]> git.ipfire.org Git - thirdparty/squid.git/blame - src/cf.data.pre
projects / thirdparty / squid.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
gindent
[thirdparty/squid.git] / src / cf.data.pre
CommitLineData
934b03fc 1NAME: http_port ascii_port
2TYPE: ushortlist
3DEFAULT: 3128
4LOC: Config.Port.http
5DOC_START
6 The port number where squid will listen for HTTP client
7 requests. Default is 3128, for httpd-accel mode use port 80.
8 May be overridden with -a on the command line.
9
10 You may specify multiple ports here, but they MUST all be on
11 a single line.
12
13http_port 3128
14DOC_END
15
16
17NAME: icp_port udp_port
18TYPE: ushort
19DEFAULT: 3130
20LOC: Config.Port.icp
21DOC_START
22 The port number where squid send and receive ICP requests to
23 and from neighbor caches. Default is 3130. To disable use
24 "0". May be overridden with -u on the command line.
25
26icp_port 3130
27DOC_END
28
29
30NAME: mcast_groups
31TYPE: wordlist
32LOC: Config.mcast_group_list
33DEFAULT: NULL
34DOC_START
35 This tag specifies a list of multicast groups which your
36 server should join to receive multicasted ICP requests.
37
38 NOTE! Be very careful what you put here! Be sure you
39 understand the difference between an ICP _query_ and an ICP
40 _reply_. This option is to be set only if you want to RECEIVE
41 multicast queries. Do NOT set this option to SEND multicast
42 ICP (use cache_host for that). ICP replies are always sent via
43 unicast, so this option does not affect whether or not you will
44 receive replies from multicast group members.
45
46 You must be very careful to NOT use a multicast address which
47 is already in use by another group of caches. NLANR has been
48 assigned a block of multicast address space for use in Web
49 Caching. Plese write to us at nlanr-cache@nlanr.net to receive
50 an address for your own use.
51
52 Usage: mcast_groups 239.128.16.128 224.0.1.20
53
54 By default, squid doesn't listen on any multicast groups.
55
56mcast_groups 239.128.16.128
57DOC_END
58
59
60NAME: tcp_incoming_address bind_address
61TYPE: address
62LOC: Config.Addrs.tcp_incoming
63DEFAULT: INADDR_ANY
64DOC_NONE
65
66NAME: tcp_outgoing_address outbound_address
67TYPE: address
68LOC: Config.Addrs.tcp_outgoing
69DEFAULT: no_addr.s_addr
70DOC_NONE
71
72NAME: udp_incoming_address
73TYPE: address
74LOC:Config.Addrs.udp_incoming
75DEFAULT: INADDR_ANY
76DOC_NONE
77
78NAME: udp_outgoing_address
79TYPE: address
80LOC: Config.Addrs.udp_outgoing
81DEFAULT: no_addr.s_addr
82DOC_START
83 Usage: tcp_incoming_address 10.20.30.40
84 udp_outgoing_address fully.qualified.domain.name
85
86 These tags have replaced 'bind_address' and 'outbound_address'
87 to provide more control for multihomed hosts.
88
89 tcp_incoming_address is used for the HTTP socket which accepts
90 connections from clients and other caches.
91 tcp_outgoing_address is used for connections made to remote
92 servers and other caches.
93 udp_incoming_address is used for the ICP socket receiving packets
94 from other caches.
95 udp_outgoing_address is used for ICP packets sent out to other
96 caches.
97
98 The defaults behaviour is to not bind to any specific address.
99
100 NOTE, udp_incoming_address and udp_outgoing_address can not have
101 the same value since they both use port 3130.
102
103tcp_incoming_address 0.0.0.0
104tcp_outgoing_address 0.0.0.0
105udp_incoming_address 0.0.0.0
106udp_outgoing_address 0.0.0.0
107DOC_END
108
109
110# OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
111#-----------------------------------------------------------------------------
112
113NAME: cache_host
114TYPE: cachehost
115DOC_START
116 To specify other caches in a hierarchy, use the format:
117
118 hostname type http_port icp_port
119
120 For example,
121
122 # proxy icp
123 # hostname type port port options
124 # -------------------- -------- ----- ----- -----------
125 cache_host bigserver.usc.edu parent 3128 3130 [proxy-only]
126 cache_host littleguy1.usc.edu sibling 3128 3130 [proxy-only]
127 cache_host littleguy1.usc.edu sibling 3128 3130 [proxy-only]
128
129 type: either 'parent', 'sibling', or 'multicast'.
130
131 proxy_port: The port number where the cache listens for proxy
132 requests.
133
134 icp_port: Used for querying neighbor caches about
135 objects. To have a non-ICP neighbor
136 specify '7' for the ICP port and make sure the
137 neighbor machine has the UDP echo port
138 enabled in its /etc/inetd.conf file.
139
140 options: proxy-only
141 weight=n
142 ttl=n
143 no-query
144 default
145 round-robin
146 multicast-responder
147
148 use 'proxy-only' to specify that objects fetched
149 from this cache should not be saved locally.
150
151 use 'weight=n' to specify a weighted parent.
152 The weight must be an integer. The default weight
153 is 1, larger weights are favored more.
154
155 use 'ttl=n' to specify a IP multicast TTL to use
156 when sending an ICP request to this address.
157 Only useful when sending to a multicast group.
158 Because we don't accept ICP replies from random
159 hosts, you must configure other group members as
160 peers with the 'multicast-responder' option below.
161
162 use 'no-query' to NOT send ICP queries to this
163 neighbor.
164
165 use 'default' if this is a parent cache which can
166 be used as a "last-resort." You should probably
167 only use 'default' in situations where you cannot
168 use ICP with your parent cache(s).
169
170 use 'round-robin' to define a set of parents which
171 should be used in a round-robin fashion in the
172 absence of any ICP queries.
173
174 'multicast-responder' indicates that the named peer
175 is a member of a multicast group. ICP queries will
176 not be sent directly to the peer, but ICP replies
177 will be accepted from it.
178
179 NOTE: non-ICP neighbors must be specified as 'parent'.
180
181cache_host hostname type 3128 3130
182DOC_END
183
184
185NAME: cache_host_domain
186TYPE: hostdomain
187DOC_START
188 Use to limit the domains for which a neighbor cache will be queried.
189 Usage:
190
191 cache_host_domain cache-host domain [domain ...]
192 cache_host_domain cache-host !domain
193
194 For example, specifying
195
196 cache_host_domain bigserver.usc.edu .edu
197
198 has the effect such that UDP query packets are sent to
199 'bigserver' only when the requested object exists on a
200 server in the .edu domain. Prefixing the domainname
201 with '!' means that the cache will be queried for objects
202 NOT in that domain.
203
204 NOTE: * Any number of domains may be given for a cache-host,
205 either on the same or separate lines.
206 * When multiple domains are given for a particular
207 cache-host, the first matched domain is applied.
208 * Cache hosts with no domain restrictions are queried
209 for all requests.
210 * There are no defaults.
211 * There is also a 'cache_host_acl' tag in the ACL
212 section.
213DOC_END
214
215
216NAME: neighbor_type_domain
217TYPE: hostdomaintype
218DOC_START
219 usage: neighbor_type_domain parent|sibling domain domain ...
220
221 Modifying the neighbor type for specific domains is now
222 possible. You can treat some domains differently than the the
223 default neighbor type specified on the 'cache_host' line.
224 Normally it should only be necessary to list domains which
225 should be treated differently because the default neighbor type
226 applies for hostnames which do not match domains listed here.
227
228EXAMPLE:
229 cache_host parent cache.foo.org 3128 3130
230 neighbor_type_domain cache.foo.org sibling .com .net
231 neighbor_type_domain cache.foo.org sibling .au .de
232DOC_END
233
234NAME: single_parent_bypass
235COMMENT: on|off
236TYPE: onoff
237DEFAULT: 0
238LOC: Config.singleParentBypass
239DOC_START
240 This tag specifies that it is okay to bypass the hierarchy
241 "Pinging" when there is only a single parent for a given URL.
242
243 Usage: single_parent_bypass on|off
244
245 Before actually sending ICP "ping" packets to parents and
246 neighbors, we figure out which hosts would be pinged based
247 on the cache_host_domain rules, etc. Often it may be the
248 case that only a single parent cache would be pinged.
249
250 Since there is only a single parent, there is a very good
251 chance that we will end up fetching the object from that
252 parent. For this reason, it may be beneficial to avoid
253 the ping and just fetch the object anyway.
254
255 However, if we avoid the ping, we will be assuming that the
256 parent host is reachable and that the cache process is running.
257 By using the ping, we can be reasonably sure that the parent
258 host will be able to handle our request. If the ping fails then
259 it may be possible to fetch the object directly from the source.
260
261 To favor the resiliency provided by the ping algorithm,
262 single_parent_bypass is 'off' by default.
263
264single_parent_bypass off
265DOC_END
266
267
268NAME: source_ping
269COMMENT: on|off
270TYPE: onoff
271DEFAULT: 0
272LOC: Config.sourcePing
273DOC_START
274 If source_ping is enabled, then squid will include the source
275 provider site in its selection algorithm. This is accomplished
276 by sending ICP "HIT" packets to the UDP echo port of the source
277 host. Note that using source_ping may send a fair amount of UDP
278 traffic out on the Internet and may irritate paranoid network
279 administrators.
280
281 Note that source_ping is incompatible with inside_firewall.
282 For hosts beyond the firewall, source_ping packets will never
283 be sent.
284
285 By default, source_ping is off.
286
287source_ping off
288DOC_END
289
290
291NAME: neighbor_timeout neighbour_timeout
292COMMENT: (seconds)
293DEFAULT: 2
294TYPE: time_sec
295LOC: Config.neighborTimeout
296DOC_START
297 This controls how long to wait for replies from neighbor caches.
298 If none of the parent or neighbor caches reply before this many
299 seconds (due to dropped packets or slow links), then the object
300 request will be satisfied from the default source. The default
301 timeout is two seconds.
302
303neighbor_timeout 2 seconds
304DOC_END
305
306
307NAME: hierarchy_stoplist
308TYPE: wordlist
309LOC: Config.hierarchy_stoplist
310DOC_START
311 A list of words which, if found in a URL, cause the object to
312 be handled directly by this cache. In other words, use this
313 to not query neighbor caches for certain objects. You may
314 list this option multiple times.
315
316 The default is to directly fetch URLs containing 'cgi-bin' or '?'.
317
318hierarchy_stoplist cgi-bin ?
319DOC_END
320
321
322NAME: cache_stoplist
323TYPE: wordlist
324LOC: Config.cache_stoplist
325DOC_START
326 A list of words which, if found in a URL, cause the object to
327 immediately removed from the cache. In other words, use this
328 to force certain objects to never be cached. You may list this
329 option multiple times.
330
331 The default is to not cache URLs containing 'cgi-bin' or '?'.
332
333cache_stoplist cgi-bin ?
334DOC_END
335
336
337NAME: cache_stoplist_pattern
338COMMENT: # case sensitive
339TYPE: regexlist
340LOC: Config.cache_stop_relist
341DOC_NONE
342
343NAME: cache_stoplist_pattern/i
344COMMENT: # case insensitive
345TYPE: regexlist_icase
346LOC: Config.cache_stop_relist
347DOC_START
348 Just like 'cache_stoplist' but you can use regular expressions
349 instead of simple string matching. There is no default.
350
351cache_stoplist_pattern
352DOC_END
353
354
355# OPTIONS WHICH AFFECT THE CACHE SIZE
356#-----------------------------------------------------------------------------
357
358
359NAME: cache_mem
360COMMENT: (in megabytes)
361TYPE: int
362DEFAULT: 8
363LOC: Config.Mem.maxSize
364DOC_START
365 Maximum amout of VM used to store objects in memory.
366 This includes:
367 in-transit objects,
368 negative-cached objects,
369 "hot" objects
370 The value of cache_mem is an upper limit on the size of the
371 "in-memory object data" pool. This is a pool of 4k pages used
372 to hold object data.
373
374 In-transit objects have priority over the others. When
375 additional space is needed for incoming data, negative-cached
376 and hot objects will be released. In other words, the
377 negative-cached and hot objects will fill up any unused space
378 not needed for in-transit objects.
379
380 The values of cache_mem_low and cache_mem_high (below) can be
381 used to tune the use of the memory pool. When the high mark is
382 reached, in-transit and hot objects will be released to clear
383 space. When an object transfer is completed, it will remain in
384 memory only if the current memory usage is below the low water
385 mark.
386
387 The default is 8 Megabytes.
388
389cache_mem 8
390DOC_END
391
392
393NAME: cache_swap_low
394COMMENT: (percent, 0-100)
395TYPE: int
396DEFAULT: 90
397LOC: Config.Swap.lowWaterMark
398DOC_NONE
399
400NAME: cache_swap_high
401COMMENT: (percent, 0-100)
402TYPE: int
403DEFAULT: 95
404LOC: Config.Swap.highWaterMark
405DOC_START
406 The low- and high-water marks for cache LRU replacement.
407 LRU replacement begins when the high-water mark is reached
408 and ends when enough objects have been removed and the low-water
409 mark is reached. Defaults are 90% and 95%.
410
411cache_swap_low 90
412cache_swap_high 95
413DOC_END
414
415
416NAME: cache_mem_low
417COMMENT: (in percent, 0-100)
418TYPE: int
419DEFAULT: 75
420LOC: Config.Mem.lowWaterMark
421DOC_NONE
422
423NAME: cache_mem_high
424COMMENT: (in percent, 0-100)
425TYPE: int
426DEFAULT: 95
427LOC: Config.Mem.highWaterMark
428DOC_START
429 The low- and high-water mark for cache memory storage. When
430 the amount of RAM used by the hot-object RAM cache reaches this
431 point, the cache starts throwing objects out of the RAM cache
432 (but they remain on disk). Defaults are 75% and 90%.
433
434cache_mem_low 75
435cache_mem_high 90
436DOC_END
437
438
439NAME: maximum_object_size
440TYPE: int
441DEFAULT: 4096
442LOC: Config.Store.maxObjectSize
443DOC_START
444 Objects larger than this size will NOT be saved on disk. The
445 value is specified in kilobytes, and the default is 4MB.
446
447maximum_object_size 4096
448DOC_END
449
450
451NAME: ipcache_size
452COMMENT: (number of entries)
453TYPE: int
454DEFAULT: 1024
455LOC: Config.ipcache.size
456DOC_NONE
457
458NAME: ipcache_low
459COMMENT: (percent)
460TYPE: int
461DEFAULT: 90
462LOC: Config.ipcache.low
463DOC_NONE
464
465NAME: ipcache_high
466COMMENT: (percent)
467TYPE: int
468DEFAULT: 95
469LOC: Config.ipcache.high
470DOC_START
471 The size, low-, and high-water marks for the IP cache.
472
473ipcache_size 1024
474ipcache_low 90
475ipcache_high 95
476DOC_END
477
478
479# LOGFILE PATHNAMES AND CACHE DIRECTORIES
480#-----------------------------------------------------------------------------
481
482NAME: cache_dir
483TYPE: cachedir
484DOC_START
485 Directory for on-disk cache storage. The cache will change into
486 this directory when running. The default is
487 /usr/local/squid/cache.
488
489 You can specify multiple cache_dir lines to spread the
490 cache among different disk partitions.
491
492cache_dir /usr/local/squid/cache 256 16 16
493DOC_END
494
495
496NAME: cache_access_log
497TYPE: string
498DEFAULT: /usr/local/squid/logs/access.log
499LOC: Config.Log.access
500DOC_START
501 Logs the client request activity. Contains an entry for
502 every HTTP and ICP request received.
503
504cache_access_log /usr/local/squid/logs/access.log
505DOC_END
506
507
508NAME: cache_log
509TYPE: string
510DEFAULT: /usr/local/squid/logs/cache.log
511LOC: Config.Log.log
512DOC_START
513 Cache logging file. Set logging levels with "debug_options" below.
514
515cache_log /usr/local/squid/logs/cache.log
516DOC_END
517
518
519NAME: cache_store_log
520TYPE: string
521DEFAULT: /usr/local/squid/logs/store.log
522LOC: Config.Log.store
523DOC_START
524 Logs the activities of the storage manager. Shows which
525 objects are ejected from the cache, and which objects are
526 saved and for how long. To disable, enter "none".
527
528cache_store_log /usr/local/squid/logs/store.log
529DOC_END
530
531
532NAME: cache_swap_log
533TYPE: string
534LOC: Config.Log.swap
535DOC_START
536 Location for the cache "swap log." This log file holds the
537 metadata of objects saved on disk. It is used to rebuild the
538 cache during startup. Normally this file resides in the first
539 'cache_dir' directory, but you may specify an alternate
540 pathname here. Note you must give a full filename, not just
541 a directory.
542
543cache_swap_log
544DOC_END
545
546
547NAME: emulate_httpd_log
548COMMENT: on|off
549TYPE: onoff
550DEFAULT: 0
551LOC: Config.commonLogFormat
552DOC_START
553 The Cache can emulate the log file format which many 'httpd'
554 programs use. To disable/enable this emulation, set
555 emulate_httpd_log to 'off' or 'on'. The default
556 is to use the native log format.
557
558emulate_httpd_log off
559DOC_END
560
561
562NAME: mime_table
563TYPE: pathname_check
564DEFAULT: /usr/local/squid/etc/mime.conf
565LOC: Config.mimeTablePathname
566DOC_START
567 Pathname to Squid's MIME table which has the format
568
569 regex content-type icon content-encoding transfer-mode
570
571mime_table /usr/local/squid/etc/mime.conf
572DOC_END
573
574
575NAME: log_mime_hdrs
576COMMENT: on|off
577TYPE: onoff
578LOC: Config.logMimeHdrs
579DOC_START
580 The Cache can record both the request and the response
581 MIME headers for each HTTP transaction. The headers are
582 encoded safely and will appear as two bracketed fields
583 at the end of the access log (for either the native
584 or httpd-emulated log formats). To enable this logging
585 set log_mime_hdrs to 'on'.
586
587 NOTE: support for this may require you to define
588 LOG_FULL_HEADERS before compiling.
589
590log_mime_hdrs off
591DOC_END
592
593
594NAME: useragent_log
595TYPE: string
596LOC: Config.Log.useragent
597DOC_START
598 If compiled with "-DUSE_USERAGENT_LOG=1" Squid will write
599 the User-Agent field from HTTP requests to the filename
600 specified here. By default useragent_log is disabled.
601
602useragent_log none
603DOC_END
604
605
606NAME: pid_filename
607TYPE: string
608DEFAULT: /usr/local/squid/logs/squid.pid
609LOC: Config.pidFilename
610DOC_START
611 A pathname to write the process-id to. To disable, enter "none".
612
613pid_filename /usr/local/squid/logs/squid.pid
614DOC_END
615
616
617NAME: debug_options
618TYPE: string_optional
619DEFAULT: ALL,1
620LOC: Config.debugOptions
621DOC_START
622 Logging options are set as section,level where each source file
623 is assigned a unique section. Lower levels result in less
624 output, Full debugging (level 9) can result in a very large
625 log file, so be careful. The magic word "ALL" sets debugging
626 levels for all sections. We recommend normally running with
627 "ALL,1".
628
629debug_options ALL,1
630DOC_END
631
632
633NAME: ident_lookup
634COMMENT: on|off
635TYPE: onoff
636DEFAULT: 0
637LOC: Config.identLookup
638DOC_START
639 If you wish to make an RFC931/ident lookup of the client username
640 for each connection, enable this. It is off by default.
641
642ident_lookup off
643DOC_END
644
645
646NAME: log_fqdn
647COMMENT: on|off
648TYPE: onoff
649DEFAULT: 0
650LOC: Config.Log.log_fqdn
651DOC_START
652 Turn this on if you wish to log fully qualified domain names
653 in the access.log.
654
655log_fqdn off
656DOC_END
657
658
659NAME: client_netmask
660TYPE: address
661LOC: Config.Addrs.client_netmask
662DOC_START
663 A netmask for client addresses in logfiles and cachemgr output.
664 Change this to protect the privacy of your cache clients.
665
666client_netmask 255.255.255.255
667DOC_END
668
669
670# OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
671#-----------------------------------------------------------------------------
672
673# TAG: ftpget_program
674# Where to find the 'ftpget' program that retrieves FTP data (HTTP
675# and Gopher protocol support are built into the cache).
676#
677# To disable ftpget and the ability to retrieve FTP objects, set
678# this to "none". Note that ftpget is automatically disabled for
679# http_accel mode.
680#
681#ftpget_program @DEFAULT_FTPGET@
682
683# TAG: ftpget_options
684# Options for the 'ftpget' program. Please run 'ftpget' without
685# any arguments to see a list of options. The default is
686# no options. An example is
687#
688# ftpget_options -n 60 -R -W
689#
690#ftpget_options
691
692NAME: ftp_user
693TYPE: string
694DEFAULT: squid@
695LOC: Config.Ftp.anon_user
696DOC_START
697 If you want the anonymous login password to be more informative
698 (and enable the use of picky ftp servers), set this to something
699 resonable for your domain, like wwwuser@somewhere.net
700
701 The reason why this is domainless by default is that the
702 request can be made on the behalf of a user in any domain,
703 depending on how the cache is used.
704 Some ftp server also validate that the email address is valid
705 (for example perl.com).
706
707ftp_user squid@
708DOC_END
709
710
711NAME: cache_dns_program
712TYPE: pathname_check
713DEFAULT: /usr/local/squid/bin/dnsserver
714LOC: Config.Program.dnsserver
715DOC_START
716 Specify the location of the executable for dnslookup process.
717
718cache_dns_program /usr/local/squid/bin/dnsserver
719DOC_END
720
721
722NAME: dns_children
723TYPE: int
724DEFAULT: 5
725LOC: Config.dnsChildren
726DOC_START
727 The number of processes spawn to service DNS name lookups.
728 For heavily loaded caches on large servers, you should
729 probably increase this value to at least 10. The maximum
730 is 32. The default is 5.
731
732 To disable dnsservers, set this to 0. NOTE, this is very
733 strongly discouraged. If you disable dnsservers your Squid
734 process will BLOCK on DNS lookups!
735
736dns_children 5
737DOC_END
738
739
740NAME: dns_defnames
741COMMENT: on|off
742TYPE: onoff
743DEFAULT: 0
744LOC: Config.Options.res_defnames
745DOC_START
746 Normally the 'dnsserver' disables the RES_DEFNAMES resolver
747 option (see res_init(3)). This prevents caches in a hierarchy
748 from interpreting single-component hostnames locally. To allow
749 dnsserver to handle single-component names, enable this
750 option.
751
752dns_defnames off
753DOC_END
754
755
756NAME: unlinkd_program
757TYPE: pathname_check
758DEFAULT: /usr/local/squid/bin/unlinkd
759LOC: Config.Program.unlinkd
760DOC_START
761 Specify the location of the executable for file deletion process.
762
763unlinkd_program /usr/local/squid/bin/unlinkd
764DOC_END
765
766
767NAME: pinger_program
768TYPE: pathname_check
769DEFAULT: /usr/local/squid/bin/pinger
770LOC: Config.Program.pinger
771DOC_START
772 Specify the location of the executable for the pinger process.
773
774pinger_program /usr/local/squid/bin/pinger
775DOC_END
776
777
778NAME: redirect_program
779TYPE: pathname_check
780LOC: Config.Program.redirect
781DOC_START
782 Specify the location of the executable for the URL redirector.
783 Currently, you must provide your own redirector program.
784 See the Release-Notes for how to write one.
785 By default, the redirector is not used.
786
787redirect_program /bin/false
788DOC_END
789
790
791NAME: redirect_children
792TYPE: int
793DEFAULT: 5
794LOC: Config.redirectChildren
795DOC_START
796 The number of redirector processes to spawn.
797
798redirect_children 5
799DOC_END
800
801
802# OPTIONS FOR TUNING THE CACHE
803#-----------------------------------------------------------------------------
804
805NAME: wais_relay
806TYPE: waisrelay
807DOC_START
808 Relay WAIS request to host (1st arg) at port (2 arg).
809
810wais_relay localhost 8000
811DOC_END
812
813
814NAME: request_size
815TYPE: int
816DEFAULT: 100
817LOC: Config.maxRequestSize
818DOC_START
819 Maximum allowed request size in kilobytes. If people are using
820 POST to upload files, then set this to the largest acceptable
821 filesize plus a few extra kbytes.
822
823request_size 100
824DOC_END
825
826
827NAME: refresh_pattern
828COMMENT: # case sensitive
829TYPE: refreshpattern
830DOC_NONE
831
832NAME: refresh_pattern/i
833COMMENT: # case insensitive
834TYPE: refreshpattern_icase
835DOC_START
836 usage: refresh_pattern regex min percent max
837
838 min and max are specified in MINUTES.
839 percent is an integer number.
840
841 Please see the file doc/Release-Notes-1.1.txt for a full
842 description of Squid's refresh algorithm. Basically a
843 cached object is:
844
845 FRESH if age < min
846 STALE if expires < now
847 STALE if age > max
848 FRESH if lm-factor < percent
849
850 The refresh_pattern lines are checked in the order listed here.
851 The first entry which matches is used. If none of the entries
852 match, then the default will be used.
853
854Default:
855refresh_pattern . 0 20% 4320
856DOC_END
857
858
859NAME: reference_age
860TYPE: time_min
861LOC: Config.referenceAge
862DOC_START
863 As a part of normal operation, Squid performs Least Recently
864 Used removal of cached objects. The LRU age for removal is
865 computed dynamically, based on the amount of disk space in
866 use. The 'reference_age' value defines the maximum LRU age.
867 For example, setting reference_age to '1 week' will cause
868 objects to be removed if they have not been accessed for a week
869 or more. If set to zero, LRU removal is disabled, and objects
870 will be removed only when disk usage is over the high water
871 mark. The default value is one year.
872
873 Specify a number here, followed by units of time. For example:
874 1 week
875 3.5 days
876 4 months
877 2.2 hours
878
879reference_age 1 year
880DOC_END
881
882
883NAME: quick_abort
884TYPE: quickabort
885DOC_START
886 By default the cache continues to retrieve objects from
887 aborted requests. This may be undesirable on slow (e.g. SLIP)
888 links and/or very busy caches. Impatient users may tie up
889 file descriptors by repeatedly aborting and re-requesting
890 non-cachable objects.
891
892 Usage: quick_abort min-kbytes percent max-kbytes
893
894 When the user aborts a request, Squid will check the
895 quick_abort values to the amount of data transfered until
896 then.
897
898 If the transfer has less than 'min-kbytes' remaining, it
899 will finish the retrieval. Setting minlength to -1 will
900 disable the quick_abort feature.
901
902 If the transfer has more than 'max-kbytes' remaining, it
903 will abort the retrieval.
904
905 If more than 'percent' of the transfer has completed, it will
906 finish the retrieval.
907
908quick_abort -1 0 0
909DOC_END
910
911
912NAME: negative_ttl
913COMMENT: (in minutes)
914TYPE: time_sec
915LOC: Config.negativeTtl
916DOC_START
917 Time-to-Live (TTL) for failed requests. Certain types of
918 failures (such as "connection refused" and "404 Not Found") are
919 negatively-cached for a small amount of time. The default is 5
920 minutes. Note that this is different from negative caching of
921 DNS lookups.
922
923negative_ttl 5 minutes
924DOC_END
925
926
927NAME: positive_dns_ttl
928COMMENT: (in minutes)
929TYPE: time_sec
930LOC: Config.positiveDnsTtl
931DOC_START
932 Time-to-Live (TTL) for positive caching of successful DNS lookups.
933 Default is 6 hours (360 minutes). If you want to minimize the
934 use of Squid's ipcache, set this to 1, not 0.
935
936positive_dns_ttl 360 minutes
937DOC_END
938
939
940NAME: negative_dns_ttl
941COMMENT: (in minutes)
942TYPE: time_sec
943LOC: Config.negativeDnsTtl
944DOC_START
945 Time-to-Live (TTL) for negative caching of failed DNS lookups.
946
947negative_dns_ttl 5 minutes
948DOC_END
949
950
951# TIMEOUTS
952#-----------------------------------------------------------------------------
953
954NAME: connect_timeout
955COMMENT: (in seconds)
956TYPE: time_sec
957LOC: Config.Timeout.connect
958DOC_START
959 Some systems (notably Linux) can not be relied upon to properly
960 time out connect(2) requests. Therefore the squid process
961 enforces its own timeout on server connections. This parameter
962 specifies how long to wait for the connect to complete. The
963 default is two minutes (120 seconds).
964
965connect_timeout 120 seconds
966DOC_END
967
968
969NAME: read_timeout
970COMMENT: (in minutes)
971TYPE: time_sec
972LOC: Config.Timeout.read
973DOC_START
974 The read_timeout is applied on server-side connections. After
975 each successful read(), the timeout will be extended by this
976 amount. If no data is read again after this amount of time,
977 the request is aborted and logged with ERR_READ_TIMEOUT. The
978 default is 15 minutes.
979
980read_timeout 15 minutes
981DOC_END
982
983
984NAME: defer_timeout
985COMMENT: (in minutes)
986TYPE: time_sec
987LOC: Config.Timeout.defer
988DOC_START
989 If your clients are behind slow (e.g. PPP/SLIP) connections,
990 then data may come in from the server-side faster than it can
991 be written to the client-side. When the server side gets too
992 far ahead of the client-side, subsequent reads will be deferred
993 until the client catches up. This timeout determines how long
994 to wait while in "deferred read mode." The default is one
995 hour.
996
997defer_timeout 60 minutes
998DOC_END
999
1000
1001NAME: request_timeout
1002TYPE: time_sec
1003LOC: Config.Timeout.request
1004DOC_START
1005 How long to wait for an HTTP request after connection
1006 establishment. For persistent connections, wait this long
1007 after the previous request completes.
1008
1009defer_timeout 30 seconds
1010DOC_END
1011
1012
1013NAME: client_lifetime
1014COMMENT: (in minutes)
1015TYPE: time_sec
1016LOC: Config.Timeout.lifetime
1017DOC_START
1018 The maximum amount of time that a client (browser) is allowed to
1019 remain connected to the cache process. This protects the Cache
1020 from having alot of sockets (and hence file descriptors) tied up
1021 in a CLOSE_WAIT state from remote clients that go away without
1022 properly shutting down (either because of a network failure or
1023 because of a poor client implementation). The default is one
1024 day, 1440 minutes.
1025
1026 NOTE: The default value is intended to be much larger than any
1027 client would ever need to be connected to your cache. You
1028 should probably change client_lifetime only as a last resort.
1029 If you seem to have many client connections tying up
1030 filedescriptors, we recommend first tuning the read_timeout,
1031 defer_timeout, and quick_abort values.
1032
1033client_lifetime 1 day
1034DOC_END
1035
1036
1037NAME: shutdown_lifetime
1038COMMENT: (in seconds)
1039TYPE: time_sec
1040LOC: Config.shutdownLifetime
1041DOC_START
1042 When SIGTERM or SIGHUP is received, the cache is put into
1043 "shutdown pending" mode until all active sockets are closed.
1044 This value is the lifetime to set for all open descriptors
1045 during shutdown mode. Any active clients after this many
1046 seconds will receive a 'timeout' message.
1047
1048shutdown_lifetime 30 seconds
1049DOC_END
1050
1051
1052# ACCESS CONTROLS
1053#-----------------------------------------------------------------------------
1054
1055NAME: acl
1056TYPE: acl
1057DOC_START
1058 Defining an Access List
1059
1060 acl aclname acltype string1 ...
1061 acl aclname acltype "file" ...
1062
1063 when using "file", the file should contain one item per line
1064
1065 acltype is one of src dst srcdomain dstdomain url_pattern
1066 urlpath_pattern time port proto method browser user
1067
1068 acl aclname src ip-address/netmask ... (clients IP address)
1069 acl aclname src addr1-addr2/netmask ... (range of addresses)
1070 acl aclname dst ip-address/netmask ... (URL host's IP address)
1071 acl aclname srcdomain foo.com ... (taken from reverse DNS lookup)
1072 acl aclname dstdomain foo.com ... (taken from the URL)
1073 acl aclname time [day-abbrevs] [h1:m1-h2:m2]
1074 day-abbrevs:
1075 S - Sunday
1076 M - Monday
1077 T - Tuesday
1078 W - Wednesday
1079 H - Thursday
1080 F - Friday
1081 A - Saturday
1082 h1:m1 must be less than h2:m2
1083 acl aclname url_regex ^http:// ... # regex matching on whole URL
1084 acl aclname urlpath_regex \.gif$ ... # regex matching on URL path only
1085 acl aclname port 80 70 21 ...
1086 acl aclname proto HTTP FTP ...
1087 acl aclname method GET POST ...
1088 acl aclname browser regexp
1089 acl aclname user username ... # string match on ident output.
1090 # use REQUIRED to accept any
1091 # non-null ident.
1092
1093acl manager proto cache_object
1094acl localhost src 127.0.0.1/255.255.255.255
1095acl all src 0.0.0.0/0.0.0.0
1096
1097acl SSL_ports port 443 563
1098acl Dangerous_ports port 7 9 19
1099acl CONNECT method CONNECT
1100DOC_END
1101
1102NAME: http_access
1103TYPE: acl_access
1104LOC: Config.accessList.HTTP
1105DOC_START
1106 Allowing or Denying access based on defined access lists
1107
1108 Access to the HTTP port:
1109 http_access allow|deny [!]aclname ...
1110
1111 Access to the ICP port:
1112 icp_access allow|deny [!]aclname ...
1113
1114 NOTE on default values:
1115
1116 If there are no "access" lines present, the default is to allow
1117 the request.
1118
1119 If none of the "access" lines cause a match, the default is the
1120 opposite of the last line in the list. If the last line was
1121 deny, then the default is allow. Conversely, if the last line
1122 is allow, the default will be deny. For these reasons, it is a
1123 good idea to have an "deny all" or "allow all" entry at the end
1124 of your access lists to avoid potential confusion.
1125
1126
1127 Only allow access to the cache manager functions from the local host.
1128http_access deny manager !localhost
1129http_access deny CONNECT !SSL_ports
1130http_access deny Dangerous_ports
1131
1132 Allow everything else
1133http_access allow all
1134DOC_END
1135
1136
1137NAME: icp_access
1138TYPE: acl_access
1139LOC: Config.accessList.ICP
1140DOC_START
1141 Reply to all ICP queries we receive
1142
1143icp_access allow all
1144DOC_END
1145
1146
1147NAME: miss_access
1148TYPE: acl_access
1149LOC: Config.accessList.MISS
1150DOC_START
1151 Use to force your neighbors to use you as a sibling instead of
1152 a parent. For example:
1153
1154 acl localclients src 172.16.0.0/16
1155 miss_access allow localclients
1156 miss_access deny !localclients
1157
1158 This means that only your local clients are allowed to fetch
1159 MISSES and all other clients can only fetch HITS.
1160
1161 By default, allow all clients who passed the http_access rules
1162 to fetch MISSES from us.
1163
1164miss_access allow all
1165DOC_END
1166
1167
1168NAME: cache_host_acl
1169TYPE: hostacl
1170DOC_START
1171 Just like 'cache_host_domain' but provides more flexibility by
1172 using ACL's.
1173
1174 cache_host_acl cache-host [!]aclname ...
1175
1176 NOTE: * Any number of ACL's may be given for a cache-host,
1177 either on the same or separate lines.
1178 * When multiple ACL's are given for a particular
1179 cache-host, the first matched ACL is applied.
1180 * Cache hosts with no domain or ACL restrictions are
1181 queried for all requests.
1182 * There are no defaults.
1183DOC_END
1184
1185
1186# ADMINISTRATIVE PARAMETERS
1187#-----------------------------------------------------------------------------
1188
1189NAME: cache_mgr
1190TYPE: string
1191DEFAULT: webmaster
1192LOC: Config.adminEmail
1193DOC_START
1194 Email-address of local cache manager who will receive
1195 mail if the cache dies. The default is "webmaster."
1196
1197cache_mgr webmaster
1198DOC_END
1199
1200
1201NAME: cache_effective_user
1202TYPE: effectiveuser
1203DOC_START
1204 If the cache is run as root, it will change its effective/real
1205 UID/GID to the UID/GID specified below. The default is not to
1206 change UID/GID.
1207
1208cache_effective_user nobody nogroup
1209DOC_END
1210
1211
1212NAME: visible_hostname
1213TYPE: string
1214LOC: Config.visibleHostname
1215DOC_START
1216 If you want to present a special hostname in error messages, etc,
1217 then define this. Otherwise, the return value of gethostname()
1218 will be used.
1219
1220visible_hostname www-cache.foo.org
1221DOC_END
1222
1223
1224# OPTIONS FOR THE CACHE REGISTRATION SERVICE
1225#-----------------------------------------------------------------------------
1226
1227# This section contains parameters for the (optional) cache
1228# announcement service. This service is provided to help
1229# cache administrators locate one another in order to join or
1230# create cache hierarchies.
1231#
1232# An 'announcement' message is sent (via UDP) to the registration
1233# service by Squid. By default, the annoucement message is NOT
1234# SENT unless you enable it with 'cache_announce' below.
1235#
1236# The announcement message includes your hostname, plus the
1237# following information from this configuration file:
1238#
1239# http_port
1240# icp_port
1241# cache_mgr
1242#
1243# All current information is processed regularly and made
1244# available on the Web at http://www.nlanr.net/Cache/Tracker/.
1245
1246
1247NAME: cache_announce
1248TYPE: cacheannounce
1249DOC_START
1250 This is how frequently to send cache announcements. The default
1251 is `0' which disables sending the announcement messages.
1252
1253 To enable announcing your cache, just uncomment the line below.
1254
1255cache_announce 24
1256DOC_END
1257
1258
1259NAME: announce_to
1260TYPE: announceto
1261DOC_START
1262 This is the hostname and portnumber where the registration message
1263 will be sent.
1264
1265 Format: announce_to host[:port] [filename]
1266
1267 Hostname will default to 'sd.cache.nlanr.net' and port will default
1268 to 3131. If the 'filename' argument is given, the contents of that
1269 file will be included in the announce message.
1270
1271announce_to sd.cache.nlanr.net:3131
1272DOC_END
1273
1274
1275# HTTPD-ACCELERATOR OPTIONS
1276#-----------------------------------------------------------------------------
1277
1278NAME: httpd_accel
1279TYPE: httpdaccel
1280DOC_START
1281 If you want to run squid as an httpd accelerator, define the
1282 host name and port number where the real HTTP server is.
1283
1284 If you want virtual host support then specify the hostname
1285 as "virtual".
1286
1287httpd_accel real_httpd_host real_httpd_port
1288DOC_END
1289
1290
1291NAME: httpd_accel_with_proxy
1292COMMENT: on|off
1293TYPE: onoff
1294DEFAULT: 0
1295LOC: Config.Accel.withProxy
1296DOC_START
1297 If you want to use squid as both a local httpd accelerator
1298 and as a proxy, change this to 'on'.
1299
1300httpd_accel_with_proxy off
1301DOC_END
1302
1303
1304NAME: httpd_accel_uses_host_header
1305COMMENT: on|off
1306TYPE: onoff
1307LOC: opt_accel_uses_host
1308DOC_START
1309 HTTP/1.1 requests include a Host: header which is basically the
1310 hostname from the URL. Squid can be an accelerator for
1311 different HTTP servers by looking at this header. However,
1312 Squid does NOT check the value of the Host header, so it opens
1313 a big security hole. We recommend that this option remain
1314 disabled unless you are sure of what you are doing.
1315
1316httpd_accel_uses_host_header off
1317DOC_END
1318
1319
1320# MISCELLANEOUS
1321#-----------------------------------------------------------------------------
1322
1323NAME: dns_testnames
1324TYPE: wordlist
1325LOC: Config.dns_testname_list
1326DOC_START
1327 The DNS tests exit as soon as the first site is successfully looked up
1328
1329 If you want to disable DNS tests, do not comment out or delete this
1330 list. Instead use the -D command line option
1331
1332dns_testnames internic.net usc.edu cs.colorado.edu mit.edu yale.edu
1333DOC_END
1334
1335
1336NAME: logfile_rotate
1337TYPE: int
1338DEFAULT: 10
1339LOC: Config.Log.rotateNumber
1340DOC_START
1341 Specifies the number of logfile rotations to make upon receiving
1342 a USR1 signal. The default is 10, which will rotate with
1343 extensions 0 through 9. Setting logfile_rotate to 0 will
1344 disable the rotation, but the logfiles are still closed and
1345 re-opened. This will enable you to rename the logfiles yourself
1346 just before sending a USR1 signal to the squid process.
1347
1348logfile_rotate 10
1349DOC_END
1350
1351
1352NAME: append_domain
1353TYPE: appenddomain
1354DOC_START
1355 Appends local domain name to hostnames without any dots in them.
1356 append_domain must begin with a period.
1357
1358append_domain .yourdomain.com
1359DOC_END
1360
1361
1362NAME: tcp_recv_bufsize
1363TYPE: int
1364DEFAULT: 0
1365LOC: Config.tcpRcvBufsz
1366DOC_START
1367 Size of receive buffer to set for TCP sockets. Probably just
1368 as easy to change your kernel's default. Set to zero to use
1369 the default buffer size.
1370
1371tcp_recv_bufsize 0
1372DOC_END
1373
1374
1375NAME: ssl_proxy
1376TYPE: proxy
1377LOC: Config.sslProxy
1378DOC_START
1379 Specify the name of a 'cache_host' listed above, or a hostname
1380 and port number where all SSL requests should be forwarded to.
1381
1382 Usage: ssl_proxy cache_host
1383 ssl_proxy host:port
1384
1385ssl_proxy
1386DOC_END
1387
1388
1389NAME: passthrough_proxy
1390TYPE: proxy
1391LOC: Config.passProxy
1392DOC_START
1393 Specify the name of a 'cache_host' listed above, or a hostname
1394 and port number where all non-GET (i.e. POST, PUT) requests
1395 should be forwarded to.
1396
1397 Usage: passthrough_proxy cache_host
1398 passthrough_proxy host:port
1399
1400passthrough_proxy
1401DOC_END
1402
1403
1404NAME: proxy_auth
1405TYPE: proxyauth
1406DOC_START
1407 Usage: proxy_auth passwd_file [ ignore-domain ]
1408
1409 'passwd_file' is an apache-style file of passwords for
1410 authenticated proxy access Looks like user:password, with the
1411 password being standard crypt() format. Proxy authentication
1412 is disabled by default.
1413
1414 'ignore-domain' is a domain name for which authorization will
1415 *not* be required.
1416
1417 NOTE, proxy_auth support is not compiled into Squid by default.
1418 To use this feature you must enable the USE_PROXY_AUTH option
1419 near the top of src/Makefile.
1420
1421proxy_auth /dev/null
1422DOC_END
1423
1424
1425NAME: err_html_text
1426TYPE: errhtml
1427DOC_START
1428 HTML text to include in error messages. Make this a "mailto"
1429 URL to your admin address, or maybe just a link to your
1430 organizations Web page.
1431
1432err_html_text
1433DOC_END
1434
1435
1436NAME: deny_info
1437TYPE: denyinfo
1438LOC: Config.denyInfoList
1439DOC_START
1440 Usage: deny_info URL acl
1441
1442 This can be used to return a HTTP redirect for requests which
1443 do not pass the 'http_access' rules. A single ACL will cause
1444 the http_access check to fail. If a 'deny_info' line exists
1445 for that ACL then Squid returns a redirect to the given URL.
1446DOC_END
1447
1448
1449NAME: udp_hit_obj
1450COMMENT: on|off
1451TYPE: onoff
1452LOC: opt_udp_hit_obj
1453DOC_START
1454 If set, Squid will request UDP_HIT_OBJ replies from its
1455 neighbors. UDP_HIT_OBJ is nice because it saves bandwidth, but
1456 it can cause some other problems. For one it complicates
1457 calculating hit rates. Also, problems arise because the ICP
1458 query does not contain any HTTP request headers which may
1459 affect the reply.
1460
1461udp_hit_obj off
1462DOC_END
1463
1464
1465NAME: udp_hit_obj_size
1466TYPE: int
1467LOC: Config.udpMaxHitObjsz
1468DOC_START
1469 If set, Squid will limit UDP_HIT_OBJ size to be less than
1470 this value. Setting this value to more than SQUID_UDP_SO_SNDBUF
1471 will not work as expected. Set to zero to select the size
1472 permited by the socket.
1473udp_hit_obj_size 0
1474DOC_END
1475
1476
1477NAME: memory_pools
1478COMMENT: on|off
1479TYPE: onoff
1480LOC: opt_mem_pools
1481DOC_START
1482 If set, Squid will keep pools of allocated (but unused) memory
1483 available for future use. If memory is a premium on your
1484 system, disable this.
1485
1486memory_pools on
1487DOC_END
1488
1489NAME: forwarded_for
1490COMMENT: on|off
1491TYPE: onoff
1492LOC: opt_forwarded_for
1493DOC_START
1494 If set, Squid will include your system's IP address or name
1495 in the HTTP requests it forwards. By default it looks like
1496 this:
1497
1498 X-Forwarded-For: 192.1.2.3
1499
1500 If you disable this, it will appear as
1501
1502 X-Forwarded-For: unknown
1503
1504forwarded_for on
1505DOC_END
1506
1507NAME: log_icp_queries
1508COMMENT: on|off
1509TYPE: onoff
1510DEFAULT: 1
1511LOC: Config.Options.log_udp
1512DOC_START
1513 If set, ICP queries are logged to access.log. ICP logging
1514 is enabled by default, so uncomment and change the line
1515 below to disable it.
1516
1517log_icp_queries on
1518DOC_END
1519
1520
1521NAME: minimum_direct_hops
1522TYPE: int
1523DEFAULT: 4
1524LOC: Config.minDirectHops
1525DOC_START
1526 If using the ICMP pinging stuff, do direct fetches for sites
1527 which are no more than this many hops away.
1528
1529minimum_direct_hops 4
1530DOC_END
1531
1532
1533NAME: cachemgr_passwd
1534TYPE: cachemgrpasswd
1535DOC_START
1536 Specify passwords for cachemgr operations.
1537
1538 Usage: cachemgr_passwd password action action ...
1539
1540 valid actions are:
1541 shutdown *
1542 info
1543 stats/objects
1544 stats/vm_objects
1545 stats/utilization
1546 stats/ipcache
1547 stats/fqdncache
1548 stats/dns
1549 stats/redirector
1550 stats/io
1551 stats/reply_headers
1552 stats/filedescriptors
1553 stats/netdb
1554 log/status *
1555 log/enable *
1556 log/disable *
1557 log/clear *
1558 log *
1559 parameter
1560 server_list
1561 client_list
1562 squid.conf *
1563
1564 * Indicates actions which will not be performed without a
1565 valid password, others can be performed if not listed here.
1566
1567 To disable an action, set the password to "disable".
1568 To allow performing an action without a password, set the
1569 password to "none".
1570
1571 Use the keyword "all" to set the same password for all actions.
1572
1573cachemgr_passwd secret shutdown
1574cachemgr_passwd lesssssssecret info stats/objects
1575cachemgr_passwd disable all
1576DOC_END
1577
1578
1579# TAG: swap_level1_dirs
1580# Number of first-level directories to create for storing cached
1581# objects. Minimum 1, maximum 256, default 16.
1582#
1583#swap_level1_dirs 16
1584
1585# TAG: swap_level2_dirs
1586# Number of sub-directories to create under each first-level
1587# directory. Minimum 1, maximum 256, default 256.
1588#
1589#swap_level2_dirs 256
1590
1591NAME: store_avg_object_size
1592TYPE: int
1593DEFAULT: 20
1594LOC: Config.Store.avgObjectSize
1595DOC_START
1596 Average object size, used to estimate number of objects your
1597 cache can hold. See doc/Release-Notes-1.1.txt. The default is
1598 20K.
1599
1600store_avg_object_size 20
1601DOC_END
1602
1603NAME: store_objects_per_bucket
1604TYPE: int
1605DEFAULT: 50
1606LOC: Config.Store.objectsPerBucket
1607DOC_START
1608 Target number of objects per bucket in the store hash table.
1609 Lowering this value increases the total number of buckets and
1610 also the storage maintenance rate. The default is 20.
1611
1612store_objects_per_bucket 20
1613DOC_END
1614
1615
1616NAME: http_anonymizer
1617TYPE: httpanonymizer
1618LOC: Config.Options.anonymizer
1619DOC_START
1620 If you want to filter out certain HTTP request headers for
1621 privacy reasons, enable this option. There are three
1622 appropriate settings:
1623 'off' All HTTP request headers are passed.
1624 'standard' Specific headers are removed
1625 'paranoid' Only specific headers are allowed.
1626 To see which headers are allowed or denied, please see the
1627 http-anon.c source file.
1628
1629http_anonymizer off
1630DOC_END
1631
1632
1633NAME: client_db
1634COMMENT: on|off
1635TYPE: onoff
1636DEFAULT: 1
1637LOC: Config.Options.client_db
1638DOC_START
1639 If you want to disable collecting per-client statistics, then
1640 turn off client_db here.
1641
1642client_db on
1643DOC_END
1644
1645
1646NAME: netdb_low
1647TYPE: int
1648DEFAULT: 900
1649LOC: Config.Netdb.low
1650DOC_NONE
1651
1652NAME: netdb_high
1653TYPE: int
1654DEFAULT: 1000
1655LOC: Config.Netdb.high
1656DOC_START
1657 The low and high water marks for the ICMP measurement
1658 database. These are counts, not percents. The defaults are
1659 900 and 1000. When the high water mark is reached, database
1660 entries will be deleted until the low mark is reached.
1661
1662netdb_low 900
1663netdb_high 1000
1664DOC_END
1665
1666
1667NAME: netdb_ping_period
1668TYPE: time_sec
1669LOC: Config.Netdb.period
1670DOC_START
1671 The minimum period for measuring a site. There will be at
1672 least this much delay between successive pings to the same
1673 network. The default is five minutes.
1674
1675netdb_ping_period 5 minutes
1676DOC_END
1677
1678
1679NAME: query_icmp
1680COMMENT: on|off
1681TYPE: onoff
1682DEFAULT: 0
1683LOC: Config.Options.query_icmp
1684DOC_START
1685 If you want to ask your peers to include ICMP data in their ICP
1686 replies, enable this option.
1687
1688 If your peer has built squid with '-DUSE_ICMP=1' then that peer
1689 will send ICMP pings to origin server sites of the URLs it
1690 receives. If you enable this option then the ICP replies from
1691 that peer will include the ICMP data (if available). Then,
1692 when choosing a parent cache, Squid will choose the parent with
1693 the minimal RTT to the origin server. When this happens, the
1694 hierarchy field of the access.log will be
1695 "CLOSEST_PARENT_MISS". This option is off by default.
1696
1697query_icmp off
1698DOC_END
1699
1700NAME: clean_rate
1701TYPE: time_sec
1702LOC: Config.cleanRate
1703DOC_START
1704 XXX need docs
1705DOC_END
1706
1707NAME: always_direct
1708TYPE: acl_access
1709LOC: Config.accessList.AlwaysDirect
1710DOC_START
1711 XXX need docs
1712DOC_END
1713
1714NAME: never_direct
1715TYPE: acl_access
1716LOC: Config.accessList.NeverDirect
1717DOC_START
1718 XXX need docs
1719DOC_END
1720
1721NAME: viz_hack_addr
1722TYPE: vizhack
1723DOC_START
1724 XXX need docs
1725DOC_END
1726
1727#NAME: proxy_auth_ignore
1728#TYPE: regexplist_icase
1729#LOC: Config.proxyAuth.IgnoreDomains
1730#DOC_START
1731# XXX need docs
1732#DOC_END
1733
1734EOF
Mirror of https://github.com/squid-cache/squid.git