot retry. It is useful
9643 in environments where opening new connections is very expensive
9644 (e.g., all connections are secured with TLS with complex client and server
9645 certificate validation) and race conditions associated with persistent
9646 connections are very rare and/or only cause minor problems.
9648 HTTP prohibits retrying unsafe and non-idempotent requests (e.g., POST).
9649 Squid limitations also prohibit retrying all requests with bodies (e.g., PUT).
9650 By default, when forwarding such "risky" requests, Squid opens a new
9651 connection to the server or cache_peer, even if there is an idle persistent
9652 connection available. When Squid is configured to risk sending a non-retriable
9653 request on a previously used persistent connection, and the server closes
9654 the connection before seeing that risky request, the user gets an error response
9655 from Squid. In most cases, that error response will be HTTP 502 (Bad Gateway)
9656 with ERR_ZERO_SIZE_OBJECT or ERR_WRITE_ERROR (peer connection reset) error detail.
9658 If an allow rule matches, Squid reuses an available idle persistent connection
9659 (if any) for the request that Squid cannot retry. If a deny rule matches, then
9660 Squid opens a new connection for the request that Squid cannot retry.
9662 This option does not affect requests that Squid can retry. They will reuse idle
9663 persistent connections (if any).
9665 This clause only supports fast acl types.
9666 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
9669 acl SpeedIsWorthTheRisk method POST
9670 server_pconn_for_nonretriable allow SpeedIsWorthTheRisk
projects / thirdparty / squid.git / blob