2 # SQUID Web Proxy Cache http://www.squid-cache.org/
3 # ----------------------------------------------------------
5 # Squid is the result of efforts by numerous individuals from
6 # the Internet community; see the CONTRIBUTORS file for full
7 # details. Many organizations have provided support for Squid's
8 # development; see the SPONSORS file for full details. Squid is
9 # Copyrighted (C) 2000 by the Regents of the University of
10 # California; see the COPYRIGHT file for full details. Squid
11 # incorporates software developed and/or copyrighted by other
12 # sources; see the CREDITS file for full details.
14 # This program is free software; you can redistribute it and/or modify
15 # it under the terms of the GNU General Public License as published by
16 # the Free Software Foundation; either version 2 of the License, or
17 # (at your option) any later version.
19 # This program is distributed in the hope that it will be useful,
20 # but WITHOUT ANY WARRANTY; without even the implied warranty of
21 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
22 # GNU General Public License for more details.
24 # You should have received a copy of the GNU General Public License
25 # along with this program; if not, write to the Free Software
26 # Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111, USA.
31 ----------------------------
33 This is the default Squid configuration file. You may wish
34 to look at the Squid home page (http://www.squid-cache.org/)
35 for the FAQ and other documentation.
37 The default Squid config file shows what the defaults for
38 various options happen to be. If you don't need to change the
39 default, you shouldn't uncomment the line. Doing so may cause
40 run-time problems. In some cases "none" refers to no default
41 setting at all, while in other cases it refers to a valid
42 option - the comments for that keyword indicate if this is the
48 Configuration options can be included using the "include" directive.
49 Include takes a list of files to include. Quoting and wildcards is
54 include /path/to/included/file/squid.acl.config
56 Includes can be nested up to a hard-coded depth of 16 levels.
57 This arbitrary restriction is to prevent recursive include references
58 from causing Squid entering an infinite loop whilst trying to load
62 Conditional configuration
64 If-statements can be used to make configuration directives
68 ... regular configuration directives ...
70 ... regular configuration directives ...]
73 The else part is optional. The keywords "if", "else", and "endif"
74 must be typed on their own lines, as if they were regular
75 configuration directives.
77 These individual conditions types are supported:
80 Always evaluates to true.
82 Always evaluates to false.
84 Equality comparison of two integer numbers.
89 The following SMP-related preprocessor macros can be used.
91 ${process_name} expands to the current Squid process "name"
92 (e.g., squid1, squid2, or cache1).
94 ${process_number} expands to the current Squid process
95 identifier, which is an integer number (e.g., 1, 2, 3) unique
96 across all Squid processes.
99 # Options Removed in 3.2
100 NAME: ignore_expect_100
103 Remove this line. The HTTP/1.1 feature is now fully supported by default.
109 Remove this line. Configure FTP page display using the CSS controls in errorpages.css instead.
112 NAME: url_rewrite_concurrency
115 Remove this line. Set the 'concurrency=' option of url_rewrite_children instead.
118 # Options Removed in 3.1
122 Remove this line. DNS is no longer tested on startup.
125 NAME: extension_methods
128 Remove this line. All valid methods for HTTP are accepted by default.
131 # 2.7 Options Removed/Replaced in 3.1
139 Remove this line. HTTP/1.1 is supported by default.
142 NAME: upgrade_http0.9
145 Remove this line. ICY/1.0 streaming protocol is supported by default.
148 NAME: zph_local zph_mode zph_option zph_parent zph_sibling
151 Alter these entries. Use the qos_flows directive instead.
154 # Options Removed in 3.0
158 Since squid-3.0 replace with request_header_access or reply_header_access
159 depending on whether you wish to match client requests or server replies.
162 NAME: httpd_accel_no_pmtu_disc
165 Since squid-3.0 use the 'disable-pmtu-discovery' flag on http_port instead.
169 OPTIONS FOR AUTHENTICATION
170 -----------------------------------------------------------------------------
178 This is used to define parameters for the various authentication
179 schemes supported by Squid.
181 format: auth_param scheme parameter [setting]
183 The order in which authentication schemes are presented to the client is
184 dependent on the order the scheme first appears in config file. IE
185 has a bug (it's not RFC 2617 compliant) in that it will use the basic
186 scheme if basic is the first entry presented, even if more secure
187 schemes are presented. For now use the order in the recommended
188 settings section below. If other browsers have difficulties (don't
189 recognize the schemes offered even if you are using basic) either
190 put basic first, or disable the other schemes (by commenting out their
193 Once an authentication scheme is fully configured, it can only be
194 shutdown by shutting squid down and restarting. Changes can be made on
195 the fly and activated with a reconfigure. I.E. You can change to a
196 different helper, but not unconfigure the helper completely.
198 Please note that while this directive defines how Squid processes
199 authentication it does not automatically activate authentication.
200 To use authentication you must in addition make use of ACLs based
201 on login name in http_access (proxy_auth, proxy_auth_regex or
202 external with %LOGIN used in the format tag). The browser will be
203 challenged for authentication on the first such acl encountered
204 in http_access processing and will also be re-challenged for new
205 login credentials if the request is being denied by a proxy_auth
208 WARNING: authentication can't be used in a transparently intercepting
209 proxy as the client then thinks it is talking to an origin server and
210 not the proxy. This is a limitation of bending the TCP/IP protocol to
211 transparently intercepting port 80, not a limitation in Squid.
212 Ports flagged 'transparent', 'intercept', or 'tproxy' have
213 authentication disabled.
215 === Parameters for the basic scheme follow. ===
218 Specify the command for the external authenticator. Such a program
219 reads a line containing "username password" and replies "OK" or
220 "ERR" in an endless loop. "ERR" responses may optionally be followed
221 by a error description available as %m in the returned error page.
222 If you use an authenticator, make sure you have 1 acl of type
225 By default, the basic authentication scheme is not used unless a
226 program is specified.
228 If you want to use the traditional NCSA proxy authentication, set
229 this line to something like
231 auth_param basic program @DEFAULT_PREFIX@/libexec/ncsa_auth @DEFAULT_PREFIX@/etc/passwd
234 HTTP uses iso-latin-1 as characterset, while some authentication
235 backends such as LDAP expects UTF-8. If this is set to on Squid will
236 translate the HTTP iso-latin-1 charset to UTF-8 before sending the
237 username & password to the helper.
239 "children" numberofchildren [startup=N] [idle=N] [concurrency=N]
240 The maximum number of authenticator processes to spawn. If you start too few
241 Squid will have to wait for them to process a backlog of credential
242 verifications, slowing it down. When password verifications are
243 done via a (slow) network you are likely to need lots of
244 authenticator processes.
246 The startup= and idle= options permit some skew in the exact amount
247 run. A minimum of startup=N will begin during startup and reconfigure.
248 Squid will start more in groups of up to idle=N in an attempt to meet
249 traffic needs and to keep idle=N free above those traffic needs up to
252 The concurrency= option sets the number of concurrent requests the
253 helper can process. The default of 0 is used for helpers who only
254 supports one request at a time. Setting this to a number greater than
255 0 changes the protocol used to include a channel number first on the
256 request/response line, allowing multiple requests to be sent to the
257 same helper in parallell without wating for the response.
258 Must not be set unless it's known the helper supports this.
260 auth_param basic children 20 startup=0 idle=1
263 Specifies the realm name which is to be reported to the
264 client for the basic proxy authentication scheme (part of
265 the text the user will see when prompted their username and
266 password). There is no default.
267 auth_param basic realm Squid proxy-caching web server
269 "credentialsttl" timetolive
270 Specifies how long squid assumes an externally validated
271 username:password pair is valid for - in other words how
272 often the helper program is called for that user. Set this
273 low to force revalidation with short lived passwords. Note
274 setting this high does not impact your susceptibility
275 to replay attacks unless you are using an one-time password
276 system (such as SecureID). If you are using such a system,
277 you will be vulnerable to replay attacks unless you also
278 use the max_user_ip ACL in an http_access rule.
280 "casesensitive" on|off
281 Specifies if usernames are case sensitive. Most user databases are
282 case insensitive allowing the same username to be spelled using both
283 lower and upper case letters, but some are case sensitive. This
284 makes a big difference for user_max_ip ACL processing and similar.
285 auth_param basic casesensitive off
287 === Parameters for the digest scheme follow ===
290 Specify the command for the external authenticator. Such
291 a program reads a line containing "username":"realm" and
292 replies with the appropriate H(A1) value hex encoded or
293 ERR if the user (or his H(A1) hash) does not exists.
294 See rfc 2616 for the definition of H(A1).
295 "ERR" responses may optionally be followed by a error description
296 available as %m in the returned error page.
298 By default, the digest authentication scheme is not used unless a
299 program is specified.
301 If you want to use a digest authenticator, set this line to
304 auth_param digest program @DEFAULT_PREFIX@/bin/digest_pw_auth @DEFAULT_PREFIX@/etc/digpass
307 HTTP uses iso-latin-1 as characterset, while some authentication
308 backends such as LDAP expects UTF-8. If this is set to on Squid will
309 translate the HTTP iso-latin-1 charset to UTF-8 before sending the
310 username & password to the helper.
312 "children" numberofchildren [startup=N] [idle=N] [concurrency=N]
313 The maximum number of authenticator processes to spawn (default 5).
314 If you start too few Squid will have to wait for them to
315 process a backlog of H(A1) calculations, slowing it down.
316 When the H(A1) calculations are done via a (slow) network
317 you are likely to need lots of authenticator processes.
319 The startup= and idle= options permit some skew in the exact amount
320 run. A minimum of startup=N will begin during startup and reconfigure.
321 Squid will start more in groups of up to idle=N in an attempt to meet
322 traffic needs and to keep idle=N free above those traffic needs up to
325 The concurrency= option sets the number of concurrent requests the
326 helper can process. The default of 0 is used for helpers who only
327 supports one request at a time. Setting this to a number greater than
328 0 changes the protocol used to include a channel number first on the
329 request/response line, allowing multiple requests to be sent to the
330 same helper in parallell without wating for the response.
331 Must not be set unless it's known the helper supports this.
333 auth_param digest children 20 startup=0 idle=1
336 Specifies the realm name which is to be reported to the
337 client for the digest proxy authentication scheme (part of
338 the text the user will see when prompted their username and
339 password). There is no default.
340 auth_param digest realm Squid proxy-caching web server
342 "nonce_garbage_interval" timeinterval
343 Specifies the interval that nonces that have been issued
344 to client_agent's are checked for validity.
346 "nonce_max_duration" timeinterval
347 Specifies the maximum length of time a given nonce will be
350 "nonce_max_count" number
351 Specifies the maximum number of times a given nonce can be
354 "nonce_strictness" on|off
355 Determines if squid requires strict increment-by-1 behavior
356 for nonce counts, or just incrementing (off - for use when
357 useragents generate nonce counts that occasionally miss 1
358 (ie, 1,2,4,6)). Default off.
360 "check_nonce_count" on|off
361 This directive if set to off can disable the nonce count check
362 completely to work around buggy digest qop implementations in
363 certain mainstream browser versions. Default on to check the
364 nonce count to protect from authentication replay attacks.
366 "post_workaround" on|off
367 This is a workaround to certain buggy browsers who sends
368 an incorrect request digest in POST requests when reusing
369 the same nonce as acquired earlier on a GET request.
371 === NTLM scheme options follow ===
374 Specify the command for the external NTLM authenticator.
375 Such a program reads exchanged NTLMSSP packets with
376 the browser via Squid until authentication is completed.
377 If you use an NTLM authenticator, make sure you have 1 acl
378 of type proxy_auth. By default, the NTLM authenticator_program
381 auth_param ntlm program @DEFAULT_PREFIX@/bin/ntlm_auth
383 "children" numberofchildren [startup=N] [idle=N]
384 The maximum number of authenticator processes to spawn (default 5).
385 If you start too few Squid will have to wait for them to
386 process a backlog of credential verifications, slowing it
387 down. When credential verifications are done via a (slow)
388 network you are likely to need lots of authenticator
391 The startup= and idle= options permit some skew in the exact amount
392 run. A minimum of startup=N will begin during startup and reconfigure.
393 Squid will start more in groups of up to idle=N in an attempt to meet
394 traffic needs and to keep idle=N free above those traffic needs up to
397 auth_param ntlm children 20 startup=0 idle=1
400 If you experience problems with PUT/POST requests when using the
401 Negotiate authentication scheme then you can try setting this to
402 off. This will cause Squid to forcibly close the connection on
403 the initial requests where the browser asks which schemes are
404 supported by the proxy.
406 auth_param ntlm keep_alive on
408 === Options for configuring the NEGOTIATE auth-scheme follow ===
411 Specify the command for the external Negotiate authenticator.
412 This protocol is used in Microsoft Active-Directory enabled setups with
413 the Microsoft Internet Explorer or Mozilla Firefox browsers.
414 Its main purpose is to exchange credentials with the Squid proxy
415 using the Kerberos mechanisms.
416 If you use a Negotiate authenticator, make sure you have at least
417 one acl of type proxy_auth active. By default, the negotiate
418 authenticator_program is not used.
419 The only supported program for this role is the ntlm_auth
420 program distributed as part of Samba, version 4 or later.
422 auth_param negotiate program @DEFAULT_PREFIX@/bin/ntlm_auth --helper-protocol=gss-spnego
424 "children" numberofchildren [startup=N] [idle=N]
425 The maximum number of authenticator processes to spawn (default 5).
426 If you start too few Squid will have to wait for them to
427 process a backlog of credential verifications, slowing it
428 down. When crendential verifications are done via a (slow)
429 network you are likely to need lots of authenticator
432 The startup= and idle= options permit some skew in the exact amount
433 run. A minimum of startup=N will begin during startup and reconfigure.
434 Squid will start more in groups of up to idle=N in an attempt to meet
435 traffic needs and to keep idle=N free above those traffic needs up to
438 auth_param negotiate children 20 startup=0 idle=1
441 If you experience problems with PUT/POST requests when using the
442 Negotiate authentication scheme then you can try setting this to
443 off. This will cause Squid to forcibly close the connection on
444 the initial requests where the browser asks which schemes are
445 supported by the proxy.
447 auth_param negotiate keep_alive on
452 #Recommended minimum configuration per scheme:
453 #auth_param negotiate program <uncomment and complete this line to activate>
454 #auth_param negotiate children 20 startup=0 idle=1
455 #auth_param negotiate keep_alive on
457 #auth_param ntlm program <uncomment and complete this line to activate>
458 #auth_param ntlm children 20 startup=0 idle=1
459 #auth_param ntlm keep_alive on
461 #auth_param digest program <uncomment and complete this line>
462 #auth_param digest children 20 startup=0 idle=1
463 #auth_param digest realm Squid proxy-caching web server
464 #auth_param digest nonce_garbage_interval 5 minutes
465 #auth_param digest nonce_max_duration 30 minutes
466 #auth_param digest nonce_max_count 50
468 #auth_param basic program <uncomment and complete this line>
469 #auth_param basic children 5 startup=5 idle=1
470 #auth_param basic realm Squid proxy-caching web server
471 #auth_param basic credentialsttl 2 hours
474 NAME: authenticate_cache_garbage_interval
477 LOC: Config.authenticateGCInterval
479 The time period between garbage collection across the username cache.
480 This is a tradeoff between memory utilization (long intervals - say
481 2 days) and CPU (short intervals - say 1 minute). Only change if you
485 NAME: authenticate_ttl
488 LOC: Config.authenticateTTL
490 The time a user & their credentials stay in the logged in
491 user cache since their last request. When the garbage
492 interval passes, all user credentials that have passed their
493 TTL are removed from memory.
496 NAME: authenticate_ip_ttl
498 LOC: Config.authenticateIpTTL
501 If you use proxy authentication and the 'max_user_ip' ACL,
502 this directive controls how long Squid remembers the IP
503 addresses associated with each user. Use a small value
504 (e.g., 60 seconds) if your users might change addresses
505 quickly, as is the case with dialups. You might be safe
506 using a larger value (e.g., 2 hours) in a corporate LAN
507 environment with relatively static address assignments.
512 -----------------------------------------------------------------------------
515 NAME: external_acl_type
516 TYPE: externalAclHelper
517 LOC: Config.externalAclHelperList
520 This option defines external acl classes using a helper program
521 to look up the status
523 external_acl_type name [options] FORMAT.. /path/to/helper [helper arguments..]
527 ttl=n TTL in seconds for cached results (defaults to 3600
530 TTL for cached negative lookups (default same
533 Maximum number of acl helper processes spawned to service
534 external acl lookups of this type. (default 20)
536 Minimum number of acl helper processes to spawn during
537 startup and reconfigure to service external acl lookups
538 of this type. (default 0)
540 Number of acl helper processes to keep ahead of traffic
541 loads. Squid will spawn this many at once whenever load
542 rises above the capabilities of existing processes.
543 Up to the value of children-max. (default 1)
544 concurrency=n concurrency level per process. Only used with helpers
545 capable of processing more than one query at a time.
546 cache=n limit the result cache size, default is unbounded.
547 grace=n Percentage remaining of TTL where a refresh of a
548 cached entry should be initiated without needing to
549 wait for a new reply. (default is for no grace period)
550 protocol=2.5 Compatibility mode for Squid-2.5 external acl helpers
551 ipv4 / ipv6 IP-mode used to communicate to this helper.
552 For compatability with older configurations and helpers
553 the default is currently 'ipv4'.
555 FORMAT specifications
557 %LOGIN Authenticated user login name
558 %EXT_USER Username from external acl
559 %IDENT Ident user name
561 %SRCPORT Client source port
564 %PROTO Requested protocol
566 %PATH Requested URL path
567 %METHOD Request method
568 %MYADDR Squid interface address
569 %MYPORT Squid http_port number
570 %PATH Requested URL-path (including query-string if any)
571 %USER_CERT SSL User certificate in PEM format
572 %USER_CERTCHAIN SSL User certificate chain in PEM format
573 %USER_CERT_xx SSL User certificate subject attribute xx
574 %USER_CA_xx SSL User certificate issuer attribute xx
576 %>{Header} HTTP request header "Header"
578 HTTP request header "Hdr" list member "member"
580 HTTP request header list member using ; as
581 list separator. ; can be any non-alphanumeric
584 %<{Header} HTTP reply header "Header"
586 HTTP reply header "Hdr" list member "member"
588 HTTP reply header list member using ; as
589 list separator. ; can be any non-alphanumeric
592 In addition to the above, any string specified in the referencing
593 acl will also be included in the helper request line, after the
594 specified formats (see the "acl external" directive)
596 The helper receives lines per the above format specification,
597 and returns lines starting with OK or ERR indicating the validity
598 of the request and optionally followed by additional keywords with
601 General result syntax:
603 OK/ERR keyword=value ...
607 user= The users name (login)
608 password= The users password (for login= cache_peer option)
609 message= Message describing the reason. Available as %o
611 tag= Apply a tag to a request (for both ERR and OK results)
612 Only sets a tag, does not alter existing tags.
613 log= String to be logged in access.log. Available as
614 %ea in logformat specifications
616 If protocol=3.0 (the default) then URL escaping is used to protect
617 each value in both requests and responses.
619 If using protocol=2.5 then all values need to be enclosed in quotes
620 if they may contain whitespace, or the whitespace escaped using \.
621 And quotes or \ characters within the keyword value must be \ escaped.
623 When using the concurrency= option the protocol is changed by
624 introducing a query channel tag infront of the request/response.
625 The query channel tag is a number between 0 and concurrency-1.
633 Defining an Access List
635 Every access list definition must begin with an aclname and acltype,
636 followed by either type-specific arguments or a quoted filename that
639 acl aclname acltype argument ...
640 acl aclname acltype "file" ...
642 When using "file", the file should contain one item per line.
644 By default, regular expressions are CASE-SENSITIVE. To make
645 them case-insensitive, use the -i option.
647 Some acl types require suspending the current request in order
648 to access some external data source.
649 Those which do are marked with the tag [slow], those which
650 don't are marked as [fast].
651 See http://wiki.squid-cache.org/SquidFaq/SquidAcl
652 for further information
654 ***** ACL TYPES AVAILABLE *****
656 acl aclname src ip-address/netmask ... # clients IP address [fast]
657 acl aclname src addr1-addr2/netmask ... # range of addresses [fast]
658 acl aclname dst ip-address/netmask ... # URL host's IP address [slow]
659 acl aclname myip ip-address/netmask ... # local socket IP address [fast]
661 acl aclname arp mac-address ... (xx:xx:xx:xx:xx:xx notation)
662 # The arp ACL requires the special configure option --enable-arp-acl.
663 # Furthermore, the ARP ACL code is not portable to all operating systems.
664 # It works on Linux, Solaris, Windows, FreeBSD, and some
665 # other *BSD variants.
668 # NOTE: Squid can only determine the MAC address for clients that are on
669 # the same subnet. If the client is on a different subnet,
670 # then Squid cannot find out its MAC address.
672 acl aclname srcdomain .foo.com ...
673 # reverse lookup, from client IP [slow]
674 acl aclname dstdomain .foo.com ...
675 # Destination server from URL [fast]
676 acl aclname srcdom_regex [-i] \.foo\.com ...
677 # regex matching client name [slow]
678 acl aclname dstdom_regex [-i] \.foo\.com ...
679 # regex matching server [fast]
681 # For dstdomain and dstdom_regex a reverse lookup is tried if a IP
682 # based URL is used and no match is found. The name "none" is used
683 # if the reverse lookup fails.
685 acl aclname src_as number ...
686 acl aclname dst_as number ...
688 # Except for access control, AS numbers can be used for
689 # routing of requests to specific caches. Here's an
690 # example for routing all requests for AS#1241 and only
691 # those to mycache.mydomain.net:
692 # acl asexample dst_as 1241
693 # cache_peer_access mycache.mydomain.net allow asexample
694 # cache_peer_access mycache_mydomain.net deny all
696 acl aclname peername myPeer ...
698 # match against a named cache_peer entry
699 # set unique name= on cache_peer lines for reliable use.
701 acl aclname time [day-abbrevs] [h1:m1-h2:m2]
711 # h1:m1 must be less than h2:m2
713 acl aclname url_regex [-i] ^http:// ...
714 # regex matching on whole URL [fast]
715 acl aclname urlpath_regex [-i] \.gif$ ...
716 # regex matching on URL path [fast]
718 acl aclname port 80 70 21 0-1024... # destination TCP port [fast]
720 acl aclname myport 3128 ... # local socket TCP port [fast]
721 acl aclname myportname 3128 ... # http(s)_port name [fast]
723 acl aclname proto HTTP FTP ... # request protocol [fast]
725 acl aclname method GET POST ... # HTTP request method [fast]
727 acl aclname http_status 200 301 500- 400-403 ...
728 # status code in reply [fast]
730 acl aclname browser [-i] regexp ...
731 # pattern match on User-Agent header (see also req_header below) [fast]
733 acl aclname referer_regex [-i] regexp ...
734 # pattern match on Referer header [fast]
735 # Referer is highly unreliable, so use with care
737 acl aclname ident username ...
738 acl aclname ident_regex [-i] pattern ...
739 # string match on ident output [slow]
740 # use REQUIRED to accept any non-null ident.
742 acl aclname proxy_auth [-i] username ...
743 acl aclname proxy_auth_regex [-i] pattern ...
744 # perform http authentication challenge to the client and match against
745 # supplied credentials [slow]
747 # takes a list of allowed usernames.
748 # use REQUIRED to accept any valid username.
750 # Will use proxy authentication in forward-proxy scenarios, and plain
751 # http authenticaiton in reverse-proxy scenarios
753 # NOTE: when a Proxy-Authentication header is sent but it is not
754 # needed during ACL checking the username is NOT logged
757 # NOTE: proxy_auth requires a EXTERNAL authentication program
758 # to check username/password combinations (see
759 # auth_param directive).
761 # NOTE: proxy_auth can't be used in a transparent/intercepting proxy
762 # as the browser needs to be configured for using a proxy in order
763 # to respond to proxy authentication.
765 acl aclname snmp_community string ...
766 # A community string to limit access to your SNMP Agent [fast]
769 # acl snmppublic snmp_community public
771 acl aclname maxconn number
772 # This will be matched when the client's IP address has
773 # more than <number> HTTP connections established. [fast]
775 acl aclname max_user_ip [-s] number
776 # This will be matched when the user attempts to log in from more
777 # than <number> different ip addresses. The authenticate_ip_ttl
778 # parameter controls the timeout on the ip entries. [fast]
779 # If -s is specified the limit is strict, denying browsing
780 # from any further IP addresses until the ttl has expired. Without
781 # -s Squid will just annoy the user by "randomly" denying requests.
782 # (the counter is reset each time the limit is reached and a
784 # NOTE: in acceleration mode or where there is mesh of child proxies,
785 # clients may appear to come from multiple addresses if they are
786 # going through proxy farms, so a limit of 1 may cause user problems.
788 acl aclname random probability
789 # Pseudo-randomly match requests. Based on the probability given.
790 # Probability may be written as a decimal (0.333), fraction (1/3)
791 # or ratio of matches:non-matches (3:5).
793 acl aclname req_mime_type [-i] mime-type ...
794 # regex match against the mime type of the request generated
795 # by the client. Can be used to detect file upload or some
796 # types HTTP tunneling requests [fast]
797 # NOTE: This does NOT match the reply. You cannot use this
798 # to match the returned file type.
800 acl aclname req_header header-name [-i] any\.regex\.here
801 # regex match against any of the known request headers. May be
802 # thought of as a superset of "browser", "referer" and "mime-type"
805 acl aclname rep_mime_type [-i] mime-type ...
806 # regex match against the mime type of the reply received by
807 # squid. Can be used to detect file download or some
808 # types HTTP tunneling requests. [fast]
809 # NOTE: This has no effect in http_access rules. It only has
810 # effect in rules that affect the reply data stream such as
813 acl aclname rep_header header-name [-i] any\.regex\.here
814 # regex match against any of the known reply headers. May be
815 # thought of as a superset of "browser", "referer" and "mime-type"
818 acl aclname external class_name [arguments...]
819 # external ACL lookup via a helper class defined by the
820 # external_acl_type directive [slow]
822 acl aclname user_cert attribute values...
823 # match against attributes in a user SSL certificate
824 # attribute is one of DN/C/O/CN/L/ST [fast]
826 acl aclname ca_cert attribute values...
827 # match against attributes a users issuing CA SSL certificate
828 # attribute is one of DN/C/O/CN/L/ST [fast]
830 acl aclname ext_user username ...
831 acl aclname ext_user_regex [-i] pattern ...
832 # string match on username returned by external acl helper [slow]
833 # use REQUIRED to accept any non-null user name.
835 acl aclname tag tagvalue ...
836 # string match on tag returned by external acl helper [slow]
838 acl aclname hier_code codename ...
839 # string match against squid hierarchy code(s); [fast]
840 # e.g., DIRECT, PARENT_HIT, NONE, etc.
842 # NOTE: This has no effect in http_access rules. It only has
843 # effect in rules that affect the reply data stream such as
847 acl macaddress arp 09:00:2b:23:45:67
848 acl myexample dst_as 1241
849 acl password proxy_auth REQUIRED
850 acl fileupload req_mime_type -i ^multipart/form-data$
851 acl javascript rep_mime_type -i ^application/x-javascript$
855 # Recommended minimum configuration:
857 acl manager proto cache_object
858 acl localhost src 127.0.0.1/32 ::1
859 acl to_localhost dst 127.0.0.0/8 0.0.0.0/32 ::1
861 # Example rule allowing access from your local networks.
862 # Adapt to list your (internal) IP networks from where browsing
864 acl localnet src 10.0.0.0/8 # RFC1918 possible internal network
865 acl localnet src 172.16.0.0/12 # RFC1918 possible internal network
866 acl localnet src 192.168.0.0/16 # RFC1918 possible internal network
867 acl localnet src fc00::/7 # RFC 4193 local private network range
868 acl localnet src fe80::/10 # RFC 4291 link-local (directly plugged) machines
870 acl SSL_ports port 443
871 acl Safe_ports port 80 # http
872 acl Safe_ports port 21 # ftp
873 acl Safe_ports port 443 # https
874 acl Safe_ports port 70 # gopher
875 acl Safe_ports port 210 # wais
876 acl Safe_ports port 1025-65535 # unregistered ports
877 acl Safe_ports port 280 # http-mgmt
878 acl Safe_ports port 488 # gss-http
879 acl Safe_ports port 591 # filemaker
880 acl Safe_ports port 777 # multiling http
881 acl CONNECT method CONNECT
885 NAME: follow_x_forwarded_for
887 IFDEF: FOLLOW_X_FORWARDED_FOR
888 LOC: Config.accessList.followXFF
889 DEFAULT_IF_NONE: deny all
891 Allowing or Denying the X-Forwarded-For header to be followed to
892 find the original source of a request.
894 Requests may pass through a chain of several other proxies
895 before reaching us. The X-Forwarded-For header will contain a
896 comma-separated list of the IP addresses in the chain, with the
897 rightmost address being the most recent.
899 If a request reaches us from a source that is allowed by this
900 configuration item, then we consult the X-Forwarded-For header
901 to see where that host received the request from. If the
902 X-Forwarded-For header contains multiple addresses, we continue
903 backtracking until we reach an address for which we are not allowed
904 to follow the X-Forwarded-For header, or until we reach the first
905 address in the list. For the purpose of ACL used in the
906 follow_x_forwarded_for directive the src ACL type always matches
907 the address we are testing and srcdomain matches its rDNS.
909 The end result of this process is an IP address that we will
910 refer to as the indirect client address. This address may
911 be treated as the client address for access control, ICAP, delay
912 pools and logging, depending on the acl_uses_indirect_client,
913 icap_uses_indirect_client, delay_pool_uses_indirect_client,
914 log_uses_indirect_client and tproxy_uses_indirect_client options.
916 This clause only supports fast acl types.
917 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
919 SECURITY CONSIDERATIONS:
921 Any host for which we follow the X-Forwarded-For header
922 can place incorrect information in the header, and Squid
923 will use the incorrect information as if it were the
924 source address of the request. This may enable remote
925 hosts to bypass any access control restrictions that are
926 based on the client's source addresses.
930 acl localhost src 127.0.0.1
931 acl my_other_proxy srcdomain .proxy.example.com
932 follow_x_forwarded_for allow localhost
933 follow_x_forwarded_for allow my_other_proxy
936 NAME: acl_uses_indirect_client
939 IFDEF: FOLLOW_X_FORWARDED_FOR
941 LOC: Config.onoff.acl_uses_indirect_client
943 Controls whether the indirect client address
944 (see follow_x_forwarded_for) is used instead of the
945 direct client address in acl matching.
948 NAME: delay_pool_uses_indirect_client
951 IFDEF: FOLLOW_X_FORWARDED_FOR&&DELAY_POOLS
953 LOC: Config.onoff.delay_pool_uses_indirect_client
955 Controls whether the indirect client address
956 (see follow_x_forwarded_for) is used instead of the
957 direct client address in delay pools.
960 NAME: log_uses_indirect_client
963 IFDEF: FOLLOW_X_FORWARDED_FOR
965 LOC: Config.onoff.log_uses_indirect_client
967 Controls whether the indirect client address
968 (see follow_x_forwarded_for) is used instead of the
969 direct client address in the access log.
972 NAME: tproxy_uses_indirect_client
975 IFDEF: FOLLOW_X_FORWARDED_FOR&&LINUX_NETFILTER
977 LOC: Config.onoff.tproxy_uses_indirect_client
979 Controls whether the indirect client address
980 (see follow_x_forwarded_for) is used instead of the
981 direct client address when spoofing the outgoing client.
983 This has no effect on requests arriving in non-tproxy
986 SECURITY WARNING: Usage of this option is dangerous
987 and should not be used trivially. Correct configuration
988 of folow_x_forewarded_for with a limited set of trusted
989 sources is required to prevent abuse of your proxy.
994 LOC: Config.accessList.http
995 DEFAULT_IF_NONE: deny all
997 Allowing or Denying access based on defined access lists
999 Access to the HTTP port:
1000 http_access allow|deny [!]aclname ...
1002 NOTE on default values:
1004 If there are no "access" lines present, the default is to deny
1007 If none of the "access" lines cause a match, the default is the
1008 opposite of the last line in the list. If the last line was
1009 deny, the default is allow. Conversely, if the last line
1010 is allow, the default will be deny. For these reasons, it is a
1011 good idea to have an "deny all" entry at the end of your access
1012 lists to avoid potential confusion.
1014 This clause supports both fast and slow acl types.
1015 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1020 # Recommended minimum Access Permission configuration:
1022 # Only allow cachemgr access from localhost
1023 http_access allow manager localhost
1024 http_access deny manager
1026 # Deny requests to certain unsafe ports
1027 http_access deny !Safe_ports
1029 # Deny CONNECT to other than secure SSL ports
1030 http_access deny CONNECT !SSL_ports
1032 # We strongly recommend the following be uncommented to protect innocent
1033 # web applications running on the proxy server who think the only
1034 # one who can access services on "localhost" is a local user
1035 #http_access deny to_localhost
1038 # INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS
1041 # Example rule allowing access from your local networks.
1042 # Adapt localnet in the ACL section to list your (internal) IP networks
1043 # from where browsing should be allowed
1044 http_access allow localnet
1045 http_access allow localhost
1047 # And finally deny all other access to this proxy
1048 http_access deny all
1052 NAME: adapted_http_access http_access2
1054 LOC: Config.accessList.adapted_http
1057 Allowing or Denying access based on defined access lists
1059 Essentially identical to http_access, but runs after redirectors
1060 and ICAP/eCAP adaptation. Allowing access control based on their
1063 If not set then only http_access is used.
1066 NAME: http_reply_access
1068 LOC: Config.accessList.reply
1071 Allow replies to client requests. This is complementary to http_access.
1073 http_reply_access allow|deny [!] aclname ...
1075 NOTE: if there are no access lines present, the default is to allow
1078 If none of the access lines cause a match the opposite of the
1079 last line will apply. Thus it is good practice to end the rules
1080 with an "allow all" or "deny all" entry.
1082 This clause supports both fast and slow acl types.
1083 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1088 LOC: Config.accessList.icp
1089 DEFAULT_IF_NONE: deny all
1091 Allowing or Denying access to the ICP port based on defined
1094 icp_access allow|deny [!]aclname ...
1096 See http_access for details
1098 This clause only supports fast acl types.
1099 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1101 # Allow ICP queries from local networks only
1102 #icp_access allow localnet
1103 #icp_access deny all
1109 LOC: Config.accessList.htcp
1110 DEFAULT_IF_NONE: deny all
1112 Allowing or Denying access to the HTCP port based on defined
1115 htcp_access allow|deny [!]aclname ...
1117 See http_access for details
1119 NOTE: The default if no htcp_access lines are present is to
1120 deny all traffic. This default may cause problems with peers
1121 using the htcp option.
1123 This clause only supports fast acl types.
1124 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1126 # Allow HTCP queries from local networks only
1127 #htcp_access allow localnet
1128 #htcp_access deny all
1131 NAME: htcp_clr_access
1134 LOC: Config.accessList.htcp_clr
1135 DEFAULT_IF_NONE: deny all
1137 Allowing or Denying access to purge content using HTCP based
1138 on defined access lists
1140 htcp_clr_access allow|deny [!]aclname ...
1142 See http_access for details
1144 This clause only supports fast acl types.
1145 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1147 # Allow HTCP CLR requests from trusted peers
1148 acl htcp_clr_peer src 172.16.1.2
1149 htcp_clr_access allow htcp_clr_peer
1154 LOC: Config.accessList.miss
1157 Use to force your neighbors to use you as a sibling instead of
1158 a parent. For example:
1160 acl localclients src 172.16.0.0/16
1161 miss_access allow localclients
1162 miss_access deny !localclients
1164 This means only your local clients are allowed to fetch
1165 MISSES and all other clients can only fetch HITS.
1167 By default, allow all clients who passed the http_access rules
1168 to fetch MISSES from us.
1170 This clause only supports fast acl types.
1171 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1174 NAME: ident_lookup_access
1177 DEFAULT_IF_NONE: deny all
1178 LOC: Ident::TheConfig.identLookup
1180 A list of ACL elements which, if matched, cause an ident
1181 (RFC 931) lookup to be performed for this request. For
1182 example, you might choose to always perform ident lookups
1183 for your main multi-user Unix boxes, but not for your Macs
1184 and PCs. By default, ident lookups are not performed for
1187 To enable ident lookups for specific client addresses, you
1188 can follow this example:
1190 acl ident_aware_hosts src 198.168.1.0/24
1191 ident_lookup_access allow ident_aware_hosts
1192 ident_lookup_access deny all
1194 Only src type ACL checks are fully supported. A srcdomain
1195 ACL might work at times, but it will not always provide
1198 This clause only supports fast acl types.
1199 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1202 NAME: reply_body_max_size
1203 COMMENT: size [acl acl...]
1206 LOC: Config.ReplyBodySize
1208 This option specifies the maximum size of a reply body. It can be
1209 used to prevent users from downloading very large files, such as
1210 MP3's and movies. When the reply headers are received, the
1211 reply_body_max_size lines are processed, and the first line where
1212 all (if any) listed ACLs are true is used as the maximum body size
1215 This size is checked twice. First when we get the reply headers,
1216 we check the content-length value. If the content length value exists
1217 and is larger than the allowed size, the request is denied and the
1218 user receives an error message that says "the request or reply
1219 is too large." If there is no content-length, and the reply
1220 size exceeds this limit, the client's connection is just closed
1221 and they will receive a partial reply.
1223 WARNING: downstream caches probably can not detect a partial reply
1224 if there is no content-length header, so they will cache
1225 partial responses and give them out as hits. You should NOT
1226 use this option if you have downstream caches.
1228 WARNING: A maximum size smaller than the size of squid's error messages
1229 will cause an infinite loop and crash squid. Ensure that the smallest
1230 non-zero value you use is greater that the maximum header size plus
1231 the size of your largest error page.
1233 If you set this parameter none (the default), there will be
1236 Configuration Format is:
1237 reply_body_max_size SIZE UNITS [acl ...]
1239 reply_body_max_size 10 MB
1245 -----------------------------------------------------------------------------
1248 NAME: http_port ascii_port
1249 TYPE: http_port_list
1251 LOC: Config.Sockaddr.http
1253 Usage: port [mode] [options]
1254 hostname:port [mode] [options]
1255 1.2.3.4:port [mode] [options]
1257 The socket addresses where Squid will listen for HTTP client
1258 requests. You may specify multiple socket addresses.
1259 There are three forms: port alone, hostname with port, and
1260 IP address with port. If you specify a hostname or IP
1261 address, Squid binds the socket to that specific
1262 address. Most likely, you do not need to bind to a specific
1263 address, so you can use the port number alone.
1265 If you are running Squid in accelerator mode, you
1266 probably want to listen on port 80 also, or instead.
1268 The -a command line option may be used to specify additional
1269 port(s) where Squid listens for proxy request. Such ports will
1270 be plain proxy ports with no options.
1272 You may specify multiple socket addresses on multiple lines.
1276 intercept Support for IP-Layer interception of
1277 outgoing requests without browser settings.
1278 NP: disables authentication and IPv6 on the port.
1280 tproxy Support Linux TPROXY for spoofing outgoing
1281 connections using the client IP address.
1282 NP: disables authentication and maybe IPv6 on the port.
1284 accel Accelerator mode. Also needs at least one of
1285 vhost / vport / defaultsite.
1287 ssl-bump Intercept each CONNECT request matching ssl_bump ACL,
1288 establish secure connection with the client and with
1289 the server, decrypt HTTP messages as they pass through
1290 Squid, and treat them as unencrypted HTTP messages,
1291 becoming the man-in-the-middle.
1293 The ssl_bump option is required to fully enable
1294 the SslBump feature.
1296 Omitting the mode flag causes default forward proxy mode to be used.
1299 Accelerator Mode Options:
1301 allow-direct Allow direct forwarding in accelerator mode. Normally
1302 accelerated requests are denied direct forwarding as if
1303 never_direct was used.
1305 defaultsite=domainname
1306 What to use for the Host: header if it is not present
1307 in a request. Determines what site (not origin server)
1308 accelerators should consider the default.
1311 vhost Using the Host header for virtual domain support.
1312 Also uses the port as specified in Host: header.
1314 vport IP based virtual host support. Using the http_port number
1315 in passed on Host: headers.
1317 vport=NN Uses the specified port number rather than the
1320 protocol= Protocol to reconstruct accelerated requests with.
1321 Defaults to http://.
1323 ignore-cc Ignore request Cache-Control headers.
1325 Warning: This option violates HTTP specifications if
1326 used in non-accelerator setups.
1329 SSL Bump Mode Options:
1331 cert= Path to SSL certificate (PEM format).
1333 key= Path to SSL private key file (PEM format)
1334 if not specified, the certificate file is
1335 assumed to be a combined certificate and
1338 version= The version of SSL/TLS supported
1339 1 automatic (default)
1344 cipher= Colon separated list of supported ciphers.
1346 options= Various SSL engine options. The most important
1348 NO_SSLv2 Disallow the use of SSLv2
1349 NO_SSLv3 Disallow the use of SSLv3
1350 NO_TLSv1 Disallow the use of TLSv1
1351 SINGLE_DH_USE Always create a new key when using
1352 temporary/ephemeral DH key exchanges
1353 See src/ssl_support.c or OpenSSL SSL_CTX_set_options
1354 documentation for a complete list of options.
1356 clientca= File containing the list of CAs to use when
1357 requesting a client certificate.
1359 cafile= File containing additional CA certificates to
1360 use when verifying client certificates. If unset
1361 clientca will be used.
1363 capath= Directory containing additional CA certificates
1364 and CRL lists to use when verifying client certificates.
1366 crlfile= File of additional CRL lists to use when verifying
1367 the client certificate, in addition to CRLs stored in
1368 the capath. Implies VERIFY_CRL flag below.
1370 dhparams= File containing DH parameters for temporary/ephemeral
1373 sslflags= Various flags modifying the use of SSL:
1375 Don't request client certificates
1376 immediately, but wait until acl processing
1377 requires a certificate (not yet implemented).
1379 Don't use the default CA lists built in
1382 Don't allow for session reuse. Each connection
1383 will result in a new SSL session.
1385 Verify CRL lists when accepting client
1388 Verify CRL lists for all certificates in the
1389 client certificate chain.
1391 sslcontext= SSL session ID context identifier.
1396 connection-auth[=on|off]
1397 use connection-auth=off to tell Squid to prevent
1398 forwarding Microsoft connection oriented authentication
1399 (NTLM, Negotiate and Kerberos)
1401 disable-pmtu-discovery=
1402 Control Path-MTU discovery usage:
1403 off lets OS decide on what to do (default).
1404 transparent disable PMTU discovery when transparent
1406 always disable always PMTU discovery.
1408 In many setups of transparently intercepting proxies
1409 Path-MTU discovery can not work on traffic towards the
1410 clients. This is the case when the intercepting device
1411 does not fully track connections and fails to forward
1412 ICMP must fragment messages to the cache server. If you
1413 have such setup and experience that certain clients
1414 sporadically hang or never complete requests set
1415 disable-pmtu-discovery option to 'transparent'.
1417 name= Specifies a internal name for the port. Defaults to
1418 the port specification (port or addr:port)
1420 tcpkeepalive[=idle,interval,timeout]
1421 Enable TCP keepalive probes of idle connections.
1422 In seconds; idle is the initial time before TCP starts
1423 probing the connection, interval how often to probe, and
1424 timeout the time before giving up.
1426 If you run Squid on a dual-homed machine with an internal
1427 and an external interface we recommend you to specify the
1428 internal address:port in http_port. This way Squid will only be
1429 visible on the internal address.
1433 # Squid normally listens to port 3128
1434 http_port @DEFAULT_HTTP_PORT@
1440 TYPE: https_port_list
1442 LOC: Config.Sockaddr.https
1444 Usage: [ip:]port cert=certificate.pem [key=key.pem] [options...]
1446 The socket address where Squid will listen for HTTPS client
1449 This is really only useful for situations where you are running
1450 squid in accelerator mode and you want to do the SSL work at the
1453 You may specify multiple socket addresses on multiple lines,
1454 each with their own SSL certificate and/or options.
1458 accel Accelerator mode. Also needs at least one of
1459 defaultsite or vhost.
1461 defaultsite= The name of the https site presented on
1462 this port. Implies accel.
1464 vhost Accelerator mode using Host header for virtual
1465 domain support. Requires a wildcard certificate
1466 or other certificate valid for more than one domain.
1469 protocol= Protocol to reconstruct accelerated requests with.
1472 cert= Path to SSL certificate (PEM format).
1474 key= Path to SSL private key file (PEM format)
1475 if not specified, the certificate file is
1476 assumed to be a combined certificate and
1479 version= The version of SSL/TLS supported
1480 1 automatic (default)
1485 cipher= Colon separated list of supported ciphers.
1487 options= Various SSL engine options. The most important
1489 NO_SSLv2 Disallow the use of SSLv2
1490 NO_SSLv3 Disallow the use of SSLv3
1491 NO_TLSv1 Disallow the use of TLSv1
1492 SINGLE_DH_USE Always create a new key when using
1493 temporary/ephemeral DH key exchanges
1494 See src/ssl_support.c or OpenSSL SSL_CTX_set_options
1495 documentation for a complete list of options.
1497 clientca= File containing the list of CAs to use when
1498 requesting a client certificate.
1500 cafile= File containing additional CA certificates to
1501 use when verifying client certificates. If unset
1502 clientca will be used.
1504 capath= Directory containing additional CA certificates
1505 and CRL lists to use when verifying client certificates.
1507 crlfile= File of additional CRL lists to use when verifying
1508 the client certificate, in addition to CRLs stored in
1509 the capath. Implies VERIFY_CRL flag below.
1511 dhparams= File containing DH parameters for temporary/ephemeral
1514 sslflags= Various flags modifying the use of SSL:
1516 Don't request client certificates
1517 immediately, but wait until acl processing
1518 requires a certificate (not yet implemented).
1520 Don't use the default CA lists built in
1523 Don't allow for session reuse. Each connection
1524 will result in a new SSL session.
1526 Verify CRL lists when accepting client
1529 Verify CRL lists for all certificates in the
1530 client certificate chain.
1532 sslcontext= SSL session ID context identifier.
1534 vport Accelerator with IP based virtual host support.
1536 vport=NN As above, but uses specified port number rather
1537 than the https_port number. Implies accel.
1539 name= Specifies a internal name for the port. Defaults to
1540 the port specification (port or addr:port)
1544 NAME: tcp_outgoing_tos tcp_outgoing_ds tcp_outgoing_dscp
1547 LOC: Ip::Qos::TheConfig.tosToServer
1549 Allows you to select a TOS/Diffserv value for packets outgoing
1550 on the server side, based on an ACL.
1552 tcp_outgoing_tos ds-field [!]aclname ...
1554 Example where normal_service_net uses the TOS value 0x00
1555 and good_service_net uses 0x20
1557 acl normal_service_net src 10.0.0.0/24
1558 acl good_service_net src 10.0.1.0/24
1559 tcp_outgoing_tos 0x00 normal_service_net
1560 tcp_outgoing_tos 0x20 good_service_net
1562 TOS/DSCP values really only have local significance - so you should
1563 know what you're specifying. For more information, see RFC2474,
1564 RFC2475, and RFC3260.
1566 The TOS/DSCP byte must be exactly that - a octet value 0 - 255, or
1567 "default" to use whatever default your host has. Note that in
1568 practice often only multiples of 4 is usable as the two rightmost bits
1569 have been redefined for use by ECN (RFC 3168 section 23.1).
1571 Processing proceeds in the order specified, and stops at first fully
1575 NAME: clientside_tos
1578 LOC: Ip::Qos::TheConfig.tosToClient
1580 Allows you to select a TOS/Diffserv value for packets being transmitted
1581 on the client-side, based on an ACL.
1583 clientside_tos ds-field [!]aclname ...
1585 Example where normal_service_net uses the TOS value 0x00
1586 and good_service_net uses 0x20
1588 acl normal_service_net src 10.0.0.0/24
1589 acl good_service_net src 10.0.1.0/24
1590 clientside_tos 0x00 normal_service_net
1591 clientside_tos 0x20 good_service_net
1593 Note: This feature is incompatible with qos_flows. Any TOS values set here
1594 will be overwritten by TOS values in qos_flows.
1597 NAME: tcp_outgoing_mark
1601 LOC: Ip::Qos::TheConfig.nfmarkToServer
1603 Allows you to apply a Netfilter mark value to outgoing packets
1604 on the server side, based on an ACL.
1606 tcp_outgoing_mark mark-value [!]aclname ...
1608 Example where normal_service_net uses the mark value 0x00
1609 and good_service_net uses 0x20
1611 acl normal_service_net src 10.0.0.0/24
1612 acl good_service_net src 10.0.1.0/24
1613 tcp_outgoing_mark 0x00 normal_service_net
1614 tcp_outgoing_mark 0x20 good_service_net
1617 NAME: clientside_mark
1621 LOC: Ip::Qos::TheConfig.nfmarkToClient
1623 Allows you to apply a Netfilter mark value to packets being transmitted
1624 on the client-side, based on an ACL.
1626 clientside_mark mark-value [!]aclname ...
1628 Example where normal_service_net uses the mark value 0x00
1629 and good_service_net uses 0x20
1631 acl normal_service_net src 10.0.0.0/24
1632 acl good_service_net src 10.0.1.0/24
1633 clientside_mark 0x00 normal_service_net
1634 clientside_mark 0x20 good_service_net
1636 Note: This feature is incompatible with qos_flows. Any mark values set here
1637 will be overwritten by mark values in qos_flows.
1644 LOC: Ip::Qos::TheConfig
1646 Allows you to select a TOS/DSCP value to mark outgoing
1647 connections with, based on where the reply was sourced. For
1648 platforms using netfilter, allows you to set a netfilter mark
1649 value instead of, or in addition to, a TOS value.
1651 TOS values really only have local significance - so you should
1652 know what you're specifying. For more information, see RFC2474,
1653 RFC2475, and RFC3260.
1655 The TOS/DSCP byte must be exactly that - a octet value 0 - 255. Note that
1656 in practice often only multiples of 4 is usable as the two rightmost bits
1657 have been redefined for use by ECN (RFC 3168 section 23.1).
1659 Mark values can be any unsigned 32-bit integer value.
1661 This setting is configured by setting the following values:
1663 tos|mark Whether to set TOS or netfilter mark values
1665 local-hit=0xFF Value to mark local cache hits.
1667 sibling-hit=0xFF Value to mark hits from sibling peers.
1669 parent-hit=0xFF Value to mark hits from parent peers.
1671 miss=0xFF Value to mark cache misses. Takes precedence
1672 over the preserve-miss feature (see below).
1674 The TOS variant of the following features are only possible on Linux
1675 and require your kernel to be patched with the TOS preserving ZPH
1676 patch, available from http://zph.bratcheda.org
1677 No patch is needed to preserve the netfilter mark, which will work
1678 with all variants of netfilter.
1680 disable-preserve-miss
1681 This option disables the preservation of the TOS or netfilter
1682 mark. By default, the existing TOS or netfilter mark value of
1683 the response coming from the remote server will be retained
1684 and masked with miss-mark.
1685 NOTE: in the case of a netfilter mark, the mark must be set on
1686 the connection (using the CONNMARK target) not on the packet
1690 Allows you to mask certain bits in the TOS or mark value
1691 received from the remote server, before copying the value to
1692 the TOS sent towards clients.
1693 Default for tos: 0xFF (TOS from server is not changed).
1694 Default for mark: 0xFFFFFFFF (mark from server is not changed).
1696 All of these features require the --enable-zph-qos compilation flag
1697 (enabled by default). Netfilter marking also requires the
1698 libnetfilter_conntrack libraries (--with-netfilter-conntrack) and
1699 libcap 2.09+ (--with-libcap).
1703 NAME: tcp_outgoing_address
1706 LOC: Config.accessList.outgoing_address
1708 Allows you to map requests to different outgoing IP addresses
1709 based on the username or source address of the user making
1712 tcp_outgoing_address ipaddr [[!]aclname] ...
1714 Example where requests from 10.0.0.0/24 will be forwarded
1715 with source address 10.1.0.1, 10.0.2.0/24 forwarded with
1716 source address 10.1.0.2 and the rest will be forwarded with
1717 source address 10.1.0.3.
1719 acl normal_service_net src 10.0.0.0/24
1720 acl good_service_net src 10.0.2.0/24
1721 tcp_outgoing_address 10.1.0.1 normal_service_net
1722 tcp_outgoing_address 10.1.0.2 good_service_net
1723 tcp_outgoing_address 10.1.0.3
1725 Processing proceeds in the order specified, and stops at first fully
1728 Note: The use of this directive using client dependent ACLs is
1729 incompatible with the use of server side persistent connections. To
1730 ensure correct results it is best to set server_persistent_connections
1731 to off when using this directive in such configurations.
1733 Note: The use of this directive to set a local IP on outgoing TCP links
1734 is incompatible with using TPROXY to set client IP out outbound TCP links.
1735 When needing to contact peers use the no-tproxy cache_peer option to
1736 re-enable normal forwarding such as this.
1740 Squid is built with a capability of bridging the IPv4 and IPv6
1742 tcp_outgoing_address as exampled above breaks this bridging by forcing
1743 all outbound traffic through a certain IPv4 which may be on the wrong
1744 side of the IPv4/IPv6 boundary.
1746 To operate with tcp_outgoing_address and keep the bridging benefits
1747 an additional ACL needs to be used which ensures the IPv6-bound traffic
1748 is never forced or permitted out the IPv4 interface.
1750 acl to_ipv6 dst ipv6
1751 tcp_outgoing_address 2002::c001 good_service_net to_ipv6
1752 tcp_outgoing_address 10.1.0.2 good_service_net !to_ipv6
1754 tcp_outgoing_address 2002::beef normal_service_net to_ipv6
1755 tcp_outgoing_address 10.1.0.1 normal_service_net !to_ipv6
1757 tcp_outgoing_address 2002::1 to_ipv6
1758 tcp_outgoing_address 10.1.0.3 !to_ipv6
1761 'dst ipv6' bases its selection assuming DIRECT access.
1762 If peers are used the peername ACL are needed to select outgoing
1763 address which can link to the peer.
1765 'dst ipv6' is a slow ACL. It will only work here if 'dst' is used
1766 previously in the http_access rules to locate the destination IP.
1767 Some more magic may be needed for that:
1768 http_access allow to_ipv6 !all
1769 (meaning, allow if to IPv6 but not from anywhere ;)
1775 -----------------------------------------------------------------------------
1778 NAME: ssl_unclean_shutdown
1782 LOC: Config.SSL.unclean_shutdown
1784 Some browsers (especially MSIE) bugs out on SSL shutdown
1791 LOC: Config.SSL.ssl_engine
1794 The OpenSSL engine to use. You will need to set this if you
1795 would like to use hardware SSL acceleration for example.
1798 NAME: sslproxy_client_certificate
1801 LOC: Config.ssl_client.cert
1804 Client SSL Certificate to use when proxying https:// URLs
1807 NAME: sslproxy_client_key
1810 LOC: Config.ssl_client.key
1813 Client SSL Key to use when proxying https:// URLs
1816 NAME: sslproxy_version
1819 LOC: Config.ssl_client.version
1822 SSL version level to use when proxying https:// URLs
1825 NAME: sslproxy_options
1828 LOC: Config.ssl_client.options
1831 SSL engine options to use when proxying https:// URLs
1833 The most important being:
1835 NO_SSLv2 Disallow the use of SSLv2
1836 NO_SSLv3 Disallow the use of SSLv3
1837 NO_TLSv1 Disallow the use of TLSv1
1839 Always create a new key when using
1840 temporary/ephemeral DH key exchanges
1842 These options vary depending on your SSL engine.
1843 See the OpenSSL SSL_CTX_set_options documentation for a
1844 complete list of possible options.
1847 NAME: sslproxy_cipher
1850 LOC: Config.ssl_client.cipher
1853 SSL cipher list to use when proxying https:// URLs
1855 Colon separated list of supported ciphers.
1858 NAME: sslproxy_cafile
1861 LOC: Config.ssl_client.cafile
1864 file containing CA certificates to use when verifying server
1865 certificates while proxying https:// URLs
1868 NAME: sslproxy_capath
1871 LOC: Config.ssl_client.capath
1874 directory containing CA certificates to use when verifying
1875 server certificates while proxying https:// URLs
1881 LOC: Config.accessList.ssl_bump
1884 This ACL controls which CONNECT requests to an http_port
1885 marked with an sslBump flag are actually "bumped". Please
1886 see the sslBump flag of an http_port option for more details
1887 about decoding proxied SSL connections.
1889 By default, no requests are bumped.
1891 See also: http_port sslBump
1893 This clause only supports fast acl types.
1894 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1897 # Example: Bump all requests except those originating from localhost and
1898 # those going to webax.com or example.com sites.
1900 acl localhost src 127.0.0.1/32
1901 acl broken_sites dstdomain .webax.com
1902 acl broken_sites dstdomain .example.com
1903 ssl_bump deny localhost
1904 ssl_bump deny broken_sites
1908 NAME: sslproxy_flags
1911 LOC: Config.ssl_client.flags
1914 Various flags modifying the use of SSL while proxying https:// URLs:
1915 DONT_VERIFY_PEER Accept certificates that fail verification.
1916 For refined control, see sslproxy_cert_error.
1917 NO_DEFAULT_CA Don't use the default CA list built in
1922 NAME: sslproxy_cert_error
1925 LOC: Config.ssl_client.cert_error
1928 Use this ACL to bypass server certificate validation errors.
1930 For example, the following lines will bypass all validation errors
1931 when talking to servers located at 172.16.0.0/16. All other
1932 validation errors will result in ERR_SECURE_CONNECT_FAIL error.
1934 acl BrokenServersAtTrustedIP dst 172.16.0.0/16
1935 sslproxy_cert_error allow BrokenServersAtTrustedIP
1936 sslproxy_cert_error deny all
1938 This clause only supports fast acl types.
1939 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
1940 Using slow acl types may result in server crashes
1942 Without this option, all server certificate validation errors
1943 terminate the transaction. Bypassing validation errors is dangerous
1944 because an error usually implies that the server cannot be trusted and
1945 the connection may be insecure.
1947 See also: sslproxy_flags and DONT_VERIFY_PEER.
1949 Default setting: sslproxy_cert_error deny all
1954 NAME: sslpassword_program
1957 LOC: Config.Program.ssl_password
1960 Specify a program used for entering SSL key passphrases
1961 when using encrypted SSL certificate keys. If not specified
1962 keys must either be unencrypted, or Squid started with the -N
1963 option to allow it to query interactively for the passphrase.
1965 The key file name is given as argument to the program allowing
1966 selection of the right password if you have multiple encrypted
1971 OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
1972 -----------------------------------------------------------------------------
1980 To specify other caches in a hierarchy, use the format:
1982 cache_peer hostname type http-port icp-port [options]
1987 # hostname type port port options
1988 # -------------------- -------- ----- ----- -----------
1989 cache_peer parent.foo.net parent 3128 3130 default
1990 cache_peer sib1.foo.net sibling 3128 3130 proxy-only
1991 cache_peer sib2.foo.net sibling 3128 3130 proxy-only
1992 cache_peer example.com parent 80 0 no-query default
1993 cache_peer cdn.example.com sibling 3128 0
1995 type: either 'parent', 'sibling', or 'multicast'.
1997 proxy-port: The port number where the peer accept HTTP requests.
1998 For other Squid proxies this is usually 3128
1999 For web servers this is usually 80
2001 icp-port: Used for querying neighbor caches about objects.
2002 Set to 0 if the peer does not support ICP or HTCP.
2003 See ICP and HTCP options below for additional details.
2006 ==== ICP OPTIONS ====
2008 You MUST also set icp_port and icp_access explicitly when using these options.
2009 The defaults will prevent peer traffic using ICP.
2012 no-query Disable ICP queries to this neighbor.
2015 Indicates the named peer is a member of a multicast group.
2016 ICP queries will not be sent directly to the peer, but ICP
2017 replies will be accepted from it.
2019 closest-only Indicates that, for ICP_OP_MISS replies, we'll only forward
2020 CLOSEST_PARENT_MISSes and never FIRST_PARENT_MISSes.
2023 To only send ICP queries to this neighbor infrequently.
2024 This is used to keep the neighbor round trip time updated
2025 and is usually used in conjunction with weighted-round-robin.
2028 ==== HTCP OPTIONS ====
2030 You MUST also set htcp_port and htcp_access explicitly when using these options.
2031 The defaults will prevent peer traffic using HTCP.
2034 htcp Send HTCP, instead of ICP, queries to the neighbor.
2035 You probably also want to set the "icp-port" to 4827
2036 instead of 3130. This directive accepts a comma separated
2037 list of options described below.
2039 htcp=oldsquid Send HTCP to old Squid versions (2.5 or earlier).
2041 htcp=no-clr Send HTCP to the neighbor but without
2042 sending any CLR requests. This cannot be used with
2045 htcp=only-clr Send HTCP to the neighbor but ONLY CLR requests.
2046 This cannot be used with no-clr.
2049 Send HTCP to the neighbor including CLRs but only when
2050 they do not result from PURGE requests.
2053 Forward any HTCP CLR requests this proxy receives to the peer.
2056 ==== PEER SELECTION METHODS ====
2058 The default peer selection method is ICP, with the first responding peer
2059 being used as source. These options can be used for better load balancing.
2062 default This is a parent cache which can be used as a "last-resort"
2063 if a peer cannot be located by any of the peer-selection methods.
2064 If specified more than once, only the first is used.
2066 round-robin Load-Balance parents which should be used in a round-robin
2067 fashion in the absence of any ICP queries.
2068 weight=N can be used to add bias.
2070 weighted-round-robin
2071 Load-Balance parents which should be used in a round-robin
2072 fashion with the frequency of each parent being based on the
2073 round trip time. Closer parents are used more often.
2074 Usually used for background-ping parents.
2075 weight=N can be used to add bias.
2077 carp Load-Balance parents which should be used as a CARP array.
2078 The requests will be distributed among the parents based on the
2079 CARP load balancing hash function based on their weight.
2081 userhash Load-balance parents based on the client proxy_auth or ident username.
2083 sourcehash Load-balance parents based on the client source IP.
2086 To be used only for cache peers of type "multicast".
2087 ALL members of this multicast group have "sibling"
2088 relationship with it, not "parent". This is to a mulicast
2089 group when the requested object would be fetched only from
2090 a "parent" cache, anyway. It's useful, e.g., when
2091 configuring a pool of redundant Squid proxies, being
2092 members of the same multicast group.
2095 ==== PEER SELECTION OPTIONS ====
2097 weight=N use to affect the selection of a peer during any weighted
2098 peer-selection mechanisms.
2099 The weight must be an integer; default is 1,
2100 larger weights are favored more.
2101 This option does not affect parent selection if a peering
2102 protocol is not in use.
2104 basetime=N Specify a base amount to be subtracted from round trip
2106 It is subtracted before division by weight in calculating
2107 which parent to fectch from. If the rtt is less than the
2108 base time the rtt is set to a minimal value.
2110 ttl=N Specify a IP multicast TTL to use when sending an ICP
2111 queries to this address.
2112 Only useful when sending to a multicast group.
2113 Because we don't accept ICP replies from random
2114 hosts, you must configure other group members as
2115 peers with the 'multicast-responder' option.
2117 no-delay To prevent access to this neighbor from influencing the
2120 digest-url=URL Tell Squid to fetch the cache digest (if digests are
2121 enabled) for this host from the specified URL rather
2122 than the Squid default location.
2125 ==== ACCELERATOR / REVERSE-PROXY OPTIONS ====
2127 originserver Causes this parent to be contacted as an origin server.
2128 Meant to be used in accelerator setups when the peer
2132 Set the Host header of requests forwarded to this peer.
2133 Useful in accelerator setups where the server (peer)
2134 expects a certain domain name but clients may request
2135 others. ie example.com or www.example.com
2137 no-digest Disable request of cache digests.
2140 Disables requesting ICMP RTT database (NetDB).
2143 ==== AUTHENTICATION OPTIONS ====
2146 If this is a personal/workgroup proxy and your parent
2147 requires proxy authentication.
2149 Note: The string can include URL escapes (i.e. %20 for
2150 spaces). This also means % must be written as %%.
2153 Send login details received from client to this peer.
2154 Both Proxy- and WWW-Authorization headers are passed
2155 without alteration to the peer.
2156 Authentication is not required by Squid for this to work.
2158 Note: This will pass any form of authentication but
2159 only Basic auth will work through a proxy unless the
2160 connection-auth options are also used.
2162 login=PASS Send login details received from client to this peer.
2163 Authentication is not required by this option.
2165 If there are no client-provided authentication headers
2166 to pass on, but username and password are available
2167 from an external ACL user= and password= result tags
2168 they may be sent instead.
2170 Note: To combine this with proxy_auth both proxies must
2171 share the same user database as HTTP only allows for
2172 a single login (one for proxy, one for origin server).
2173 Also be warned this will expose your users proxy
2174 password to the peer. USE WITH CAUTION
2177 Send the username to the upstream cache, but with a
2178 fixed password. This is meant to be used when the peer
2179 is in another administrative domain, but it is still
2180 needed to identify each user.
2181 The star can optionally be followed by some extra
2182 information which is added to the username. This can
2183 be used to identify this proxy to the peer, similar to
2184 the login=username:password option above.
2187 If this is a personal/workgroup proxy and your parent
2188 requires a secure proxy authentication.
2189 The first principal from the default keytab or defined by
2190 the environment variable KRB5_KTNAME will be used.
2192 login=NEGOTIATE:principal_name
2193 If this is a personal/workgroup proxy and your parent
2194 requires a secure proxy authentication.
2195 The principal principal_name from the default keytab or
2196 defined by the environment variable KRB5_KTNAME will be
2199 connection-auth=on|off
2200 Tell Squid that this peer does or not support Microsoft
2201 connection oriented authentication, and any such
2202 challenges received from there should be ignored.
2203 Default is auto to automatically determine the status
2207 ==== SSL / HTTPS / TLS OPTIONS ====
2209 ssl Encrypt connections to this peer with SSL/TLS.
2211 sslcert=/path/to/ssl/certificate
2212 A client SSL certificate to use when connecting to
2215 sslkey=/path/to/ssl/key
2216 The private SSL key corresponding to sslcert above.
2217 If 'sslkey' is not specified 'sslcert' is assumed to
2218 reference a combined file containing both the
2219 certificate and the key.
2222 The SSL version to use when connecting to this peer
2223 1 = automatic (default)
2228 sslcipher=... The list of valid SSL ciphers to use when connecting
2231 ssloptions=... Specify various SSL engine options:
2232 NO_SSLv2 Disallow the use of SSLv2
2233 NO_SSLv3 Disallow the use of SSLv3
2234 NO_TLSv1 Disallow the use of TLSv1
2235 See src/ssl_support.c or the OpenSSL documentation for
2236 a more complete list.
2238 sslcafile=... A file containing additional CA certificates to use
2239 when verifying the peer certificate.
2241 sslcapath=... A directory containing additional CA certificates to
2242 use when verifying the peer certificate.
2244 sslcrlfile=... A certificate revocation list file to use when
2245 verifying the peer certificate.
2247 sslflags=... Specify various flags modifying the SSL implementation:
2250 Accept certificates even if they fail to
2253 Don't use the default CA list built in
2256 Don't verify the peer certificate
2257 matches the server name
2259 ssldomain= The peer name as advertised in it's certificate.
2260 Used for verifying the correctness of the received peer
2261 certificate. If not specified the peer hostname will be
2265 Enable the "Front-End-Https: On" header needed when
2266 using Squid as a SSL frontend in front of Microsoft OWA.
2267 See MS KB document Q307347 for details on this header.
2268 If set to auto the header will only be added if the
2269 request is forwarded as a https:// URL.
2272 ==== GENERAL OPTIONS ====
2275 A peer-specific connect timeout.
2276 Also see the peer_connect_timeout directive.
2278 connect-fail-limit=N
2279 How many times connecting to a peer must fail before
2280 it is marked as down. Default is 10.
2282 allow-miss Disable Squid's use of only-if-cached when forwarding
2283 requests to siblings. This is primarily useful when
2284 icp_hit_stale is used by the sibling. To extensive use
2285 of this option may result in forwarding loops, and you
2286 should avoid having two-way peerings with this option.
2287 For example to deny peer usage on requests from peer
2288 by denying cache_peer_access if the source is a peer.
2290 max-conn=N Limit the amount of connections Squid may open to this
2293 name=xxx Unique name for the peer.
2294 Required if you have multiple peers on the same host
2295 but different ports.
2296 This name can be used in cache_peer_access and similar
2297 directives to dentify the peer.
2298 Can be used by outgoing access controls through the
2301 no-tproxy Do not use the client-spoof TPROXY support when forwarding
2302 requests to this peer. Use normal address selection instead.
2304 proxy-only objects fetched from the peer will not be stored locally.
2308 NAME: cache_peer_domain cache_host_domain
2313 Use to limit the domains for which a neighbor cache will be
2316 cache_peer_domain cache-host domain [domain ...]
2317 cache_peer_domain cache-host !domain
2319 For example, specifying
2321 cache_peer_domain parent.foo.net .edu
2323 has the effect such that UDP query packets are sent to
2324 'bigserver' only when the requested object exists on a
2325 server in the .edu domain. Prefixing the domainname
2326 with '!' means the cache will be queried for objects
2329 NOTE: * Any number of domains may be given for a cache-host,
2330 either on the same or separate lines.
2331 * When multiple domains are given for a particular
2332 cache-host, the first matched domain is applied.
2333 * Cache hosts with no domain restrictions are queried
2335 * There are no defaults.
2336 * There is also a 'cache_peer_access' tag in the ACL
2340 NAME: cache_peer_access
2345 Similar to 'cache_peer_domain' but provides more flexibility by
2348 cache_peer_access cache-host allow|deny [!]aclname ...
2350 The syntax is identical to 'http_access' and the other lists of
2351 ACL elements. See the comments for 'http_access' below, or
2352 the Squid FAQ (http://wiki.squid-cache.org/SquidFaq/SquidAcl).
2355 NAME: neighbor_type_domain
2356 TYPE: hostdomaintype
2360 usage: neighbor_type_domain neighbor parent|sibling domain domain ...
2362 Modifying the neighbor type for specific domains is now
2363 possible. You can treat some domains differently than the
2364 default neighbor type specified on the 'cache_peer' line.
2365 Normally it should only be necessary to list domains which
2366 should be treated differently because the default neighbor type
2367 applies for hostnames which do not match domains listed here.
2370 cache_peer cache.foo.org parent 3128 3130
2371 neighbor_type_domain cache.foo.org sibling .com .net
2372 neighbor_type_domain cache.foo.org sibling .au .de
2375 NAME: dead_peer_timeout
2379 LOC: Config.Timeout.deadPeer
2381 This controls how long Squid waits to declare a peer cache
2382 as "dead." If there are no ICP replies received in this
2383 amount of time, Squid will declare the peer dead and not
2384 expect to receive any further ICP replies. However, it
2385 continues to send ICP queries, and will mark the peer as
2386 alive upon receipt of the first subsequent ICP reply.
2388 This timeout also affects when Squid expects to receive ICP
2389 replies from peers. If more than 'dead_peer' seconds have
2390 passed since the last ICP reply was received, Squid will not
2391 expect to receive an ICP reply on the next query. Thus, if
2392 your time between requests is greater than this timeout, you
2393 will see a lot of requests sent DIRECT to origin servers
2394 instead of to your parents.
2397 NAME: forward_max_tries
2400 LOC: Config.forward_max_tries
2402 Controls how many different forward paths Squid will try
2403 before giving up. See also forward_timeout.
2406 NAME: hierarchy_stoplist
2409 LOC: Config.hierarchy_stoplist
2411 A list of words which, if found in a URL, cause the object to
2412 be handled directly by this cache. In other words, use this
2413 to not query neighbor caches for certain objects. You may
2414 list this option multiple times.
2415 Note: never_direct overrides this option.
2418 # We recommend you to use at least the following line.
2419 hierarchy_stoplist cgi-bin ?
2424 MEMORY CACHE OPTIONS
2425 -----------------------------------------------------------------------------
2432 LOC: Config.memMaxSize
2434 NOTE: THIS PARAMETER DOES NOT SPECIFY THE MAXIMUM PROCESS SIZE.
2435 IT ONLY PLACES A LIMIT ON HOW MUCH ADDITIONAL MEMORY SQUID WILL
2436 USE AS A MEMORY CACHE OF OBJECTS. SQUID USES MEMORY FOR OTHER
2437 THINGS AS WELL. SEE THE SQUID FAQ SECTION 8 FOR DETAILS.
2439 'cache_mem' specifies the ideal amount of memory to be used
2441 * In-Transit objects
2443 * Negative-Cached objects
2445 Data for these objects are stored in 4 KB blocks. This
2446 parameter specifies the ideal upper limit on the total size of
2447 4 KB blocks allocated. In-Transit objects take the highest
2450 In-transit objects have priority over the others. When
2451 additional space is needed for incoming data, negative-cached
2452 and hot objects will be released. In other words, the
2453 negative-cached and hot objects will fill up any unused space
2454 not needed for in-transit objects.
2456 If circumstances require, this limit will be exceeded.
2457 Specifically, if your incoming request rate requires more than
2458 'cache_mem' of memory to hold in-transit objects, Squid will
2459 exceed this limit to satisfy the new requests. When the load
2460 decreases, blocks will be freed until the high-water mark is
2461 reached. Thereafter, blocks will be used to store hot
2465 NAME: maximum_object_size_in_memory
2469 LOC: Config.Store.maxInMemObjSize
2471 Objects greater than this size will not be attempted to kept in
2472 the memory cache. This should be set high enough to keep objects
2473 accessed frequently in memory to improve performance whilst low
2474 enough to keep larger objects from hoarding cache_mem.
2477 NAME: memory_cache_mode
2482 Controls which objects to keep in the memory cache (cache_mem)
2484 always Keep most recently fetched objects in memory (default)
2486 disk Only disk cache hits are kept in memory, which means
2487 an object must first be cached on disk and then hit
2488 a second time before cached in memory.
2490 network Only objects fetched from network is kept in memory
2493 NAME: memory_replacement_policy
2495 LOC: Config.memPolicy
2498 The memory replacement policy parameter determines which
2499 objects are purged from memory when memory space is needed.
2501 See cache_replacement_policy for details.
2506 -----------------------------------------------------------------------------
2509 NAME: cache_replacement_policy
2511 LOC: Config.replPolicy
2514 The cache replacement policy parameter determines which
2515 objects are evicted (replaced) when disk space is needed.
2517 lru : Squid's original list based LRU policy
2518 heap GDSF : Greedy-Dual Size Frequency
2519 heap LFUDA: Least Frequently Used with Dynamic Aging
2520 heap LRU : LRU policy implemented using a heap
2522 Applies to any cache_dir lines listed below this.
2524 The LRU policies keeps recently referenced objects.
2526 The heap GDSF policy optimizes object hit rate by keeping smaller
2527 popular objects in cache so it has a better chance of getting a
2528 hit. It achieves a lower byte hit rate than LFUDA though since
2529 it evicts larger (possibly popular) objects.
2531 The heap LFUDA policy keeps popular objects in cache regardless of
2532 their size and thus optimizes byte hit rate at the expense of
2533 hit rate since one large, popular object will prevent many
2534 smaller, slightly less popular objects from being cached.
2536 Both policies utilize a dynamic aging mechanism that prevents
2537 cache pollution that can otherwise occur with frequency-based
2538 replacement policies.
2540 NOTE: if using the LFUDA replacement policy you should increase
2541 the value of maximum_object_size above its default of 4096 KB to
2542 to maximize the potential byte hit rate improvement of LFUDA.
2544 For more information about the GDSF and LFUDA cache replacement
2545 policies see http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html
2546 and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html.
2552 LOC: Config.cacheSwap
2556 cache_dir Type Directory-Name Fs-specific-data [options]
2558 You can specify multiple cache_dir lines to spread the
2559 cache among different disk partitions.
2561 Type specifies the kind of storage system to use. Only "ufs"
2562 is built by default. To enable any of the other storage systems
2563 see the --enable-storeio configure option.
2565 'Directory' is a top-level directory where cache swap
2566 files will be stored. If you want to use an entire disk
2567 for caching, this can be the mount-point directory.
2568 The directory must exist and be writable by the Squid
2569 process. Squid will NOT create this directory for you.
2571 In SMP configurations, cache_dir must not precede the workers option
2572 and should use configuration macros or conditionals to give each
2573 worker interested in disk caching a dedicated cache directory.
2577 "ufs" is the old well-known Squid storage format that has always
2580 cache_dir ufs Directory-Name Mbytes L1 L2 [options]
2582 'Mbytes' is the amount of disk space (MB) to use under this
2583 directory. The default is 100 MB. Change this to suit your
2584 configuration. Do NOT put the size of your disk drive here.
2585 Instead, if you want Squid to use the entire disk drive,
2586 subtract 20% and use that value.
2588 'L1' is the number of first-level subdirectories which
2589 will be created under the 'Directory'. The default is 16.
2591 'L2' is the number of second-level subdirectories which
2592 will be created under each first-level directory. The default
2595 The aufs store type:
2597 "aufs" uses the same storage format as "ufs", utilizing
2598 POSIX-threads to avoid blocking the main Squid process on
2599 disk-I/O. This was formerly known in Squid as async-io.
2601 cache_dir aufs Directory-Name Mbytes L1 L2 [options]
2603 see argument descriptions under ufs above
2605 The diskd store type:
2607 "diskd" uses the same storage format as "ufs", utilizing a
2608 separate process to avoid blocking the main Squid process on
2611 cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]
2613 see argument descriptions under ufs above
2615 Q1 specifies the number of unacknowledged I/O requests when Squid
2616 stops opening new files. If this many messages are in the queues,
2617 Squid won't open new files. Default is 64
2619 Q2 specifies the number of unacknowledged messages when Squid
2620 starts blocking. If this many messages are in the queues,
2621 Squid blocks until it receives some replies. Default is 72
2623 When Q1 < Q2 (the default), the cache directory is optimized
2624 for lower response time at the expense of a decrease in hit
2625 ratio. If Q1 > Q2, the cache directory is optimized for
2626 higher hit ratio at the expense of an increase in response
2629 The coss store type:
2631 NP: COSS filesystem in Squid-3 has been deemed too unstable for
2632 production use and has thus been removed from this release.
2633 We hope that it can be made usable again soon.
2635 block-size=n defines the "block size" for COSS cache_dir's.
2636 Squid uses file numbers as block numbers. Since file numbers
2637 are limited to 24 bits, the block size determines the maximum
2638 size of the COSS partition. The default is 512 bytes, which
2639 leads to a maximum cache_dir size of 512<<24, or 8 GB. Note
2640 you should not change the coss block size after Squid
2641 has written some objects to the cache_dir.
2643 The coss file store has changed from 2.5. Now it uses a file
2644 called 'stripe' in the directory names in the config - and
2645 this will be created by squid -z.
2649 no-store, no new objects should be stored to this cache_dir
2651 max-size=n, refers to the max object size this storedir supports.
2652 It is used to initially choose the storedir to dump the object.
2653 Note: To make optimal use of the max-size limits you should order
2654 the cache_dir lines with the smallest max-size value first and the
2655 ones with no max-size specification last.
2657 Note for coss, max-size must be less than COSS_MEMBUF_SZ,
2658 which can be changed with the --with-coss-membuf-size=N configure
2662 # Uncomment and adjust the following to add a disk cache directory.
2663 #cache_dir ufs @DEFAULT_SWAP_DIR@ 100 16 256
2667 NAME: store_dir_select_algorithm
2669 LOC: Config.store_dir_select_algorithm
2672 Set this to 'round-robin' as an alternative.
2675 NAME: max_open_disk_fds
2677 LOC: Config.max_open_disk_fds
2680 To avoid having disk as the I/O bottleneck Squid can optionally
2681 bypass the on-disk cache if more than this amount of disk file
2682 descriptors are open.
2684 A value of 0 indicates no limit.
2687 NAME: minimum_object_size
2691 LOC: Config.Store.minObjectSize
2693 Objects smaller than this size will NOT be saved on disk. The
2694 value is specified in kilobytes, and the default is 0 KB, which
2695 means there is no minimum.
2698 NAME: maximum_object_size
2702 LOC: Config.Store.maxObjectSize
2704 Objects larger than this size will NOT be saved on disk. The
2705 value is specified in kilobytes, and the default is 4MB. If
2706 you wish to get a high BYTES hit ratio, you should probably
2707 increase this (one 32 MB object hit counts for 3200 10KB
2708 hits). If you wish to increase speed more than your want to
2709 save bandwidth you should leave this low.
2711 NOTE: if using the LFUDA replacement policy you should increase
2712 this value to maximize the byte hit rate improvement of LFUDA!
2713 See replacement_policy below for a discussion of this policy.
2716 NAME: cache_swap_low
2717 COMMENT: (percent, 0-100)
2720 LOC: Config.Swap.lowWaterMark
2723 NAME: cache_swap_high
2724 COMMENT: (percent, 0-100)
2727 LOC: Config.Swap.highWaterMark
2730 The low- and high-water marks for cache object replacement.
2731 Replacement begins when the swap (disk) usage is above the
2732 low-water mark and attempts to maintain utilization near the
2733 low-water mark. As swap utilization gets close to high-water
2734 mark object eviction becomes more aggressive. If utilization is
2735 close to the low-water mark less replacement is done each time.
2737 Defaults are 90% and 95%. If you have a large cache, 5% could be
2738 hundreds of MB. If this is the case you may wish to set these
2739 numbers closer together.
2744 -----------------------------------------------------------------------------
2749 LOC: Config.Log.logformats
2754 logformat <name> <format specification>
2756 Defines an access log format.
2758 The <format specification> is a string with embedded % format codes
2760 % format codes all follow the same basic structure where all but
2761 the formatcode is optional. Output strings are automatically escaped
2762 as required according to their context and the output format
2763 modifiers are usually not needed, but can be specified if an explicit
2764 output format is desired.
2766 % ["|[|'|#] [-] [[0]width] [{argument}] formatcode
2768 " output in quoted string format
2769 [ output in squid text log format as used by log_mime_hdrs
2770 # output in URL quoted format
2774 width field width. If starting with 0 the
2775 output is zero padded
2776 {arg} argument such as header name etc
2780 % a literal % character
2781 sn Unique sequence number per log line entry
2782 err_code The ID of an error response served by Squid or
2783 a similar internal error identifier.
2784 err_detail Additional err_code-dependent error information.
2786 Connection related format codes:
2788 >a Client source IP address
2790 >p Client source port
2791 <A Server IP address or peer name
2792 la Local IP address (http_port)
2793 lp Local port number (http_port)
2794 <lp Local port number of the last server or peer connection
2796 Time related format codes:
2798 ts Seconds since epoch
2799 tu subsecond time (milliseconds)
2800 tl Local time. Optional strftime format argument
2801 default %d/%b/%Y:%H:%M:%S %z
2802 tg GMT time. Optional strftime format argument
2803 default %d/%b/%Y:%H:%M:%S %z
2804 tr Response time (milliseconds)
2805 dt Total time spent making DNS lookups (milliseconds)
2807 HTTP cache related format codes:
2809 [http::]>h Original request header. Optional header name argument
2810 on the format header[:[separator]element]
2811 [http::]>ha The HTTP request headers after adaptation and redirection.
2812 Optional header name argument as for >h
2813 [http::]<h Reply header. Optional header name argument
2815 [http::]un User name
2816 [http::]ul User name from authentication
2817 [http::]ui User name from ident
2818 [http::]us User name from SSL
2819 [http::]ue User name from external acl helper
2820 [http::]>Hs HTTP status code sent to the client
2821 [http::]<Hs HTTP status code received from the next hop
2822 [http::]<bs Number of HTTP-equivalent message body bytes
2823 received from the next hop, excluding chunked
2824 transfer encoding and control messages.
2825 Generated FTP/Gopher listings are treated as
2827 [http::]Ss Squid request status (TCP_MISS etc)
2828 [http::]Sh Squid hierarchy status (DEFAULT_PARENT etc)
2829 [http::]mt MIME content type
2830 [http::]rm Request method (GET/POST etc)
2831 [http::]ru Request URL
2832 [http::]rp Request URL-Path excluding hostname
2833 [http::]rv Request protocol version
2834 [http::]et Tag returned by external acl
2835 [http::]ea Log string returned by external acl
2836 [http::]<st Sent reply size including HTTP headers
2837 [http::]>st Received request size including HTTP headers. In the
2838 case of chunked requests the chunked encoding metadata
2840 [http::]>sh Received HTTP request headers size
2841 [http::]<sh Sent HTTP reply headers size
2842 [http::]st Request+Reply size including HTTP headers
2843 [http::]<sH Reply high offset sent
2844 [http::]<sS Upstream object size
2845 [http::]<pt Peer response time in milliseconds. The timer starts
2846 when the last request byte is sent to the next hop
2847 and stops when the last response byte is received.
2848 [http::]<tt Total server-side time in milliseconds. The timer
2849 starts with the first connect request (or write I/O)
2850 sent to the first selected peer. The timer stops
2851 with the last I/O with the last peer.
2853 If ICAP is enabled, the following two codes become available (as
2854 well as ICAP log codes documented with the icap_log option):
2856 icap::tt Total ICAP processing time for the HTTP
2857 transaction. The timer ticks when ICAP
2858 ACLs are checked and when ICAP
2859 transaction is in progress.
2861 icap::<last_h The header of the last ICAP response
2862 related to the HTTP transaction. Like
2863 <h, accepts an optional header name
2864 argument. Will not change semantics
2865 when multiple ICAP transactions per HTTP
2866 transaction are supported.
2868 If adaptation is enabled the following two codes become available:
2870 adapt::sum_trs Summed adaptation transaction response
2871 times recorded as a comma-separated list in
2872 the order of transaction start time. Each time
2873 value is recorded as an integer number,
2874 representing response time of one or more
2875 adaptation (ICAP or eCAP) transaction in
2876 milliseconds. When a failed transaction is
2877 being retried or repeated, its time is not
2878 logged individually but added to the
2879 replacement (next) transaction. See also:
2882 adapt::all_trs All adaptation transaction response times.
2883 Same as adaptation_strs but response times of
2884 individual transactions are never added
2885 together. Instead, all transaction response
2886 times are recorded individually.
2888 You can prefix adapt::*_trs format codes with adaptation
2889 service name in curly braces to record response time(s) specific
2890 to that service. For example: %{my_service}adapt::sum_trs
2892 The default formats available (which do not need re-defining) are:
2894 logformat squid %ts.%03tu %6tr %>a %Ss/%03>Hs %<st %rm %ru %un %Sh/%<A %mt
2895 logformat squidmime %ts.%03tu %6tr %>a %Ss/%03>Hs %<st %rm %ru %un %Sh/%<A %mt [%>h] [%<h]
2896 logformat common %>a %ui %un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st %Ss:%Sh
2897 logformat combined %>a %ui %un [%tl] "%rm %ru HTTP/%rv" %>Hs %<st "%{Referer}>h" "%{User-Agent}>h" %Ss:%Sh
2900 NAME: access_log cache_access_log
2902 LOC: Config.Log.accesslogs
2903 DEFAULT_IF_NONE: daemon:@DEFAULT_ACCESS_LOG@ squid
2905 These files log client request activities. Has a line every HTTP or
2906 ICP request. The format is:
2907 access_log <module>:<place> [<logformat name> [acl acl ...]]
2908 access_log none [acl acl ...]]
2910 Will log to the specified module:place using the specified format (which
2911 must be defined in a logformat directive) those entries which match
2912 ALL the acl's specified (which must be defined in acl clauses).
2913 If no acl is specified, all requests will be logged to this destination.
2915 ===== Modules Currently available =====
2917 none Do not log any requests matching these ACL.
2918 Do not specify Place or logformat name.
2920 stdio Write each log line to disk immediately at the completion of
2922 Place: the filename and path to be written.
2924 daemon Very similar to stdio. But instead of writing to disk the log
2925 line is passed to a daemon helper for asychronous handling instead.
2926 Place: varies depending on the daemon.
2928 log_file_daemon Place: the file name and path to be written.
2930 syslog To log each request via syslog facility.
2931 Place: The syslog facility and priority level for these entries.
2932 Place Format: facility.priority
2934 where facility could be any of:
2935 authpriv, daemon, local0 ... local7 or user.
2937 And priority could be any of:
2938 err, warning, notice, info, debug.
2940 udp To send each log line as text data to a UDP receiver.
2941 Place: The destination host name or IP and port.
2942 Place Format: \\host:port
2944 tcp To send each log line as text data to a TCP receiver.
2945 Place: The destination host name or IP and port.
2946 Place Format: \\host:port
2949 access_log daemon:@DEFAULT_ACCESS_LOG@ squid
2955 LOC: Config.Log.icaplogs
2958 ICAP log files record ICAP transaction summaries, one line per
2961 The icap_log option format is:
2962 icap_log <filepath> [<logformat name> [acl acl ...]]
2963 icap_log none [acl acl ...]]
2965 Please see access_log option documentation for details. The two
2966 kinds of logs share the overall configuration approach and many
2969 ICAP processing of a single HTTP message or transaction may
2970 require multiple ICAP transactions. In such cases, multiple
2971 ICAP transaction log lines will correspond to a single access
2974 ICAP log uses logformat codes that make sense for an ICAP
2975 transaction. Header-related codes are applied to the HTTP header
2976 embedded in an ICAP server response, with the following caveats:
2977 For REQMOD, there is no HTTP response header unless the ICAP
2978 server performed request satisfaction. For RESPMOD, the HTTP
2979 request header is the header sent to the ICAP server. For
2980 OPTIONS, there are no HTTP headers.
2982 The following format codes are also available for ICAP logs:
2984 icap::<A ICAP server IP address. Similar to <A.
2986 icap::<service_name ICAP service name from the icap_service
2987 option in Squid configuration file.
2989 icap::ru ICAP Request-URI. Similar to ru.
2991 icap::rm ICAP request method (REQMOD, RESPMOD, or
2992 OPTIONS). Similar to existing rm.
2994 icap::>st Bytes sent to the ICAP server (TCP payload
2995 only; i.e., what Squid writes to the socket).
2997 icap::<st Bytes received from the ICAP server (TCP
2998 payload only; i.e., what Squid reads from
3001 icap::<bs Number of message body bytes received from the
3002 ICAP server. ICAP message body, if any, usually
3003 includes encapsulated HTTP message headers and
3004 possibly encapsulated HTTP message body. The
3005 HTTP body part is dechunked before its size is
3008 icap::tr Transaction response time (in
3009 milliseconds). The timer starts when
3010 the ICAP transaction is created and
3011 stops when the transaction is completed.
3014 icap::tio Transaction I/O time (in milliseconds). The
3015 timer starts when the first ICAP request
3016 byte is scheduled for sending. The timers
3017 stops when the last byte of the ICAP response
3020 icap::to Transaction outcome: ICAP_ERR* for all
3021 transaction errors, ICAP_OPT for OPTION
3022 transactions, ICAP_ECHO for 204
3023 responses, ICAP_MOD for message
3024 modification, and ICAP_SAT for request
3025 satisfaction. Similar to Ss.
3027 icap::Hs ICAP response status code. Similar to Hs.
3029 icap::>h ICAP request header(s). Similar to >h.
3031 icap::<h ICAP response header(s). Similar to <h.
3033 The default ICAP log format, which can be used without an explicit
3034 definition, is called icap_squid:
3036 logformat icap_squid %ts.%03tu %6icap::tr %>a %icap::to/%03icap::Hs %icap::<size %icap::rm %icap::ru% %un -/%icap::<A -
3038 See also: logformat, log_icap, and %icap::<last_h
3041 NAME: logfile_daemon
3043 DEFAULT: @DEFAULT_LOGFILED@
3044 LOC: Log::TheConfig.logfile_daemon
3046 Specify the path to the logfile-writing daemon. This daemon is
3047 used to write the access and store logs, if configured.
3049 Squid sends a number of commands to the log daemon:
3050 L<data>\n - logfile data
3055 r<n>\n - set rotate count to <n>
3056 b<n>\n - 1 = buffer output, 0 = don't buffer output
3058 No responses is expected.
3063 LOC: Config.accessList.log
3065 COMMENT: allow|deny acl acl...
3067 This options allows you to control which requests gets logged
3068 to access.log (see access_log directive). Requests denied for
3069 logging will also not be accounted for in performance counters.
3071 This clause only supports fast acl types.
3072 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
3078 LOC: Config.accessList.icap
3081 This options allows you to control which requests get logged
3082 to icap.log. See the icap_log directive for ICAP log details.
3085 NAME: cache_store_log
3088 LOC: Config.Log.store
3090 Logs the activities of the storage manager. Shows which
3091 objects are ejected from the cache, and which objects are
3092 saved and for how long. To disable, enter "none" or remove the line.
3093 There are not really utilities to analyze this data, so you can safely
3097 cache_store_log @DEFAULT_STORE_LOG@
3100 NAME: cache_swap_state cache_swap_log
3102 LOC: Config.Log.swap
3105 Location for the cache "swap.state" file. This index file holds
3106 the metadata of objects saved on disk. It is used to rebuild
3107 the cache during startup. Normally this file resides in each
3108 'cache_dir' directory, but you may specify an alternate
3109 pathname here. Note you must give a full filename, not just
3110 a directory. Since this is the index for the whole object
3111 list you CANNOT periodically rotate it!
3113 If %s can be used in the file name it will be replaced with a
3114 a representation of the cache_dir name where each / is replaced
3115 with '.'. This is needed to allow adding/removing cache_dir
3116 lines when cache_swap_log is being used.
3118 If have more than one 'cache_dir', and %s is not used in the name
3119 these swap logs will have names such as:
3125 The numbered extension (which is added automatically)
3126 corresponds to the order of the 'cache_dir' lines in this
3127 configuration file. If you change the order of the 'cache_dir'
3128 lines in this file, these index files will NOT correspond to
3129 the correct 'cache_dir' entry (unless you manually rename
3130 them). We recommend you do NOT use this option. It is
3131 better to keep these index files in each 'cache_dir' directory.
3134 NAME: logfile_rotate
3137 LOC: Config.Log.rotateNumber
3139 Specifies the number of logfile rotations to make when you
3140 type 'squid -k rotate'. The default is 10, which will rotate
3141 with extensions 0 through 9. Setting logfile_rotate to 0 will
3142 disable the file name rotation, but the logfiles are still closed
3143 and re-opened. This will enable you to rename the logfiles
3144 yourself just before sending the rotate signal.
3146 Note, the 'squid -k rotate' command normally sends a USR1
3147 signal to the running squid process. In certain situations
3148 (e.g. on Linux with Async I/O), USR1 is used for other
3149 purposes, so -k rotate uses another signal. It is best to get
3150 in the habit of using 'squid -k rotate' instead of 'kill -USR1
3153 Note, from Squid-3.1 this option has no effect on the cache.log,
3154 that log can be rotated separately by using debug_options
3157 NAME: emulate_httpd_log
3161 LOC: Config.onoff.common_log
3163 The Cache can emulate the log file format which many 'httpd'
3164 programs use. To disable/enable this emulation, set
3165 emulate_httpd_log to 'off' or 'on'. The default
3166 is to use the native log format since it includes useful
3167 information Squid-specific log analyzers use.
3170 NAME: log_ip_on_direct
3174 LOC: Config.onoff.log_ip_on_direct
3176 Log the destination IP address in the hierarchy log tag when going
3177 direct. Earlier Squid versions logged the hostname here. If you
3178 prefer the old way set this to off.
3183 DEFAULT: @DEFAULT_MIME_TABLE@
3184 LOC: Config.mimeTablePathname
3186 Pathname to Squid's MIME table. You shouldn't need to change
3187 this, but the default file contains examples and formatting
3188 information if you do.
3194 LOC: Config.onoff.log_mime_hdrs
3197 The Cache can record both the request and the response MIME
3198 headers for each HTTP transaction. The headers are encoded
3199 safely and will appear as two bracketed fields at the end of
3200 the access log (for either the native or httpd-emulated log
3201 formats). To enable this logging set log_mime_hdrs to 'on'.
3206 LOC: Config.Log.useragent
3208 IFDEF: USE_USERAGENT_LOG
3210 Squid will write the User-Agent field from HTTP requests
3211 to the filename specified here. By default useragent_log
3215 NAME: referer_log referrer_log
3217 LOC: Config.Log.referer
3219 IFDEF: USE_REFERER_LOG
3221 Squid will write the Referer field from HTTP requests to the
3222 filename specified here. By default referer_log is disabled.
3223 Note that "referer" is actually a misspelling of "referrer"
3224 however the misspelt version has been accepted into the HTTP RFCs
3230 DEFAULT: @DEFAULT_PID_FILE@
3231 LOC: Config.pidFilename
3233 A filename to write the process-id to. To disable, enter "none".
3239 Remove this option from your config. To log FQDN use %>A in the log format.
3242 NAME: client_netmask
3244 LOC: Config.Addrs.client_netmask
3247 A netmask for client addresses in logfiles and cachemgr output.
3248 Change this to protect the privacy of your cache clients.
3249 A netmask of 255.255.255.0 will log all IP's in that range with
3250 the last digit set to '0'.
3257 LOC: Config.Log.forward
3259 Logs the server-side requests.
3261 This is currently work in progress.
3264 NAME: strip_query_terms
3266 LOC: Config.onoff.strip_query_terms
3269 By default, Squid strips query terms from requested URLs before
3270 logging. This protects your user's privacy.
3277 LOC: Config.onoff.buffered_logs
3279 cache.log log file is written with stdio functions, and as such
3280 it can be buffered or unbuffered. By default it will be unbuffered.
3281 Buffering it can speed up the writing slightly (though you are
3282 unlikely to need to worry unless you run with tons of debugging
3283 enabled in which case performance will suffer badly anyway..).
3286 NAME: netdb_filename
3288 DEFAULT: @DEFAULT_NETDB_FILE@
3289 LOC: Config.netdbFilename
3292 A filename where Squid stores it's netdb state between restarts.
3293 To disable, enter "none".
3297 OPTIONS FOR TROUBLESHOOTING
3298 -----------------------------------------------------------------------------
3303 DEFAULT_IF_NONE: @DEFAULT_CACHE_LOG@
3304 LOC: Debug::cache_log
3306 Cache logging file. This is where general information about
3307 your cache's behavior goes. You can increase the amount of data
3308 logged to this file and how often its rotated with "debug_options"
3314 LOC: Debug::debugOptions
3316 Logging options are set as section,level where each source file
3317 is assigned a unique section. Lower levels result in less
3318 output, Full debugging (level 9) can result in a very large
3319 log file, so be careful.
3321 The magic word "ALL" sets debugging levels for all sections.
3322 We recommend normally running with "ALL,1".
3324 The rotate=N option can be used to keep more or less of these logs
3325 than would otherwise be kept by logfile_rotate.
3326 For most uses a single log should be enough to monitor current
3327 events affecting Squid.
3332 LOC: Config.coredump_dir
3333 DEFAULT_IF_NONE: none
3335 By default Squid leaves core files in the directory from where
3336 it was started. If you set 'coredump_dir' to a directory
3337 that exists, Squid will chdir() to that directory at startup
3338 and coredump files will be left there.
3342 # Leave coredumps in the first cache dir
3343 coredump_dir @DEFAULT_SWAP_DIR@
3349 OPTIONS FOR FTP GATEWAYING
3350 -----------------------------------------------------------------------------
3356 LOC: Config.Ftp.anon_user
3358 If you want the anonymous login password to be more informative
3359 (and enable the use of picky ftp servers), set this to something
3360 reasonable for your domain, like wwwuser@somewhere.net
3362 The reason why this is domainless by default is the
3363 request can be made on the behalf of a user in any domain,
3364 depending on how the cache is used.
3365 Some ftp server also validate the email address is valid
3366 (for example perl.com).
3372 LOC: Config.Ftp.passive
3374 If your firewall does not allow Squid to use passive
3375 connections, turn off this option.
3377 Use of ftp_epsv_all option requires this to be ON.
3383 LOC: Config.Ftp.epsv_all
3385 FTP Protocol extensions permit the use of a special "EPSV ALL" command.
3387 NATs may be able to put the connection on a "fast path" through the
3388 translator, as the EPRT command will never be used and therefore,
3389 translation of the data portion of the segments will never be needed.
3391 When a client only expects to do two-way FTP transfers this may be
3393 If squid finds that it must do a three-way FTP transfer after issuing
3394 an EPSV ALL command, the FTP session will fail.
3396 If you have any doubts about this option do not use it.
3397 Squid will nicely attempt all other connection methods.
3399 Requires ftp_passive to be ON (default) for any effect.
3405 LOC: Config.Ftp.epsv
3407 FTP Protocol extensions permit the use of a special "EPSV" command.
3409 NATs may be able to put the connection on a "fast path" through the
3410 translator using EPSV, as the EPRT command will never be used
3411 and therefore, translation of the data portion of the segments
3412 will never be needed.
3414 Turning this OFF will prevent EPSV being attempted.
3415 WARNING: Doing so will convert Squid back to the old behavior with all
3416 the related problems with external NAT devices/layers.
3418 Requires ftp_passive to be ON (default) for any effect.
3421 NAME: ftp_sanitycheck
3424 LOC: Config.Ftp.sanitycheck
3426 For security and data integrity reasons Squid by default performs
3427 sanity checks of the addresses of FTP data connections ensure the
3428 data connection is to the requested server. If you need to allow
3429 FTP connections to servers using another IP address for the data
3430 connection turn this off.
3433 NAME: ftp_telnet_protocol
3436 LOC: Config.Ftp.telnet
3438 The FTP protocol is officially defined to use the telnet protocol
3439 as transport channel for the control connection. However, many
3440 implementations are broken and does not respect this aspect of
3443 If you have trouble accessing files with ASCII code 255 in the
3444 path or similar problems involving this ASCII code you can
3445 try setting this directive to off. If that helps, report to the
3446 operator of the FTP server in question that their FTP server
3447 is broken and does not follow the FTP standard.
3451 OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
3452 -----------------------------------------------------------------------------
3457 DEFAULT: @DEFAULT_DISKD@
3458 LOC: Config.Program.diskd
3460 Specify the location of the diskd executable.
3461 Note this is only useful if you have compiled in
3462 diskd as one of the store io modules.
3465 NAME: unlinkd_program
3468 DEFAULT: @DEFAULT_UNLINKD@
3469 LOC: Config.Program.unlinkd
3471 Specify the location of the executable for file deletion process.
3474 NAME: pinger_program
3476 DEFAULT: @DEFAULT_PINGER@
3477 LOC: Config.pinger.program
3480 Specify the location of the executable for the pinger process.
3486 LOC: Config.pinger.enable
3489 Control whether the pinger is active at run-time.
3490 Enables turning ICMP pinger on and off with a simple
3491 squid -k reconfigure.
3496 OPTIONS FOR URL REWRITING
3497 -----------------------------------------------------------------------------
3500 NAME: url_rewrite_program redirect_program
3502 LOC: Config.Program.redirect
3505 Specify the location of the executable for the URL rewriter.
3506 Since they can perform almost any function there isn't one included.
3508 For each requested URL rewriter will receive on line with the format
3510 URL <SP> client_ip "/" fqdn <SP> user <SP> method [<SP> kvpairs]<NL>
3512 In the future, the rewriter interface will be extended with
3513 key=value pairs ("kvpairs" shown above). Rewriter programs
3514 should be prepared to receive and possibly ignore additional
3515 whitespace-separated tokens on each input line.
3517 And the rewriter may return a rewritten URL. The other components of
3518 the request line does not need to be returned (ignored if they are).
3520 The rewriter can also indicate that a client-side redirect should
3521 be performed to the new URL. This is done by prefixing the returned
3522 URL with "301:" (moved permanently) or 302: (moved temporarily).
3524 By default, a URL rewriter is not used.
3527 NAME: url_rewrite_children redirect_children
3528 TYPE: HelperChildConfig
3529 DEFAULT: 20 startup=0 idle=1 concurrency=0
3530 LOC: Config.redirectChildren
3532 The maximum number of redirector processes to spawn. If you limit
3533 it too few Squid will have to wait for them to process a backlog of
3534 URLs, slowing it down. If you allow too many they will use RAM
3535 and other system resources noticably.
3537 The startup= and idle= options allow some measure of skew in your
3542 Sets a minimum of how many processes are to be spawned when Squid
3543 starts or reconfigures. When set to zero the first request will
3544 cause spawning of the first child process to handle it.
3546 Starting too few will cause an initial slowdown in traffic as Squid
3547 attempts to simultaneously spawn enough processes to cope.
3551 Sets a minimum of how many processes Squid is to try and keep available
3552 at all times. When traffic begins to rise above what the existing
3553 processes can handle this many more will be spawned up to the maximum
3554 configured. A minimum setting of 1 is required.
3558 The number of requests each redirector helper can handle in
3559 parallel. Defaults to 0 which indicates the redirector
3560 is a old-style single threaded redirector.
3562 When this directive is set to a value >= 1 then the protocol
3563 used to communicate with the helper is modified to include
3564 a request ID in front of the request/response. The request
3565 ID from the request must be echoed back with the response
3569 NAME: url_rewrite_host_header redirect_rewrites_host_header
3572 LOC: Config.onoff.redir_rewrites_host
3574 By default Squid rewrites any Host: header in redirected
3575 requests. If you are running an accelerator this may
3576 not be a wanted effect of a redirector.
3578 WARNING: Entries are cached on the result of the URL rewriting
3579 process, so be careful if you have domain-virtual hosts.
3582 NAME: url_rewrite_access redirector_access
3585 LOC: Config.accessList.redirector
3587 If defined, this access list specifies which requests are
3588 sent to the redirector processes. By default all requests
3591 This clause supports both fast and slow acl types.
3592 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
3595 NAME: url_rewrite_bypass redirector_bypass
3597 LOC: Config.onoff.redirector_bypass
3600 When this is 'on', a request will not go through the
3601 redirector if all redirectors are busy. If this is 'off'
3602 and the redirector queue grows too large, Squid will exit
3603 with a FATAL error and ask you to increase the number of
3604 redirectors. You should only enable this if the redirectors
3605 are not critical to your caching system. If you use
3606 redirectors for access control, and you enable this option,
3607 users may have access to pages they should not
3608 be allowed to request.
3612 OPTIONS FOR TUNING THE CACHE
3613 -----------------------------------------------------------------------------
3616 NAME: cache no_cache
3619 LOC: Config.accessList.noCache
3621 A list of ACL elements which, if matched and denied, cause the request to
3622 not be satisfied from the cache and the reply to not be cached.
3623 In other words, use this to force certain objects to never be cached.
3625 You must use the words 'allow' or 'deny' to indicate whether items
3626 matching the ACL should be allowed or denied into the cache.
3628 Default is to allow all to be cached.
3630 This clause supports both fast and slow acl types.
3631 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
3634 NAME: refresh_pattern
3635 TYPE: refreshpattern
3639 usage: refresh_pattern [-i] regex min percent max [options]
3641 By default, regular expressions are CASE-SENSITIVE. To make
3642 them case-insensitive, use the -i option.
3644 'Min' is the time (in minutes) an object without an explicit
3645 expiry time should be considered fresh. The recommended
3646 value is 0, any higher values may cause dynamic applications
3647 to be erroneously cached unless the application designer
3648 has taken the appropriate actions.
3650 'Percent' is a percentage of the objects age (time since last
3651 modification age) an object without explicit expiry time
3652 will be considered fresh.
3654 'Max' is an upper limit on how long objects without an explicit
3655 expiry time will be considered fresh.
3657 options: override-expire
3663 ignore-must-revalidate
3669 override-expire enforces min age even if the server
3670 sent an explicit expiry time (e.g., with the
3671 Expires: header or Cache-Control: max-age). Doing this
3672 VIOLATES the HTTP standard. Enabling this feature
3673 could make you liable for problems which it causes.
3675 Note: override-expire does not enforce staleness - it only extends
3676 freshness / min. If the server returns a Expires time which
3677 is longer than your max time, Squid will still consider
3678 the object fresh for that period of time.
3680 override-lastmod enforces min age even on objects
3681 that were modified recently.
3683 reload-into-ims changes client no-cache or ``reload''
3684 to If-Modified-Since requests. Doing this VIOLATES the
3685 HTTP standard. Enabling this feature could make you
3686 liable for problems which it causes.
3688 ignore-reload ignores a client no-cache or ``reload''
3689 header. Doing this VIOLATES the HTTP standard. Enabling
3690 this feature could make you liable for problems which
3693 ignore-no-cache ignores any ``Pragma: no-cache'' and
3694 ``Cache-control: no-cache'' headers received from a server.
3695 The HTTP RFC never allows the use of this (Pragma) header
3696 from a server, only a client, though plenty of servers
3699 ignore-no-store ignores any ``Cache-control: no-store''
3700 headers received from a server. Doing this VIOLATES
3701 the HTTP standard. Enabling this feature could make you
3702 liable for problems which it causes.
3704 ignore-must-revalidate ignores any ``Cache-Control: must-revalidate``
3705 headers received from a server. Doing this VIOLATES
3706 the HTTP standard. Enabling this feature could make you
3707 liable for problems which it causes.
3709 ignore-private ignores any ``Cache-control: private''
3710 headers received from a server. Doing this VIOLATES
3711 the HTTP standard. Enabling this feature could make you
3712 liable for problems which it causes.
3714 ignore-auth caches responses to requests with authorization,
3715 as if the originserver had sent ``Cache-control: public''
3716 in the response header. Doing this VIOLATES the HTTP standard.
3717 Enabling this feature could make you liable for problems which
3720 refresh-ims causes squid to contact the origin server
3721 when a client issues an If-Modified-Since request. This
3722 ensures that the client will receive an updated version
3723 if one is available.
3725 store-stale stores responses even if they don't have explicit
3726 freshness or a validator (i.e., Last-Modified or an ETag)
3727 present, or if they're already stale. By default, Squid will
3728 not cache such responses because they usually can't be
3729 reused. Note that such responses will be stale by default.
3731 Basically a cached object is:
3733 FRESH if expires < now, else STALE
3735 FRESH if lm-factor < percent, else STALE
3739 The refresh_pattern lines are checked in the order listed here.
3740 The first entry which matches is used. If none of the entries
3741 match the default will be used.
3743 Note, you must uncomment all the default lines if you want
3744 to change one. The default setting is only active if none is
3749 # Add any of your own refresh_pattern entries above these.
3750 refresh_pattern ^ftp: 1440 20% 10080
3751 refresh_pattern ^gopher: 1440 0% 1440
3752 refresh_pattern -i (/cgi-bin/|\?) 0 0% 0
3753 refresh_pattern . 0 20% 4320
3757 NAME: quick_abort_min
3761 LOC: Config.quickAbort.min
3764 NAME: quick_abort_max
3768 LOC: Config.quickAbort.max
3771 NAME: quick_abort_pct
3775 LOC: Config.quickAbort.pct
3777 The cache by default continues downloading aborted requests
3778 which are almost completed (less than 16 KB remaining). This
3779 may be undesirable on slow (e.g. SLIP) links and/or very busy
3780 caches. Impatient users may tie up file descriptors and
3781 bandwidth by repeatedly requesting and immediately aborting
3784 When the user aborts a request, Squid will check the
3785 quick_abort values to the amount of data transfered until
3788 If the transfer has less than 'quick_abort_min' KB remaining,
3789 it will finish the retrieval.
3791 If the transfer has more than 'quick_abort_max' KB remaining,
3792 it will abort the retrieval.
3794 If more than 'quick_abort_pct' of the transfer has completed,
3795 it will finish the retrieval.
3797 If you do not want any retrieval to continue after the client
3798 has aborted, set both 'quick_abort_min' and 'quick_abort_max'
3801 If you want retrievals to always continue if they are being
3802 cached set 'quick_abort_min' to '-1 KB'.
3805 NAME: read_ahead_gap
3806 COMMENT: buffer-size
3808 LOC: Config.readAheadGap
3811 The amount of data the cache will buffer ahead of what has been
3812 sent to the client when retrieving an object from another server.
3816 IFDEF: USE_HTTP_VIOLATIONS
3819 LOC: Config.negativeTtl
3822 Set the Default Time-to-Live (TTL) for failed requests.
3823 Certain types of failures (such as "connection refused" and
3824 "404 Not Found") are able to be negatively-cached for a short time.
3825 Modern web servers should provide Expires: header, however if they
3826 do not this can provide a minimum TTL.
3827 The default is not to cache errors with unknown expiry details.
3829 Note that this is different from negative caching of DNS lookups.
3831 WARNING: Doing this VIOLATES the HTTP standard. Enabling
3832 this feature could make you liable for problems which it
3836 NAME: positive_dns_ttl
3839 LOC: Config.positiveDnsTtl
3842 Upper limit on how long Squid will cache positive DNS responses.
3843 Default is 6 hours (360 minutes). This directive must be set
3844 larger than negative_dns_ttl.
3847 NAME: negative_dns_ttl
3850 LOC: Config.negativeDnsTtl
3853 Time-to-Live (TTL) for negative caching of failed DNS lookups.
3854 This also sets the lower cache limit on positive lookups.
3855 Minimum value is 1 second, and it is not recommendable to go
3856 much below 10 seconds.
3859 NAME: range_offset_limit
3860 COMMENT: size [acl acl...]
3862 LOC: Config.rangeOffsetLimit
3865 usage: (size) [units] [[!]aclname]
3867 Sets an upper limit on how far (number of bytes) into the file
3868 a Range request may be to cause Squid to prefetch the whole file.
3869 If beyond this limit, Squid forwards the Range request as it is and
3870 the result is NOT cached.
3872 This is to stop a far ahead range request (lets say start at 17MB)
3873 from making Squid fetch the whole object up to that point before
3874 sending anything to the client.
3876 Multiple range_offset_limit lines may be specified, and they will
3877 be searched from top to bottom on each request until a match is found.
3878 The first match found will be used. If no line matches a request, the
3879 default limit of 0 bytes will be used.
3881 'size' is the limit specified as a number of units.
3883 'units' specifies whether to use bytes, KB, MB, etc.
3884 If no units are specified bytes are assumed.
3886 A size of 0 causes Squid to never fetch more than the
3887 client requested. (default)
3889 A size of 'none' causes Squid to always fetch the object from the
3890 beginning so it may cache the result. (2.0 style)
3892 'aclname' is the name of a defined ACL.
3894 NP: Using 'none' as the byte value here will override any quick_abort settings
3895 that may otherwise apply to the range request. The range request will
3896 be fully fetched from start to finish regardless of the client
3897 actions. This affects bandwidth usage.
3900 NAME: minimum_expiry_time
3903 LOC: Config.minimum_expiry_time
3906 The minimum caching time according to (Expires - Date)
3907 Headers Squid honors if the object can't be revalidated
3908 defaults to 60 seconds. In reverse proxy environments it
3909 might be desirable to honor shorter object lifetimes. It
3910 is most likely better to make your server return a
3911 meaningful Last-Modified header however. In ESI environments
3912 where page fragments often have short lifetimes, this will
3913 often be best set to 0.
3916 NAME: store_avg_object_size
3920 LOC: Config.Store.avgObjectSize
3922 Average object size, used to estimate number of objects your
3923 cache can hold. The default is 13 KB.
3926 NAME: store_objects_per_bucket
3929 LOC: Config.Store.objectsPerBucket
3931 Target number of objects per bucket in the store hash table.
3932 Lowering this value increases the total number of buckets and
3933 also the storage maintenance rate. The default is 20.
3938 -----------------------------------------------------------------------------
3941 NAME: request_header_max_size
3945 LOC: Config.maxRequestHeaderSize
3947 This specifies the maximum size for HTTP headers in a request.
3948 Request headers are usually relatively small (about 512 bytes).
3949 Placing a limit on the request header size will catch certain
3950 bugs (for example with persistent connections) and possibly
3951 buffer-overflow or denial-of-service attacks.
3954 NAME: reply_header_max_size
3958 LOC: Config.maxReplyHeaderSize
3960 This specifies the maximum size for HTTP headers in a reply.
3961 Reply headers are usually relatively small (about 512 bytes).
3962 Placing a limit on the reply header size will catch certain
3963 bugs (for example with persistent connections) and possibly
3964 buffer-overflow or denial-of-service attacks.
3967 NAME: request_body_max_size
3971 LOC: Config.maxRequestBodySize
3973 This specifies the maximum size for an HTTP request body.
3974 In other words, the maximum size of a PUT/POST request.
3975 A user who attempts to send a request with a body larger
3976 than this limit receives an "Invalid Request" error message.
3977 If you set this parameter to a zero (the default), there will
3978 be no limit imposed.
3981 NAME: chunked_request_body_max_size
3985 LOC: Config.maxChunkedRequestBodySize
3987 A broken or confused HTTP/1.1 client may send a chunked HTTP
3988 request to Squid. Squid does not have full support for that
3989 feature yet. To cope with such requests, Squid buffers the
3990 entire request and then dechunks request body to create a
3991 plain HTTP/1.0 request with a known content length. The plain
3992 request is then used by the rest of Squid code as usual.
3994 The option value specifies the maximum size of the buffer used
3995 to hold the request before the conversion. If the chunked
3996 request size exceeds the specified limit, the conversion
3997 fails, and the client receives an "unsupported request" error,
3998 as if dechunking was disabled.
4000 Dechunking is enabled by default. To disable conversion of
4001 chunked requests, set the maximum to zero.
4003 Request dechunking feature and this option in particular are a
4004 temporary hack. When chunking requests and responses are fully
4005 supported, there will be no need to buffer a chunked request.
4009 IFDEF: USE_HTTP_VIOLATIONS
4012 LOC: Config.accessList.brokenPosts
4014 A list of ACL elements which, if matched, causes Squid to send
4015 an extra CRLF pair after the body of a PUT/POST request.
4017 Some HTTP servers has broken implementations of PUT/POST,
4018 and rely on an extra CRLF pair sent by some WWW clients.
4020 Quote from RFC2616 section 4.1 on this matter:
4022 Note: certain buggy HTTP/1.0 client implementations generate an
4023 extra CRLF's after a POST request. To restate what is explicitly
4024 forbidden by the BNF, an HTTP/1.1 client must not preface or follow
4025 a request with an extra CRLF.
4027 This clause only supports fast acl types.
4028 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
4031 acl buggy_server url_regex ^http://....
4032 broken_posts allow buggy_server
4035 NAME: icap_uses_indirect_client
4038 IFDEF: FOLLOW_X_FORWARDED_FOR&&ICAP_CLIENT
4040 LOC: Adaptation::Icap::TheConfig.icap_uses_indirect_client
4042 Controls whether the indirect client address
4043 (see follow_x_forwarded_for) instead of the
4044 direct client address is passed to an ICAP
4045 server as "X-Client-IP".
4049 IFDEF: USE_HTTP_VIOLATIONS
4053 LOC: Config.onoff.via
4055 If set (default), Squid will include a Via header in requests and
4056 replies as required by RFC2616.
4062 LOC: Config.onoff.ie_refresh
4065 Microsoft Internet Explorer up until version 5.5 Service
4066 Pack 1 has an issue with transparent proxies, wherein it
4067 is impossible to force a refresh. Turning this on provides
4068 a partial fix to the problem, by causing all IMS-REFRESH
4069 requests from older IE versions to check the origin server
4070 for fresh content. This reduces hit ratio by some amount
4071 (~10% in my experience), but allows users to actually get
4072 fresh content when they want it. Note because Squid
4073 cannot tell if the user is using 5.5 or 5.5SP1, the behavior
4074 of 5.5 is unchanged from old versions of Squid (i.e. a
4075 forced refresh is impossible). Newer versions of IE will,
4076 hopefully, continue to have the new behavior and will be
4077 handled based on that assumption. This option defaults to
4078 the old Squid behavior, which is better for hit ratios but
4079 worse for clients using IE, if they need to be able to
4080 force fresh content.
4083 NAME: vary_ignore_expire
4086 LOC: Config.onoff.vary_ignore_expire
4089 Many HTTP servers supporting Vary gives such objects
4090 immediate expiry time with no cache-control header
4091 when requested by a HTTP/1.0 client. This option
4092 enables Squid to ignore such expiry times until
4093 HTTP/1.1 is fully implemented.
4095 WARNING: If turned on this may eventually cause some
4096 varying objects not intended for caching to get cached.
4099 NAME: request_entities
4101 LOC: Config.onoff.request_entities
4104 Squid defaults to deny GET and HEAD requests with request entities,
4105 as the meaning of such requests are undefined in the HTTP standard
4106 even if not explicitly forbidden.
4108 Set this directive to on if you have clients which insists
4109 on sending request entities in GET or HEAD requests. But be warned
4110 that there is server software (both proxies and web servers) which
4111 can fail to properly process this kind of request which may make you
4112 vulnerable to cache pollution attacks if enabled.
4115 NAME: request_header_access
4116 IFDEF: USE_HTTP_VIOLATIONS
4117 TYPE: http_header_access[]
4118 LOC: Config.request_header_access
4121 Usage: request_header_access header_name allow|deny [!]aclname ...
4123 WARNING: Doing this VIOLATES the HTTP standard. Enabling
4124 this feature could make you liable for problems which it
4127 This option replaces the old 'anonymize_headers' and the
4128 older 'http_anonymizer' option with something that is much
4129 more configurable. This new method creates a list of ACLs
4130 for each header, allowing you very fine-tuned header
4133 This option only applies to request headers, i.e., from the
4134 client to the server.
4136 You can only specify known headers for the header name.
4137 Other headers are reclassified as 'Other'. You can also
4138 refer to all the headers with 'All'.
4140 For example, to achieve the same behavior as the old
4141 'http_anonymizer standard' option, you should use:
4143 request_header_access From deny all
4144 request_header_access Referer deny all
4145 request_header_access Server deny all
4146 request_header_access User-Agent deny all
4147 request_header_access WWW-Authenticate deny all
4148 request_header_access Link deny all
4150 Or, to reproduce the old 'http_anonymizer paranoid' feature
4153 request_header_access Allow allow all
4154 request_header_access Authorization allow all
4155 request_header_access WWW-Authenticate allow all
4156 request_header_access Proxy-Authorization allow all
4157 request_header_access Proxy-Authenticate allow all
4158 request_header_access Cache-Control allow all
4159 request_header_access Content-Encoding allow all
4160 request_header_access Content-Length allow all
4161 request_header_access Content-Type allow all
4162 request_header_access Date allow all
4163 request_header_access Expires allow all
4164 request_header_access Host allow all
4165 request_header_access If-Modified-Since allow all
4166 request_header_access Last-Modified allow all
4167 request_header_access Location allow all
4168 request_header_access Pragma allow all
4169 request_header_access Accept allow all
4170 request_header_access Accept-Charset allow all
4171 request_header_access Accept-Encoding allow all
4172 request_header_access Accept-Language allow all
4173 request_header_access Content-Language allow all
4174 request_header_access Mime-Version allow all
4175 request_header_access Retry-After allow all
4176 request_header_access Title allow all
4177 request_header_access Connection allow all
4178 request_header_access All deny all
4180 although many of those are HTTP reply headers, and so should be
4181 controlled with the reply_header_access directive.
4183 By default, all headers are allowed (no anonymizing is
4187 NAME: reply_header_access
4188 IFDEF: USE_HTTP_VIOLATIONS
4189 TYPE: http_header_access[]
4190 LOC: Config.reply_header_access
4193 Usage: reply_header_access header_name allow|deny [!]aclname ...
4195 WARNING: Doing this VIOLATES the HTTP standard. Enabling
4196 this feature could make you liable for problems which it
4199 This option only applies to reply headers, i.e., from the
4200 server to the client.
4202 This is the same as request_header_access, but in the other
4205 This option replaces the old 'anonymize_headers' and the
4206 older 'http_anonymizer' option with something that is much
4207 more configurable. This new method creates a list of ACLs
4208 for each header, allowing you very fine-tuned header
4211 You can only specify known headers for the header name.
4212 Other headers are reclassified as 'Other'. You can also
4213 refer to all the headers with 'All'.
4215 For example, to achieve the same behavior as the old
4216 'http_anonymizer standard' option, you should use:
4218 reply_header_access From deny all
4219 reply_header_access Referer deny all
4220 reply_header_access Server deny all
4221 reply_header_access User-Agent deny all
4222 reply_header_access WWW-Authenticate deny all
4223 reply_header_access Link deny all
4225 Or, to reproduce the old 'http_anonymizer paranoid' feature
4228 reply_header_access Allow allow all
4229 reply_header_access Authorization allow all
4230 reply_header_access WWW-Authenticate allow all
4231 reply_header_access Proxy-Authorization allow all
4232 reply_header_access Proxy-Authenticate allow all
4233 reply_header_access Cache-Control allow all
4234 reply_header_access Content-Encoding allow all
4235 reply_header_access Content-Length allow all
4236 reply_header_access Content-Type allow all
4237 reply_header_access Date allow all
4238 reply_header_access Expires allow all
4239 reply_header_access Host allow all
4240 reply_header_access If-Modified-Since allow all
4241 reply_header_access Last-Modified allow all
4242 reply_header_access Location allow all
4243 reply_header_access Pragma allow all
4244 reply_header_access Accept allow all
4245 reply_header_access Accept-Charset allow all
4246 reply_header_access Accept-Encoding allow all
4247 reply_header_access Accept-Language allow all
4248 reply_header_access Content-Language allow all
4249 reply_header_access Mime-Version allow all
4250 reply_header_access Retry-After allow all
4251 reply_header_access Title allow all
4252 reply_header_access Connection allow all
4253 reply_header_access All deny all
4255 although the HTTP request headers won't be usefully controlled
4256 by this directive -- see request_header_access for details.
4258 By default, all headers are allowed (no anonymizing is
4262 NAME: header_replace
4263 IFDEF: USE_HTTP_VIOLATIONS
4264 TYPE: http_header_replace[]
4265 LOC: Config.request_header_access
4268 Usage: header_replace header_name message
4269 Example: header_replace User-Agent Nutscrape/1.0 (CP/M; 8-bit)
4271 This option allows you to change the contents of headers
4272 denied with header_access above, by replacing them with
4273 some fixed string. This replaces the old fake_user_agent
4276 This only applies to request headers, not reply headers.
4278 By default, headers are removed if denied.
4281 NAME: relaxed_header_parser
4282 COMMENT: on|off|warn
4284 LOC: Config.onoff.relaxed_header_parser
4287 In the default "on" setting Squid accepts certain forms
4288 of non-compliant HTTP messages where it is unambiguous
4289 what the sending application intended even if the message
4290 is not correctly formatted. The messages is then normalized
4291 to the correct form when forwarded by Squid.
4293 If set to "warn" then a warning will be emitted in cache.log
4294 each time such HTTP error is encountered.
4296 If set to "off" then such HTTP errors will cause the request
4297 or response to be rejected.
4302 -----------------------------------------------------------------------------
4305 NAME: forward_timeout
4308 LOC: Config.Timeout.forward
4311 This parameter specifies how long Squid should at most attempt in
4312 finding a forwarding path for the request before giving up.
4315 NAME: connect_timeout
4318 LOC: Config.Timeout.connect
4321 This parameter specifies how long to wait for the TCP connect to
4322 the requested server or peer to complete before Squid should
4323 attempt to find another path where to forward the request.
4326 NAME: peer_connect_timeout
4329 LOC: Config.Timeout.peer_connect
4332 This parameter specifies how long to wait for a pending TCP
4333 connection to a peer cache. The default is 30 seconds. You
4334 may also set different timeout values for individual neighbors
4335 with the 'connect-timeout' option on a 'cache_peer' line.
4341 LOC: Config.Timeout.read
4344 The read_timeout is applied on server-side connections. After
4345 each successful read(), the timeout will be extended by this
4346 amount. If no data is read again after this amount of time,
4347 the request is aborted and logged with ERR_READ_TIMEOUT. The
4348 default is 15 minutes.
4354 LOC: Config.Timeout.write
4357 This timeout is tracked for all connections that have data
4358 available for writing and are waiting for the socket to become
4359 ready. After each successful write, the timeout is extended by
4360 the configured amount. If Squid has data to write but the
4361 connection is not ready for the configured duration, the
4362 transaction associated with the connection is terminated. The
4363 default is 15 minutes.
4366 NAME: request_timeout
4368 LOC: Config.Timeout.request
4371 How long to wait for an HTTP request after initial
4372 connection establishment.
4375 NAME: persistent_request_timeout
4377 LOC: Config.Timeout.persistent_request
4380 How long to wait for the next HTTP request on a persistent
4381 connection after the previous request completes.
4384 NAME: client_lifetime
4387 LOC: Config.Timeout.lifetime
4390 The maximum amount of time a client (browser) is allowed to
4391 remain connected to the cache process. This protects the Cache
4392 from having a lot of sockets (and hence file descriptors) tied up
4393 in a CLOSE_WAIT state from remote clients that go away without
4394 properly shutting down (either because of a network failure or
4395 because of a poor client implementation). The default is one
4398 NOTE: The default value is intended to be much larger than any
4399 client would ever need to be connected to your cache. You
4400 should probably change client_lifetime only as a last resort.
4401 If you seem to have many client connections tying up
4402 filedescriptors, we recommend first tuning the read_timeout,
4403 request_timeout, persistent_request_timeout and quick_abort values.
4406 NAME: half_closed_clients
4408 LOC: Config.onoff.half_closed_clients
4411 Some clients may shutdown the sending side of their TCP
4412 connections, while leaving their receiving sides open. Sometimes,
4413 Squid can not tell the difference between a half-closed and a
4414 fully-closed TCP connection.
4416 By default, Squid will immediately close client connections when
4417 read(2) returns "no more data to read."
4419 Change this option to 'on' and Squid will keep open connections
4420 until a read(2) or write(2) on the socket returns an error.
4421 This may show some benefits for reverse proxies. But if not
4422 it is recommended to leave OFF.
4427 LOC: Config.Timeout.pconn
4430 Timeout for idle persistent connections to servers and other
4437 LOC: Ident::TheConfig.timeout
4440 Maximum time to wait for IDENT lookups to complete.
4442 If this is too high, and you enabled IDENT lookups from untrusted
4443 users, you might be susceptible to denial-of-service by having
4444 many ident requests going at once.
4447 NAME: shutdown_lifetime
4450 LOC: Config.shutdownLifetime
4453 When SIGTERM or SIGHUP is received, the cache is put into
4454 "shutdown pending" mode until all active sockets are closed.
4455 This value is the lifetime to set for all open descriptors
4456 during shutdown mode. Any active clients after this many
4457 seconds will receive a 'timeout' message.
4461 ADMINISTRATIVE PARAMETERS
4462 -----------------------------------------------------------------------------
4468 LOC: Config.adminEmail
4470 Email-address of local cache manager who will receive
4471 mail if the cache dies. The default is "webmaster."
4477 LOC: Config.EmailFrom
4479 From: email-address for mail sent when the cache dies.
4480 The default is to use 'appname@unique_hostname'.
4481 Default appname value is "squid", can be changed into
4482 src/globals.h before building squid.
4488 LOC: Config.EmailProgram
4490 Email program used to send mail if the cache dies.
4491 The default is "mail". The specified program must comply
4492 with the standard Unix mail syntax:
4493 mail-program recipient < mailfile
4495 Optional command line options can be specified.
4498 NAME: cache_effective_user
4500 DEFAULT: @DEFAULT_CACHE_EFFECTIVE_USER@
4501 LOC: Config.effectiveUser
4503 If you start Squid as root, it will change its effective/real
4504 UID/GID to the user specified below. The default is to change
4505 to UID of @DEFAULT_CACHE_EFFECTIVE_USER@.
4506 see also; cache_effective_group
4509 NAME: cache_effective_group
4512 LOC: Config.effectiveGroup
4514 Squid sets the GID to the effective user's default group ID
4515 (taken from the password file) and supplementary group list
4516 from the groups membership.
4518 If you want Squid to run with a specific GID regardless of
4519 the group memberships of the effective user then set this
4520 to the group (or GID) you want Squid to run as. When set
4521 all other group privileges of the effective user are ignored
4522 and only this GID is effective. If Squid is not started as
4523 root the user starting Squid MUST be member of the specified
4526 This option is not recommended by the Squid Team.
4527 Our preference is for administrators to configure a secure
4528 user account for squid with UID/GID matching system policies.
4531 NAME: httpd_suppress_version_string
4535 LOC: Config.onoff.httpd_suppress_version_string
4537 Suppress Squid version string info in HTTP headers and HTML error pages.
4540 NAME: visible_hostname
4542 LOC: Config.visibleHostname
4545 If you want to present a special hostname in error messages, etc,
4546 define this. Otherwise, the return value of gethostname()
4547 will be used. If you have multiple caches in a cluster and
4548 get errors about IP-forwarding you must set them to have individual
4549 names with this setting.
4552 NAME: unique_hostname
4554 LOC: Config.uniqueHostname
4557 If you want to have multiple machines with the same
4558 'visible_hostname' you must give each machine a different
4559 'unique_hostname' so forwarding loops can be detected.
4562 NAME: hostname_aliases
4564 LOC: Config.hostnameAliases
4567 A list of other DNS names your cache has.
4575 Minimum umask which should be enforced while the proxy
4576 is running, in addition to the umask set at startup.
4578 For a traditional octal representation of umasks, start
4583 OPTIONS FOR THE CACHE REGISTRATION SERVICE
4584 -----------------------------------------------------------------------------
4586 This section contains parameters for the (optional) cache
4587 announcement service. This service is provided to help
4588 cache administrators locate one another in order to join or
4589 create cache hierarchies.
4591 An 'announcement' message is sent (via UDP) to the registration
4592 service by Squid. By default, the announcement message is NOT
4593 SENT unless you enable it with 'announce_period' below.
4595 The announcement message includes your hostname, plus the
4596 following information from this configuration file:
4602 All current information is processed regularly and made
4603 available on the Web at http://www.ircache.net/Cache/Tracker/.
4606 NAME: announce_period
4608 LOC: Config.Announce.period
4611 This is how frequently to send cache announcements. The
4612 default is `0' which disables sending the announcement
4615 To enable announcing your cache, just set an announce period.
4618 announce_period 1 day
4623 DEFAULT: tracker.ircache.net
4624 LOC: Config.Announce.host
4630 LOC: Config.Announce.file
4636 LOC: Config.Announce.port
4638 announce_host and announce_port set the hostname and port
4639 number where the registration message will be sent.
4641 Hostname will default to 'tracker.ircache.net' and port will
4642 default default to 3131. If the 'filename' argument is given,
4643 the contents of that file will be included in the announce
4648 HTTPD-ACCELERATOR OPTIONS
4649 -----------------------------------------------------------------------------
4652 NAME: httpd_accel_surrogate_id
4655 LOC: Config.Accel.surrogate_id
4657 Surrogates (http://www.esi.org/architecture_spec_1.0.html)
4658 need an identification token to allow control targeting. Because
4659 a farm of surrogates may all perform the same tasks, they may share
4660 an identification token.
4662 The default ID is the visible_hostname
4665 NAME: http_accel_surrogate_remote
4669 LOC: Config.onoff.surrogate_is_remote
4671 Remote surrogates (such as those in a CDN) honour Surrogate-Control: no-store-remote.
4672 Set this to on to have squid behave as a remote surrogate.
4676 IFDEF: USE_SQUID_ESI
4677 COMMENT: libxml2|expat|custom
4679 LOC: ESIParser::Type
4682 ESI markup is not strictly XML compatible. The custom ESI parser
4683 will give higher performance, but cannot handle non ASCII character
4688 DELAY POOL PARAMETERS
4689 -----------------------------------------------------------------------------
4693 TYPE: delay_pool_count
4698 This represents the number of delay pools to be used. For example,
4699 if you have one class 2 delay pool and one class 3 delays pool, you
4700 have a total of 2 delay pools.
4704 TYPE: delay_pool_class
4709 This defines the class of each delay pool. There must be exactly one
4710 delay_class line for each delay pool. For example, to define two
4711 delay pools, one of class 2 and one of class 3, the settings above
4715 delay_pools 4 # 4 delay pools
4716 delay_class 1 2 # pool 1 is a class 2 pool
4717 delay_class 2 3 # pool 2 is a class 3 pool
4718 delay_class 3 4 # pool 3 is a class 4 pool
4719 delay_class 4 5 # pool 4 is a class 5 pool
4721 The delay pool classes are:
4723 class 1 Everything is limited by a single aggregate
4726 class 2 Everything is limited by a single aggregate
4727 bucket as well as an "individual" bucket chosen
4728 from bits 25 through 32 of the IPv4 address.
4730 class 3 Everything is limited by a single aggregate
4731 bucket as well as a "network" bucket chosen
4732 from bits 17 through 24 of the IP address and a
4733 "individual" bucket chosen from bits 17 through
4734 32 of the IPv4 address.
4736 class 4 Everything in a class 3 delay pool, with an
4737 additional limit on a per user basis. This
4738 only takes effect if the username is established
4739 in advance - by forcing authentication in your
4742 class 5 Requests are grouped according their tag (see
4743 external_acl's tag= reply).
4745 NOTE: If an IP address is a.b.c.d
4746 -> bits 25 through 32 are "d"
4747 -> bits 17 through 24 are "c"
4748 -> bits 17 through 32 are "c * 256 + d"
4750 NOTE-2: Due to the use of bitmasks in class 2,3,4 pools they only apply to
4751 IPv4 traffic. Class 1 and 5 pools may be used with IPv6 traffic.
4755 TYPE: delay_pool_access
4760 This is used to determine which delay pool a request falls into.
4762 delay_access is sorted per pool and the matching starts with pool 1,
4763 then pool 2, ..., and finally pool N. The first delay pool where the
4764 request is allowed is selected for the request. If it does not allow
4765 the request to any pool then the request is not delayed (default).
4767 For example, if you want some_big_clients in delay
4768 pool 1 and lotsa_little_clients in delay pool 2:
4771 delay_access 1 allow some_big_clients
4772 delay_access 1 deny all
4773 delay_access 2 allow lotsa_little_clients
4774 delay_access 2 deny all
4775 delay_access 3 allow authenticated_clients
4778 NAME: delay_parameters
4779 TYPE: delay_pool_rates
4784 This defines the parameters for a delay pool. Each delay pool has
4785 a number of "buckets" associated with it, as explained in the
4786 description of delay_class. For a class 1 delay pool, the syntax is:
4788 delay_parameters pool aggregate
4790 For a class 2 delay pool:
4792 delay_parameters pool aggregate individual
4794 For a class 3 delay pool:
4796 delay_parameters pool aggregate network individual
4798 For a class 4 delay pool:
4800 delay_parameters pool aggregate network individual user
4802 For a class 5 delay pool:
4804 delay_parameters pool tag
4806 The variables here are:
4808 pool a pool number - ie, a number between 1 and the
4809 number specified in delay_pools as used in
4812 aggregate the "delay parameters" for the aggregate bucket
4815 individual the "delay parameters" for the individual
4816 buckets (class 2, 3).
4818 network the "delay parameters" for the network buckets
4821 user the delay parameters for the user buckets
4824 tag the delay parameters for the tag buckets
4827 A pair of delay parameters is written restore/maximum, where restore is
4828 the number of bytes (not bits - modem and network speeds are usually
4829 quoted in bits) per second placed into the bucket, and maximum is the
4830 maximum number of bytes which can be in the bucket at any time.
4832 For example, if delay pool number 1 is a class 2 delay pool as in the
4833 above example, and is being used to strictly limit each host to 64kbps
4834 (plus overheads), with no overall limit, the line is:
4836 delay_parameters 1 -1/-1 8000/8000
4838 Note that the figure -1 is used to represent "unlimited".
4840 And, if delay pool number 2 is a class 3 delay pool as in the above
4841 example, and you want to limit it to a total of 256kbps (strict limit)
4842 with each 8-bit network permitted 64kbps (strict limit) and each
4843 individual host permitted 4800bps with a bucket maximum size of 64kb
4844 to permit a decent web page to be downloaded at a decent speed
4845 (if the network is not being limited due to overuse) but slow down
4846 large downloads more significantly:
4848 delay_parameters 2 32000/32000 8000/8000 600/8000
4850 There must be one delay_parameters line for each delay pool.
4852 Finally, for a class 4 delay pool as in the example - each user will
4853 be limited to 128Kb no matter how many workstations they are logged into.:
4855 delay_parameters 4 32000/32000 8000/8000 600/64000 16000/16000
4858 NAME: delay_initial_bucket_level
4859 COMMENT: (percent, 0-100)
4863 LOC: Config.Delay.initial
4865 The initial bucket percentage is used to determine how much is put
4866 in each bucket when squid starts, is reconfigured, or first notices
4867 a host accessing it (in class 2 and class 3, individual hosts and
4868 networks only have buckets associated with them once they have been
4873 CLIENT DELAY POOL PARAMETERS
4874 -----------------------------------------------------------------------------
4877 NAME: client_delay_pools
4878 TYPE: client_delay_pool_count
4881 LOC: Config.ClientDelay
4883 This option specifies the number of client delay pools used. It must
4884 preceed other client_delay_* options.
4887 client_delay_pools 2
4890 NAME: client_delay_initial_bucket_level
4891 COMMENT: (percent, 0-no_limit)
4895 LOC: Config.ClientDelay.initial
4897 This option determines the initial bucket size as a percentage of
4898 max_bucket_size from client_delay_parameters. Buckets are created
4899 at the time of the "first" connection from the matching IP. Idle
4900 buckets are periodically deleted up.
4902 You can specify more than 100 percent but note that such "oversized"
4903 buckets are not refilled until their size goes down to max_bucket_size
4904 from client_delay_parameters.
4907 client_delay_initial_bucket_level 50
4910 NAME: client_delay_parameters
4911 TYPE: client_delay_pool_rates
4914 LOC: Config.ClientDelay
4917 This option configures client-side bandwidth limits using the
4920 client_delay_parameters pool speed_limit max_bucket_size
4922 pool is an integer ID used for client_delay_access matching.
4924 speed_limit is bytes added to the bucket per second.
4926 max_bucket_size is the maximum size of a bucket, enforced after any
4927 speed_limit additions.
4929 Please see the delay_parameters option for more information and
4933 client_delay_parameters 1 1024 2048
4934 client_delay_parameters 2 51200 16384
4937 NAME: client_delay_access
4938 TYPE: client_delay_pool_access
4941 LOC: Config.ClientDelay
4944 This option determines the client-side delay pool for the
4947 client_delay_access pool_ID allow|deny acl_name
4949 All client_delay_access options are checked in their pool ID
4950 order, starting with pool 1. The first checked pool with allowed
4951 request is selected for the request. If no ACL matches or there
4952 are no client_delay_access options, the request bandwidth is not
4955 The ACL-selected pool is then used to find the
4956 client_delay_parameters for the request. Client-side pools are
4957 not used to aggregate clients. Clients are always aggregated
4958 based on their source IP addresses (one bucket per source IP).
4960 Please see delay_access for more examples.
4963 client_delay_access 1 allow low_rate_network
4964 client_delay_access 2 allow vips_network
4968 WCCPv1 AND WCCPv2 CONFIGURATION OPTIONS
4969 -----------------------------------------------------------------------------
4974 LOC: Config.Wccp.router
4978 Use this option to define your WCCP ``home'' router for
4981 wccp_router supports a single WCCP(v1) router
4983 wccp2_router supports multiple WCCPv2 routers
4985 only one of the two may be used at the same time and defines
4986 which version of WCCP to use.
4990 TYPE: IpAddress_list
4991 LOC: Config.Wccp2.router
4995 Use this option to define your WCCP ``home'' router for
4998 wccp_router supports a single WCCP(v1) router
5000 wccp2_router supports multiple WCCPv2 routers
5002 only one of the two may be used at the same time and defines
5003 which version of WCCP to use.
5008 LOC: Config.Wccp.version
5012 This directive is only relevant if you need to set up WCCP(v1)
5013 to some very old and end-of-life Cisco routers. In all other
5014 setups it must be left unset or at the default setting.
5015 It defines an internal version in the WCCP(v1) protocol,
5016 with version 4 being the officially documented protocol.
5018 According to some users, Cisco IOS 11.2 and earlier only
5019 support WCCP version 3. If you're using that or an earlier
5020 version of IOS, you may need to change this value to 3, otherwise
5021 do not specify this parameter.
5024 NAME: wccp2_rebuild_wait
5026 LOC: Config.Wccp2.rebuildwait
5030 If this is enabled Squid will wait for the cache dir rebuild to finish
5031 before sending the first wccp2 HereIAm packet
5034 NAME: wccp2_forwarding_method
5036 LOC: Config.Wccp2.forwarding_method
5040 WCCP2 allows the setting of forwarding methods between the
5041 router/switch and the cache. Valid values are as follows:
5043 gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
5044 l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
5046 Currently (as of IOS 12.4) cisco routers only support GRE.
5047 Cisco switches only support the L2 redirect assignment method.
5050 NAME: wccp2_return_method
5052 LOC: Config.Wccp2.return_method
5056 WCCP2 allows the setting of return methods between the
5057 router/switch and the cache for packets that the cache
5058 decides not to handle. Valid values are as follows:
5060 gre - GRE encapsulation (forward the packet in a GRE/WCCP tunnel)
5061 l2 - L2 redirect (forward the packet using Layer 2/MAC rewriting)
5063 Currently (as of IOS 12.4) cisco routers only support GRE.
5064 Cisco switches only support the L2 redirect assignment.
5066 If the "ip wccp redirect exclude in" command has been
5067 enabled on the cache interface, then it is still safe for
5068 the proxy server to use a l2 redirect method even if this
5069 option is set to GRE.
5072 NAME: wccp2_assignment_method
5074 LOC: Config.Wccp2.assignment_method
5078 WCCP2 allows the setting of methods to assign the WCCP hash
5079 Valid values are as follows:
5081 hash - Hash assignment
5082 mask - Mask assignment
5084 As a general rule, cisco routers support the hash assignment method
5085 and cisco switches support the mask assignment method.
5090 LOC: Config.Wccp2.info
5091 DEFAULT_IF_NONE: standard 0
5094 WCCP2 allows for multiple traffic services. There are two
5095 types: "standard" and "dynamic". The standard type defines
5096 one service id - http (id 0). The dynamic service ids can be from
5097 51 to 255 inclusive. In order to use a dynamic service id
5098 one must define the type of traffic to be redirected; this is done
5099 using the wccp2_service_info option.
5101 The "standard" type does not require a wccp2_service_info option,
5102 just specifying the service id will suffice.
5104 MD5 service authentication can be enabled by adding
5105 "password=<password>" to the end of this service declaration.
5109 wccp2_service standard 0 # for the 'web-cache' standard service
5110 wccp2_service dynamic 80 # a dynamic service type which will be
5111 # fleshed out with subsequent options.
5112 wccp2_service standard 0 password=foo
5115 NAME: wccp2_service_info
5116 TYPE: wccp2_service_info
5117 LOC: Config.Wccp2.info
5121 Dynamic WCCPv2 services require further information to define the
5122 traffic you wish to have diverted.
5126 wccp2_service_info <id> protocol=<protocol> flags=<flag>,<flag>..
5127 priority=<priority> ports=<port>,<port>..
5129 The relevant WCCPv2 flags:
5130 + src_ip_hash, dst_ip_hash
5131 + source_port_hash, dst_port_hash
5132 + src_ip_alt_hash, dst_ip_alt_hash
5133 + src_port_alt_hash, dst_port_alt_hash
5136 The port list can be one to eight entries.
5140 wccp2_service_info 80 protocol=tcp flags=src_ip_hash,ports_source
5141 priority=240 ports=80
5143 Note: the service id must have been defined by a previous
5144 'wccp2_service dynamic <id>' entry.
5149 LOC: Config.Wccp2.weight
5153 Each cache server gets assigned a set of the destination
5154 hash proportional to their weight.
5159 LOC: Config.Wccp.address
5166 LOC: Config.Wccp2.address
5170 Use this option if you require WCCP to use a specific
5173 The default behavior is to not bind to any specific address.
5177 PERSISTENT CONNECTION HANDLING
5178 -----------------------------------------------------------------------------
5180 Also see "pconn_timeout" in the TIMEOUTS section
5183 NAME: client_persistent_connections
5185 LOC: Config.onoff.client_pconns
5189 NAME: server_persistent_connections
5191 LOC: Config.onoff.server_pconns
5194 Persistent connection support for clients and servers. By
5195 default, Squid uses persistent connections (when allowed)
5196 with its clients and servers. You can use these options to
5197 disable persistent connections with clients and/or servers.
5200 NAME: persistent_connection_after_error
5202 LOC: Config.onoff.error_pconns
5205 With this directive the use of persistent connections after
5206 HTTP errors can be disabled. Useful if you have clients
5207 who fail to handle errors on persistent connections proper.
5210 NAME: detect_broken_pconn
5212 LOC: Config.onoff.detect_broken_server_pconns
5215 Some servers have been found to incorrectly signal the use
5216 of HTTP/1.0 persistent connections even on replies not
5217 compatible, causing significant delays. This server problem
5218 has mostly been seen on redirects.
5220 By enabling this directive Squid attempts to detect such
5221 broken replies and automatically assume the reply is finished
5222 after 10 seconds timeout.
5226 CACHE DIGEST OPTIONS
5227 -----------------------------------------------------------------------------
5230 NAME: digest_generation
5231 IFDEF: USE_CACHE_DIGESTS
5233 LOC: Config.onoff.digest_generation
5236 This controls whether the server will generate a Cache Digest
5237 of its contents. By default, Cache Digest generation is
5238 enabled if Squid is compiled with --enable-cache-digests defined.
5241 NAME: digest_bits_per_entry
5242 IFDEF: USE_CACHE_DIGESTS
5244 LOC: Config.digest.bits_per_entry
5247 This is the number of bits of the server's Cache Digest which
5248 will be associated with the Digest entry for a given HTTP
5249 Method and URL (public key) combination. The default is 5.
5252 NAME: digest_rebuild_period
5253 IFDEF: USE_CACHE_DIGESTS
5256 LOC: Config.digest.rebuild_period
5259 This is the wait time between Cache Digest rebuilds.
5262 NAME: digest_rewrite_period
5264 IFDEF: USE_CACHE_DIGESTS
5266 LOC: Config.digest.rewrite_period
5269 This is the wait time between Cache Digest writes to
5273 NAME: digest_swapout_chunk_size
5276 IFDEF: USE_CACHE_DIGESTS
5277 LOC: Config.digest.swapout_chunk_size
5280 This is the number of bytes of the Cache Digest to write to
5281 disk at a time. It defaults to 4096 bytes (4KB), the Squid
5285 NAME: digest_rebuild_chunk_percentage
5286 COMMENT: (percent, 0-100)
5287 IFDEF: USE_CACHE_DIGESTS
5289 LOC: Config.digest.rebuild_chunk_percentage
5292 This is the percentage of the Cache Digest to be scanned at a
5293 time. By default it is set to 10% of the Cache Digest.
5298 -----------------------------------------------------------------------------
5303 LOC: Config.Port.snmp
5307 The port number where Squid listens for SNMP requests. To enable
5308 SNMP support set this to a suitable port number. Port number
5309 3401 is often used for the Squid SNMP agent. By default it's
5310 set to "0" (disabled)
5318 LOC: Config.accessList.snmp
5319 DEFAULT_IF_NONE: deny all
5322 Allowing or denying access to the SNMP port.
5324 All access to the agent is denied by default.
5327 snmp_access allow|deny [!]aclname ...
5329 This clause only supports fast acl types.
5330 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5332 snmp_access allow snmppublic localhost
5333 snmp_access deny all
5336 NAME: snmp_incoming_address
5338 LOC: Config.Addrs.snmp_incoming
5343 NAME: snmp_outgoing_address
5345 LOC: Config.Addrs.snmp_outgoing
5349 Just like 'udp_incoming_address', but for the SNMP port.
5351 snmp_incoming_address is used for the SNMP socket receiving
5352 messages from SNMP agents.
5353 snmp_outgoing_address is used for SNMP packets returned to SNMP
5356 The default snmp_incoming_address is to listen on all
5357 available network interfaces.
5359 If snmp_outgoing_address is not set it will use the same socket
5360 as snmp_incoming_address. Only change this if you want to have
5361 SNMP replies sent using another address than where this Squid
5362 listens for SNMP queries.
5364 NOTE, snmp_incoming_address and snmp_outgoing_address can not have
5365 the same value since they both use port 3401.
5370 -----------------------------------------------------------------------------
5373 NAME: icp_port udp_port
5376 LOC: Config.Port.icp
5378 The port number where Squid sends and receives ICP queries to
5379 and from neighbor caches. The standard UDP port for ICP is 3130.
5380 Default is disabled (0).
5383 icp_port @DEFAULT_ICP_PORT@
5390 LOC: Config.Port.htcp
5392 The port number where Squid sends and receives HTCP queries to
5393 and from neighbor caches. To turn it on you want to set it to
5394 4827. By default it is set to "0" (disabled).
5400 NAME: log_icp_queries
5404 LOC: Config.onoff.log_udp
5406 If set, ICP queries are logged to access.log. You may wish
5407 do disable this if your ICP load is VERY high to speed things
5408 up or to simplify log analysis.
5411 NAME: udp_incoming_address
5413 LOC:Config.Addrs.udp_incoming
5416 udp_incoming_address is used for UDP packets received from other
5419 The default behavior is to not bind to any specific address.
5421 Only change this if you want to have all UDP queries received on
5422 a specific interface/address.
5424 NOTE: udp_incoming_address is used by the ICP, HTCP, and DNS
5425 modules. Altering it will affect all of them in the same manner.
5427 see also; udp_outgoing_address
5429 NOTE, udp_incoming_address and udp_outgoing_address can not
5430 have the same value since they both use the same port.
5433 NAME: udp_outgoing_address
5435 LOC: Config.Addrs.udp_outgoing
5438 udp_outgoing_address is used for UDP packets sent out to other
5441 The default behavior is to not bind to any specific address.
5443 Instead it will use the same socket as udp_incoming_address.
5444 Only change this if you want to have UDP queries sent using another
5445 address than where this Squid listens for UDP queries from other
5448 NOTE: udp_outgoing_address is used by the ICP, HTCP, and DNS
5449 modules. Altering it will affect all of them in the same manner.
5451 see also; udp_incoming_address
5453 NOTE, udp_incoming_address and udp_outgoing_address can not
5454 have the same value since they both use the same port.
5461 LOC: Config.onoff.icp_hit_stale
5463 If you want to return ICP_HIT for stale cache objects, set this
5464 option to 'on'. If you have sibling relationships with caches
5465 in other administrative domains, this should be 'off'. If you only
5466 have sibling relationships with caches under your control,
5467 it is probably okay to set this to 'on'.
5468 If set to 'on', your siblings should use the option "allow-miss"
5469 on their cache_peer lines for connecting to you.
5472 NAME: minimum_direct_hops
5475 LOC: Config.minDirectHops
5477 If using the ICMP pinging stuff, do direct fetches for sites
5478 which are no more than this many hops away.
5481 NAME: minimum_direct_rtt
5484 LOC: Config.minDirectRtt
5486 If using the ICMP pinging stuff, do direct fetches for sites
5487 which are no more than this many rtt milliseconds away.
5493 LOC: Config.Netdb.low
5499 LOC: Config.Netdb.high
5501 The low and high water marks for the ICMP measurement
5502 database. These are counts, not percents. The defaults are
5503 900 and 1000. When the high water mark is reached, database
5504 entries will be deleted until the low mark is reached.
5507 NAME: netdb_ping_period
5509 LOC: Config.Netdb.period
5512 The minimum period for measuring a site. There will be at
5513 least this much delay between successive pings to the same
5514 network. The default is five minutes.
5521 LOC: Config.onoff.query_icmp
5523 If you want to ask your peers to include ICMP data in their ICP
5524 replies, enable this option.
5526 If your peer has configured Squid (during compilation) with
5527 '--enable-icmp' that peer will send ICMP pings to origin server
5528 sites of the URLs it receives. If you enable this option the
5529 ICP replies from that peer will include the ICMP data (if available).
5530 Then, when choosing a parent cache, Squid will choose the parent with
5531 the minimal RTT to the origin server. When this happens, the
5532 hierarchy field of the access.log will be
5533 "CLOSEST_PARENT_MISS". This option is off by default.
5536 NAME: test_reachability
5540 LOC: Config.onoff.test_reachability
5542 When this is 'on', ICP MISS replies will be ICP_MISS_NOFETCH
5543 instead of ICP_MISS if the target host is NOT in the ICMP
5544 database, or has a zero RTT.
5547 NAME: icp_query_timeout
5551 LOC: Config.Timeout.icp_query
5553 Normally Squid will automatically determine an optimal ICP
5554 query timeout value based on the round-trip-time of recent ICP
5555 queries. If you want to override the value determined by
5556 Squid, set this 'icp_query_timeout' to a non-zero value. This
5557 value is specified in MILLISECONDS, so, to use a 2-second
5558 timeout (the old default), you would write:
5560 icp_query_timeout 2000
5563 NAME: maximum_icp_query_timeout
5567 LOC: Config.Timeout.icp_query_max
5569 Normally the ICP query timeout is determined dynamically. But
5570 sometimes it can lead to very large values (say 5 seconds).
5571 Use this option to put an upper limit on the dynamic timeout
5572 value. Do NOT use this option to always use a fixed (instead
5573 of a dynamic) timeout value. To set a fixed timeout see the
5574 'icp_query_timeout' directive.
5577 NAME: minimum_icp_query_timeout
5581 LOC: Config.Timeout.icp_query_min
5583 Normally the ICP query timeout is determined dynamically. But
5584 sometimes it can lead to very small timeouts, even lower than
5585 the normal latency variance on your link due to traffic.
5586 Use this option to put an lower limit on the dynamic timeout
5587 value. Do NOT use this option to always use a fixed (instead
5588 of a dynamic) timeout value. To set a fixed timeout see the
5589 'icp_query_timeout' directive.
5592 NAME: background_ping_rate
5596 LOC: Config.backgroundPingRate
5598 Controls how often the ICP pings are sent to siblings that
5599 have background-ping set.
5603 MULTICAST ICP OPTIONS
5604 -----------------------------------------------------------------------------
5609 LOC: Config.mcast_group_list
5612 This tag specifies a list of multicast groups which your server
5613 should join to receive multicasted ICP queries.
5615 NOTE! Be very careful what you put here! Be sure you
5616 understand the difference between an ICP _query_ and an ICP
5617 _reply_. This option is to be set only if you want to RECEIVE
5618 multicast queries. Do NOT set this option to SEND multicast
5619 ICP (use cache_peer for that). ICP replies are always sent via
5620 unicast, so this option does not affect whether or not you will
5621 receive replies from multicast group members.
5623 You must be very careful to NOT use a multicast address which
5624 is already in use by another group of caches.
5626 If you are unsure about multicast, please read the Multicast
5627 chapter in the Squid FAQ (http://www.squid-cache.org/FAQ/).
5629 Usage: mcast_groups 239.128.16.128 224.0.1.20
5631 By default, Squid doesn't listen on any multicast groups.
5634 NAME: mcast_miss_addr
5635 IFDEF: MULTICAST_MISS_STREAM
5637 LOC: Config.mcast_miss.addr
5640 If you enable this option, every "cache miss" URL will
5641 be sent out on the specified multicast address.
5643 Do not enable this option unless you are are absolutely
5644 certain you understand what you are doing.
5647 NAME: mcast_miss_ttl
5648 IFDEF: MULTICAST_MISS_STREAM
5650 LOC: Config.mcast_miss.ttl
5653 This is the time-to-live value for packets multicasted
5654 when multicasting off cache miss URLs is enabled. By
5655 default this is set to 'site scope', i.e. 16.
5658 NAME: mcast_miss_port
5659 IFDEF: MULTICAST_MISS_STREAM
5661 LOC: Config.mcast_miss.port
5664 This is the port number to be used in conjunction with
5668 NAME: mcast_miss_encode_key
5669 IFDEF: MULTICAST_MISS_STREAM
5671 LOC: Config.mcast_miss.encode_key
5672 DEFAULT: XXXXXXXXXXXXXXXX
5674 The URLs that are sent in the multicast miss stream are
5675 encrypted. This is the encryption key.
5678 NAME: mcast_icp_query_timeout
5682 LOC: Config.Timeout.mcast_icp_query
5684 For multicast peers, Squid regularly sends out ICP "probes" to
5685 count how many other peers are listening on the given multicast
5686 address. This value specifies how long Squid should wait to
5687 count all the replies. The default is 2000 msec, or 2
5692 INTERNAL ICON OPTIONS
5693 -----------------------------------------------------------------------------
5696 NAME: icon_directory
5698 LOC: Config.icons.directory
5699 DEFAULT: @DEFAULT_ICON_DIR@
5701 Where the icons are stored. These are normally kept in
5705 NAME: global_internal_static
5707 LOC: Config.onoff.global_internal_static
5710 This directive controls is Squid should intercept all requests for
5711 /squid-internal-static/ no matter which host the URL is requesting
5712 (default on setting), or if nothing special should be done for
5713 such URLs (off setting). The purpose of this directive is to make
5714 icons etc work better in complex cache hierarchies where it may
5715 not always be possible for all corners in the cache mesh to reach
5716 the server generating a directory listing.
5719 NAME: short_icon_urls
5721 LOC: Config.icons.use_short_names
5724 If this is enabled Squid will use short URLs for icons.
5725 If disabled it will revert to the old behavior of including
5726 it's own name and port in the URL.
5728 If you run a complex cache hierarchy with a mix of Squid and
5729 other proxies you may need to disable this directive.
5734 -----------------------------------------------------------------------------
5737 NAME: error_directory
5739 LOC: Config.errorDirectory
5742 If you wish to create your own versions of the default
5743 error files to customize them to suit your company copy
5744 the error/template files to another directory and point
5747 WARNING: This option will disable multi-language support
5748 on error pages if used.
5750 The squid developers are interested in making squid available in
5751 a wide variety of languages. If you are making translations for a
5752 language that Squid does not currently provide please consider
5753 contributing your translation back to the project.
5754 http://wiki.squid-cache.org/Translations
5756 The squid developers working on translations are happy to supply drop-in
5757 translated error files in exchange for any new language contributions.
5760 NAME: error_default_language
5761 IFDEF: USE_ERR_LOCALES
5763 LOC: Config.errorDefaultLanguage
5766 Set the default language which squid will send error pages in
5767 if no existing translation matches the clients language
5770 If unset (default) generic English will be used.
5772 The squid developers are interested in making squid available in
5773 a wide variety of languages. If you are interested in making
5774 translations for any language see the squid wiki for details.
5775 http://wiki.squid-cache.org/Translations
5778 NAME: error_log_languages
5779 IFDEF: USE_ERR_LOCALES
5781 LOC: Config.errorLogMissingLanguages
5784 Log to cache.log what languages users are attempting to
5785 auto-negotiate for translations.
5787 Successful negotiations are not logged. Only failures
5788 have meaning to indicate that Squid may need an upgrade
5789 of its error page translations.
5792 NAME: err_page_stylesheet
5794 LOC: Config.errorStylesheet
5795 DEFAULT: @DEFAULT_CONFIG_DIR@/errorpage.css
5797 CSS Stylesheet to pattern the display of Squid default error pages.
5799 For information on CSS see http://www.w3.org/Style/CSS/
5804 LOC: Config.errHtmlText
5807 HTML text to include in error messages. Make this a "mailto"
5808 URL to your admin address, or maybe just a link to your
5809 organizations Web page.
5811 To include this in your error messages, you must rewrite
5812 the error template files (found in the "errors" directory).
5813 Wherever you want the 'err_html_text' line to appear,
5814 insert a %L tag in the error template file.
5817 NAME: email_err_data
5820 LOC: Config.onoff.emailErrData
5823 If enabled, information about the occurred error will be
5824 included in the mailto links of the ERR pages (if %W is set)
5825 so that the email body contains the data.
5826 Syntax is <A HREF="mailto:%w%W">%w</A>
5831 LOC: Config.denyInfoList
5834 Usage: deny_info err_page_name acl
5835 or deny_info http://... acl
5836 or deny_info TCP_RESET acl
5838 This can be used to return a ERR_ page for requests which
5839 do not pass the 'http_access' rules. Squid remembers the last
5840 acl it evaluated in http_access, and if a 'deny_info' line exists
5841 for that ACL Squid returns a corresponding error page.
5843 The acl is typically the last acl on the http_access deny line which
5844 denied access. The exceptions to this rule are:
5845 - When Squid needs to request authentication credentials. It's then
5846 the first authentication related acl encountered
5847 - When none of the http_access lines matches. It's then the last
5848 acl processed on the last http_access line.
5850 NP: If providing your own custom error pages with error_directory
5851 you may also specify them by your custom file name:
5852 Example: deny_info ERR_CUSTOM_ACCESS_DENIED bad_guys
5854 Alternatively you can tell Squid to reset the TCP connection
5855 by specifying TCP_RESET.
5857 Or you can specify an error URL or URL pattern. The browsers will
5858 get redirected (302) to the specified URL after formatting tags have
5862 %a - username (if available. Password NOT included)
5865 %E - Error description
5867 %H - Request domain name
5868 %i - Client IP Address
5870 %o - Message result from external ACL helper
5871 %p - Request Port number
5872 %P - Request Protocol name
5873 %R - Request URL path
5874 %T - Timestamp in RFC 1123 format
5875 %U - Full canonical URL from client
5876 (HTTPS URLs terminate with *)
5877 %u - Full canonical URL from client
5878 %w - Admin email from squid.conf
5879 %% - Literal percent (%) code
5884 OPTIONS INFLUENCING REQUEST FORWARDING
5885 -----------------------------------------------------------------------------
5888 NAME: nonhierarchical_direct
5890 LOC: Config.onoff.nonhierarchical_direct
5893 By default, Squid will send any non-hierarchical requests
5894 (matching hierarchy_stoplist or not cacheable request type) direct
5897 If you set this to off, Squid will prefer to send these
5898 requests to parents.
5900 Note that in most configurations, by turning this off you will only
5901 add latency to these request without any improvement in global hit
5904 If you are inside an firewall see never_direct instead of
5910 LOC: Config.onoff.prefer_direct
5913 Normally Squid tries to use parents for most requests. If you for some
5914 reason like it to first try going direct and only use a parent if
5915 going direct fails set this to on.
5917 By combining nonhierarchical_direct off and prefer_direct on you
5918 can set up Squid to use a parent as a backup path if going direct
5921 Note: If you want Squid to use parents for all requests see
5922 the never_direct directive. prefer_direct only modifies how Squid
5923 acts on cacheable requests.
5928 LOC: Config.accessList.AlwaysDirect
5931 Usage: always_direct allow|deny [!]aclname ...
5933 Here you can use ACL elements to specify requests which should
5934 ALWAYS be forwarded by Squid to the origin servers without using
5935 any peers. For example, to always directly forward requests for
5936 local servers ignoring any parents or siblings you may have use
5939 acl local-servers dstdomain my.domain.net
5940 always_direct allow local-servers
5942 To always forward FTP requests directly, use
5945 always_direct allow FTP
5947 NOTE: There is a similar, but opposite option named
5948 'never_direct'. You need to be aware that "always_direct deny
5949 foo" is NOT the same thing as "never_direct allow foo". You
5950 may need to use a deny rule to exclude a more-specific case of
5951 some other rule. Example:
5953 acl local-external dstdomain external.foo.net
5954 acl local-servers dstdomain .foo.net
5955 always_direct deny local-external
5956 always_direct allow local-servers
5958 NOTE: If your goal is to make the client forward the request
5959 directly to the origin server bypassing Squid then this needs
5960 to be done in the client configuration. Squid configuration
5961 can only tell Squid how Squid should fetch the object.
5963 NOTE: This directive is not related to caching. The replies
5964 is cached as usual even if you use always_direct. To not cache
5965 the replies see the 'cache' directive.
5967 This clause supports both fast and slow acl types.
5968 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
5973 LOC: Config.accessList.NeverDirect
5976 Usage: never_direct allow|deny [!]aclname ...
5978 never_direct is the opposite of always_direct. Please read
5979 the description for always_direct if you have not already.
5981 With 'never_direct' you can use ACL elements to specify
5982 requests which should NEVER be forwarded directly to origin
5983 servers. For example, to force the use of a proxy for all
5984 requests, except those in your local domain use something like:
5986 acl local-servers dstdomain .foo.net
5987 never_direct deny local-servers
5988 never_direct allow all
5990 or if Squid is inside a firewall and there are local intranet
5991 servers inside the firewall use something like:
5993 acl local-intranet dstdomain .foo.net
5994 acl local-external dstdomain external.foo.net
5995 always_direct deny local-external
5996 always_direct allow local-intranet
5997 never_direct allow all
5999 This clause supports both fast and slow acl types.
6000 See http://wiki.squid-cache.org/SquidFaq/SquidAcl for details.
6004 ADVANCED NETWORKING OPTIONS
6005 -----------------------------------------------------------------------------
6008 NAME: incoming_icp_average
6011 LOC: Config.comm_incoming.icp_average
6014 NAME: incoming_http_average
6017 LOC: Config.comm_incoming.http_average
6020 NAME: incoming_dns_average
6023 LOC: Config.comm_incoming.dns_average
6026 NAME: min_icp_poll_cnt
6029 LOC: Config.comm_incoming.icp_min_poll
6032 NAME: min_dns_poll_cnt
6035 LOC: Config.comm_incoming.dns_min_poll
6038 NAME: min_http_poll_cnt
6041 LOC: Config.comm_incoming.http_min_poll
6043 Heavy voodoo here. I can't even believe you are reading this.
6044 Are you crazy? Don't even think about adjusting these unless
6045 you understand the algorithms in comm_select.c first!
6051 LOC: Config.accept_filter
6055 The name of an accept(2) filter to install on Squid's
6056 listen socket(s). This feature is perhaps specific to
6057 FreeBSD and requires support in the kernel.
6059 The 'httpready' filter delays delivering new connections
6060 to Squid until a full HTTP request has been received.
6061 See the accf_http(9) man page for details.
6063 The 'dataready' filter delays delivering new connections
6064 to Squid until there is some data to process.
6065 See the accf_dataready(9) man page for details.
6069 The 'data' filter delays delivering of new connections
6070 to Squid until there is some data to process by TCP_ACCEPT_DEFER.
6071 You may optionally specify a number of seconds to wait by
6072 'data=N' where N is the number of seconds. Defaults to 30
6073 if not specified. See the tcp(7) man page for details.
6076 accept_filter httpready
6081 NAME: client_ip_max_connections
6083 LOC: Config.client_ip_max_connections
6086 Set an absolute limit on the number of connections a single
6087 client IP can use. Any more than this and Squid will begin to drop
6088 new connections from the client until it closes some links.
6090 Note that this is a global limit. It affects all HTTP, HTCP, Gopher and FTP
6091 connections from the client. For finer control use the ACL access controls.
6093 Requires client_db to be enabled (the default).
6095 WARNING: This may noticably slow down traffic received via external proxies
6096 or NAT devices and cause them to rebound error messages back to their clients.
6099 NAME: tcp_recv_bufsize
6103 LOC: Config.tcpRcvBufsz
6105 Size of receive buffer to set for TCP sockets. Probably just
6106 as easy to change your kernel's default. Set to zero to use
6107 the default buffer size.
6112 -----------------------------------------------------------------------------
6119 LOC: Adaptation::Icap::TheConfig.onoff
6122 If you want to enable the ICAP module support, set this to on.
6125 NAME: icap_connect_timeout
6128 LOC: Adaptation::Icap::TheConfig.connect_timeout_raw
6131 This parameter specifies how long to wait for the TCP connect to
6132 the requested ICAP server to complete before giving up and either
6133 terminating the HTTP transaction or bypassing the failure.
6135 The default for optional services is peer_connect_timeout.
6136 The default for essential services is connect_timeout.
6137 If this option is explicitly set, its value applies to all services.
6140 NAME: icap_io_timeout
6144 LOC: Adaptation::Icap::TheConfig.io_timeout_raw
6147 This parameter specifies how long to wait for an I/O activity on
6148 an established, active ICAP connection before giving up and
6149 either terminating the HTTP transaction or bypassing the
6152 The default is read_timeout.
6155 NAME: icap_service_failure_limit
6156 COMMENT: limit [in memory-depth time-units]
6157 TYPE: icap_service_failure_limit
6159 LOC: Adaptation::Icap::TheConfig
6162 The limit specifies the number of failures that Squid tolerates
6163 when establishing a new TCP connection with an ICAP service. If
6164 the number of failures exceeds the limit, the ICAP service is
6165 not used for new ICAP requests until it is time to refresh its
6168 A negative value disables the limit. Without the limit, an ICAP
6169 service will not be considered down due to connectivity failures
6170 between ICAP OPTIONS requests.
6172 Squid forgets ICAP service failures older than the specified
6173 value of memory-depth. The memory fading algorithm
6174 is approximate because Squid does not remember individual
6175 errors but groups them instead, splitting the option
6176 value into ten time slots of equal length.
6178 When memory-depth is 0 and by default this option has no
6179 effect on service failure expiration.
6181 Squid always forgets failures when updating service settings
6182 using an ICAP OPTIONS transaction, regardless of this option
6186 # suspend service usage after 10 failures in 5 seconds:
6187 icap_service_failure_limit 10 in 5 seconds
6190 NAME: icap_service_revival_delay
6193 LOC: Adaptation::Icap::TheConfig.service_revival_delay
6196 The delay specifies the number of seconds to wait after an ICAP
6197 OPTIONS request failure before requesting the options again. The
6198 failed ICAP service is considered "down" until fresh OPTIONS are
6201 The actual delay cannot be smaller than the hardcoded minimum
6202 delay of 30 seconds.
6205 NAME: icap_preview_enable
6209 LOC: Adaptation::Icap::TheConfig.preview_enable
6212 The ICAP Preview feature allows the ICAP server to handle the
6213 HTTP message by looking only at the beginning of the message body
6214 or even without receiving the body at all. In some environments,
6215 previews greatly speedup ICAP processing.
6217 During an ICAP OPTIONS transaction, the server may tell Squid what
6218 HTTP messages should be previewed and how big the preview should be.
6219 Squid will not use Preview if the server did not request one.
6221 To disable ICAP Preview for all ICAP services, regardless of
6222 individual ICAP server OPTIONS responses, set this option to "off".
6224 icap_preview_enable off
6227 NAME: icap_preview_size
6230 LOC: Adaptation::Icap::TheConfig.preview_size
6233 The default size of preview data to be sent to the ICAP server.
6234 -1 means no preview. This value might be overwritten on a per server
6235 basis by OPTIONS requests.
6238 NAME: icap_206_enable
6242 LOC: Adaptation::Icap::TheConfig.allow206_enable
6245 206 (Partial Content) responses is an ICAP extension that allows the
6246 ICAP agents to optionally combine adapted and original HTTP message
6247 content. The decision to combine is postponed until the end of the
6248 ICAP response. Squid supports Partial Content extension by default.
6250 Activation of the Partial Content extension is negotiated with each
6251 ICAP service during OPTIONS exchange. Most ICAP servers should handle
6252 negotation correctly even if they do not support the extension, but
6253 some might fail. To disable Partial Content support for all ICAP
6254 services and to avoid any negotiation, set this option to "off".
6260 NAME: icap_default_options_ttl
6263 LOC: Adaptation::Icap::TheConfig.default_options_ttl
6266 The default TTL value for ICAP OPTIONS responses that don't have
6267 an Options-TTL header.
6270 NAME: icap_persistent_connections
6274 LOC: Adaptation::Icap::TheConfig.reuse_connections
6277 Whether or not Squid should use persistent connections to
6281 NAME: icap_send_client_ip
6285 LOC: Adaptation::Icap::TheConfig.send_client_ip
6288 This adds the header "X-Client-IP" to ICAP requests.
6291 NAME: icap_send_client_username
6295 LOC: Adaptation::Icap::TheConfig.send_client_username
6298 This sends authenticated HTTP client username (if available) to
6299 the ICAP service. The username value is encoded based on the
6300 icap_client_username_encode option and is sent using the header
6301 specified by the icap_client_username_header option.
6304 NAME: icap_client_username_header
6307 LOC: Adaptation::Icap::TheConfig.client_username_header
6308 DEFAULT: X-Client-Username
6310 ICAP request header name to use for send_client_username.
6313 NAME: icap_client_username_encode
6317 LOC: Adaptation::Icap::TheConfig.client_username_encode
6320 Whether to base64 encode the authenticated client username.
6324 TYPE: icap_service_type
6326 LOC: Adaptation::Icap::TheConfig
6329 Defines a single ICAP service using the following format:
6331 icap_service service_name vectoring_point [options] service_url
6334 an opaque identifier which must be unique in squid.conf
6336 vectoring_point: reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
6337 This specifies at which point of transaction processing the
6338 ICAP service should be activated. *_postcache vectoring points
6339 are not yet supported.
6341 service_url: icap://servername:port/servicepath
6342 ICAP server and service location.
6344 ICAP does not allow a single service to handle both REQMOD and RESPMOD
6345 transactions. Squid does not enforce that requirement. You can specify
6346 services with the same service_url and different vectoring_points. You
6347 can even specify multiple identical services as long as their
6348 service_names differ.
6351 Service options are separated by white space. ICAP services support
6352 the following name=value options:
6355 If set to 'on' or '1', the ICAP service is treated as
6356 optional. If the service cannot be reached or malfunctions,
6357 Squid will try to ignore any errors and process the message as
6358 if the service was not enabled. No all ICAP errors can be
6359 bypassed. If set to 0, the ICAP service is treated as
6360 essential and all ICAP errors will result in an error page
6361 returned to the HTTP client.
6363 Bypass is off by default: services are treated as essential.
6366 If set to 'on' or '1', the ICAP service is allowed to
6367 dynamically change the current message adaptation plan by
6368 returning a chain of services to be used next. The services
6369 are specified using the X-Next-Services ICAP response header
6370 value, formatted as a comma-separated list of service names.
6371 Each named service should be configured in squid.conf and
6372 should have the same method and vectoring point as the current
6373 ICAP transaction. Services violating these rules are ignored.
6374 An empty X-Next-Services value results in an empty plan which
6375 ends the current adaptation.
6377 Routing is not allowed by default: the ICAP X-Next-Services
6378 response header is ignored.
6381 Only has effect on split-stack systems. The default on those systems
6382 is to use IPv4-only connections. When set to 'on' this option will
6383 make Squid use IPv6-only connections to contact this ICAP service.
6385 Older icap_service format without optional named parameters is
6386 deprecated but supported for backward compatibility.
6389 icap_service svcBlocker reqmod_precache bypass=0 icap://icap1.mydomain.net:1344/reqmod
6390 icap_service svcLogger reqmod_precache routing=on icap://icap2.mydomain.net:1344/respmod
6394 TYPE: icap_class_type
6399 This deprecated option was documented to define an ICAP service
6400 chain, even though it actually defined a set of similar, redundant
6401 services, and the chains were not supported.
6403 To define a set of redundant services, please use the
6404 adaptation_service_set directive. For service chains, use
6405 adaptation_service_chain.
6409 TYPE: icap_access_type
6414 This option is deprecated. Please use adaptation_access, which
6415 has the same ICAP functionality, but comes with better
6416 documentation, and eCAP support.
6421 -----------------------------------------------------------------------------
6428 LOC: Adaptation::Ecap::TheConfig.onoff
6431 Controls whether eCAP support is enabled.
6435 TYPE: ecap_service_type
6437 LOC: Adaptation::Ecap::TheConfig
6440 Defines a single eCAP service
6442 ecap_service servicename vectoring_point bypass service_url
6444 vectoring_point = reqmod_precache|reqmod_postcache|respmod_precache|respmod_postcache
6445 This specifies at which point of transaction processing the
6446 eCAP service should be activated. *_postcache vectoring points
6447 are not yet supported.
6449 If set to 1, the eCAP service is treated as optional. If the
6450 service cannot be reached or malfunctions, Squid will try to
6451 ignore any errors and process the message as if the service
6452 was not enabled. No all eCAP errors can be bypassed.
6453 If set to 0, the eCAP service is treated as essential and all
6454 eCAP errors will result in an error page returned to the
6456 service_url = ecap://vendor/service_name?custom&cgi=style¶meters=optional
6459 ecap_service service_1 reqmod_precache 0 ecap://filters-R-us/leakDetector?on_error=block
6460 ecap_service service_2 respmod_precache 1 icap://filters-R-us/virusFilter?config=/etc/vf.cfg
6463 NAME: loadable_modules
6465 IFDEF: USE_LOADABLE_MODULES
6466 LOC: Config.loadable_module_names
6469 Instructs Squid to load the specified dynamic module(s) or activate
6470 preloaded module(s).
6472 loadable_modules @DEFAULT_PREFIX@/lib/MinimalAdapter.so
6476 MESSAGE ADAPTATION OPTIONS
6477 -----------------------------------------------------------------------------
6480 NAME: adaptation_service_set
6481 TYPE: adaptation_service_set_type
6482 IFDEF: USE_ADAPTATION
6487 Configures an ordered set of similar, redundant services. This is
6488 useful when hot standby or backup adaptation servers are available.
6490 adaptation_service_set set_name service_name1 service_name2 ...
6492 The named services are used in the set declaration order. The first
6493 applicable adaptation service from the set is used first. The next
6494 applicable service is tried if and only if the transaction with the
6495 previous service fails and the message waiting to be adapted is still
6498 When adaptation starts, broken services are ignored as if they were
6499 not a part of the set. A broken service is a down optional service.
6501 The services in a set must be attached to the same vectoring point
6502 (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
6504 If all services in a set are optional then adaptation failures are
6505 bypassable. If all services in the set are essential, then a
6506 transaction failure with one service may still be retried using
6507 another service from the set, but when all services fail, the master
6508 transaction fails as well.
6510 A set may contain a mix of optional and essential services, but that
6511 is likely to lead to surprising results because broken services become
6512 ignored (see above), making previously bypassable failures fatal.
6513 Technically, it is the bypassability of the last failed service that
6516 See also: adaptation_access adaptation_service_chain
6519 adaptation_service_set svcBlocker urlFilterPrimary urlFilterBackup
6520 adaptation service_set svcLogger loggerLocal loggerRemote
6523 NAME: adaptation_service_chain
6524 TYPE: adaptation_service_chain_type
6525 IFDEF: USE_ADAPTATION
6530 Configures a list of complementary services that will be applied
6531 one-by-one, forming an adaptation chain or pipeline. This is useful
6532 when Squid must perform different adaptations on the same message.
6534 adaptation_service_chain chain_name service_name1 svc_name2 ...
6536 The named services are used in the chain declaration order. The first
6537 applicable adaptation service from the chain is used first. The next
6538 applicable service is applied to the successful adaptation results of
6539 the previous service in the chain.
6541 When adaptation starts, broken services are ignored as if they were
6542 not a part of the chain. A broken service is a down optional service.
6544 Request satisfaction terminates the adaptation chain because Squid
6545 does not currently allow declaration of RESPMOD services at the
6546 "reqmod_precache" vectoring point (see icap_service or ecap_service).
6548 The services in a chain must be attached to the same vectoring point
6549 (e.g., pre-cache) and use the same adaptation method (e.g., REQMOD).
6551 A chain may contain a mix of optional and essential services. If an
6552 essential adaptation fails (or the failure cannot be bypassed for
6553 other reasons), the master transaction fails. Otherwise, the failure
6554 is bypassed as if the failed adaptation service was not in the chain.
6556 See also: adaptation_access adaptation_service_set
6559 adaptation_service_chain svcRequest requestLogger urlFilter leakDetector
6562 NAME: adaptation_access
6563 TYPE: adaptation_access_type
6564 IFDEF: USE_ADAPTATION
6568 Sends an HTTP transaction to an ICAP or eCAP adaptation service.
6570 adaptation_access service_name allow|deny [!]aclname...
6571 adaptation_access set_name allow|deny [!]aclname...
6573 At each supported vectoring point, the adaptation_access
6574 statements are processed in the order they appear in this
6575 configuration file. Statements pointing to the following services
6576 are ignored (i.e., skipped without checking their ACL):
6578 - services serving different vectoring points
6579 - "broken-but-bypassable" services
6580 - "up" services configured to ignore such transactions
6581 (e.g., based on the ICAP Transfer-Ignore header).
6583 When a set_name is used, all services in the set are checked
6584 using the same rules, to find the first applicable one. See
6585 adaptation_service_set for details.
6587 If an access list is checked and there is a match, the
6588 processing stops: For an "allow" rule, the corresponding
6589 adaptation service is used for the transaction. For a "deny"
6590 rule, no adaptation service is activated.
6592 It is currently not possible to apply more than one adaptation
6593 service at the same vectoring point to the same HTTP transaction.
6595 See also: icap_service and ecap_service
6598 adaptation_access service_1 allow all
6601 NAME: adaptation_service_iteration_limit
6603 IFDEF: USE_ADAPTATION
6604 LOC: Adaptation::Config::service_iteration_limit
6607 Limits the number of iterations allowed when applying adaptation
6608 services to a message. If your longest adaptation set or chain
6609 may have more than 16 services, increase the limit beyond its
6610 default value of 16. If detecting infinite iteration loops sooner
6611 is critical, make the iteration limit match the actual number
6612 of services in your longest adaptation set or chain.
6614 Infinite adaptation loops are most likely with routing services.
6616 See also: icap_service routing=1
6619 NAME: adaptation_masterx_shared_names
6621 IFDEF: USE_ADAPTATION
6622 LOC: Adaptation::Config::masterx_shared_name
6625 For each master transaction (i.e., the HTTP request and response
6626 sequence, including all related ICAP and eCAP exchanges), Squid
6627 maintains a table of metadata. The table entries are (name, value)
6628 pairs shared among eCAP and ICAP exchanges. The table is destroyed
6629 with the master transaction.
6631 This option specifies the table entry names that Squid must accept
6632 from and forward to the adaptation transactions.
6634 An ICAP REQMOD or RESPMOD transaction may set an entry in the
6635 shared table by returning an ICAP header field with a name
6636 specified in adaptation_masterx_shared_names. Squid will store
6637 and forward that ICAP header field to subsequent ICAP
6638 transactions within the same master transaction scope.
6640 Only one shared entry name is supported at this time.
6643 # share authentication information among ICAP services
6644 adaptation_masterx_shared_names X-Subscriber-ID
6650 LOC: Adaptation::Icap::TheConfig.repeat
6651 DEFAULT_IF_NONE: deny all
6653 This ACL determines which retriable ICAP transactions are
6654 retried. Transactions that received a complete ICAP response
6655 and did not have to consume or produce HTTP bodies to receive
6656 that response are usually retriable.
6658 icap_retry allow|deny [!]aclname ...
6660 Squid automatically retries some ICAP I/O timeouts and errors
6661 due to persistent connection race conditions.
6663 See also: icap_retry_limit
6666 NAME: icap_retry_limit
6669 LOC: Adaptation::Icap::TheConfig.repeat_limit
6672 Limits the number of retries allowed. When set to zero (default),
6673 no retries are allowed.
6675 Communication errors due to persistent connection race
6676 conditions are unavoidable, automatically retried, and do not
6677 count against this limit.
6679 See also: icap_retry
6685 -----------------------------------------------------------------------------
6688 NAME: check_hostnames
6691 LOC: Config.onoff.check_hostnames
6693 For security and stability reasons Squid can check
6694 hostnames for Internet standard RFC compliance. If you want
6695 Squid to perform these checks turn this directive on.
6698 NAME: allow_underscore
6701 LOC: Config.onoff.allow_underscore
6703 Underscore characters is not strictly allowed in Internet hostnames
6704 but nevertheless used by many sites. Set this to off if you want
6705 Squid to be strict about the standard.
6706 This check is performed only when check_hostnames is set to on.
6709 NAME: cache_dns_program
6711 IFDEF: USE_DNSSERVERS
6712 DEFAULT: @DEFAULT_DNSSERVER@
6713 LOC: Config.Program.dnsserver
6715 Specify the location of the executable for dnslookup process.
6719 TYPE: HelperChildConfig
6720 IFDEF: USE_DNSSERVERS
6721 DEFAULT: 32 startup=1 idle=1
6722 LOC: Config.dnsChildren
6724 The maximum number of processes spawn to service DNS name lookups.
6725 If you limit it too few Squid will have to wait for them to process
6726 a backlog of requests, slowing it down. If you allow too many they
6727 will use RAM and other system resources noticably.
6728 The maximum this may be safely set to is 32.
6730 The startup= and idle= options allow some measure of skew in your
6735 Sets a minimum of how many processes are to be spawned when Squid
6736 starts or reconfigures. When set to zero the first request will
6737 cause spawning of the first child process to handle it.
6739 Starting too few will cause an initial slowdown in traffic as Squid
6740 attempts to simultaneously spawn enough processes to cope.
6744 Sets a minimum of how many processes Squid is to try and keep available
6745 at all times. When traffic begins to rise above what the existing
6746 processes can handle this many more will be spawned up to the maximum
6747 configured. A minimum setting of 1 is required.
6750 NAME: dns_retransmit_interval
6753 LOC: Config.Timeout.idns_retransmit
6754 IFDEF: !USE_DNSSERVERS
6756 Initial retransmit interval for DNS queries. The interval is
6757 doubled each time all configured DNS servers have been tried.
6763 LOC: Config.Timeout.idns_query
6764 IFDEF: !USE_DNSSERVERS
6766 DNS Query timeout. If no response is received to a DNS query
6767 within this time all DNS servers for the queried domain
6768 are assumed to be unavailable.
6771 NAME: dns_packet_max
6774 LOC: Config.dns.packet_max
6775 IFDEF: !USE_DNSSERVERS
6777 Maximum number of bytes packet size to advertise via EDNS.
6778 Set to "none" to disable EDNS large packet support.
6780 For legacy reasons DNS UDP replies will default to 512 bytes which
6781 is too small for many responses. EDNS provides a means for Squid to
6782 negotiate receiving larger responses back immediately without having
6783 to failover with repeat requests. Responses larger than this limit
6784 will retain the old behaviour of failover to TCP DNS.
6786 Squid has no real fixed limit internally, but allowing packet sizes
6787 over 1500 bytes requires network jumbogram support and is usually not
6790 WARNING: The RFC also indicates that some older resolvers will reply
6791 with failure of the whole request if the extension is added. Some
6792 resolvers have already been identified which will reply with mangled
6793 EDNS response on occasion. Usually in response to many-KB jumbogram
6794 sizes being advertised by Squid.
6795 Squid will currently treat these both as an unable-to-resolve domain
6796 even if it would be resolvable without EDNS.
6803 LOC: Config.onoff.res_defnames
6805 Normally the RES_DEFNAMES resolver option is disabled
6806 (see res_init(3)). This prevents caches in a hierarchy
6807 from interpreting single-component hostnames locally. To allow
6808 Squid to handle single-component names, enable this option.
6811 NAME: dns_nameservers
6814 LOC: Config.dns_nameservers
6816 Use this if you want to specify a list of DNS name servers
6817 (IP addresses) to use instead of those given in your
6818 /etc/resolv.conf file.
6819 On Windows platforms, if no value is specified here or in
6820 the /etc/resolv.conf file, the list of DNS name servers are
6821 taken from the Windows registry, both static and dynamic DHCP
6822 configurations are supported.
6824 Example: dns_nameservers 10.0.0.1 192.172.0.4
6829 DEFAULT: @DEFAULT_HOSTS@
6830 LOC: Config.etcHostsPath
6832 Location of the host-local IP name-address associations
6833 database. Most Operating Systems have such a file on different
6835 - Un*X & Linux: /etc/hosts
6836 - Windows NT/2000: %SystemRoot%\system32\drivers\etc\hosts
6837 (%SystemRoot% value install default is c:\winnt)
6838 - Windows XP/2003: %SystemRoot%\system32\drivers\etc\hosts
6839 (%SystemRoot% value install default is c:\windows)
6840 - Windows 9x/Me: %windir%\hosts
6841 (%windir% value is usually c:\windows)
6842 - Cygwin: /etc/hosts
6844 The file contains newline-separated definitions, in the
6845 form ip_address_in_dotted_form name [name ...] names are
6846 whitespace-separated. Lines beginning with an hash (#)
6847 character are comments.
6849 The file is checked at startup and upon configuration.
6850 If set to 'none', it won't be checked.
6851 If append_domain is used, that domain will be added to
6852 domain-local (i.e. not containing any dot character) host
6858 LOC: Config.appendDomain
6861 Appends local domain name to hostnames without any dots in
6862 them. append_domain must begin with a period.
6864 Be warned there are now Internet names with no dots in
6865 them using only top-domain names, so setting this may
6866 cause some Internet sites to become unavailable.
6869 append_domain .yourdomain.com
6872 NAME: ignore_unknown_nameservers
6874 LOC: Config.onoff.ignore_unknown_nameservers
6877 By default Squid checks that DNS responses are received
6878 from the same IP addresses they are sent to. If they
6879 don't match, Squid ignores the response and writes a warning
6880 message to cache.log. You can allow responses from unknown
6881 nameservers by setting this option to 'off'.
6884 NAME: dns_v4_fallback
6887 LOC: Config.onoff.dns_require_A
6889 Standard practice with DNS is to lookup either A or AAAA records
6890 and use the results if it succeeds. Only looking up the other if
6891 the first attempt fails or otherwise produces no results.
6893 That policy however will cause squid to produce error pages for some
6894 servers that advertise AAAA but are unreachable over IPv6.
6896 If this is ON squid will always lookup both AAAA and A, using both.
6897 If this is OFF squid will lookup AAAA and only try A if none found.
6899 WARNING: There are some possibly unwanted side-effects with this on:
6900 *) Doubles the load placed by squid on the DNS network.
6901 *) May negatively impact connection delay times.
6905 COMMENT: (number of entries)
6908 LOC: Config.ipcache.size
6915 LOC: Config.ipcache.low
6922 LOC: Config.ipcache.high
6924 The size, low-, and high-water marks for the IP cache.
6927 NAME: fqdncache_size
6928 COMMENT: (number of entries)
6931 LOC: Config.fqdncache.size
6933 Maximum number of FQDN cache entries.
6938 -----------------------------------------------------------------------------
6945 LOC: Config.onoff.mem_pools
6947 If set, Squid will keep pools of allocated (but unused) memory
6948 available for future use. If memory is a premium on your
6949 system and you believe your malloc library outperforms Squid
6950 routines, disable this.
6953 NAME: memory_pools_limit
6957 LOC: Config.MemPools.limit
6959 Used only with memory_pools on:
6960 memory_pools_limit 50 MB
6962 If set to a non-zero value, Squid will keep at most the specified
6963 limit of allocated (but unused) memory in memory pools. All free()
6964 requests that exceed this limit will be handled by your malloc
6965 library. Squid does not pre-allocate any memory, just safe-keeps
6966 objects that otherwise would be free()d. Thus, it is safe to set
6967 memory_pools_limit to a reasonably high value even if your
6968 configuration will use less memory.
6970 If set to none, Squid will keep all memory it can. That is, there
6971 will be no limit on the total amount of memory used for safe-keeping.
6973 To disable memory allocation optimization, do not set
6974 memory_pools_limit to 0 or none. Set memory_pools to "off" instead.
6976 An overhead for maintaining memory pools is not taken into account
6977 when the limit is checked. This overhead is close to four bytes per
6978 object kept. However, pools may actually _save_ memory because of
6979 reduced memory thrashing in your malloc library.
6983 COMMENT: on|off|transparent|truncate|delete
6986 LOC: opt_forwarded_for
6988 If set to "on", Squid will append your client's IP address
6989 in the HTTP requests it forwards. By default it looks like:
6991 X-Forwarded-For: 192.1.2.3
6993 If set to "off", it will appear as
6995 X-Forwarded-For: unknown
6997 If set to "transparent", Squid will not alter the
6998 X-Forwarded-For header in any way.
7000 If set to "delete", Squid will delete the entire
7001 X-Forwarded-For header.
7003 If set to "truncate", Squid will remove all existing
7004 X-Forwarded-For entries, and place itself as the sole entry.
7007 NAME: cachemgr_passwd
7008 TYPE: cachemgrpasswd
7010 LOC: Config.passwd_list
7012 Specify passwords for cachemgr operations.
7014 Usage: cachemgr_passwd password action action ...
7016 Some valid actions are (see cache manager menu for a full list):
7056 * Indicates actions which will not be performed without a
7057 valid password, others can be performed if not listed here.
7059 To disable an action, set the password to "disable".
7060 To allow performing an action without a password, set the
7063 Use the keyword "all" to set the same password for all actions.
7066 cachemgr_passwd secret shutdown
7067 cachemgr_passwd lesssssssecret info stats/objects
7068 cachemgr_passwd disable all
7075 LOC: Config.onoff.client_db
7077 If you want to disable collecting per-client statistics,
7078 turn off client_db here.
7081 NAME: refresh_all_ims
7085 LOC: Config.onoff.refresh_all_ims
7087 When you enable this option, squid will always check
7088 the origin server for an update when a client sends an
7089 If-Modified-Since request. Many browsers use IMS
7090 requests when the user requests a reload, and this
7091 ensures those clients receive the latest version.
7093 By default (off), squid may return a Not Modified response
7094 based on the age of the cached version.
7097 NAME: reload_into_ims
7098 IFDEF: USE_HTTP_VIOLATIONS
7102 LOC: Config.onoff.reload_into_ims
7104 When you enable this option, client no-cache or ``reload''
7105 requests will be changed to If-Modified-Since requests.
7106 Doing this VIOLATES the HTTP standard. Enabling this
7107 feature could make you liable for problems which it
7110 see also refresh_pattern for a more selective approach.
7113 NAME: maximum_single_addr_tries
7115 LOC: Config.retry.maxtries
7118 This sets the maximum number of connection attempts for a
7119 host that only has one address (for multiple-address hosts,
7120 each address is tried once).
7122 The default value is one attempt, the (not recommended)
7123 maximum is 255 tries. A warning message will be generated
7124 if it is set to a value greater than ten.
7126 Note: This is in addition to the request re-forwarding which
7127 takes place if Squid fails to get a satisfying response.
7130 NAME: retry_on_error
7132 LOC: Config.retry.onerror
7135 If set to on Squid will automatically retry requests when
7136 receiving an error response. This is mainly useful if you
7137 are in a complex cache hierarchy to work around access
7141 NAME: as_whois_server
7143 LOC: Config.as_whois_server
7144 DEFAULT: whois.ra.net
7146 WHOIS server to query for AS numbers. NOTE: AS numbers are
7147 queried only when Squid starts up, not for every request.
7152 LOC: Config.onoff.offline
7155 Enable this option and Squid will never try to validate cached
7159 NAME: uri_whitespace
7160 TYPE: uri_whitespace
7161 LOC: Config.uri_whitespace
7164 What to do with requests that have whitespace characters in the
7167 strip: The whitespace characters are stripped out of the URL.
7168 This is the behavior recommended by RFC2396.
7169 deny: The request is denied. The user receives an "Invalid
7171 allow: The request is allowed and the URI is not changed. The
7172 whitespace characters remain in the URI. Note the
7173 whitespace is passed to redirector processes if they
7175 encode: The request is allowed and the whitespace characters are
7176 encoded according to RFC1738. This could be considered
7177 a violation of the HTTP/1.1
7178 RFC because proxies are not allowed to rewrite URI's.
7179 chop: The request is allowed and the URI is chopped at the
7180 first whitespace. This might also be considered a
7186 LOC: Config.chroot_dir
7189 Specifies a directory where Squid should do a chroot() while
7190 initializing. This also causes Squid to fully drop root
7191 privileges after initializing. This means, for example, if you
7192 use a HTTP port less than 1024 and try to reconfigure, you may
7193 get an error saying that Squid can not open the port.
7196 NAME: balance_on_multiple_ip
7198 LOC: Config.onoff.balance_on_multiple_ip
7201 Modern IP resolvers in squid sort lookup results by preferred access.
7202 By default squid will use these IP in order and only rotates to
7203 the next listed when the most preffered fails.
7205 Some load balancing servers based on round robin DNS have been
7206 found not to preserve user session state across requests
7207 to different IP addresses.
7209 Enabling this directive Squid rotates IP's per request.
7212 NAME: pipeline_prefetch
7214 LOC: Config.onoff.pipeline_prefetch
7217 To boost the performance of pipelined requests to closer
7218 match that of a non-proxied environment Squid can try to fetch
7219 up to two requests in parallel from a pipeline.
7221 Defaults to off for bandwidth management and access logging
7225 NAME: high_response_time_warning
7228 LOC: Config.warnings.high_rptm
7231 If the one-minute median response time exceeds this value,
7232 Squid prints a WARNING with debug level 0 to get the
7233 administrators attention. The value is in milliseconds.
7236 NAME: high_page_fault_warning
7238 LOC: Config.warnings.high_pf
7241 If the one-minute average page fault rate exceeds this
7242 value, Squid prints a WARNING with debug level 0 to get
7243 the administrators attention. The value is in page faults
7247 NAME: high_memory_warning
7249 LOC: Config.warnings.high_memory
7252 If the memory usage (as determined by mallinfo) exceeds
7253 this amount, Squid prints a WARNING with debug level 0 to get
7254 the administrators attention.
7257 NAME: sleep_after_fork
7258 COMMENT: (microseconds)
7260 LOC: Config.sleep_after_fork
7263 When this is set to a non-zero value, the main Squid process
7264 sleeps the specified number of microseconds after a fork()
7265 system call. This sleep may help the situation where your
7266 system reports fork() failures due to lack of (virtual)
7267 memory. Note, however, if you have a lot of child
7268 processes, these sleep delays will add up and your
7269 Squid will not service requests for some amount of time
7270 until all the child processes have been started.
7271 On Windows value less then 1000 (1 milliseconds) are
7275 NAME: windows_ipaddrchangemonitor
7276 IFDEF: _SQUID_MSWIN_
7280 LOC: Config.onoff.WIN32_IpAddrChangeMonitor
7282 On Windows Squid by default will monitor IP address changes and will
7283 reconfigure itself after any detected event. This is very useful for
7284 proxies connected to internet with dial-up interfaces.
7285 In some cases (a Proxy server acting as VPN gateway is one) it could be
7286 desiderable to disable this behaviour setting this to 'off'.
7287 Note: after changing this, Squid service must be restarted.
7292 IFDEF: USE_SQUID_EUI
7294 LOC: Eui::TheConfig.euiLookup
7296 Whether to lookup the EUI or MAC address of a connected client.
7299 NAME: max_filedescriptors max_filedesc
7302 LOC: Config.max_filedescriptors
7304 The maximum number of filedescriptors supported.
7306 The default "0" means Squid inherits the current ulimit setting.
7308 Note: Changing this requires a restart of Squid. Also
7309 not all comm loops supports large values.
7317 Number of main Squid processes or "workers" to fork and maintain.
7318 0: "no daemon" mode, like running "squid -N ..."
7319 1: "no SMP" mode, start one main Squid process daemon (default)
7320 N: start N main Squid process daemons (i.e., SMP mode)
7322 In SMP mode, each worker does nearly all what a single Squid daemon
7323 does (e.g., listen on http_port and forward HTTP requests).
7326 NAME: cpu_affinity_map
7327 TYPE: CpuAffinityMap
7328 LOC: Config.cpuAffinityMap
7331 Usage: cpu_affinity_map process_numbers=P1,P2,... cores=C1,C2,...
7333 Sets 1:1 mapping between Squid processes and CPU cores. For example,
7335 cpu_affinity_map process_numbers=1,2,3,4 cores=1,3,5,7
7337 affects processes 1 through 4 only and places them on the first
7338 four even cores, starting with core #1.
7340 CPU cores are numbered starting from 1. Requires support for
7341 sched_getaffinity(2) and sched_setaffinity(2) system calls.
7343 Multiple cpu_affinity_map options are merged.
projects / thirdparty / squid.git / blob