]> git.ipfire.org Git - thirdparty/squid.git/blame - src/cf.data.pre
Removed warnings for dec alpha cc on snmp code
[thirdparty/squid.git] / src / cf.data.pre
CommitLineData
3a278cb8 1
0f74202c 2COMMENT_START
3 WELCOME TO SQUID 1.2
4 --------------------
3a278cb8 5
cf5cc17e 6 This is the default Squid configuration file. You may wish
7 to look at http://cache.is.co.za/Squid/ for documentation,
8 or the Squid home page (http://Squid.nlanr.net/) for the FAQ.
3a278cb8 9
0f74202c 10COMMENT_END
3a278cb8 11
0f74202c 12COMMENT_START
3a278cb8 13 NETWORK OPTIONS
14 -----------------------------------------------------------------------------
0f74202c 15COMMENT_END
3a278cb8 16
934b03fc 17NAME: http_port ascii_port
18TYPE: ushortlist
f53b06f9 19DEFAULT: none
20DEFAULT_IF_NONE: 3128
934b03fc 21LOC: Config.Port.http
22DOC_START
cf5cc17e 23 The port number where Squid will listen for HTTP client
934b03fc 24 requests. Default is 3128, for httpd-accel mode use port 80.
25 May be overridden with -a on the command line.
26
27 You may specify multiple ports here, but they MUST all be on
28 a single line.
29
30http_port 3128
31DOC_END
32
33
34NAME: icp_port udp_port
35TYPE: ushort
36DEFAULT: 3130
37LOC: Config.Port.icp
38DOC_START
cf5cc17e 39 The port number where Squid send and receive ICP requests to
934b03fc 40 and from neighbor caches. Default is 3130. To disable use
41 "0". May be overridden with -u on the command line.
42
43icp_port 3130
44DOC_END
45
46
47NAME: mcast_groups
48TYPE: wordlist
49LOC: Config.mcast_group_list
1273d501 50DEFAULT: none
934b03fc 51DOC_START
cf5cc17e 52 This tag specifies a list of multicast groups which your server
53 should join to receive multicasted ICP requests.
934b03fc 54
55 NOTE! Be very careful what you put here! Be sure you
56 understand the difference between an ICP _query_ and an ICP
57 _reply_. This option is to be set only if you want to RECEIVE
58 multicast queries. Do NOT set this option to SEND multicast
a95856a0 59 ICP (use cache_peer for that). ICP replies are always sent via
934b03fc 60 unicast, so this option does not affect whether or not you will
61 receive replies from multicast group members.
62
63 You must be very careful to NOT use a multicast address which
64 is already in use by another group of caches. NLANR has been
65 assigned a block of multicast address space for use in Web
66 Caching. Plese write to us at nlanr-cache@nlanr.net to receive
67 an address for your own use.
68
cf5cc17e 69 If you are unsure about multicast, please read the Multicast
70 chapter in the Squid FAQ (http://squid.nlanr.net/Squid/FAQ/).
71
934b03fc 72 Usage: mcast_groups 239.128.16.128 224.0.1.20
73
cf5cc17e 74 By default, Squid doesn't listen on any multicast groups.
934b03fc 75
76mcast_groups 239.128.16.128
77DOC_END
78
79
80NAME: tcp_incoming_address bind_address
81TYPE: address
82LOC: Config.Addrs.tcp_incoming
270b86af 83DEFAULT: 0.0.0.0
934b03fc 84DOC_NONE
85
86NAME: tcp_outgoing_address outbound_address
87TYPE: address
88LOC: Config.Addrs.tcp_outgoing
270b86af 89DEFAULT: 255.255.255.255
934b03fc 90DOC_NONE
91
92NAME: udp_incoming_address
93TYPE: address
94LOC:Config.Addrs.udp_incoming
270b86af 95DEFAULT: 0.0.0.0
934b03fc 96DOC_NONE
97
98NAME: udp_outgoing_address
99TYPE: address
100LOC: Config.Addrs.udp_outgoing
270b86af 101DEFAULT: 255.255.255.255
934b03fc 102DOC_START
103 Usage: tcp_incoming_address 10.20.30.40
104 udp_outgoing_address fully.qualified.domain.name
105
934b03fc 106 tcp_incoming_address is used for the HTTP socket which accepts
107 connections from clients and other caches.
108 tcp_outgoing_address is used for connections made to remote
109 servers and other caches.
110 udp_incoming_address is used for the ICP socket receiving packets
111 from other caches.
112 udp_outgoing_address is used for ICP packets sent out to other
113 caches.
114
115 The defaults behaviour is to not bind to any specific address.
116
117 NOTE, udp_incoming_address and udp_outgoing_address can not have
118 the same value since they both use port 3130.
119
120tcp_incoming_address 0.0.0.0
121tcp_outgoing_address 0.0.0.0
122udp_incoming_address 0.0.0.0
123udp_outgoing_address 0.0.0.0
124DOC_END
125
0f74202c 126COMMENT_START
3a278cb8 127 OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
128 -----------------------------------------------------------------------------
0f74202c 129COMMENT_END
934b03fc 130
40a1495e 131NAME: cache_peer
132TYPE: peer
1273d501 133DEFAULT: none
0153d498 134LOC: Config.peers
934b03fc 135DOC_START
136 To specify other caches in a hierarchy, use the format:
137
138 hostname type http_port icp_port
139
140 For example,
141
142 # proxy icp
143 # hostname type port port options
144 # -------------------- -------- ----- ----- -----------
cf5cc17e 145 cache_peer parent.foo.net parent 3128 3130 [proxy-only]
146 cache_peer sib1.foo.net sibling 3128 3130 [proxy-only]
147 cache_peer sib2.foo.net sibling 3128 3130 [proxy-only]
934b03fc 148
149 type: either 'parent', 'sibling', or 'multicast'.
150
151 proxy_port: The port number where the cache listens for proxy
152 requests.
153
154 icp_port: Used for querying neighbor caches about
155 objects. To have a non-ICP neighbor
156 specify '7' for the ICP port and make sure the
157 neighbor machine has the UDP echo port
158 enabled in its /etc/inetd.conf file.
159
160 options: proxy-only
161 weight=n
162 ttl=n
163 no-query
164 default
165 round-robin
166 multicast-responder
b3264694 167 closest-only
934b03fc 168
169 use 'proxy-only' to specify that objects fetched
170 from this cache should not be saved locally.
171
172 use 'weight=n' to specify a weighted parent.
173 The weight must be an integer. The default weight
174 is 1, larger weights are favored more.
175
176 use 'ttl=n' to specify a IP multicast TTL to use
177 when sending an ICP request to this address.
178 Only useful when sending to a multicast group.
179 Because we don't accept ICP replies from random
180 hosts, you must configure other group members as
181 peers with the 'multicast-responder' option below.
182
183 use 'no-query' to NOT send ICP queries to this
184 neighbor.
185
186 use 'default' if this is a parent cache which can
187 be used as a "last-resort." You should probably
188 only use 'default' in situations where you cannot
189 use ICP with your parent cache(s).
190
191 use 'round-robin' to define a set of parents which
192 should be used in a round-robin fashion in the
193 absence of any ICP queries.
194
195 'multicast-responder' indicates that the named peer
196 is a member of a multicast group. ICP queries will
197 not be sent directly to the peer, but ICP replies
198 will be accepted from it.
199
b3264694 200 'closest-only' indicates that, for ICP_OP_MISS
201 replies, we'll only forward CLOSEST_PARENT_MISSes
202 and never FIRST_PARENT_MISSes.
203
934b03fc 204 NOTE: non-ICP neighbors must be specified as 'parent'.
205
a95856a0 206cache_peer hostname type 3128 3130
934b03fc 207DOC_END
208
209
210NAME: cache_host_domain
211TYPE: hostdomain
f1dc9b30 212DEFAULT: none
213LOC: none
934b03fc 214DOC_START
cf5cc17e 215 Use to limit the domains for which a neighbor cache will be
216 queried. Usage:
934b03fc 217
218 cache_host_domain cache-host domain [domain ...]
219 cache_host_domain cache-host !domain
220
221 For example, specifying
222
cf5cc17e 223 cache_host_domain parent.foo.net .edu
934b03fc 224
225 has the effect such that UDP query packets are sent to
226 'bigserver' only when the requested object exists on a
227 server in the .edu domain. Prefixing the domainname
228 with '!' means that the cache will be queried for objects
229 NOT in that domain.
230
231 NOTE: * Any number of domains may be given for a cache-host,
232 either on the same or separate lines.
233 * When multiple domains are given for a particular
234 cache-host, the first matched domain is applied.
235 * Cache hosts with no domain restrictions are queried
236 for all requests.
237 * There are no defaults.
238 * There is also a 'cache_host_acl' tag in the ACL
239 section.
240DOC_END
241
242
243NAME: neighbor_type_domain
244TYPE: hostdomaintype
f1dc9b30 245DEFAULT: none
246LOC: none
934b03fc 247DOC_START
248 usage: neighbor_type_domain parent|sibling domain domain ...
249
250 Modifying the neighbor type for specific domains is now
251 possible. You can treat some domains differently than the the
a95856a0 252 default neighbor type specified on the 'cache_peer' line.
934b03fc 253 Normally it should only be necessary to list domains which
254 should be treated differently because the default neighbor type
255 applies for hostnames which do not match domains listed here.
256
257EXAMPLE:
a95856a0 258 cache_peer parent cache.foo.org 3128 3130
934b03fc 259 neighbor_type_domain cache.foo.org sibling .com .net
260 neighbor_type_domain cache.foo.org sibling .au .de
261DOC_END
262
263NAME: single_parent_bypass
264COMMENT: on|off
265TYPE: onoff
f1dc9b30 266DEFAULT: off
17a0a4ee 267LOC: Config.onoff.single_parent_bypass
934b03fc 268DOC_START
269 This tag specifies that it is okay to bypass the hierarchy
270 "Pinging" when there is only a single parent for a given URL.
271
272 Usage: single_parent_bypass on|off
273
274 Before actually sending ICP "ping" packets to parents and
275 neighbors, we figure out which hosts would be pinged based
276 on the cache_host_domain rules, etc. Often it may be the
277 case that only a single parent cache would be pinged.
278
279 Since there is only a single parent, there is a very good
280 chance that we will end up fetching the object from that
281 parent. For this reason, it may be beneficial to avoid
282 the ping and just fetch the object anyway.
283
284 However, if we avoid the ping, we will be assuming that the
285 parent host is reachable and that the cache process is running.
286 By using the ping, we can be reasonably sure that the parent
287 host will be able to handle our request. If the ping fails then
288 it may be possible to fetch the object directly from the source.
289
290 To favor the resiliency provided by the ping algorithm,
291 single_parent_bypass is 'off' by default.
292
293single_parent_bypass off
294DOC_END
295
296
297NAME: source_ping
298COMMENT: on|off
299TYPE: onoff
f1dc9b30 300DEFAULT: off
17a0a4ee 301LOC: Config.onoff.source_ping
934b03fc 302DOC_START
cf5cc17e 303 If source_ping is enabled, then Squid will include the source
934b03fc 304 provider site in its selection algorithm. This is accomplished
305 by sending ICP "HIT" packets to the UDP echo port of the source
306 host. Note that using source_ping may send a fair amount of UDP
307 traffic out on the Internet and may irritate paranoid network
308 administrators.
309
310 Note that source_ping is incompatible with inside_firewall.
311 For hosts beyond the firewall, source_ping packets will never
312 be sent.
313
314 By default, source_ping is off.
315
316source_ping off
317DOC_END
318
934b03fc 319NAME: neighbor_timeout neighbour_timeout
320COMMENT: (seconds)
9e975e4e 321DEFAULT: 2 seconds
f1dc9b30 322TYPE: time_t
934b03fc 323LOC: Config.neighborTimeout
324DOC_START
325 This controls how long to wait for replies from neighbor caches.
326 If none of the parent or neighbor caches reply before this many
327 seconds (due to dropped packets or slow links), then the object
328 request will be satisfied from the default source. The default
329 timeout is two seconds.
330
331neighbor_timeout 2 seconds
332DOC_END
333
dc835977 334NAME: dead_peer_timeout
335COMMENT: (seconds)
336DEFAULT: 10 seconds
337TYPE: time_t
338LOC: Config.Timeout.deadPeer
339DOC_START
340 This controls how long Squid waits to declare a peer cache
341 as "dead." If there are no ICP replies received in this
342 amount of time, Squid will declare the peer dead and not
343 expect to receive any further ICP replies. However, it
344 continues to send ICP queries, and will mark the peer as
345 alive upon receipt of the first subsequent ICP reply.
346
347dead_peer_timeout 10 seconds
348DOC_END
349
934b03fc 350
351NAME: hierarchy_stoplist
352TYPE: wordlist
1273d501 353DEFAULT: none
934b03fc 354LOC: Config.hierarchy_stoplist
355DOC_START
356 A list of words which, if found in a URL, cause the object to
357 be handled directly by this cache. In other words, use this
358 to not query neighbor caches for certain objects. You may
359 list this option multiple times.
360
361 The default is to directly fetch URLs containing 'cgi-bin' or '?'.
362
363hierarchy_stoplist cgi-bin ?
364DOC_END
365
366
367NAME: cache_stoplist
368TYPE: wordlist
1273d501 369DEFAULT: none
934b03fc 370LOC: Config.cache_stoplist
371DOC_START
372 A list of words which, if found in a URL, cause the object to
373 immediately removed from the cache. In other words, use this
374 to force certain objects to never be cached. You may list this
375 option multiple times.
376
377 The default is to not cache URLs containing 'cgi-bin' or '?'.
378
379cache_stoplist cgi-bin ?
380DOC_END
381
382
383NAME: cache_stoplist_pattern
934b03fc 384TYPE: regexlist
385LOC: Config.cache_stop_relist
1273d501 386DEFAULT: none
934b03fc 387DOC_START
388 Just like 'cache_stoplist' but you can use regular expressions
389 instead of simple string matching. There is no default.
0153d498 390 Insert -i to get case-insensitive regular expressions.
934b03fc 391
392cache_stoplist_pattern
393DOC_END
394
395
0f74202c 396COMMENT_START
3a278cb8 397 OPTIONS WHICH AFFECT THE CACHE SIZE
398 -----------------------------------------------------------------------------
0f74202c 399COMMENT_END
934b03fc 400
401NAME: cache_mem
9906e724 402COMMENT: (bytes)
1b635117 403TYPE: b_size_t
9906e724 404DEFAULT: 8 MB
934b03fc 405LOC: Config.Mem.maxSize
406DOC_START
407 Maximum amout of VM used to store objects in memory.
408 This includes:
409 in-transit objects,
410 negative-cached objects,
411 "hot" objects
412 The value of cache_mem is an upper limit on the size of the
413 "in-memory object data" pool. This is a pool of 4k pages used
414 to hold object data.
415
416 In-transit objects have priority over the others. When
417 additional space is needed for incoming data, negative-cached
418 and hot objects will be released. In other words, the
419 negative-cached and hot objects will fill up any unused space
420 not needed for in-transit objects.
421
422 The values of cache_mem_low and cache_mem_high (below) can be
423 used to tune the use of the memory pool. When the high mark is
424 reached, in-transit and hot objects will be released to clear
425 space. When an object transfer is completed, it will remain in
426 memory only if the current memory usage is below the low water
427 mark.
428
429 The default is 8 Megabytes.
430
9906e724 431cache_mem 8 MB
934b03fc 432DOC_END
433
434
435NAME: cache_swap_low
436COMMENT: (percent, 0-100)
437TYPE: int
438DEFAULT: 90
439LOC: Config.Swap.lowWaterMark
440DOC_NONE
441
442NAME: cache_swap_high
443COMMENT: (percent, 0-100)
444TYPE: int
445DEFAULT: 95
446LOC: Config.Swap.highWaterMark
447DOC_START
cf5cc17e 448 The low- and high-water marks for cache LRU replacement. LRU
449 replacement begins when the high-water mark is reached and ends
450 when enough objects have been removed and the low-water mark is
451 reached. Defaults are 90% and 95%. If you have a large cache, 5%
452 could be hundreds of MB. If this is the case you may wish to
453 set these numbers closer together.
934b03fc 454
455cache_swap_low 90
456cache_swap_high 95
457DOC_END
458
459
460NAME: cache_mem_low
461COMMENT: (in percent, 0-100)
462TYPE: int
463DEFAULT: 75
464LOC: Config.Mem.lowWaterMark
465DOC_NONE
466
467NAME: cache_mem_high
468COMMENT: (in percent, 0-100)
469TYPE: int
470DEFAULT: 95
471LOC: Config.Mem.highWaterMark
472DOC_START
473 The low- and high-water mark for cache memory storage. When
474 the amount of RAM used by the hot-object RAM cache reaches this
475 point, the cache starts throwing objects out of the RAM cache
476 (but they remain on disk). Defaults are 75% and 90%.
477
478cache_mem_low 75
479cache_mem_high 90
480DOC_END
481
482
483NAME: maximum_object_size
9e975e4e 484COMMENT: (bytes)
1b635117 485TYPE: b_size_t
9906e724 486DEFAULT: 4096 KB
934b03fc 487LOC: Config.Store.maxObjectSize
488DOC_START
489 Objects larger than this size will NOT be saved on disk. The
cf5cc17e 490 value is specified in kilobytes, and the default is 4MB. If
491 you wish to get a high BYTES hit ratio, you should probably
492 increase this (one 32 MB object hit counts for 3200 10KB
493 hits). If you wish to increase speed more than your want to
494 save bandwidth you should leave this low.
934b03fc 495
9906e724 496maximum_object_size 4096 KB
934b03fc 497DOC_END
498
499
500NAME: ipcache_size
501COMMENT: (number of entries)
502TYPE: int
503DEFAULT: 1024
504LOC: Config.ipcache.size
505DOC_NONE
506
507NAME: ipcache_low
508COMMENT: (percent)
509TYPE: int
510DEFAULT: 90
511LOC: Config.ipcache.low
512DOC_NONE
513
514NAME: ipcache_high
515COMMENT: (percent)
516TYPE: int
517DEFAULT: 95
518LOC: Config.ipcache.high
519DOC_START
520 The size, low-, and high-water marks for the IP cache.
521
522ipcache_size 1024
523ipcache_low 90
524ipcache_high 95
525DOC_END
526
0f74202c 527COMMENT_START
3a278cb8 528 LOGFILE PATHNAMES AND CACHE DIRECTORIES
529 -----------------------------------------------------------------------------
0f74202c 530COMMENT_END
934b03fc 531
532NAME: cache_dir
533TYPE: cachedir
f1dc9b30 534DEFAULT: none
0108d71f 535DEFAULT_IF_NONE: @DEFAULT_SWAP_DIR@ 100 16 256
f1dc9b30 536LOC: Config.cacheSwap
934b03fc 537DOC_START
cf5cc17e 538 Usage:
539
540 cache_dir Directory-Name Mbytes Level-1 Level2
934b03fc 541
542 You can specify multiple cache_dir lines to spread the
543 cache among different disk partitions.
544
cf5cc17e 545 'Directory' is a top-level directory where cache swap
546 files will be stored. If you want to use an entire disk
547 for caching, then this can be the mount-point directory.
548 The directory must exist and be writable by the Squid
549 process. Squid will NOT create this directory for you.
550
551 If no 'cache_dir' lines are specified, the following
552 default will be used: @DEFAULT_SWAP_DIR@.
553
554 'Mbytes' is the amount of disk space (MB) to use under this
555 directory. The default is 100 MB. Change this to suit your
556 configuration.
557
558 'Level-1' is the number of first-level subdirectories which
559 will be created under the 'Directory'. The default is 16.
560
561 'Level-2' is the number of second-level subdirectories which
562 will be created under each first-level directory. The default
563 is 256.
564
5375428c 565cache_dir @DEFAULT_SWAP_DIR@ 100 16 256
934b03fc 566DOC_END
567
568
569NAME: cache_access_log
570TYPE: string
d0b98f84 571DEFAULT: @DEFAULT_ACCESS_LOG@
934b03fc 572LOC: Config.Log.access
573DOC_START
574 Logs the client request activity. Contains an entry for
575 every HTTP and ICP request received.
576
5375428c 577cache_access_log @DEFAULT_ACCESS_LOG@
934b03fc 578DOC_END
579
580
581NAME: cache_log
582TYPE: string
0153d498 583DEFAULT: @DEFAULT_CACHE_LOG@
934b03fc 584LOC: Config.Log.log
585DOC_START
cf5cc17e 586 Cache logging file. This is where general information about
587 your cache's behaviour goes. You can increase the amount of data
588 logged to this file with the "debug_options" tag below.
934b03fc 589
0153d498 590cache_log @DEFAULT_CACHE_LOG@
934b03fc 591DOC_END
592
593
594NAME: cache_store_log
595TYPE: string
0153d498 596DEFAULT: @DEFAULT_STORE_LOG@
934b03fc 597LOC: Config.Log.store
598DOC_START
599 Logs the activities of the storage manager. Shows which
600 objects are ejected from the cache, and which objects are
cf5cc17e 601 saved and for how long. To disable, enter "none". There are
602 not really utilities to analyse this data, so you can safely
603 disable it.
934b03fc 604
0153d498 605cache_store_log @DEFAULT_STORE_LOG@
934b03fc 606DOC_END
607
608
609NAME: cache_swap_log
610TYPE: string
611LOC: Config.Log.swap
1273d501 612DEFAULT: none
934b03fc 613DOC_START
d0d3ec94 614 Location for the cache "swap.log." This log file holds the
934b03fc 615 metadata of objects saved on disk. It is used to rebuild the
616 cache during startup. Normally this file resides in the first
617 'cache_dir' directory, but you may specify an alternate
618 pathname here. Note you must give a full filename, not just
cf5cc17e 619 a directory. Since this is the index for the whole object
620 list you CANNOT periodically rotate it!
934b03fc 621
622cache_swap_log
623DOC_END
624
625
626NAME: emulate_httpd_log
627COMMENT: on|off
628TYPE: onoff
f1dc9b30 629DEFAULT: off
17a0a4ee 630LOC: Config.onoff.common_log
934b03fc 631DOC_START
632 The Cache can emulate the log file format which many 'httpd'
633 programs use. To disable/enable this emulation, set
634 emulate_httpd_log to 'off' or 'on'. The default
cf5cc17e 635 is to use the native log format since it includes useful
636 information that Squid-specific log analysers use.
934b03fc 637
638emulate_httpd_log off
639DOC_END
640
641
642NAME: mime_table
f0b19334 643TYPE: string
0153d498 644DEFAULT: @DEFAULT_MIME_TABLE@
934b03fc 645LOC: Config.mimeTablePathname
646DOC_START
cf5cc17e 647 Pathname to Squid's MIME table. You shouldn't need to change
648 this, but the default file contains examples and formatting
649 information if you do.
934b03fc 650
0153d498 651mime_table @DEFAULT_MIME_TABLE@
934b03fc 652DOC_END
653
654
655NAME: log_mime_hdrs
656COMMENT: on|off
657TYPE: onoff
17a0a4ee 658LOC: Config.onoff.log_mime_hdrs
f1dc9b30 659DEFAULT: off
934b03fc 660DOC_START
cf5cc17e 661 The Cache can record both the request and the response MIME
662 headers for each HTTP transaction. The headers are encoded
663 safely and will appear as two bracketed fields at the end of
664 the access log (for either the native or httpd-emulated log
665 formats). To enable this logging set log_mime_hdrs to 'on'.
934b03fc 666
667log_mime_hdrs off
668DOC_END
669
670
671NAME: useragent_log
672TYPE: string
673LOC: Config.Log.useragent
f1dc9b30 674DEFAULT: none
934b03fc 675DOC_START
cf5cc17e 676 If configured with the "--enable-useragent_log" configure
677 option, Squid will write the User-Agent field from HTTP
678 requests to the filename specified here. By default
679 useragent_log is disabled.
934b03fc 680
681useragent_log none
682DOC_END
683
684
685NAME: pid_filename
686TYPE: string
0153d498 687DEFAULT: @DEFAULT_PID_FILE@
934b03fc 688LOC: Config.pidFilename
689DOC_START
cf5cc17e 690 A filename to write the process-id to. To disable, enter "none".
934b03fc 691
0153d498 692pid_filename @DEFAULT_PID_FILE@
934b03fc 693DOC_END
694
695
696NAME: debug_options
f1dc9b30 697TYPE: eol
934b03fc 698DEFAULT: ALL,1
699LOC: Config.debugOptions
700DOC_START
701 Logging options are set as section,level where each source file
702 is assigned a unique section. Lower levels result in less
703 output, Full debugging (level 9) can result in a very large
704 log file, so be careful. The magic word "ALL" sets debugging
705 levels for all sections. We recommend normally running with
706 "ALL,1".
707
708debug_options ALL,1
709DOC_END
710
711
712NAME: ident_lookup
713COMMENT: on|off
714TYPE: onoff
f1dc9b30 715DEFAULT: off
17a0a4ee 716LOC: Config.onoff.ident_lookup
934b03fc 717DOC_START
cf5cc17e 718 If you wish to make an RFC931/ident lookup of the client
719 username for each connection, enable this. It is off by
720 default.
934b03fc 721
722ident_lookup off
723DOC_END
724
725
726NAME: log_fqdn
727COMMENT: on|off
728TYPE: onoff
f1dc9b30 729DEFAULT: off
17a0a4ee 730LOC: Config.onoff.log_fqdn
934b03fc 731DOC_START
732 Turn this on if you wish to log fully qualified domain names
cf5cc17e 733 in the access.log. To do this Squid does a DNS lookup of all
734 IP's connecting to it. This can (in some situations) increase
735 latency, which makes your cache seem slower for interactive
736 browsing.
934b03fc 737
738log_fqdn off
739DOC_END
740
741
742NAME: client_netmask
743TYPE: address
744LOC: Config.Addrs.client_netmask
f1dc9b30 745DEFAULT: 255.255.255.255
934b03fc 746DOC_START
747 A netmask for client addresses in logfiles and cachemgr output.
748 Change this to protect the privacy of your cache clients.
cf5cc17e 749 A netmask of 255.255.255.0 will log all IP's in that range with
750 the last digit set to '0'.
934b03fc 751
752client_netmask 255.255.255.255
753DOC_END
754
755
0f74202c 756COMMENT_START
3a278cb8 757 OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
758 -----------------------------------------------------------------------------
0f74202c 759COMMENT_END
934b03fc 760
761# TAG: ftpget_program
762# Where to find the 'ftpget' program that retrieves FTP data (HTTP
763# and Gopher protocol support are built into the cache).
764#
765# To disable ftpget and the ability to retrieve FTP objects, set
766# this to "none". Note that ftpget is automatically disabled for
767# http_accel mode.
768#
769#ftpget_program @DEFAULT_FTPGET@
770
771# TAG: ftpget_options
772# Options for the 'ftpget' program. Please run 'ftpget' without
773# any arguments to see a list of options. The default is
774# no options. An example is
775#
776# ftpget_options -n 60 -R -W
777#
778#ftpget_options
779
780NAME: ftp_user
781TYPE: string
cf5cc17e 782DEFAULT: Squid@
934b03fc 783LOC: Config.Ftp.anon_user
784DOC_START
785 If you want the anonymous login password to be more informative
786 (and enable the use of picky ftp servers), set this to something
787 resonable for your domain, like wwwuser@somewhere.net
788
789 The reason why this is domainless by default is that the
790 request can be made on the behalf of a user in any domain,
791 depending on how the cache is used.
792 Some ftp server also validate that the email address is valid
793 (for example perl.com).
794
cf5cc17e 795ftp_user Squid@
934b03fc 796DOC_END
797
365cb147 798NAME: ftp_list_width
1b635117 799TYPE: size_t
365cb147 800DEFAULT: 32
801LOC: Config.Ftp.list_width
802DOC_START
cf5cc17e 803 Sets the width of ftp listings. This should be set to fit in
804 the width of a standard browser. Setting this too small
805 can cut off long filenames when browsing ftp sites.
806
365cb147 807ftp_list_width 32
808DOC_END
809
934b03fc 810NAME: cache_dns_program
f0b19334 811TYPE: string
0153d498 812DEFAULT: @DEFAULT_DNSSERVER@
934b03fc 813LOC: Config.Program.dnsserver
814DOC_START
815 Specify the location of the executable for dnslookup process.
816
0153d498 817cache_dns_program @DEFAULT_DNSSERVER@
934b03fc 818DOC_END
819
934b03fc 820NAME: dns_children
821TYPE: int
822DEFAULT: 5
823LOC: Config.dnsChildren
824DOC_START
825 The number of processes spawn to service DNS name lookups.
826 For heavily loaded caches on large servers, you should
827 probably increase this value to at least 10. The maximum
828 is 32. The default is 5.
829
830 To disable dnsservers, set this to 0. NOTE, this is very
831 strongly discouraged. If you disable dnsservers your Squid
832 process will BLOCK on DNS lookups!
833
834dns_children 5
835DOC_END
836
837
838NAME: dns_defnames
839COMMENT: on|off
840TYPE: onoff
f1dc9b30 841DEFAULT: off
17a0a4ee 842LOC: Config.onoff.res_defnames
934b03fc 843DOC_START
844 Normally the 'dnsserver' disables the RES_DEFNAMES resolver
845 option (see res_init(3)). This prevents caches in a hierarchy
846 from interpreting single-component hostnames locally. To allow
847 dnsserver to handle single-component names, enable this
848 option.
849
850dns_defnames off
851DOC_END
852
853
854NAME: unlinkd_program
f0b19334 855TYPE: string
0153d498 856DEFAULT: @DEFAULT_UNLINKD@
934b03fc 857LOC: Config.Program.unlinkd
858DOC_START
859 Specify the location of the executable for file deletion process.
cf5cc17e 860 This isn't needed if you are using async-io since it's handled by
861 a thread.
934b03fc 862
0153d498 863unlinkd_program @DEFAULT_UNLINKD@
934b03fc 864DOC_END
865
866
867NAME: pinger_program
a95856a0 868TYPE: string
0153d498 869DEFAULT: @DEFAULT_PINGER@
934b03fc 870LOC: Config.Program.pinger
871DOC_START
872 Specify the location of the executable for the pinger process.
cf5cc17e 873 This is only useful if you configured Squid (during compliation)
874 with the '--enable-icmp' option.
934b03fc 875
0153d498 876pinger_program @DEFAULT_PINGER@
934b03fc 877DOC_END
878
879
880NAME: redirect_program
f0b19334 881TYPE: string
934b03fc 882LOC: Config.Program.redirect
62607543 883DEFAULT: none
934b03fc 884DOC_START
885 Specify the location of the executable for the URL redirector.
cf5cc17e 886 Since they can perform almost any function there isn't one included.
887 See the Release-Notes for information on how to write one.
888 By default, a redirector is not used.
934b03fc 889
f0b19334 890redirect_program none
934b03fc 891DOC_END
892
893
894NAME: redirect_children
895TYPE: int
896DEFAULT: 5
897LOC: Config.redirectChildren
898DOC_START
cf5cc17e 899 The number of redirector processes to spawn. If you start
900 too few Squid will have to wait for them to process a backlog of
901 URLs, slowing it down. If you start too many they will use RAM
902 and other system resources.
934b03fc 903
904redirect_children 5
905DOC_END
906
0f74202c 907COMMENT_START
3a278cb8 908 OPTIONS FOR TUNING THE CACHE
909 -----------------------------------------------------------------------------
0f74202c 910COMMENT_END
934b03fc 911
0153d498 912NAME: wais_relay_host
913TYPE: string
1273d501 914DEFAULT: none
0153d498 915LOC: Config.Wais.relayHost
916DOC_NONE
934b03fc 917
0153d498 918NAME: wais_relay_port
919TYPE: ushort
920DEFAULT: 0
921LOC: Config.Wais.relayPort
934b03fc 922DOC_START
923 Relay WAIS request to host (1st arg) at port (2 arg).
924
0153d498 925wais_relay_host localhost
926wais_relay_port 8000
934b03fc 927DOC_END
928
929
930NAME: request_size
9906e724 931COMMENT: (KB)
518da0bd 932TYPE: b_size_t
9906e724 933DEFAULT: 100 KB
934b03fc 934LOC: Config.maxRequestSize
935DOC_START
936 Maximum allowed request size in kilobytes. If people are using
937 POST to upload files, then set this to the largest acceptable
938 filesize plus a few extra kbytes.
939
9906e724 940request_size 100 KB
934b03fc 941DOC_END
942
943
944NAME: refresh_pattern
934b03fc 945TYPE: refreshpattern
f1dc9b30 946LOC: Config.Refresh
1273d501 947DEFAULT: none
934b03fc 948DOC_START
949 usage: refresh_pattern regex min percent max
950
951 min and max are specified in MINUTES.
952 percent is an integer number.
953
954 Please see the file doc/Release-Notes-1.1.txt for a full
955 description of Squid's refresh algorithm. Basically a
956 cached object is:
957
958 FRESH if age < min
959 STALE if expires < now
960 STALE if age > max
961 FRESH if lm-factor < percent
962
963 The refresh_pattern lines are checked in the order listed here.
964 The first entry which matches is used. If none of the entries
965 match, then the default will be used.
966
967Default:
968refresh_pattern . 0 20% 4320
969DOC_END
970
971
972NAME: reference_age
f1dc9b30 973TYPE: time_t
934b03fc 974LOC: Config.referenceAge
f1dc9b30 975DEFAULT: 1 year
934b03fc 976DOC_START
977 As a part of normal operation, Squid performs Least Recently
978 Used removal of cached objects. The LRU age for removal is
979 computed dynamically, based on the amount of disk space in
980 use. The 'reference_age' value defines the maximum LRU age.
981 For example, setting reference_age to '1 week' will cause
982 objects to be removed if they have not been accessed for a week
983 or more. If set to zero, LRU removal is disabled, and objects
984 will be removed only when disk usage is over the high water
cf5cc17e 985 mark. The default value is one month.
934b03fc 986
987 Specify a number here, followed by units of time. For example:
988 1 week
989 3.5 days
990 4 months
991 2.2 hours
992
f1dc9b30 993reference_age 1 month
934b03fc 994DOC_END
995
996
0153d498 997NAME: quick_abort_min
9906e724 998COMMENT: (KB)
999TYPE: kb_size_t
9e975e4e 1000DEFAULT: -1 kb
0153d498 1001LOC: Config.quickAbort.min
1002DOC_NONE
1003
1004NAME: quick_abort_pct
1005COMMENT: (percent)
1006TYPE: int
1007DEFAULT: 0
1008LOC: Config.quickAbort.pct
1009DOC_NONE
1010
1011NAME: quick_abort_max
9906e724 1012COMMENT: (KB)
1013TYPE: kb_size_t
9e975e4e 1014DEFAULT: 0 kb
0153d498 1015LOC: Config.quickAbort.max
934b03fc 1016DOC_START
cf5cc17e 1017 The cache can be configured to continue downloading aborted
1018 requests. This may be undesirable on slow (e.g. SLIP) links
1019 and/or very busy caches. Impatient users may tie up file
1020 descriptors and bandwidth by repeatedly requesting and
1021 immediately aborting downloads.
934b03fc 1022
1023 When the user aborts a request, Squid will check the
1024 quick_abort values to the amount of data transfered until
1025 then.
1026
cf5cc17e 1027 If the transfer has less than 'quick_abort_min' KB remaining,
1028 it will finish the retrieval. Setting 'quick_abort_min' to -1
1029 will disable the quick_abort feature.
934b03fc 1030
cf5cc17e 1031 If more than 'quick_abort_pct' of the transfer has completed,
1032 it will finish the retrieval.
934b03fc 1033
cf5cc17e 1034 If the transfer has more than 'quick_abort_max' KB remaining,
1035 it will abort the retrieval.
934b03fc 1036
cf5cc17e 1037quick_abort_min -1 KB
0153d498 1038quick_abort_pct 0
cf5cc17e 1039quick_abort_max 0 KB
934b03fc 1040DOC_END
1041
1042
1043NAME: negative_ttl
bc0eb004 1044COMMENT: time-units
f1dc9b30 1045TYPE: time_t
934b03fc 1046LOC: Config.negativeTtl
9e975e4e 1047DEFAULT: 5 minutes
934b03fc 1048DOC_START
1049 Time-to-Live (TTL) for failed requests. Certain types of
1050 failures (such as "connection refused" and "404 Not Found") are
cf5cc17e 1051 negatively-cached for a configurable amount of time. The
1052 default is 5 minutes. Note that this is different from
1053 negative caching of DNS lookups.
934b03fc 1054
1055negative_ttl 5 minutes
1056DOC_END
1057
1058
1059NAME: positive_dns_ttl
bc0eb004 1060COMMENT: time-units
f1dc9b30 1061TYPE: time_t
934b03fc 1062LOC: Config.positiveDnsTtl
9e975e4e 1063DEFAULT: 6 hours
934b03fc 1064DOC_START
1065 Time-to-Live (TTL) for positive caching of successful DNS lookups.
1066 Default is 6 hours (360 minutes). If you want to minimize the
1067 use of Squid's ipcache, set this to 1, not 0.
1068
9e975e4e 1069positive_dns_ttl 6 hours
934b03fc 1070DOC_END
1071
1072
1073NAME: negative_dns_ttl
bc0eb004 1074COMMENT: time-units
f1dc9b30 1075TYPE: time_t
934b03fc 1076LOC: Config.negativeDnsTtl
9e975e4e 1077DEFAULT: 5 minutes
934b03fc 1078DOC_START
1079 Time-to-Live (TTL) for negative caching of failed DNS lookups.
1080
1081negative_dns_ttl 5 minutes
1082DOC_END
1083
0f74202c 1084COMMENT_START
3a278cb8 1085 TIMEOUTS
1086 -----------------------------------------------------------------------------
0f74202c 1087COMMENT_END
934b03fc 1088
1089NAME: connect_timeout
bc0eb004 1090COMMENT: time-units
f1dc9b30 1091TYPE: time_t
934b03fc 1092LOC: Config.Timeout.connect
9e975e4e 1093DEFAULT: 2 minutes
934b03fc 1094DOC_START
1095 Some systems (notably Linux) can not be relied upon to properly
cf5cc17e 1096 time out connect(2) requests. Therefore the Squid process
934b03fc 1097 enforces its own timeout on server connections. This parameter
1098 specifies how long to wait for the connect to complete. The
1099 default is two minutes (120 seconds).
1100
1101connect_timeout 120 seconds
1102DOC_END
1103
23d92c64 1104NAME: siteselect_timeout
1105COMMENT: time-units
1106TYPE: time_t
1107LOC: Config.Timeout.siteSelect
1108DEFAULT: 4 seconds
1109DOC_START
1110 For URN to multiple URL's URL selection
1111
1112siteselect_timeout 4 seconds
1113DOC_END
934b03fc 1114
1115NAME: read_timeout
bc0eb004 1116COMMENT: time-units
f1dc9b30 1117TYPE: time_t
934b03fc 1118LOC: Config.Timeout.read
9e975e4e 1119DEFAULT: 15 minutes
934b03fc 1120DOC_START
1121 The read_timeout is applied on server-side connections. After
1122 each successful read(), the timeout will be extended by this
1123 amount. If no data is read again after this amount of time,
1124 the request is aborted and logged with ERR_READ_TIMEOUT. The
1125 default is 15 minutes.
1126
1127read_timeout 15 minutes
1128DOC_END
1129
1130
934b03fc 1131NAME: request_timeout
f1dc9b30 1132TYPE: time_t
934b03fc 1133LOC: Config.Timeout.request
9e975e4e 1134DEFAULT: 30 seconds
934b03fc 1135DOC_START
1136 How long to wait for an HTTP request after connection
1137 establishment. For persistent connections, wait this long
1138 after the previous request completes.
1139
1140defer_timeout 30 seconds
1141DOC_END
1142
1143
1144NAME: client_lifetime
bc0eb004 1145COMMENT: time-units
f1dc9b30 1146TYPE: time_t
934b03fc 1147LOC: Config.Timeout.lifetime
9e975e4e 1148DEFAULT: 1 day
934b03fc 1149DOC_START
1150 The maximum amount of time that a client (browser) is allowed to
1151 remain connected to the cache process. This protects the Cache
1152 from having alot of sockets (and hence file descriptors) tied up
1153 in a CLOSE_WAIT state from remote clients that go away without
1154 properly shutting down (either because of a network failure or
1155 because of a poor client implementation). The default is one
1156 day, 1440 minutes.
1157
1158 NOTE: The default value is intended to be much larger than any
1159 client would ever need to be connected to your cache. You
1160 should probably change client_lifetime only as a last resort.
1161 If you seem to have many client connections tying up
1162 filedescriptors, we recommend first tuning the read_timeout,
cf5cc17e 1163 defer_timeout, pconn_timeout and quick_abort values.
934b03fc 1164
1165client_lifetime 1 day
1166DOC_END
1167
603a02fd 1168NAME: pconn_timeout
1169TYPE: time_t
1170LOC: Config.Timeout.pconn
1171DEFAULT: 120 seconds
1172DOC_START
1173 Timeout for idle persistent connections to servers and other
1174 proxies.
1175pconn_timeout 120 seconds
1176DOC_END
1177
934b03fc 1178
1179NAME: shutdown_lifetime
bc0eb004 1180COMMENT: time-units
f1dc9b30 1181TYPE: time_t
934b03fc 1182LOC: Config.shutdownLifetime
9e975e4e 1183DEFAULT: 30 seconds
934b03fc 1184DOC_START
1185 When SIGTERM or SIGHUP is received, the cache is put into
1186 "shutdown pending" mode until all active sockets are closed.
1187 This value is the lifetime to set for all open descriptors
1188 during shutdown mode. Any active clients after this many
1189 seconds will receive a 'timeout' message.
1190
1191shutdown_lifetime 30 seconds
1192DOC_END
1193
0f74202c 1194COMMENT_START
3a278cb8 1195 ACCESS CONTROLS
1196 -----------------------------------------------------------------------------
0f74202c 1197COMMENT_END
934b03fc 1198
1199NAME: acl
1200TYPE: acl
f1dc9b30 1201LOC: Config.aclList
1202DEFAULT: none
934b03fc 1203DOC_START
1204 Defining an Access List
1205
1206 acl aclname acltype string1 ...
1207 acl aclname acltype "file" ...
1208
1209 when using "file", the file should contain one item per line
1210
1211 acltype is one of src dst srcdomain dstdomain url_pattern
1212 urlpath_pattern time port proto method browser user
1213
1214 acl aclname src ip-address/netmask ... (clients IP address)
1215 acl aclname src addr1-addr2/netmask ... (range of addresses)
1216 acl aclname dst ip-address/netmask ... (URL host's IP address)
1217 acl aclname srcdomain foo.com ... (taken from reverse DNS lookup)
1218 acl aclname dstdomain foo.com ... (taken from the URL)
1219 acl aclname time [day-abbrevs] [h1:m1-h2:m2]
1220 day-abbrevs:
1221 S - Sunday
1222 M - Monday
1223 T - Tuesday
1224 W - Wednesday
1225 H - Thursday
1226 F - Friday
1227 A - Saturday
1228 h1:m1 must be less than h2:m2
1229 acl aclname url_regex ^http:// ... # regex matching on whole URL
1230 acl aclname urlpath_regex \.gif$ ... # regex matching on URL path only
1231 acl aclname port 80 70 21 ...
1232 acl aclname proto HTTP FTP ...
1233 acl aclname method GET POST ...
1234 acl aclname browser regexp
1235 acl aclname user username ... # string match on ident output.
1236 # use REQUIRED to accept any
1237 # non-null ident.
6468fe10 1238 acl aclname src_as number ...
1239 acl aclname dst_as number ...
1240 # Except for access control, AS numbers can be used for
1241 # routing of requests to specific caches. Here's an
1242 # example for routing all requests for AS#1241 and only
1243 # those to mycache.mydomain.net:
1244 # acl asexample dst_as 1241
1245 # cache_host_acl mycache.mydomain.net asexample
1246 # cache_host_acl mycache_mydomain.net !all
1247
afe95a7e 1248 acl aclname proxy_auth passwd_file [ refresh ]
1249 # 'passwd_file' is an Apache-style file of passwords for
1250 # authenticated proxy access. Looks like user:password, with
1251 # the password being standard crypt() format. 'refresh' is
1252 # the time in seconds to check for a changes in the file
1253 # (default = 300 secs). When using a proxy_auth ACL in an
1254 # ACL list, make sure it is the *last* in the list and the
1255 # only proxy_auth ACL in the list. NOTE: when a
1256 # Proxy-Authentication header is sent but it is not needed
1257 # during ACL checking the username is NOT logged in
1258 # access.log.
934b03fc 1259
1260acl manager proto cache_object
1261acl localhost src 127.0.0.1/255.255.255.255
1262acl all src 0.0.0.0/0.0.0.0
6468fe10 1263acl myexample dst_as 1241
934b03fc 1264acl SSL_ports port 443 563
1265acl Dangerous_ports port 7 9 19
1266acl CONNECT method CONNECT
1267DOC_END
1268
1269NAME: http_access
1270TYPE: acl_access
f1dc9b30 1271LOC: Config.accessList.http
1272DEFAULT: none
934b03fc 1273DOC_START
1274 Allowing or Denying access based on defined access lists
1275
1276 Access to the HTTP port:
1277 http_access allow|deny [!]aclname ...
1278
1279 Access to the ICP port:
1280 icp_access allow|deny [!]aclname ...
1281
1282 NOTE on default values:
1283
1284 If there are no "access" lines present, the default is to allow
1285 the request.
1286
1287 If none of the "access" lines cause a match, the default is the
1288 opposite of the last line in the list. If the last line was
1289 deny, then the default is allow. Conversely, if the last line
1290 is allow, the default will be deny. For these reasons, it is a
1291 good idea to have an "deny all" or "allow all" entry at the end
1292 of your access lists to avoid potential confusion.
1293
1294
1295 Only allow access to the cache manager functions from the local host.
1296http_access deny manager !localhost
1297http_access deny CONNECT !SSL_ports
1298http_access deny Dangerous_ports
1299
1300 Allow everything else
1301http_access allow all
1302DOC_END
1303
1304
1305NAME: icp_access
1306TYPE: acl_access
f1dc9b30 1307LOC: Config.accessList.icp
1308DEFAULT: none
934b03fc 1309DOC_START
1310 Reply to all ICP queries we receive
1311
1312icp_access allow all
1313DOC_END
1314
1315
1316NAME: miss_access
1317TYPE: acl_access
f1dc9b30 1318LOC: Config.accessList.miss
1319DEFAULT: none
934b03fc 1320DOC_START
1321 Use to force your neighbors to use you as a sibling instead of
1322 a parent. For example:
1323
1324 acl localclients src 172.16.0.0/16
1325 miss_access allow localclients
1326 miss_access deny !localclients
1327
1328 This means that only your local clients are allowed to fetch
1329 MISSES and all other clients can only fetch HITS.
1330
1331 By default, allow all clients who passed the http_access rules
1332 to fetch MISSES from us.
1333
1334miss_access allow all
1335DOC_END
1336
1337
1338NAME: cache_host_acl
f1dc9b30 1339TYPE: peeracl
1340DEFAULT: none
1341LOC: none
934b03fc 1342DOC_START
1343 Just like 'cache_host_domain' but provides more flexibility by
1344 using ACL's.
1345
1346 cache_host_acl cache-host [!]aclname ...
1347
1348 NOTE: * Any number of ACL's may be given for a cache-host,
1349 either on the same or separate lines.
1350 * When multiple ACL's are given for a particular
1351 cache-host, the first matched ACL is applied.
1352 * Cache hosts with no domain or ACL restrictions are
1353 queried for all requests.
1354 * There are no defaults.
1355DOC_END
1356
0f74202c 1357COMMENT_START
3a278cb8 1358 ADMINISTRATIVE PARAMETERS
1359 -----------------------------------------------------------------------------
0f74202c 1360COMMENT_END
934b03fc 1361
1362NAME: cache_mgr
1363TYPE: string
1364DEFAULT: webmaster
1365LOC: Config.adminEmail
1366DOC_START
1367 Email-address of local cache manager who will receive
1368 mail if the cache dies. The default is "webmaster."
1369
1370cache_mgr webmaster
1371DOC_END
1372
1373
1374NAME: cache_effective_user
0153d498 1375TYPE: string
a95856a0 1376DEFAULT: nobody
0153d498 1377LOC: Config.effectiveUser
1378DOC_NONE
1379
1380NAME: cache_effective_group
1381TYPE: string
a95856a0 1382DEFAULT: nogroup
0153d498 1383LOC: Config.effectiveGroup
934b03fc 1384DOC_START
1385 If the cache is run as root, it will change its effective/real
1386 UID/GID to the UID/GID specified below. The default is not to
1387 change UID/GID.
1388
0153d498 1389cache_effective_user nobody
1390cache_effective_group nogroup
934b03fc 1391DOC_END
1392
1393
1394NAME: visible_hostname
1395TYPE: string
1396LOC: Config.visibleHostname
f1dc9b30 1397DEFAULT: none
934b03fc 1398DOC_START
1399 If you want to present a special hostname in error messages, etc,
1400 then define this. Otherwise, the return value of gethostname()
cf5cc17e 1401 will be used. If you have multiple caches in a cluster and
1402 get errors about IP-forwarding you must set them to have individual
1403 names with this setting.
934b03fc 1404
1405visible_hostname www-cache.foo.org
1406DOC_END
1407
0f74202c 1408COMMENT_START
3a278cb8 1409 OPTIONS FOR THE CACHE REGISTRATION SERVICE
1410 -----------------------------------------------------------------------------
934b03fc 1411
3a278cb8 1412 This section contains parameters for the (optional) cache
1413 announcement service. This service is provided to help
1414 cache administrators locate one another in order to join or
1415 create cache hierarchies.
934b03fc 1416
3a278cb8 1417 An 'announcement' message is sent (via UDP) to the registration
1418 service by Squid. By default, the annoucement message is NOT
1419 SENT unless you enable it with 'cache_announce' below.
1420
1421 The announcement message includes your hostname, plus the
1422 following information from this configuration file:
934b03fc 1423
3a278cb8 1424 http_port
1425 icp_port
1426 cache_mgr
1427
1428 All current information is processed regularly and made
1429 available on the Web at http://www.nlanr.net/Cache/Tracker/.
0f74202c 1430COMMENT_END
934b03fc 1431
f1dc9b30 1432NAME: announce_period
1433TYPE: time_t
1434LOC: Config.Announce.period
9e975e4e 1435DEFAULT: 1 day
934b03fc 1436DOC_START
1437 This is how frequently to send cache announcements. The default
1438 is `0' which disables sending the announcement messages.
1439
1440 To enable announcing your cache, just uncomment the line below.
1441
9e975e4e 1442announce_period 1 day
934b03fc 1443DOC_END
1444
1445
f1dc9b30 1446NAME: announce_host
1447TYPE: string
1448DEFAULT: sd.cache.nlanr.net
1449LOC: Config.Announce.host
1450DOC_NONE
1451
cf5cc17e 1452NAME: announce_file
1453TYPE: string
1454DEFAULT: /dev/null
1455LOC: Config.Announce.file
1456DOC_NONE
1457
f1dc9b30 1458NAME: announce_port
1459TYPE: ushort
1460DEFAULT: 3131
1461LOC: Config.Announce.port
934b03fc 1462DOC_START
cf5cc17e 1463 announce_host and announce_port set the hostname and portnumber where
1464 the registration message will be sent.
934b03fc 1465
1466 Hostname will default to 'sd.cache.nlanr.net' and port will default
1467 to 3131. If the 'filename' argument is given, the contents of that
1468 file will be included in the announce message.
1469
f1dc9b30 1470announce_host sd.cache.nlanr.net
1471announce_port 3131
cf5cc17e 1472announce_file /dev/null
934b03fc 1473DOC_END
1474
0f74202c 1475COMMENT_START
3a278cb8 1476 HTTPD-ACCELERATOR OPTIONS
1477 -----------------------------------------------------------------------------
0f74202c 1478COMMENT_END
934b03fc 1479
f1dc9b30 1480NAME: httpd_accel_host
1481TYPE: string
1482LOC: Config.Accel.host
1483DEFAULT: none
1484DOC_NONE
1485
1486NAME: httpd_accel_port
1487TYPE: ushort
1488LOC: Config.Accel.port
5b68a4d3 1489DEFAULT: 80
934b03fc 1490DOC_START
cf5cc17e 1491 If you want to run Squid as an httpd accelerator, define the
934b03fc 1492 host name and port number where the real HTTP server is.
1493
1494 If you want virtual host support then specify the hostname
1495 as "virtual".
1496
f1dc9b30 1497httpd_accel_host hostname
1498httpd_accel_port port
934b03fc 1499DOC_END
1500
1501
1502NAME: httpd_accel_with_proxy
1503COMMENT: on|off
1504TYPE: onoff
f1dc9b30 1505DEFAULT: off
17a0a4ee 1506LOC: Config.onoff.accel_with_proxy
934b03fc 1507DOC_START
cf5cc17e 1508 If you want to use Squid as both a local httpd accelerator
934b03fc 1509 and as a proxy, change this to 'on'.
1510
1511httpd_accel_with_proxy off
1512DOC_END
1513
1514
1515NAME: httpd_accel_uses_host_header
1516COMMENT: on|off
1517TYPE: onoff
f1dc9b30 1518DEFAULT: off
934b03fc 1519LOC: opt_accel_uses_host
1520DOC_START
1521 HTTP/1.1 requests include a Host: header which is basically the
1522 hostname from the URL. Squid can be an accelerator for
1523 different HTTP servers by looking at this header. However,
1524 Squid does NOT check the value of the Host header, so it opens
1525 a big security hole. We recommend that this option remain
1526 disabled unless you are sure of what you are doing.
1527
1528httpd_accel_uses_host_header off
1529DOC_END
1530
0f74202c 1531COMMENT_START
3a278cb8 1532 MISCELLANEOUS
1533 -----------------------------------------------------------------------------
0f74202c 1534COMMENT_END
934b03fc 1535
1536NAME: dns_testnames
1537TYPE: wordlist
1538LOC: Config.dns_testname_list
f1dc9b30 1539DEFAULT: none
934b03fc 1540DOC_START
1541 The DNS tests exit as soon as the first site is successfully looked up
1542
1543 If you want to disable DNS tests, do not comment out or delete this
1544 list. Instead use the -D command line option
1545
cf5cc17e 1546dns_testnames netscape.com internic.net nlanr.net microsoft.com
934b03fc 1547DOC_END
1548
1549
1550NAME: logfile_rotate
1551TYPE: int
1552DEFAULT: 10
1553LOC: Config.Log.rotateNumber
1554DOC_START
cf5cc17e 1555 Specifies the number of logfile rotations to make upon
1556 receiving a USR1 signal. The default is 10, which will rotate
1557 with extensions 0 through 9. Setting logfile_rotate to 0 will
934b03fc 1558 disable the rotation, but the logfiles are still closed and
cf5cc17e 1559 re-opened. This will enable you to rename the logfiles
1560 yourself just before sending a USR1 signal to the Squid
1561 process.
934b03fc 1562
1563logfile_rotate 10
1564DOC_END
1565
1566
1567NAME: append_domain
f1dc9b30 1568TYPE: string
1569LOC: Config.appendDomain
1570DEFAULT: none
934b03fc 1571DOC_START
cf5cc17e 1572 Appends local domain name to hostnames without any dots in
1573 them. append_domain must begin with a period.
934b03fc 1574
1575append_domain .yourdomain.com
1576DOC_END
1577
1578
1579NAME: tcp_recv_bufsize
89de058c 1580COMMENT: (bytes)
1b635117 1581TYPE: b_size_t
89de058c 1582DEFAULT: 0 bytes
934b03fc 1583LOC: Config.tcpRcvBufsz
1584DOC_START
1585 Size of receive buffer to set for TCP sockets. Probably just
1586 as easy to change your kernel's default. Set to zero to use
1587 the default buffer size.
1588
89de058c 1589tcp_recv_bufsize 0 bytes
934b03fc 1590DOC_END
1591
934b03fc 1592NAME: err_html_text
f1dc9b30 1593TYPE: eol
1594LOC: Config.errHtmlText
1595DEFAULT: none
934b03fc 1596DOC_START
1597 HTML text to include in error messages. Make this a "mailto"
1598 URL to your admin address, or maybe just a link to your
1599 organizations Web page.
1600
1601err_html_text
1602DOC_END
1603
1604
1605NAME: deny_info
1606TYPE: denyinfo
1607LOC: Config.denyInfoList
f1dc9b30 1608DEFAULT: none
934b03fc 1609DOC_START
1610 Usage: deny_info URL acl
1611
1612 This can be used to return a HTTP redirect for requests which
1613 do not pass the 'http_access' rules. A single ACL will cause
1614 the http_access check to fail. If a 'deny_info' line exists
1615 for that ACL then Squid returns a redirect to the given URL.
cf5cc17e 1616
1617 Be sure to make an exception for the site you are referring
1618 people to with the 'acl dstdomain' function, otherwise they
1619 are going to be refused access to the redirected URL too.
934b03fc 1620DOC_END
1621
934b03fc 1622NAME: memory_pools
1623COMMENT: on|off
1624TYPE: onoff
f1dc9b30 1625DEFAULT: on
3f6c0fb2 1626LOC: Config.onoff.mem_pools
934b03fc 1627DOC_START
1628 If set, Squid will keep pools of allocated (but unused) memory
1629 available for future use. If memory is a premium on your
1630 system, disable this.
1631
1632memory_pools on
1633DOC_END
1634
1635NAME: forwarded_for
1636COMMENT: on|off
1637TYPE: onoff
f1dc9b30 1638DEFAULT: on
934b03fc 1639LOC: opt_forwarded_for
1640DOC_START
1641 If set, Squid will include your system's IP address or name
1642 in the HTTP requests it forwards. By default it looks like
1643 this:
1644
1645 X-Forwarded-For: 192.1.2.3
1646
1647 If you disable this, it will appear as
1648
1649 X-Forwarded-For: unknown
1650
1651forwarded_for on
1652DOC_END
1653
1654NAME: log_icp_queries
1655COMMENT: on|off
1656TYPE: onoff
f1dc9b30 1657DEFAULT: on
17a0a4ee 1658LOC: Config.onoff.log_udp
934b03fc 1659DOC_START
cf5cc17e 1660 If set, ICP queries are logged to access.log. You may wish
1661 do disable this if your ICP load is VERY high to speed things
1662 up or to simplify log analysis.
934b03fc 1663
1664log_icp_queries on
1665DOC_END
1666
88738790 1667NAME: icp_hit_stale
1668COMMENT: on|off
1669TYPE: onoff
1670DEFAULT: off
17a0a4ee 1671LOC: Config.onoff.icp_hit_stale
88738790 1672DOC_START
1673 If you want to return ICP_HIT for stale cache objects, set this
1674 option to 'on'. If you have sibling relationships with caches
1675 in other administrative domains, this should be 'off'. If you only
1676 have sibling relationships with caches under your control, then
1677 it is probably okay to set this to 'on'.
1678
1679icp_hit_stale off
1680DOC_END
1681
934b03fc 1682
1683NAME: minimum_direct_hops
1684TYPE: int
1685DEFAULT: 4
1686LOC: Config.minDirectHops
1687DOC_START
1688 If using the ICMP pinging stuff, do direct fetches for sites
1689 which are no more than this many hops away.
1690
1691minimum_direct_hops 4
1692DOC_END
1693
1694
1695NAME: cachemgr_passwd
1696TYPE: cachemgrpasswd
86101e40 1697DEFAULT: none
f1dc9b30 1698LOC: Config.passwd_list
934b03fc 1699DOC_START
1700 Specify passwords for cachemgr operations.
1701
1702 Usage: cachemgr_passwd password action action ...
1703
1704 valid actions are:
1705 shutdown *
1706 info
cf5cc17e 1707 objects
1708 vm_objects
1709 utilization
1710 ipcache
1711 fqdncache
1712 dns
1713 redirector
1714 io
1715 reply_headers
1716 filedescriptors
1717 netdb
934b03fc 1718 server_list
1719 client_list
cf5cc17e 1720 config *
934b03fc 1721
1722 * Indicates actions which will not be performed without a
1723 valid password, others can be performed if not listed here.
1724
1725 To disable an action, set the password to "disable".
1726 To allow performing an action without a password, set the
1727 password to "none".
1728
1729 Use the keyword "all" to set the same password for all actions.
1730
1731cachemgr_passwd secret shutdown
1732cachemgr_passwd lesssssssecret info stats/objects
1733cachemgr_passwd disable all
1734DOC_END
1735
934b03fc 1736NAME: store_avg_object_size
86101e40 1737COMMENT: (kbytes)
1738TYPE: kb_size_t
2bf99296 1739DEFAULT: 13 KB
934b03fc 1740LOC: Config.Store.avgObjectSize
1741DOC_START
1742 Average object size, used to estimate number of objects your
1743 cache can hold. See doc/Release-Notes-1.1.txt. The default is
2bf99296 1744 13 KB.
934b03fc 1745
2bf99296 1746store_avg_object_size 13 KB
934b03fc 1747DOC_END
1748
1749NAME: store_objects_per_bucket
1750TYPE: int
1751DEFAULT: 50
1752LOC: Config.Store.objectsPerBucket
1753DOC_START
1754 Target number of objects per bucket in the store hash table.
1755 Lowering this value increases the total number of buckets and
1756 also the storage maintenance rate. The default is 20.
1757
1758store_objects_per_bucket 20
1759DOC_END
1760
1761
1762NAME: http_anonymizer
1763TYPE: httpanonymizer
17a0a4ee 1764LOC: Config.onoff.anonymizer
f1dc9b30 1765DEFAULT: off
934b03fc 1766DOC_START
1767 If you want to filter out certain HTTP request headers for
1768 privacy reasons, enable this option. There are three
1769 appropriate settings:
1770 'off' All HTTP request headers are passed.
1771 'standard' Specific headers are removed
1772 'paranoid' Only specific headers are allowed.
1773 To see which headers are allowed or denied, please see the
1774 http-anon.c source file.
1775
1776http_anonymizer off
1777DOC_END
1778
1779
1780NAME: client_db
1781COMMENT: on|off
1782TYPE: onoff
f1dc9b30 1783DEFAULT: on
17a0a4ee 1784LOC: Config.onoff.client_db
934b03fc 1785DOC_START
1786 If you want to disable collecting per-client statistics, then
1787 turn off client_db here.
1788
1789client_db on
1790DOC_END
1791
1792
1793NAME: netdb_low
1794TYPE: int
1795DEFAULT: 900
1796LOC: Config.Netdb.low
1797DOC_NONE
1798
1799NAME: netdb_high
1800TYPE: int
1801DEFAULT: 1000
1802LOC: Config.Netdb.high
1803DOC_START
1804 The low and high water marks for the ICMP measurement
1805 database. These are counts, not percents. The defaults are
1806 900 and 1000. When the high water mark is reached, database
1807 entries will be deleted until the low mark is reached.
1808
1809netdb_low 900
1810netdb_high 1000
1811DOC_END
1812
1813
1814NAME: netdb_ping_period
f1dc9b30 1815TYPE: time_t
934b03fc 1816LOC: Config.Netdb.period
9e975e4e 1817DEFAULT: 5 minutes
934b03fc 1818DOC_START
1819 The minimum period for measuring a site. There will be at
1820 least this much delay between successive pings to the same
1821 network. The default is five minutes.
1822
1823netdb_ping_period 5 minutes
1824DOC_END
1825
1826
1827NAME: query_icmp
1828COMMENT: on|off
1829TYPE: onoff
f1dc9b30 1830DEFAULT: off
17a0a4ee 1831LOC: Config.onoff.query_icmp
934b03fc 1832DOC_START
1833 If you want to ask your peers to include ICMP data in their ICP
1834 replies, enable this option.
1835
cf5cc17e 1836 If your peer has configured Squid (during compilation) with
1837 '--enable-icmp' then that peer will send ICMP pings to origin server
1838 sites of the URLs it receives. If you enable this option then the
1839 ICP replies from that peer will include the ICMP data (if available).
1840 Then, when choosing a parent cache, Squid will choose the parent with
934b03fc 1841 the minimal RTT to the origin server. When this happens, the
1842 hierarchy field of the access.log will be
1843 "CLOSEST_PARENT_MISS". This option is off by default.
1844
1845query_icmp off
1846DOC_END
1847
78f1250a 1848NAME: buffered_logs
1849COMMENT: on|off
1850TYPE: onoff
1851DEFAULT: off
17a0a4ee 1852LOC: Config.onoff.buffered_logs
78f1250a 1853DOC_START
1854 Some log files (cache.log, useragent.log) are written with
1855 stdio functions, and as such they can be buffered or
cf5cc17e 1856 unbuffered. By default they will be unbuffered. Buffering them
1857 can speed up the writing slightly (though you are unlikely to
1858 need to worry).
78f1250a 1859buffered_logs off
1860DOC_END
1861
934b03fc 1862NAME: always_direct
1863TYPE: acl_access
1864LOC: Config.accessList.AlwaysDirect
f1dc9b30 1865DEFAULT: none
934b03fc 1866DOC_START
1867 XXX need docs
1868DOC_END
1869
1870NAME: never_direct
1871TYPE: acl_access
1872LOC: Config.accessList.NeverDirect
f1dc9b30 1873DEFAULT: none
934b03fc 1874DOC_START
1875 XXX need docs
1876DOC_END
1877
1878#NAME: proxy_auth_ignore
1879#TYPE: regexplist_icase
1880#LOC: Config.proxyAuth.IgnoreDomains
1881#DOC_START
1882# XXX need docs
1883#DOC_END
1884
88738790 1885NAME: fake_user_agent
1886TYPE: eol
1887LOC: Config.fake_ua
1888DEFAULT: none
1889DOC_START
1890 If you use the paranoid http_anonymizer setting, Squid will strip
1891 your User-agent string from the request. Some Web servers will
1892 refuse your request without a User-agent string. Use this to
1893 fake one up. For example:
1894
1895 fake_user_agent Nutscrape/1.0 (CP/M; 8-bit)
1896 (credit to Paul Southworth pauls@etext.org for this one!)
1897
1898fake_user_agent none
1899DOC_END
1900
365cb147 1901NAME: icon_directory
f0b19334 1902TYPE: string
365cb147 1903LOC: Config.icons.directory
1904DEFAULT: @DEFAULT_ICON_DIR@
1905DOC_START
cf5cc17e 1906 Where the icons are stored. These are normally kept in
1907 @DEFAULT_ICON_DIR@
365cb147 1908DOC_END
1909
cf5cc17e 1910NAME: error_directory
365cb147 1911TYPE: string
cf5cc17e 1912LOC: Config.errorDirectory
1913DEFAULT: @DEFAULT_ERROR_DIR@
365cb147 1914DOC_START
cf5cc17e 1915 If you wish to create your own versions of the default
1916 (English) error files, either to customise them to suit your
1917 language or company copy the template english files to anther
1918 directory and point this tag at them.
365cb147 1919DOC_END
88738790 1920
22c653cd 1921NAME: minimum_retry_timeout
1922COMMENT: (seconds)
1923TYPE: time_t
1924LOC: Config.retry.timeout
1925DEFAULT: 5 seconds
1926DOC_START
1927 This specifies the minimum connect timeout, for when the
1928 connect timeout is reduced to compensate for the availability
1929 of multiple IP addresses.
1930
cf5cc17e 1931 When a connection to a host is initiated, and that host has
1932 several IP addresses, the default connection timeout is reduced
1933 by dividing it by the number of addresses. So, a site with 15
1934 addresses would then have a timeout of 8 seconds for each
1935 address attempted. To avoid having the timeout reduced to the
1936 point where even a working host would not have a chance to
1937 respond, this setting is provided. The default, and the
1938 minimum value, is five seconds, and the maximum value is sixty
1939 seconds, or half of connect_timeout, whichever is greater and
1940 less than connect_timeout.
22c653cd 1941
cf5cc17e 1942minimum_retry_timeout 5 seconds
22c653cd 1943DOC_END
1944
1945NAME: maximum_single_addr_tries
1946TYPE: int
1947LOC: Config.retry.maxtries
1948DEFAULT: 3
1949DOC_START
1950 This sets the maximum number of connection attempts for a
1951 host that only has one address (for multiple-address hosts,
1952 each address is tried once).
1953
1954 The default value is three tries, the (not recommended)
1955 maximum is 255 tries. A warning message will be generated
1956 if it is set to a value greater than ten.
1957
1958maximum_single_addr_tries 3
1959DOC_END
6d1c0d53 1960
a97cfa48 1961NAME: snmp_port
1962TYPE: ushort
1963LOC: Config.Port.snmp
1964DEFAULT: 3401
1df370e3 1965IFDEF: SQUID_SNMP
a97cfa48 1966DOC_START
cf5cc17e 1967 Squid can now serve statistics and status information via SNMP.
1968 By default it listens to port 3401 on the machine. If you don't
1969 wish to use SNMP, set this to '-1'.
1970
1971 NOTE: SNMP support requires use the --enable-snmp configure
1972 command line option.
a97cfa48 1973DOC_END
1974
a97cfa48 1975NAME: snmp_do_queueing
1976TYPE: onoff
1977LOC: Config.Snmp.do_queueing
1978DEFAULT: on
1df370e3 1979IFDEF: SQUID_SNMP
a97cfa48 1980DOC_START
cf5cc17e 1981 If disabled, snmp packets will not be queued but delivered
1982 immediately. This could be useful when you want to monitor a
1983 cache in trouble, but this could also make Squid block, slowing
1984 connections and possibly worsening the cache status.
a97cfa48 1985DOC_END
1986
1987NAME: forward_snmpd_port
1988TYPE: ushort
1989LOC: Config.Snmp.localPort
1990DEFAULT: 0
1df370e3 1991IFDEF: SQUID_SNMP
a97cfa48 1992DOC_START
1993 This configures whether we should be forwarding SNMP requests
cf5cc17e 1994 to another snmpd. The reason for putting this piece of
1995 functionality into Squid was to enable access to the system's
1996 installed snmpd with minimal changes. This option is turned
1997 off by default, check with your /etc/services for your system's
1998 snmp port (usually 161). We do not use getservbyname() to
1999 allow you to set Squid into port 161 and your system's snmpd to
2000 another port by changing /etc/services.
2001
2002 WARNING: Because of Squid acting as a proxy snmpd for system
a97cfa48 2003 you have to do security checks on THIS snmpd for all objects.
cf5cc17e 2004 Check your snmp_config_file.
a97cfa48 2005DOC_END
2006
2007NAME: snmp_mib_path
2008TYPE: string
2009LOC: Config.Snmp.mibPath
2010DEFAULT: @DEFAULT_MIB_PATH@
1df370e3 2011IFDEF: SQUID_SNMP
a97cfa48 2012DOC_START
cf5cc17e 2013 The location of Squid's mib.
a97cfa48 2014DOC_END
2015
a97cfa48 2016NAME: trap_sink
4feb7b2e 2017TYPE: string
a97cfa48 2018LOC: Config.Snmp.trap_sink
2019DEFAULT: 127.0.0.1
1df370e3 2020IFDEF: SQUID_SNMP
4feb7b2e 2021DOC_START
a97cfa48 2022 Hostname or ip address of trap sink for snmp
4feb7b2e 2023DOC_END
2024
a97cfa48 2025NAME: snmp_trap_community
2026TYPE: string
2027LOC: Config.Snmp.trap_community
2028DEFAULT: public
1df370e3 2029IFDEF: SQUID_SNMP
a97cfa48 2030DOC_START
cf5cc17e 2031 Community name for traps.
a97cfa48 2032DOC_END
2033
2034NAME: snmp_enable_authen_traps
2035TYPE: onoff
2036LOC: Config.Snmp.conf_authtraps
2037DEFAULT: off
1df370e3 2038IFDEF: SQUID_SNMP
a97cfa48 2039DOC_START
cf5cc17e 2040 Enable SNMP authenticated traps. Set to 'off' or 'on'.
a97cfa48 2041DOC_END
2042
2043NAME: snmp_agent_conf
2044TYPE: snmp_conf
2045LOC: Config.Snmp.snmpconf
2046DEFAULT: none
1df370e3 2047IFDEF: SQUID_SNMP
a97cfa48 2048DOC_START
2049 Define snmp views, users and communities
2050 Example:
2051 snmp_agent_conf view all .1.3.6 included
2052 snmp_agent_conf view squid .1.3.6 included
2053 snmp_agent_conf user squid - all all public
2054 snmp_agent_conf user all all all all squid
2055 snmp_agent_conf community public squid squid
2056 snmp_agent_conf community readwrite all all
2057DOC_END
4feb7b2e 2058
53ad48e6 2059NAME: as_whois_server
2060TYPE: string
2061LOC: Config.as_whois_server
2062DEFAULT: whois.ra.net
2063DEFAULT_IF_NONE: whois.ra.net
2064DOC_START
cf5cc17e 2065 WHOIS server to query for AS numbers. NOTE: AS numbers are
2066 queried only when Squid starts up, not for every request.
53ad48e6 2067DOC_END
2068
6d1c0d53 2069EOF