2 # WELCOME TO SQUID 2
3 # ------------------
4 #
5 # This is the default Squid configuration file. You may wish
6 # to look at the Squid home page (http://www.squid-cache.org/)
7 # for the FAQ and other documentation.
8 #
9 # The default Squid config file shows what the defaults for
10 # various options happen to be. If you don't need to change the
11 # default, you shouldn't uncomment the line. Doing so may cause
12 # run-time problems. In some cases "none" refers to no default
13 # setting at all, while in other cases it refers to a valid
14 # option - the comments for that keyword indicate if this is the
15 # case.
16 #
19 # NETWORK OPTIONS
20 # -----------------------------------------------------------------------------
22 # TAG: http_port
23 # Usage: port
24 # hostname:port
25 # 1.2.3.4:port
26 #
27 # The socket addresses where Squid will listen for HTTP client
28 # requests. You may specify multiple socket addresses.
29 # There are three forms: port alone, hostname with port, and
30 # IP address with port. If you specify a hostname or IP
31 # address, then Squid binds the socket to that specific
32 # address. This replaces the old 'tcp_incoming_address'
33 # option. Most likely, you do not need to bind to a specific
34 # address, so you can use the port number alone.
35 #
36 # The default port number is 3128.
37 #
38 # If you are running Squid in accelerator mode, then you
39 # probably want to listen on port 80 also, or instead.
40 #
41 # The -a command line option will override the *first* port
42 # number listed here. That option will NOT override an IP
43 # address, however.
44 #
45 # You may specify multiple socket addresses on multiple lines.
46 #
47 # If you run Squid on a dual-homed machine with an internal
48 # and an external interface then we recommend you to specify the
49 # internal address:port in http_port. This way Squid will only be
50 # visible on the internal address.
51 #
52 #Default:
53 # http_port 3128
55 # TAG: https_port
56 # Usage: [ip:]port cert=certificate.pem [key=key.pem] [options...]
57 #
58 # The socket address where Squid will listen for HTTPS client
59 # requests.
60 #
61 # This is really only useful for situations where you are running
62 # squid in accelerator mode and you want to do the SSL work at the
63 # accelerator level.
64 #
65 # You may specify multiple socket addresses on multiple lines,
66 # each with their own SSL certificate and/or options.
67 #
68 # Options:
69 #
70 # cert= Path to SSL certificate (PEM format)
71 #
72 # key= Path to SSL private key file (PEM format)
73 # if not specified, the certificate file is
74 # assumed to be a combined certificate and
75 # key file
76 #
77 # version= The version of SSL/TLS supported
78 # 1 automatic (default)
79 # 2 SSLv2 only
80 # 3 SSLv3 only
81 # 4 TLSv1 only
82 #
83 # cipher= Colon separated list of supported ciphers
84 #
85 # options= Varions SSL engine options. The most important
86 # being:
87 # NO_SSLv2 Disallow the use of SSLv2
88 # NO_SSLv3 Disallow the use of SSLv3
89 # NO_TLSv1 Disallow the use of TLSv1
90 # See src/ssl_support.c or OpenSSL documentation
91 # for a more complete list.
92 #
93 #Default:
94 # none
96 # TAG: ssl_unclean_shutdown
97 # Some browsers (especially MSIE) bugs out on SSL shutdown
98 # messages.
99 #
100 #Default:
101 # ssl_unclean_shutdown off
103 # TAG: icp_port
104 # The port number where Squid sends and receives ICP queries to
105 # and from neighbor caches. Default is 3130. To disable use
106 # "0". May be overridden with -u on the command line.
107 #
108 #Default:
109 # icp_port 3130
111 # TAG: htcp_port
112 # Note: This option is only available if Squid is rebuilt with the
113 # --enable-htcp option
114 #
115 # The port number where Squid sends and receives HTCP queries to
116 # and from neighbor caches. Default is 4827. To disable use
117 # "0".
118 #
119 #Default:
120 # htcp_port 4827
122 # TAG: mcast_groups
123 # This tag specifies a list of multicast groups which your server
124 # should join to receive multicasted ICP queries.
125 #
126 # NOTE! Be very careful what you put here! Be sure you
127 # understand the difference between an ICP _query_ and an ICP
128 # _reply_. This option is to be set only if you want to RECEIVE
129 # multicast queries. Do NOT set this option to SEND multicast
130 # ICP (use cache_peer for that). ICP replies are always sent via
131 # unicast, so this option does not affect whether or not you will
132 # receive replies from multicast group members.
133 #
134 # You must be very careful to NOT use a multicast address which
135 # is already in use by another group of caches.
136 #
137 # If you are unsure about multicast, please read the Multicast
138 # chapter in the Squid FAQ (http://www.squid-cache.org/FAQ/).
139 #
140 # Usage: mcast_groups 239.128.16.128 224.0.1.20
141 #
142 # By default, Squid doesn't listen on any multicast groups.
143 #
144 #Default:
145 # none
147 # TAG: udp_incoming_address
148 # TAG: udp_outgoing_address
149 # udp_incoming_address is used for the ICP socket receiving packets
150 # from other caches.
151 # udp_outgoing_address is used for ICP packets sent out to other
152 # caches.
153 #
154 # The default behavior is to not bind to any specific address.
155 #
156 # A udp_incoming_address value of 0.0.0.0 indicates that Squid should
157 # listen for UDP messages on all available interfaces.
158 #
159 # If udp_outgoing_address is set to 255.255.255.255 (the default)
160 # then it will use the same socket as udp_incoming_address. Only
161 # change this if you want to have ICP queries sent using another
162 # address than where this Squid listens for ICP queries from other
163 # caches.
164 #
165 # NOTE, udp_incoming_address and udp_outgoing_address can not
166 # have the same value since they both use port 3130.
167 #
168 #Default:
169 # udp_incoming_address 0.0.0.0
170 # udp_outgoing_address 255.255.255.255
173 # OPTIONS WHICH AFFECT THE NEIGHBOR SELECTION ALGORITHM
174 # -----------------------------------------------------------------------------
176 # TAG: cache_peer
177 # To specify other caches in a hierarchy, use the format:
178 #
179 # cache_peer hostname type http_port icp_port
180 #
181 # For example,
182 #
183 # # proxy icp
184 # # hostname type port port options
185 # # -------------------- -------- ----- ----- -----------
186 # cache_peer parent.foo.net parent 3128 3130 [proxy-only]
187 # cache_peer sib1.foo.net sibling 3128 3130 [proxy-only]
188 # cache_peer sib2.foo.net sibling 3128 3130 [proxy-only]
189 #
190 # type: either 'parent', 'sibling', or 'multicast'.
191 #
192 # proxy_port: The port number where the cache listens for proxy
193 # requests.
194 #
195 # icp_port: Used for querying neighbor caches about
196 # objects. To have a non-ICP neighbor
197 # specify '7' for the ICP port and make sure the
198 # neighbor machine has the UDP echo port
199 # enabled in its /etc/inetd.conf file.
200 #
201 # options: proxy-only
202 # weight=n
203 # ttl=n
204 # no-query
205 # default
206 # round-robin
207 # multicast-responder
208 # closest-only
209 # no-digest
210 # no-netdb-exchange
211 # no-delay
212 # login=user:password | PASS | *:password
213 # connect-timeout=nn
214 # digest-url=url
215 # allow-miss
216 # max-conn
217 # htcp
218 # carp-load-factor
219 #
220 # use 'proxy-only' to specify that objects fetched
221 # from this cache should not be saved locally.
222 #
223 # use 'weight=n' to specify a weighted parent.
224 # The weight must be an integer. The default weight
225 # is 1, larger weights are favored more.
226 #
227 # use 'ttl=n' to specify a IP multicast TTL to use
228 # when sending an ICP queries to this address.
229 # Only useful when sending to a multicast group.
230 # Because we don't accept ICP replies from random
231 # hosts, you must configure other group members as
232 # peers with the 'multicast-responder' option below.
233 #
234 # use 'no-query' to NOT send ICP queries to this
235 # neighbor.
236 #
237 # use 'default' if this is a parent cache which can
238 # be used as a "last-resort." You should probably
239 # only use 'default' in situations where you cannot
240 # use ICP with your parent cache(s).
241 #
242 # use 'round-robin' to define a set of parents which
243 # should be used in a round-robin fashion in the
244 # absence of any ICP queries.
245 #
246 # 'multicast-responder' indicates that the named peer
247 # is a member of a multicast group. ICP queries will
248 # not be sent directly to the peer, but ICP replies
249 # will be accepted from it.
250 #
251 # 'closest-only' indicates that, for ICP_OP_MISS
252 # replies, we'll only forward CLOSEST_PARENT_MISSes
253 # and never FIRST_PARENT_MISSes.
254 #
255 # use 'no-digest' to NOT request cache digests from
256 # this neighbor.
257 #
258 # 'no-netdb-exchange' disables requesting ICMP
259 # RTT database (NetDB) from the neighbor.
260 #
261 # use 'no-delay' to prevent access to this neighbor
262 # from influencing the delay pools.
263 #
264 # use 'login=user:password' if this is a personal/workgroup
265 # proxy and your parent requires proxy authentication.
266 # Note: The string can include URL escapes (i.e. %20 for
267 # spaces). This also means that % must be written as %%.
268 #
269 # use 'login=PASS' if users must authenticate against
270 # the upstream proxy. This will pass the users credentials
271 # as they are to the peer proxy. This only works for the
272 # Basic HTTP authentication sheme. Note: To combine this
273 # with proxy_auth both proxies must share the same user
274 # database as HTTP only allows for one proxy login.
275 # Also be warned that this will expose your users proxy
276 # password to the peer. USE WITH CAUTION
277 #
278 # use 'login=*:password' to pass the username to the
279 # upstream cache, but with a fixed password. This is meant
280 # to be used when the peer is in another administrative
281 # domain, but it is still needed to identify each user.
282 # The star can optionally be followed by some extra
283 # information which is added to the username. This can
284 # be used to identify this proxy to the peer, similar to
285 # the login=username:password option above.
286 #
287 # use 'connect-timeout=nn' to specify a peer
288 # specific connect timeout (also see the
289 # peer_connect_timeout directive)
290 #
291 # use 'digest-url=url' to tell Squid to fetch the cache
292 # digest (if digests are enabled) for this host from
293 # the specified URL rather than the Squid default
294 # location.
295 #
296 # use 'allow-miss' to disable Squid's use of only-if-cached
297 # when forwarding requests to siblings. This is primarily
298 # useful when icp_hit_stale is used by the sibling. To
299 # extensive use of this option may result in forwarding
300 # loops, and you should avoid having two-way peerings
301 # with this option. (for example to deny peer usage on
302 # requests from peer by denying cache_peer_access if the
303 # source is a peer)
304 #
305 # use 'max-conn' to limit the amount of connections Squid
306 # may open to this peer.
307 #
308 # use 'htcp' to send HTCP, instead of ICP, queries
309 # to the neighbor. You probably also want to
310 # set the "icp port" to 4827 instead of 3130.
311 #
312 # use 'carp-load-factor=f' to define a parent
313 # cache as one participating in a CARP array.
314 # The 'f' values for all CARP parents must add
315 # up to 1.0.
316 #
317 #
318 # NOTE: non-ICP/HTCP neighbors must be specified as 'parent'.
319 #
320 #Default:
321 # none
323 # TAG: cache_peer_domain
324 # Use to limit the domains for which a neighbor cache will be
325 # queried. Usage:
326 #
327 # cache_peer_domain cache-host domain [domain ...]
328 # cache_peer_domain cache-host !domain
329 #
330 # For example, specifying
331 #
332 # cache_peer_domain parent.foo.net .edu
333 #
334 # has the effect such that UDP query packets are sent to
335 # 'bigserver' only when the requested object exists on a
336 # server in the .edu domain. Prefixing the domainname
337 # with '!' means that the cache will be queried for objects
338 # NOT in that domain.
339 #
340 # NOTE: * Any number of domains may be given for a cache-host,
341 # either on the same or separate lines.
342 # * When multiple domains are given for a particular
343 # cache-host, the first matched domain is applied.
344 # * Cache hosts with no domain restrictions are queried
345 # for all requests.
346 # * There are no defaults.
347 # * There is also a 'cache_peer_access' tag in the ACL
348 # section.
349 #
350 #Default:
351 # none
353 # TAG: neighbor_type_domain
354 # usage: neighbor_type_domain neighbor parent|sibling domain domain ...
355 #
356 # Modifying the neighbor type for specific domains is now
357 # possible. You can treat some domains differently than the the
358 # default neighbor type specified on the 'cache_peer' line.
359 # Normally it should only be necessary to list domains which
360 # should be treated differently because the default neighbor type
361 # applies for hostnames which do not match domains listed here.
362 #
363 #EXAMPLE:
364 # cache_peer parent cache.foo.org 3128 3130
365 # neighbor_type_domain cache.foo.org sibling .com .net
366 # neighbor_type_domain cache.foo.org sibling .au .de
367 #
368 #Default:
369 # none
371 # TAG: icp_query_timeout (msec)
372 # Normally Squid will automatically determine an optimal ICP
373 # query timeout value based on the round-trip-time of recent ICP
374 # queries. If you want to override the value determined by
375 # Squid, set this 'icp_query_timeout' to a non-zero value. This
376 # value is specified in MILLISECONDS, so, to use a 2-second
377 # timeout (the old default), you would write:
378 #
379 # icp_query_timeout 2000
380 #
381 #Default:
382 # icp_query_timeout 0
384 # TAG: maximum_icp_query_timeout (msec)
385 # Normally the ICP query timeout is determined dynamically. But
386 # sometimes it can lead to very large values (say 5 seconds).
387 # Use this option to put an upper limit on the dynamic timeout
388 # value. Do NOT use this option to always use a fixed (instead
389 # of a dynamic) timeout value. To set a fixed timeout see the
390 # 'icp_query_timeout' directive.
391 #
392 #Default:
393 # maximum_icp_query_timeout 2000
395 # TAG: mcast_icp_query_timeout (msec)
396 # For Multicast peers, Squid regularly sends out ICP "probes" to
397 # count how many other peers are listening on the given multicast
398 # address. This value specifies how long Squid should wait to
399 # count all the replies. The default is 2000 msec, or 2
400 # seconds.
401 #
402 #Default:
403 # mcast_icp_query_timeout 2000
405 # TAG: dead_peer_timeout (seconds)
406 # This controls how long Squid waits to declare a peer cache
407 # as "dead." If there are no ICP replies received in this
408 # amount of time, Squid will declare the peer dead and not
409 # expect to receive any further ICP replies. However, it
410 # continues to send ICP queries, and will mark the peer as
411 # alive upon receipt of the first subsequent ICP reply.
412 #
413 # This timeout also affects when Squid expects to receive ICP
414 # replies from peers. If more than 'dead_peer' seconds have
415 # passed since the last ICP reply was received, Squid will not
416 # expect to receive an ICP reply on the next query. Thus, if
417 # your time between requests is greater than this timeout, you
418 # will see a lot of requests sent DIRECT to origin servers
419 # instead of to your parents.
420 #
421 #Default:
422 # dead_peer_timeout 10 seconds
424 # TAG: hierarchy_stoplist
425 # A list of words which, if found in a URL, cause the object to
426 # be handled directly by this cache. In other words, use this
427 # to not query neighbor caches for certain objects. You may
428 # list this option multiple times.
429 #We recommend you to use at least the following line.
430 hierarchy_stoplist cgi-bin ?
432 # TAG: no_cache
433 # A list of ACL elements which, if matched, cause the request to
434 # not be satisfied from the cache and the reply to not be cached.
435 # In other words, use this to force certain objects to never be cached.
436 #
437 # You must use the word 'DENY' to indicate the ACL names which should
438 # NOT be cached.
439 #
440 #We recommend you to use the following two lines.
441 acl QUERY urlpath_regex cgi-bin \?
442 no_cache deny QUERY
445 # OPTIONS WHICH AFFECT THE CACHE SIZE
446 # -----------------------------------------------------------------------------
448 # TAG: cache_mem (bytes)
449 # NOTE: THIS PARAMETER DOES NOT SPECIFY THE MAXIMUM PROCESS SIZE.
450 # IT ONLY PLACES A LIMIT ON HOW MUCH ADDITIONAL MEMORY SQUID WILL
451 # USE AS A MEMORY CACHE OF OBJECTS. SQUID USES MEMORY FOR OTHER
452 # THINGS AS WELL. SEE THE SQUID FAQ SECTION 8 FOR DETAILS.
453 #
454 # 'cache_mem' specifies the ideal amount of memory to be used
455 # for:
456 # * In-Transit objects
457 # * Hot Objects
458 # * Negative-Cached objects
459 #
460 # Data for these objects are stored in 4 KB blocks. This
461 # parameter specifies the ideal upper limit on the total size of
462 # 4 KB blocks allocated. In-Transit objects take the highest
463 # priority.
464 #
465 # In-transit objects have priority over the others. When
466 # additional space is needed for incoming data, negative-cached
467 # and hot objects will be released. In other words, the
468 # negative-cached and hot objects will fill up any unused space
469 # not needed for in-transit objects.
470 #
471 # If circumstances require, this limit will be exceeded.
472 # Specifically, if your incoming request rate requires more than
473 # 'cache_mem' of memory to hold in-transit objects, Squid will
474 # exceed this limit to satisfy the new requests. When the load
475 # decreases, blocks will be freed until the high-water mark is
476 # reached. Thereafter, blocks will be used to store hot
477 # objects.
478 #
479 #Default:
480 # cache_mem 8 MB
482 # TAG: cache_swap_low (percent, 0-100)
483 # TAG: cache_swap_high (percent, 0-100)
484 #
485 # The low- and high-water marks for cache object replacement.
486 # Replacement begins when the swap (disk) usage is above the
487 # low-water mark and attempts to maintain utilization near the
488 # low-water mark. As swap utilization gets close to high-water
489 # mark object eviction becomes more aggressive. If utilization is
490 # close to the low-water mark less replacement is done each time.
491 #
492 # Defaults are 90% and 95%. If you have a large cache, 5% could be
493 # hundreds of MB. If this is the case you may wish to set these
494 # numbers closer together.
495 #
496 #Default:
497 # cache_swap_low 90
498 # cache_swap_high 95
500 # TAG: maximum_object_size (bytes)
501 # Objects larger than this size will NOT be saved on disk. The
502 # value is specified in kilobytes, and the default is 4MB. If
503 # you wish to get a high BYTES hit ratio, you should probably
504 # increase this (one 32 MB object hit counts for 3200 10KB
505 # hits). If you wish to increase speed more than your want to
506 # save bandwidth you should leave this low.
507 #
508 # NOTE: if using the LFUDA replacement policy you should increase
509 # this value to maximize the byte hit rate improvement of LFUDA!
510 # See replacement_policy below for a discussion of this policy.
511 #
512 #Default:
513 # maximum_object_size 4096 KB
515 # TAG: minimum_object_size (bytes)
516 # Objects smaller than this size will NOT be saved on disk. The
517 # value is specified in kilobytes, and the default is 0 KB, which
518 # means there is no minimum.
519 #
520 #Default:
521 # minimum_object_size 0 KB
523 # TAG: maximum_object_size_in_memory (bytes)
524 # Objects greater than this size will not be attempted to kept in
525 # the memory cache. This should be set high enough to keep objects
526 # accessed frequently in memory to improve performance whilst low
527 # enough to keep larger objects from hoarding cache_mem .
528 #
529 #Default:
530 # maximum_object_size_in_memory 8 KB
532 # TAG: ipcache_size (number of entries)
533 # TAG: ipcache_low (percent)
534 # TAG: ipcache_high (percent)
535 # The size, low-, and high-water marks for the IP cache.
536 #
537 #Default:
538 # ipcache_size 1024
539 # ipcache_low 90
540 # ipcache_high 95
542 # TAG: fqdncache_size (number of entries)
543 # Maximum number of FQDN cache entries.
544 #
545 #Default:
546 # fqdncache_size 1024
548 # TAG: cache_replacement_policy
549 # The cache replacement policy parameter determines which
550 # objects are evicted (replaced) when disk space is needed.
551 #
552 # lru : Squid's original list based LRU policy
553 # heap GDSF : Greedy-Dual Size Frequency
554 # heap LFUDA: Least Frequently Used with Dynamic Aging
555 # heap LRU : LRU policy implemented using a heap
556 #
557 # Applies to any cache_dir lines listed below this.
558 #
559 # The LRU policies keeps recently referenced objects.
560 #
561 # The heap GDSF policy optimizes object hit rate by keeping smaller
562 # popular objects in cache so it has a better chance of getting a
563 # hit. It achieves a lower byte hit rate than LFUDA though since
564 # it evicts larger (possibly popular) objects.
565 #
566 # The heap LFUDA policy keeps popular objects in cache regardless of
567 # their size and thus optimizes byte hit rate at the expense of
568 # hit rate since one large, popular object will prevent many
569 # smaller, slightly less popular objects from being cached.
570 #
571 # Both policies utilize a dynamic aging mechanism that prevents
572 # cache pollution that can otherwise occur with frequency-based
573 # replacement policies.
574 #
575 # NOTE: if using the LFUDA replacement policy you should increase
576 # the value of maximum_object_size above its default of 4096 KB to
577 # to maximize the potential byte hit rate improvement of LFUDA.
578 #
579 # For more information about the GDSF and LFUDA cache replacement
580 # policies see http://www.hpl.hp.com/techreports/1999/HPL-1999-69.html
581 # and http://fog.hpl.external.hp.com/techreports/98/HPL-98-173.html.
582 #
583 #Default:
584 # cache_replacement_policy lru
586 # TAG: memory_replacement_policy
587 # The memory replacement policy parameter determines which
588 # objects are purged from memory when memory space is needed.
589 #
590 # See cache_replacement_policy for details.
591 #
592 #Default:
593 # memory_replacement_policy lru
596 # LOGFILE PATHNAMES AND CACHE DIRECTORIES
597 # -----------------------------------------------------------------------------
599 # TAG: cache_dir
600 # Usage:
601 #
602 # cache_dir Type Directory-Name Fs-specific-data [options]
603 #
604 # You can specify multiple cache_dir lines to spread the
605 # cache among different disk partitions.
606 #
607 # Type specifies the kind of storage system to use. Only "ufs"
608 # is built by default. To eanble any of the other storage systems
609 # see the --enable-storeio configure option.
610 #
611 # 'Directory' is a top-level directory where cache swap
612 # files will be stored. If you want to use an entire disk
613 # for caching, then this can be the mount-point directory.
614 # The directory must exist and be writable by the Squid
615 # process. Squid will NOT create this directory for you.
616 #
617 # The ufs store type:
618 #
619 # "ufs" is the old well-known Squid storage format that has always
620 # been there.
621 #
622 # cache_dir ufs Directory-Name Mbytes L1 L2 [options]
623 #
624 # 'Mbytes' is the amount of disk space (MB) to use under this
625 # directory. The default is 100 MB. Change this to suit your
626 # configuration. Do NOT put the size of your disk drive here.
627 # Instead, if you want Squid to use the entire disk drive,
628 # subtract 20% and use that value.
629 #
630 # 'Level-1' is the number of first-level subdirectories which
631 # will be created under the 'Directory'. The default is 16.
632 #
633 # 'Level-2' is the number of second-level subdirectories which
634 # will be created under each first-level directory. The default
635 # is 256.
636 #
637 # The aufs store type:
638 #
639 # "aufs" uses the same storage format as "ufs", utilizing
640 # POSIX-threads to avoid blocking the main Squid process on
641 # disk-I/O. This was formerly known in Squid as async-io.
642 #
643 # cache_dir aufs Directory-Name Mbytes L1 L2 [options]
644 #
645 # see argument descriptions under ufs above
646 #
647 # The diskd store type:
648 #
649 # "diskd" uses the same storage format as "ufs", utilizing a
650 # separate process to avoid blocking the main Squid process on
651 # disk-I/O.
652 #
653 # cache_dir diskd Directory-Name Mbytes L1 L2 [options] [Q1=n] [Q2=n]
654 #
655 # see argument descriptions under ufs above
656 #
657 # Q1 specifies the number of unacknowledged I/O requests when Squid
658 # stops opening new files. If this many messages are in the queues,
659 # Squid won't open new files. Default is 64
660 #
661 # Q2 specifies the number of unacknowledged messages when Squid
662 # starts blocking. If this many messages are in the queues,
663 # Squid blocks until it recevies some replies. Default is 72
664 #
665 # The coss store type:
666 #
667 # block-size=n defines the "block size" for COSS cache_dir's.
668 # Squid uses file numbers as block numbers. Since file numbers
669 # are limited to 24 bits, the block size determines the maximum
670 # size of the COSS partition. The default is 512 bytes, which
671 # leads to a maximum cache_dir size of 512<<24, or 8 GB. Note
672 # that you should not change the coss block size after Squid
673 # has written some objects to the cache_dir.
674 #
675 # Common options:
676 #
677 # read-only, this cache_dir is read only.
678 #
679 # max-size=n, refers to the max object size this storedir supports.
680 # It is used to initially choose the storedir to dump the object.
681 # Note: To make optimal use of the max-size limits you should order
682 # the cache_dir lines with the smallest max-size value first and the
683 # ones with no max-size specification last.
684 #
685 # Note that for coss, max-size must be less than COSS_MEMBUF_SZ
686 # (hard coded at 1 MB).
687 #
688 #Default:
689 # cache_dir ufs /var/spool/squid 100 16 256
691 # TAG: cache_access_log
692 # Logs the client request activity. Contains an entry for
693 # every HTTP and ICP queries received. To disable, enter "none".
694 #
695 #Default:
696 # cache_access_log /var/log/squid/access.log
698 # TAG: cache_log
699 # Cache logging file. This is where general information about
700 # your cache's behavior goes. You can increase the amount of data
701 # logged to this file with the "debug_options" tag below.
702 #
703 #Default:
704 # cache_log /var/log/squid/cache.log
706 # TAG: cache_store_log
707 # Logs the activities of the storage manager. Shows which
708 # objects are ejected from the cache, and which objects are
709 # saved and for how long. To disable, enter "none". There are
710 # not really utilities to analyze this data, so you can safely
711 # disable it.
712 #
713 #Default:
714 # cache_store_log /var/log/squid/store.log
716 # TAG: cache_swap_log
717 # Location for the cache "swap.log." This log file holds the
718 # metadata of objects saved on disk. It is used to rebuild the
719 # cache during startup. Normally this file resides in each
720 # 'cache_dir' directory, but you may specify an alternate
721 # pathname here. Note you must give a full filename, not just
722 # a directory. Since this is the index for the whole object
723 # list you CANNOT periodically rotate it!
724 #
725 # If %s can be used in the file name then it will be replaced with a
726 # a representation of the cache_dir name where each / is replaced
727 # with '.'. This is needed to allow adding/removing cache_dir
728 # lines when cache_swap_log is being used.
729 #
730 # If have more than one 'cache_dir', and %s is not used in the name
731 # then these swap logs will have names such as:
732 #
733 # cache_swap_log.00
734 # cache_swap_log.01
735 # cache_swap_log.02
736 #
737 # The numbered extension (which is added automatically)
738 # corresponds to the order of the 'cache_dir' lines in this
739 # configuration file. If you change the order of the 'cache_dir'
740 # lines in this file, then these log files will NOT correspond to
741 # the correct 'cache_dir' entry (unless you manually rename
742 # them). We recommend that you do NOT use this option. It is
743 # better to keep these log files in each 'cache_dir' directory.
744 #
745 #Default:
746 # none
748 # TAG: emulate_httpd_log on|off
749 # The Cache can emulate the log file format which many 'httpd'
750 # programs use. To disable/enable this emulation, set
751 # emulate_httpd_log to 'off' or 'on'. The default
752 # is to use the native log format since it includes useful
753 # information that Squid-specific log analyzers use.
754 #
755 #Default:
756 # emulate_httpd_log off
758 # TAG: log_ip_on_direct on|off
759 # Log the destination IP address in the hierarchy log tag when going
760 # direct. Earlier Squid versions logged the hostname here. If you
761 # prefer the old way set this to off.
762 #
763 #Default:
764 # log_ip_on_direct on
766 # TAG: mime_table
767 # Pathname to Squid's MIME table. You shouldn't need to change
768 # this, but the default file contains examples and formatting
769 # information if you do.
770 #
771 #Default:
772 # mime_table /etc/squid/mime.conf
774 # TAG: log_mime_hdrs on|off
775 # The Cache can record both the request and the response MIME
776 # headers for each HTTP transaction. The headers are encoded
777 # safely and will appear as two bracketed fields at the end of
778 # the access log (for either the native or httpd-emulated log
779 # formats). To enable this logging set log_mime_hdrs to 'on'.
780 #
781 #Default:
782 # log_mime_hdrs off
784 # TAG: useragent_log
785 # Squid will write the User-Agent field from HTTP requests
786 # to the filename specified here. By default useragent_log
787 # is disabled.
788 #
789 #Default:
790 # none
792 # TAG: referer_log
793 # Note: This option is only available if Squid is rebuilt with the
794 # --enable-referer-log option
795 #
796 # Squid will write the Referer field from HTTP requests to the
797 # filename specified here. By default referer_log is disabled.
798 #
799 #Default:
800 # none
802 # TAG: pid_filename
803 # A filename to write the process-id to. To disable, enter "none".
804 #
805 #Default:
806 # pid_filename /var/run/squid.pid
808 # TAG: debug_options
809 # Logging options are set as section,level where each source file
810 # is assigned a unique section. Lower levels result in less
811 # output, Full debugging (level 9) can result in a very large
812 # log file, so be careful. The magic word "ALL" sets debugging
813 # levels for all sections. We recommend normally running with
814 # "ALL,1".
815 #
816 #Default:
817 # debug_options ALL,1
819 # TAG: log_fqdn on|off
820 # Turn this on if you wish to log fully qualified domain names
821 # in the access.log. To do this Squid does a DNS lookup of all
822 # IP's connecting to it. This can (in some situations) increase
823 # latency, which makes your cache seem slower for interactive
824 # browsing.
825 #
826 #Default:
827 # log_fqdn off
829 # TAG: client_netmask
830 # A netmask for client addresses in logfiles and cachemgr output.
831 # Change this to protect the privacy of your cache clients.
832 # A netmask of 255.255.255.0 will log all IP's in that range with
833 # the last digit set to '0'.
834 #
835 #Default:
836 # client_netmask 255.255.255.255
839 # OPTIONS FOR EXTERNAL SUPPORT PROGRAMS
840 # -----------------------------------------------------------------------------
842 # TAG: ftp_user
843 # If you want the anonymous login password to be more informative
844 # (and enable the use of picky ftp servers), set this to something
845 # reasonable for your domain, like wwwuser@somewhere.net
846 #
847 # The reason why this is domainless by default is that the
848 # request can be made on the behalf of a user in any domain,
849 # depending on how the cache is used.
850 # Some ftp server also validate that the email address is valid
851 # (for example perl.com).
852 #
853 #Default:
854 # ftp_user Squid@
856 # TAG: ftp_list_width
857 # Sets the width of ftp listings. This should be set to fit in
858 # the width of a standard browser. Setting this too small
859 # can cut off long filenames when browsing ftp sites.
860 #
861 #Default:
862 # ftp_list_width 32
864 # TAG: ftp_passive
865 # If your firewall does not allow Squid to use passive
866 # connections, then turn off this option.
867 #
868 #Default:
869 # ftp_passive on
871 # TAG: ftp_sanitycheck
872 # For security and data integrity reasons Squid by default performs
873 # sanity checks of the addresses of FTP data connections ensure the
874 # data connection is to the requested server. If you need to allow
875 # FTP connections to servers using another IP address for the data
876 # connection then turn this off.
877 #
878 #Default:
879 # ftp_sanitycheck on
881 # TAG: ftp_telnet_protocol
882 # The FTP protocol is officially defined to use the telnet protocol
883 # as transport channel for the control connection. However, many
884 # implemenations are broken and does not respect this aspect of
885 # the FTP protocol.
886 #
887 # If you have trouble accessing files with ASCII code 255 in the
888 # path or similar problems involving this ASCII code then you can
889 # try setting this directive to off. If that helps report to the
890 # operator of the FTP server in question that their FTP server
891 # is broken and does not follow the FTP standard.
892 #
893 #Default:
894 # ftp_telnet_protocol on
896 # TAG: cache_dns_program
897 # Note: This option is only available if Squid is rebuilt with the
898 # --disable-internal-dns option
899 #
900 # Specify the location of the executable for dnslookup process.
901 #
902 #Default:
903 # cache_dns_program /usr/lib/squid/dnsserver
905 # TAG: dns_children
906 # Note: This option is only available if Squid is rebuilt with the
907 # --disable-internal-dns option
908 #
909 # The number of processes spawn to service DNS name lookups.
910 # For heavily loaded caches on large servers, you should
911 # probably increase this value to at least 10. The maximum
912 # is 32. The default is 5.
913 #
914 # You must have at least one dnsserver process.
915 #
916 #Default:
917 # dns_children 5
919 # TAG: dns_retransmit_interval
920 # Initial retransmit interval for DNS queries. The interval is
921 # doubled each time all configured DNS servers have been tried.
922 #
923 #
924 #Default:
925 # dns_retransmit_interval 5 seconds
927 # TAG: dns_timeout
928 # DNS Query timeout. If no response is received to a DNS query
929 # within this time then all DNS servers for the queried domain
930 # is assumed to be unavailable.
931 #
932 #Default:
933 # dns_timeout 2 minutes
935 # TAG: dns_defnames on|off
936 # Note: This option is only available if Squid is rebuilt with the
937 # --disable-internal-dns option
938 #
939 # Normally the 'dnsserver' disables the RES_DEFNAMES resolver
940 # option (see res_init(3)). This prevents caches in a hierarchy
941 # from interpreting single-component hostnames locally. To allow
942 # dnsserver to handle single-component names, enable this
943 # option.
944 #
945 #Default:
946 # dns_defnames off
948 # TAG: dns_nameservers
949 # Use this if you want to specify a list of DNS name servers
950 # (IP addresses) to use instead of those given in your
951 # /etc/resolv.conf file.
952 # On Windows platforms, if no value is specified here or in
953 # the /etc/resolv.conf file, the list of DNS name servers are
954 # taken from the Windows registry, both static and dynamic DHCP
955 # configurations are supported.
956 #
957 # Example: dns_nameservers 10.0.0.1 192.172.0.4
958 #
959 #Default:
960 # none
962 # TAG: hosts_file
963 # Location of the host-local IP name-address associations
964 # database. Most Operating Systems have such a file: under
965 # Un*X it's by default in /etc/hosts MS-Windows NT/2000 places
966 # that in %SystemRoot%(by default
967 # c:\winnt)\system32\drivers\etc\hosts, while Windows 9x/ME
968 # places that in %windir%(usually c:\windows)\hosts
969 #
970 # The file contains newline-separated definitions, in the
971 # form ip_address_in_dotted_form name [name ...] names are
972 # whitespace-separated. lines beginnng with an hash (#)
973 # character are comments.
974 #
975 # The file is checked at startup and upon configuration. If
976 # set to 'none', it won't be checked. If append_domain is
977 # used, that domain will be added to domain-local (i.e. not
978 # containing any dot character) host definitions.
979 #
980 #Default:
981 # hosts_file /etc/hosts
983 # TAG: diskd_program
984 # Specify the location of the diskd executable.
985 # Note that this is only useful if you have compiled in
986 # diskd as one of the store io modules.
987 #
988 #Default:
989 # diskd_program /usr/lib/squid/diskd
991 # TAG: unlinkd_program
992 # Specify the location of the executable for file deletion process.
993 #
994 #Default:
995 # unlinkd_program /usr/lib/squid/unlinkd
997 # TAG: pinger_program
998 # Specify the location of the executable for the pinger process.
999 #
1000 #Default:
1001 # pinger_program /usr/lib/squid/pinger
1003 # TAG: redirect_program
1004 # Specify the location of the executable for the URL redirector.
1005 # Since they can perform almost any function there isn't one included.
1006 # See the FAQ (section 15) for information on how to write one.
1007 # By default, a redirector is not used.
1008 #
1009 #Default:
1010 # none
1012 # TAG: redirect_children
1013 # The number of redirector processes to spawn. If you start
1014 # too few Squid will have to wait for them to process a backlog of
1015 # URLs, slowing it down. If you start too many they will use RAM
1016 # and other system resources.
1017 #
1018 #Default:
1019 # redirect_children 5
1021 # TAG: redirect_rewrites_host_header
1022 # By default Squid rewrites any Host: header in redirected
1023 # requests. If you are running an accelerator then this may
1024 # not be a wanted effect of a redirector.
1025 #
1026 #Default:
1027 # redirect_rewrites_host_header on
1029 # TAG: redirector_access
1030 # If defined, this access list specifies which requests are
1031 # sent to the redirector processes. By default all requests
1032 # are sent.
1033 #
1034 #Default:
1035 # none
1037 # TAG: auth_param
1038 # This is used to define parameters for the various authentication
1039 # schemes supported by Squid.
1040 #
1041 # format: auth_param scheme parameter [setting]
1042 #
1043 # The order that authentication schemes are presented to the client is
1044 # dependant on the order the scheme first appears in config file. IE
1045 # has a bug (it's not rfc 2617 compliant) in that it will use the basic
1046 # scheme if basic is the first entry presented, even if more secure
1047 # schemes are presented. For now use the order in the recommended
1048 # settings section below. If other browsers have difficulties (don't
1049 # recognise the schemes offered even if you are using basic) then either
1050 # put basic first, or disable the other schemes (by commenting out their
1051 # program entry).
1052 #
1053 # Once an authentication scheme is fully configured, it can only be
1054 # shutdown by shutting squid down and restarting. Changes can be made on
1055 # the fly and activated with a reconfigure. I.E. You can change to a
1056 # different helper, but not unconfigure the helper completely.
1057 #
1058 # Please note that while this directive defines how Squid processes
1059 # authentication it does not automatically activate authentication.
1060 # To use authenticaiton you must in addition make use of acls based
1061 # on login name in http_access (proxy_auth, proxy_auth_regex or
1062 # external with %LOGIN used in the format tag). The browser will be
1063 # challenged for authentication on the first such acl encountered
1064 # in http_access processing and will also be rechallenged for new
1065 # login credentials if the request is being denied by a proxy_auth
1066 # type acl.
1067 #
1068 # === Parameters for the basic scheme follow. ===
1069 #
1070 # "program" cmdline
1071 # Specify the command for the external authenticator. Such a program
1072 # reads a line containing "username password" and replies "OK" or
1073 # "ERR" in an endless loop.
1074 #
1075 # By default, the basic authentication sheme is not used unless a
1076 # program is specified.
1077 #
1078 # If you want to use the traditional proxy authentication, jump over to
1079 # the helpers/basic_auth/NCSA directory and type:
1080 # % make
1081 # % make install
1082 #
1083 # Then, set this line to something like
1084 #
1085 # auth_param basic program /usr/libexec/ncsa_auth /usr/etc/passwd
1086 #
1087 # "children" numberofchildren
1088 # The number of authenticator processes to spawn.
1089 # If you start too few Squid will have to wait for them to process a
1090 # backlog of usercode/password verifications, slowing it down. When
1091 # password verifications are done via a (slow) network you are likely to
1092 # need lots of authenticator processes.
1093 # auth_param basic children 5
1094 #
1095 # "realm" realmstring
1096 # Specifies the realm name which is to be reported to the client for
1097 # the basic proxy authentication scheme (part of the text the user
1098 # will see when prompted their username and password).
1099 # auth_param basic realm Squid proxy-caching web server
1100 #
1101 # "credentialsttl" timetolive
1102 # Specifies how long squid assumes an externally validated
1103 # username:password pair is valid for - in other words how often the
1104 # helper program is called for that user. Set this low to force
1105 # revalidation with short lived passwords. Note that setting this high
1106 # does not impact your susceptability to replay attacks unless you are
1107 # using an one-time password system (such as SecureID). If you are using
1108 # such a system, you will be vulnerable to replay attacks unless you
1109 # also use the max_user_ip ACL in an http_access rule.
1110 # auth_param basic credentialsttl 2 hours
1111 #
1112 # === Parameters for the digest scheme follow ===
1113 #
1114 # "program" cmdline
1115 # Specify the command for the external authenticator. Such a program
1116 # reads a line containing "username":"realm" and replies with the
1117 # appropriate H(A1) value base64 encoded. See rfc 2616 for the
1118 # definition of H(A1).
1119 #
1120 # By default, the digest authentication scheme is not used unless a
1121 # program is specified.
1122 #
1123 # If you want to use a digest authenticator, jump over to the
1124 # helpers/digest_auth/ directory and choose the authenticator to use.
1125 # It it's directory type
1126 # % make
1127 # % make install
1128 #
1129 # Then, set this line to something like
1130 #
1131 # auth_param digest program /usr/libexec/digest_auth_pw /usr/etc/digpass
1132 #
1133 #
1134 # "children" numberofchildren
1135 # The number of authenticator processes to spawn (no default). If you
1136 # start too few Squid will have to wait for them to process a backlog of
1137 # H(A1) calculations, slowing it down. When the H(A1) calculations are
1138 # done via a (slow) network you are likely to need lots of authenticator
1139 # processes.
1140 # auth_param digest children 5
1141 #
1142 # "realm" realmstring
1143 # Specifies the realm name which is to be reported to the client for the
1144 # digest proxy authentication scheme (part of the text the user will see
1145 # when prompted their username and password).
1146 # auth_param digest realm Squid proxy-caching web server
1147 #
1148 # "nonce_garbage_interval" timeinterval
1149 # Specifies the interval that nonces that have been issued to clients are
1150 # checked for validity.
1151 # auth_param digest nonce_garbage_interval 5 minutes
1152 #
1153 # "nonce_max_duration" timeinterval
1154 # Specifies the maximum length of time a given nonce will be valid for.
1155 # auth_param digest nonce_max_duration 30 minutes
1156 #
1157 # "nonce_max_count" number
1158 # Specifies the maximum number of times a given nonce can be used.
1159 # auth_param digest nonce_max_count 50
1160 #
1161 # "nonce_strictness" on|off
1162 # Determines if squid requires strict increment-by-1 behaviour for nonce
1163 # counts, or just incrementing (off - for use when useragents generate
1164 # nonce counts that occasionally miss 1 (ie, 1,2,4,6)).
1165 # auth_param digest nonce_strictness off
1166 #
1167 # "check_nonce_count" on|off
1168 # This directive if set to off can disable the nonce count check
1169 # completely to work around buggy digest qop implementations in certain
1170 # mainstream browser versions. Default on to check the nonce count to
1171 # protect from authentication replay attacks.
1172 # auth_param digest check_nonce_count on
1173 #
1174 # "post_workaround" on|off
1175 # This is a workaround to certain buggy browsers who sends an incorrect
1176 # request digest in POST requests when reusing the same nonce as aquired
1177 # earlier in response to a GET request.
1178 # auth_param digest post_workaround off
1179 #
1180 # === NTLM scheme options follow ===
1181 #
1182 # "program" cmdline
1183 # Specify the command for the external ntlm authenticator. Such a
1184 # program participates in the NTLMSSP exchanges between Squid and the
1185 # client and reads commands according to the Squid ntlmssp helper
1186 # protocol. See helpers/ntlm_auth/ for details. Recommended ntlm
1187 # authenticator is ntlm_auth from Samba-3.X, but a number of other
1188 # ntlm authenticators is available.
1189 #
1190 # By default, the ntlm authentication scheme is not used unless a
1191 # program is specified.
1192 #
1193 # auth_param ntlm program /path/to/samba/bin/ntlm_auth --helper-protocol=squid-2.5-ntlmssp
1194 #
1195 # "children" numberofchildren
1196 # The number of authenticator processes to spawn (no default). If you
1197 # start too few Squid will have to wait for them to process a backlog
1198 # of credential verifications, slowing it down. When crendential
1199 # verifications are done via a (slow) network you are likely to need
1200 # lots of authenticator processes.
1201 # auth_param ntlm children 5
1202 #
1203 # "max_challenge_reuses" number
1204 # The maximum number of times a challenge given by a ntlm authentication
1205 # helper can be reused. Increasing this number increases your exposure
1206 # to replay attacks on your network. 0 (the default) means use the
1207 # challenge is used only once. See also the max_ntlm_challenge_lifetime
1208 # directive if enabling challenge reuses.
1209 # auth_param ntlm max_challenge_reuses 0
1210 #
1211 # "max_challenge_lifetime" timespan
1212 # The maximum time period that a ntlm challenge is reused over. The
1213 # actual period will be the minimum of this time AND the number of
1214 # reused challenges.
1215 # auth_param ntlm max_challenge_lifetime 2 minutes
1216 #
1217 # "use_ntlm_negotiate" on|off
1218 # Enables support for NTLM NEGOTIATE packet exchanges with the helper.
1219 # The configured ntlm authenticator must be able to handle NTLM
1220 # NEGOTIATE packet. See the authenticator programs documentation if
1221 # unsure. ntlm_auth from Samba-3.0.2 or later supports the use of this
1222 # option.
1223 # The NEGOTIATE packet is required to support NTLMv2 and a
1224 # number of other negotiable NTLMSSP options, and also makes it
1225 # more likely the negotiation is successful. Enabling this parameter
1226 # will also solve problems encountered when NT domain policies
1227 # restrict users to access only certain workstations. When this is off,
1228 # all users must be allowed to log on the proxy servers too, or they'll
1229 # get "invalid workstation" errors - and access denied - when trying to
1230 # use Squid's services.
1231 # Use of ntlm NEGOTIATE is incompatible with challenge reuse, so
1232 # enabling this parameter will OVERRIDE the max_challenge_reuses and
1233 # max_challenge_lifetime parameters and set them to 0.
1234 # auth_param ntlm use_ntlm_negotiate off
1235 #
1236 #Recommended minimum configuration:
1237 #auth_param digest program <uncomment and complete this line>
1238 #auth_param digest children 5
1239 #auth_param digest realm Squid proxy-caching web server
1240 #auth_param digest nonce_garbage_interval 5 minutes
1241 #auth_param digest nonce_max_duration 30 minutes
1242 #auth_param digest nonce_max_count 50
1243 #auth_param ntlm program /usr/lib/squid/ntlm_auth IPH\\PDC
1244 #auth_param ntlm children 5
1245 #auth_param ntlm max_challenge_reuses 0
1246 #auth_param ntlm max_challenge_lifetime 2 minutes
1247 #auth_param ntlm use_ntlm_negotiate off
1248 auth_param basic program /usr/lib/squid/squid_ldap_auth -b ou=People,dc=example,dc=com -f (&(uid=%s)(objectClass=gosaProxyAccount))
1249 auth_param basic children 5
1250 auth_param basic realm Squid proxy-caching web server
1251 auth_param basic credentialsttl 2 hours
1253 # TAG: authenticate_cache_garbage_interval
1254 # The time period between garbage collection across the username cache.
1255 # This is a tradeoff between memory utilisation (long intervals - say
1256 # 2 days) and CPU (short intervals - say 1 minute). Only change if you
1257 # have good reason to.
1258 #
1259 #Default:
1260 # authenticate_cache_garbage_interval 1 hour
1262 # TAG: authenticate_ttl
1263 # The time a user & their credentials stay in the logged in user cache
1264 # since their last request. When the garbage interval passes, all user
1265 # credentials that have passed their TTL are removed from memory.
1266 #
1267 #Default:
1268 # authenticate_ttl 1 hour
1270 # TAG: authenticate_ip_ttl
1271 # If you use proxy authentication and the 'max_user_ip' ACL, this
1272 # directive controls how long Squid remembers the IP addresses
1273 # associated with each user. Use a small value (e.g., 60 seconds) if
1274 # your users might change addresses quickly, as is the case with
1275 # dialups. You might be safe using a larger value (e.g., 2 hours) in a
1276 # corporate LAN environment with relatively static address assignments.
1277 #
1278 #Default:
1279 # authenticate_ip_ttl 0 seconds
1281 # TAG: external_acl_type
1282 # This option defines external acl classes using a helper program to
1283 # look up the status
1284 #
1285 # external_acl_type name [options] FORMAT.. /path/to/helper [helper arguments..]
1286 #
1287 # Options:
1288 #
1289 # ttl=n TTL in seconds for cached results (defaults to 3600
1290 # for 1 hour)
1291 # negative_ttl=n
1292 # TTL for cached negative lookups (default same
1293 # as ttl)
1294 # children=n Concurrency level / number of processes spawn
1295 # to service external acl lookups of this type.
1296 # Note: see compatibility note below
1297 # cache=n result cache size, 0 is unbounded (default)
1298 #
1299 # FORMAT specifications
1300 #
1301 # %LOGIN Authenticated user login name
1302 # %IDENT Ident user name
1303 # %SRC Client IP
1304 # %DST Requested host
1305 # %PROTO Requested protocol
1306 # %PORT Requested port
1307 # %METHOD Request method
1308 # %{Header} HTTP request header
1309 # %{Hdr:member} HTTP request header list member
1310 # %{Hdr:;member}
1311 # HTTP request header list member using ; as
1312 # list separator. ; can be any non-alphanumeric
1313 # character.
1314 #
1315 # In addition, any string specified in the referencing acl will
1316 # also be included in the helper request line, after the specified
1317 # formats (see the "acl external" directive)
1318 #
1319 # The helper receives lines per the above format specification,
1320 # and returns lines starting with OK or ERR indicating the validity
1321 # of the request and optionally followed by additional keywords with
1322 # more details.
1323 #
1324 # General result syntax:
1325 #
1326 # OK/ERR keyword=value ...
1327 #
1328 # Defined keywords:
1329 #
1330 # user= The users name (login)
1331 # error= Error description (only defined for ERR results)
1332 #
1333 # Keyword values need to be enclosed in quotes if they may contain
1334 # whitespace, or the whitespace escaped using \. Any quotes or \
1335 # characters within the keyword value must be \ escaped.
1336 #
1337 # Compatibility Note: The children= option was named concurrency= in
1338 # Squid-2.5.STABLE3 and earlier and such syntax is still accepted to
1339 # keep compatibility within the Squid-2.5 release. However, the meaning
1340 # of concurrency= option has changed in Squid-3 and the old syntax of
1341 # the directive is therefore depreated from Squid-2.5.STABLE4 and later.
1342 # If you want to be able to easily downgrade to earlier Squid-2.5
1343 # releases then you may want to continue using the old name, if not
1344 # please use the new name.
1345 #
1346 #Default:
1347 # none
1350 # OPTIONS FOR TUNING THE CACHE
1351 # -----------------------------------------------------------------------------
1353 # TAG: wais_relay_host
1354 # TAG: wais_relay_port
1355 # Relay WAIS request to host (1st arg) at port (2 arg).
1356 #
1357 #Default:
1358 # wais_relay_port 0
1360 # TAG: request_header_max_size (KB)
1361 # This specifies the maximum size for HTTP headers in a request.
1362 # Request headers are usually relatively small (about 512 bytes).
1363 # Placing a limit on the request header size will catch certain
1364 # bugs (for example with persistent connections) and possibly
1365 # buffer-overflow or denial-of-service attacks.
1366 #
1367 #Default:
1368 # request_header_max_size 10 KB
1370 # TAG: request_body_max_size (KB)
1371 # This specifies the maximum size for an HTTP request body.
1372 # In other words, the maximum size of a PUT/POST request.
1373 # A user who attempts to send a request with a body larger
1374 # than this limit receives an "Invalid Request" error message.
1375 # If you set this parameter to a zero (the default), there will
1376 # be no limit imposed.
1377 #
1378 #Default:
1379 # request_body_max_size 0 KB
1381 # TAG: refresh_pattern
1382 # usage: refresh_pattern [-i] regex min percent max [options]
1383 #
1384 # By default, regular expressions are CASE-SENSITIVE. To make
1385 # them case-insensitive, use the -i option.
1386 #
1387 # 'Min' is the time (in minutes) an object without an explicit
1388 # expiry time should be considered fresh. The recommended
1389 # value is 0, any higher values may cause dynamic applications
1390 # to be erroneously cached unless the application designer
1391 # has taken the appropriate actions.
1392 #
1393 # 'Percent' is a percentage of the objects age (time since last
1394 # modification age) an object without explicit expiry time
1395 # will be considered fresh.
1396 #
1397 # 'Max' is an upper limit on how long objects without an explicit
1398 # expiry time will be considered fresh.
1399 #
1400 # options: override-expire
1401 # override-lastmod
1402 # reload-into-ims
1403 # ignore-reload
1404 #
1405 # override-expire enforces min age even if the server
1406 # sent a Expires: header. Doing this VIOLATES the HTTP
1407 # standard. Enabling this feature could make you liable
1408 # for problems which it causes.
1409 #
1410 # override-lastmod enforces min age even on objects
1411 # that was modified recently.
1412 #
1413 # reload-into-ims changes client no-cache or ``reload''
1414 # to If-Modified-Since requests. Doing this VIOLATES the
1415 # HTTP standard. Enabling this feature could make you
1416 # liable for problems which it causes.
1417 #
1418 # ignore-reload ignores a client no-cache or ``reload''
1419 # header. Doing this VIOLATES the HTTP standard. Enabling
1420 # this feature could make you liable for problems which
1421 # it causes.
1422 #
1423 # Basically a cached object is:
1424 #
1425 # FRESH if expires < now, else STALE
1426 # STALE if age > max
1427 # FRESH if lm-factor < percent, else STALE
1428 # FRESH if age < min
1429 # else STALE
1430 #
1431 # The refresh_pattern lines are checked in the order listed here.
1432 # The first entry which matches is used. If none of the entries
1433 # match, then the default will be used.
1434 #
1435 # Note, you must uncomment all the default lines if you want
1436 # to change one. The default setting is only active if none is
1437 # used.
1438 #
1439 #Suggested default:
1440 refresh_pattern ^ftp: 1440 20% 10080
1441 refresh_pattern ^gopher: 1440 0% 1440
1442 refresh_pattern . 0 20% 4320
1444 # TAG: quick_abort_min (KB)
1445 # TAG: quick_abort_max (KB)
1446 # TAG: quick_abort_pct (percent)
1447 # The cache by default continues downloading aborted requests
1448 # which are almost completed (less than 16 KB remaining). This
1449 # may be undesirable on slow (e.g. SLIP) links and/or very busy
1450 # caches. Impatient users may tie up file descriptors and
1451 # bandwidth by repeatedly requesting and immediately aborting
1452 # downloads.
1453 #
1454 # When the user aborts a request, Squid will check the
1455 # quick_abort values to the amount of data transfered until
1456 # then.
1457 #
1458 # If the transfer has less than 'quick_abort_min' KB remaining,
1459 # it will finish the retrieval.
1460 #
1461 # If the transfer has more than 'quick_abort_max' KB remaining,
1462 # it will abort the retrieval.
1463 #
1464 # If more than 'quick_abort_pct' of the transfer has completed,
1465 # it will finish the retrieval.
1466 #
1467 # If you do not want any retrieval to continue after the client
1468 # has aborted, set both 'quick_abort_min' and 'quick_abort_max'
1469 # to '0 KB'.
1470 #
1471 # If you want retrievals to always continue if they are being
1472 # cached then set 'quick_abort_min' to '-1 KB'.
1473 #
1474 #Default:
1475 # quick_abort_min 16 KB
1476 # quick_abort_max 16 KB
1477 # quick_abort_pct 95
1479 # TAG: negative_ttl time-units
1480 # Time-to-Live (TTL) for failed requests. Certain types of
1481 # failures (such as "connection refused" and "404 Not Found") are
1482 # negatively-cached for a configurable amount of time. The
1483 # default is 5 minutes. Note that this is different from
1484 # negative caching of DNS lookups.
1485 #
1486 #Default:
1487 # negative_ttl 5 minutes
1489 # TAG: positive_dns_ttl time-units
1490 # Upper limit on how long Squid will cache positive DNS responses.
1491 # Default is 6 hours (360 minutes). This directive must be set
1492 # larger than negative_dns_ttl.
1493 #
1494 #Default:
1495 # positive_dns_ttl 6 hours
1497 # TAG: negative_dns_ttl time-units
1498 # Time-to-Live (TTL) for negative caching of failed DNS lookups.
1499 # This also makes sets the lower cache limit on positive lookups.
1500 # Minimum value is 1 second, and it is not recommendable to go
1501 # much below 10 seconds.
1502 #
1503 #Default:
1504 # negative_dns_ttl 1 minute
1506 # TAG: range_offset_limit (bytes)
1507 # Sets a upper limit on how far into the the file a Range request
1508 # may be to cause Squid to prefetch the whole file. If beyond this
1509 # limit then Squid forwards the Range request as it is and the result
1510 # is NOT cached.
1511 #
1512 # This is to stop a far ahead range request (lets say start at 17MB)
1513 # from making Squid fetch the whole object up to that point before
1514 # sending anything to the client.
1515 #
1516 # A value of -1 causes Squid to always fetch the object from the
1517 # beginning so that it may cache the result. (2.0 style)
1518 #
1519 # A value of 0 causes Squid to never fetch more than the
1520 # client requested. (default)
1521 #
1522 #Default:
1523 # range_offset_limit 0 KB
1526 # TIMEOUTS
1527 # -----------------------------------------------------------------------------
1529 # TAG: forward_timeout time-units
1530 # This parameter specifies how long Squid should at most attempt in
1531 # finding a forwarding path for the request before giving up.
1532 #
1533 #Default:
1534 # forward_timeout 4 minutes
1536 # TAG: connect_timeout time-units
1537 # This parameter specifies how long to wait for the TCP connect to
1538 # the requested server or peer to complete before Squid should
1539 # attempt to find another path where to forward the request.
1540 #
1541 #Default:
1542 # connect_timeout 1 minute
1544 # TAG: peer_connect_timeout time-units
1545 # This parameter specifies how long to wait for a pending TCP
1546 # connection to a peer cache. The default is 30 seconds. You
1547 # may also set different timeout values for individual neighbors
1548 # with the 'connect-timeout' option on a 'cache_peer' line.
1549 #
1550 #Default:
1551 # peer_connect_timeout 30 seconds
1553 # TAG: read_timeout time-units
1554 # The read_timeout is applied on server-side connections. After
1555 # each successful read(), the timeout will be extended by this
1556 # amount. If no data is read again after this amount of time,
1557 # the request is aborted and logged with ERR_READ_TIMEOUT. The
1558 # default is 15 minutes.
1559 #
1560 #Default:
1561 # read_timeout 15 minutes
1563 # TAG: request_timeout
1564 # How long to wait for an HTTP request after initial
1565 # connection establishment.
1566 #
1567 #Default:
1568 # request_timeout 5 minutes
1570 # TAG: persistent_request_timeout
1571 # How long to wait for the next HTTP request on a persistent
1572 # connection after the previous request completes.
1573 #
1574 #Default:
1575 # persistent_request_timeout 1 minute
1577 # TAG: client_lifetime time-units
1578 # The maximum amount of time that a client (browser) is allowed to
1579 # remain connected to the cache process. This protects the Cache
1580 # from having a lot of sockets (and hence file descriptors) tied up
1581 # in a CLOSE_WAIT state from remote clients that go away without
1582 # properly shutting down (either because of a network failure or
1583 # because of a poor client implementation). The default is one
1584 # day, 1440 minutes.
1585 #
1586 # NOTE: The default value is intended to be much larger than any
1587 # client would ever need to be connected to your cache. You
1588 # should probably change client_lifetime only as a last resort.
1589 # If you seem to have many client connections tying up
1590 # filedescriptors, we recommend first tuning the read_timeout,
1591 # request_timeout, persistent_request_timeout and quick_abort values.
1592 #
1593 #Default:
1594 # client_lifetime 1 day
1596 # TAG: half_closed_clients
1597 # Some clients may shutdown the sending side of their TCP
1598 # connections, while leaving their receiving sides open. Sometimes,
1599 # Squid can not tell the difference between a half-closed and a
1600 # fully-closed TCP connection. By default, half-closed client
1601 # connections are kept open until a read(2) or write(2) on the
1602 # socket returns an error. Change this option to 'off' and Squid
1603 # will immediately close client connections when read(2) returns
1604 # "no more data to read."
1605 #
1606 #Default:
1607 # half_closed_clients on
1609 # TAG: pconn_timeout
1610 # Timeout for idle persistent connections to servers and other
1611 # proxies.
1612 #
1613 #Default:
1614 # pconn_timeout 120 seconds
1616 # TAG: ident_timeout
1617 # Maximum time to wait for IDENT lookups to complete.
1618 #
1619 # If this is too high, and you enabled IDENT lookups from untrusted
1620 # users, then you might be susceptible to denial-of-service by having
1621 # many ident requests going at once.
1622 #
1623 #Default:
1624 # ident_timeout 10 seconds
1626 # TAG: shutdown_lifetime time-units
1627 # When SIGTERM or SIGHUP is received, the cache is put into
1628 # "shutdown pending" mode until all active sockets are closed.
1629 # This value is the lifetime to set for all open descriptors
1630 # during shutdown mode. Any active clients after this many
1631 # seconds will receive a 'timeout' message.
1632 #
1633 #Default:
1634 # shutdown_lifetime 30 seconds
1637 # ACCESS CONTROLS
1638 # -----------------------------------------------------------------------------
1640 # TAG: acl
1641 # Defining an Access List
1642 #
1643 # acl aclname acltype string1 ...
1644 # acl aclname acltype "file" ...
1645 #
1646 # when using "file", the file should contain one item per line
1647 #
1648 # acltype is one of the types described below
1649 #
1650 # By default, regular expressions are CASE-SENSITIVE. To make
1651 # them case-insensitive, use the -i option.
1652 #
1653 # acl aclname src ip-address/netmask ... (clients IP address)
1654 # acl aclname src addr1-addr2/netmask ... (range of addresses)
1655 # acl aclname dst ip-address/netmask ... (URL host's IP address)
1656 # acl aclname myip ip-address/netmask ... (local socket IP address)
1657 #
1658 # acl aclname srcdomain .foo.com ... # reverse lookup, client IP
1659 # acl aclname dstdomain .foo.com ... # Destination server from URL
1660 # acl aclname srcdom_regex [-i] xxx ... # regex matching client name
1661 # acl aclname dstdom_regex [-i] xxx ... # regex matching server
1662 # # For dstdomain and dstdom_regex a reverse lookup is tried if a IP
1663 # # based URL is used. The name "none" is used if the reverse lookup
1664 # # fails.
1665 #
1666 # acl aclname time [day-abbrevs] [h1:m1-h2:m2]
1667 # day-abbrevs:
1668 # S - Sunday
1669 # M - Monday
1670 # T - Tuesday
1671 # W - Wednesday
1672 # H - Thursday
1673 # F - Friday
1674 # A - Saturday
1675 # h1:m1 must be less than h2:m2
1676 # acl aclname url_regex [-i] ^http:// ... # regex matching on whole URL
1677 # acl aclname urlpath_regex [-i] \.gif$ ... # regex matching on URL path
1678 # acl aclname urllogin [-i] [^a-zA-Z0-9] ... # regex matching on URL login field
1679 # acl aclname port 80 70 21 ...
1680 # acl aclname port 0-1024 ... # ranges allowed
1681 # acl aclname myport 3128 ... # (local socket TCP port)
1682 # acl aclname proto HTTP FTP ...
1683 # acl aclname method GET POST ...
1684 # acl aclname browser [-i] regexp ...
1685 # # pattern match on User-Agent header
1686 # acl aclname referer_regex [-i] regexp ...
1687 # # pattern match on Referer header
1688 # # Referer is highly unreliable, so use with care
1689 # acl aclname ident username ...
1690 # acl aclname ident_regex [-i] pattern ...
1691 # # string match on ident output.
1692 # # use REQUIRED to accept any non-null ident.
1693 # acl aclname src_as number ...
1694 # acl aclname dst_as number ...
1695 # # Except for access control, AS numbers can be used for
1696 # # routing of requests to specific caches. Here's an
1697 # # example for routing all requests for AS#1241 and only
1698 # # those to mycache.mydomain.net:
1699 # # acl asexample dst_as 1241
1700 # # cache_peer_access mycache.mydomain.net allow asexample
1701 # # cache_peer_access mycache_mydomain.net deny all
1702 #
1703 # acl aclname proxy_auth username ...
1704 # acl aclname proxy_auth_regex [-i] pattern ...
1705 # # list of valid usernames
1706 # # use REQUIRED to accept any valid username.
1707 # #
1708 # # NOTE: when a Proxy-Authentication header is sent but it is not
1709 # # needed during ACL checking the username is NOT logged
1710 # # in access.log.
1711 # #
1712 # # NOTE: proxy_auth requires a EXTERNAL authentication program
1713 # # to check username/password combinations (see
1714 # # auth_param directive).
1715 # #
1716 # # WARNING: proxy_auth can't be used in a transparent proxy. It
1717 # # collides with any authentication done by origin servers. It may
1718 # # seem like it works at first, but it doesn't.
1719 #
1720 # acl aclname snmp_community string ...
1721 # # A community string to limit access to your SNMP Agent
1722 # # Example:
1723 # #
1724 # # acl snmppublic snmp_community public
1725 #
1726 # acl aclname maxconn number
1727 # # This will be matched when the client's IP address has
1728 # # more than <number> HTTP connections established.
1729 #
1730 # acl aclname max_user_ip [-s] number
1731 # # This will be matched when the user attempts to log in from more
1732 # # than <number> different ip addresses. The authenticate_ip_ttl
1733 # # parameter controls the timeout on the ip entries.
1734 # # If -s is specified then the limit is strict, denying browsing
1735 # # from any further IP addresses until the ttl has expired. Without
1736 # # -s Squid will just annoy the user by "randomly" denying requests.
1737 # # (the counter is then reset each time the limit is reached and a
1738 # # request is denied)
1739 # # NOTE: in acceleration mode or where there is mesh of child proxies,
1740 # # clients may appear to come from multiple addresses if they are
1741 # # going through proxy farms, so a limit of 1 may cause user problems.
1742 #
1743 # acl aclname req_mime_type mime-type1 ...
1744 # # regex match agains the mime type of the request generated
1745 # # by the client. Can be used to detect file upload or some
1746 # # types HTTP tunelling requests.
1747 # # NOTE: This does NOT match the reply. You cannot use this
1748 # # to match the returned file type.
1749 #
1750 # acl aclname rep_mime_type mime-type1 ...
1751 # # regex match against the mime type of the reply recieved by
1752 # # squid. Can be used to detect file download or some
1753 # # types HTTP tunelling requests.
1754 # # NOTE: This has no effect in http_access rules. It only has
1755 # # effect in rules that affect the reply data stream such as
1756 # # http_reply_access.
1757 #
1758 # acl acl_name external class_name [arguments...]
1759 # # external ACL lookup via a helper class defined by the
1760 # # external_acl_type directive.
1761 #
1762 #Examples:
1763 #acl myexample dst_as 1241
1764 #acl password proxy_auth REQUIRED
1765 #acl fileupload req_mime_type -i ^multipart/form-data$
1766 #acl javascript rep_mime_type -i ^application/x-javascript$
1767 #
1768 #Recommended minimum configuration:
1769 acl all src 0.0.0.0/0.0.0.0
1770 acl manager proto cache_object
1771 acl localhost src 127.0.0.1/255.255.255.255
1772 acl to_localhost dst 127.0.0.0/8
1773 acl SSL_ports port 443 563
1774 acl Jabber_ports port 5222
1775 acl Safe_ports port 80 # http
1776 acl Safe_ports port 21 # ftp
1777 acl Safe_ports port 443 563 # https, snews
1778 acl Safe_ports port 70 # gopher
1779 acl Safe_ports port 210 # wais
1780 acl Safe_ports port 1025-65535 # unregistered ports
1781 acl Safe_ports port 280 # http-mgmt
1782 acl Safe_ports port 488 # gss-http
1783 acl Safe_ports port 591 # filemaker
1784 acl Safe_ports port 777 # multiling http
1785 acl CONNECT method CONNECT
1787 # TAG: http_access
1788 # Allowing or Denying access based on defined access lists
1789 #
1790 # Access to the HTTP port:
1791 # http_access allow|deny [!]aclname ...
1792 #
1793 # NOTE on default values:
1794 #
1795 # If there are no "access" lines present, the default is to deny
1796 # the request.
1797 #
1798 # If none of the "access" lines cause a match, the default is the
1799 # opposite of the last line in the list. If the last line was
1800 # deny, then the default is allow. Conversely, if the last line
1801 # is allow, the default will be deny. For these reasons, it is a
1802 # good idea to have an "deny all" or "allow all" entry at the end
1803 # of your access lists to avoid potential confusion.
1804 #
1805 #Default:
1806 # http_access deny all
1807 #
1808 #Recommended minimum configuration:
1809 #
1810 # Only allow cachemgr access from localhost
1811 http_access allow manager localhost
1812 http_access deny manager
1813 # Deny requests to unknown ports
1814 http_access deny !Safe_ports
1815 # Deny CONNECT to other than SSL ports
1816 http_access deny CONNECT !SSL_ports !Jabber_ports
1817 #
1818 # We strongly recommend to uncomment the following to protect innocent
1819 # web applications running on the proxy server who think that the only
1820 # one who can access services on "localhost" is a local user
1821 #http_access deny to_localhost
1822 #
1823 # INSERT YOUR OWN RULE(S) HERE TO ALLOW ACCESS FROM YOUR CLIENTS
1825 # Example rule allowing access from your local networks. Adapt
1826 # to list your (internal) IP networks from where browsing should
1827 # be allowed
1828 acl password proxy_auth REQUIRED
1830 http_access allow password
1833 # And finally deny all other access to this proxy
1834 http_access allow localhost
1835 http_access deny all
1837 # TAG: http_reply_access
1838 # Allow replies to client requests. This is complementary to http_access.
1839 #
1840 # http_reply_access allow|deny [!] aclname ...
1841 #
1842 # NOTE: if there are no access lines present, the default is to allow
1843 # all replies
1844 #
1845 # If none of the access lines cause a match, then the opposite of the
1846 # last line will apply. Thus it is good practice to end the rules
1847 # with an "allow all" or "deny all" entry.
1848 #
1849 #Default:
1850 # http_reply_access allow all
1851 #
1852 #Recommended minimum configuration:
1853 #
1854 # Insert your own rules here.
1855 #
1856 #
1857 # and finally allow by default
1858 http_reply_access allow all
1860 # TAG: icp_access
1861 # Allowing or Denying access to the ICP port based on defined
1862 # access lists
1863 #
1864 # icp_access allow|deny [!]aclname ...
1865 #
1866 # See http_access for details
1867 #
1868 #Default:
1869 # icp_access deny all
1870 #
1871 #Allow ICP queries from everyone
1872 #icp_access allow all
1874 # TAG: miss_access
1875 # Use to force your neighbors to use you as a sibling instead of
1876 # a parent. For example:
1877 #
1878 # acl localclients src 172.16.0.0/16
1879 # miss_access allow localclients
1880 # miss_access deny !localclients
1881 #
1882 # This means that only your local clients are allowed to fetch
1883 # MISSES and all other clients can only fetch HITS.
1884 #
1885 # By default, allow all clients who passed the http_access rules
1886 # to fetch MISSES from us.
1887 #
1888 #Default setting:
1889 # miss_access allow all
1891 # TAG: cache_peer_access
1892 # Similar to 'cache_peer_domain' but provides more flexibility by
1893 # using ACL elements.
1894 #
1895 # cache_peer_access cache-host allow|deny [!]aclname ...
1896 #
1897 # The syntax is identical to 'http_access' and the other lists of
1898 # ACL elements. See the comments for 'http_access' below, or
1899 # the Squid FAQ (http://www.squid-cache.org/FAQ/FAQ-10.html).
1900 #
1901 #Default:
1902 # none
1904 # TAG: ident_lookup_access
1905 # A list of ACL elements which, if matched, cause an ident
1906 # (RFC 931) lookup to be performed for this request. For
1907 # example, you might choose to always perform ident lookups
1908 # for your main multi-user Unix boxes, but not for your Macs
1909 # and PCs. By default, ident lookups are not performed for
1910 # any requests.
1911 #
1912 # To enable ident lookups for specific client addresses, you
1913 # can follow this example:
1914 #
1915 # acl ident_aware_hosts src 198.168.1.0/255.255.255.0
1916 # ident_lookup_access allow ident_aware_hosts
1917 # ident_lookup_access deny all
1918 #
1919 # Only src type ACL checks are fully supported. A src_domain
1920 # ACL might work at times, but it will not always provide
1921 # the correct result.
1922 #
1923 #Default:
1924 # ident_lookup_access deny all
1926 # TAG: tcp_outgoing_tos
1927 # Allows you to select a TOS/Diffserv value to mark outgoing
1928 # connections with, based on the username or source address
1929 # making the request.
1930 #
1931 # tcp_outgoing_tos ds-field [!]aclname ...
1932 #
1933 # Example where normal_service_net uses the TOS value 0x00
1934 # and normal_service_net uses 0x20
1935 #
1936 # acl normal_service_net src 10.0.0.0/255.255.255.0
1937 # acl good_service_net src 10.0.1.0/255.255.255.0
1938 # tcp_outgoing_tos 0x00 normal_service_net 0x00
1939 # tcp_outgoing_tos 0x20 good_service_net
1940 #
1941 # TOS/DSCP values really only have local significance - so you should
1942 # know what you're specifying. For more, see RFC 2474
1943 #
1944 # The TOS/DSCP byte must be exactly that - a byte, value 0 - 255, or
1945 # "default" to use whatever default your host has.
1946 #
1947 # Processing proceeds in the order specified, and stops at first fully
1948 # matching line.
1949 #
1950 #Default:
1951 # none
1953 # TAG: tcp_outgoing_address
1954 # Allows you to map requests to different outgoing IP addresses
1955 # based on the username or sourceaddress of the user making
1956 # the request.
1957 #
1958 # tcp_outgoing_address ipaddr [[!]aclname] ...
1959 #
1960 # Example where requests from 10.0.0.0/24 will be forwareded
1961 # with source address 10.1.0.1, 10.0.2.0/24 forwarded with
1962 # source address 10.1.0.2 and the rest will be forwarded with
1963 # source address 10.1.0.3.
1964 #
1965 # acl normal_service_net src 10.0.0.0/255.255.255.0
1966 # acl good_service_net src 10.0.1.0/255.255.255.0
1967 # tcp_outgoing_address 10.0.0.1 normal_service_net
1968 # tcp_outgoing_address 10.0.0.2 good_service_net
1969 # tcp_outgoing_address 10.0.0.3
1970 #
1971 # Processing proceeds in the order specified, and stops at first fully
1972 # matching line.
1973 #
1974 #Default:
1975 # none
1977 # TAG: reply_body_max_size bytes allow|deny acl acl...
1978 # This option specifies the maximum size of a reply body in bytes.
1979 # It can be used to prevent users from downloading very large files,
1980 # such as MP3's and movies. When the reply headers are recieved,
1981 # the reply_body_max_size lines are processed, and the first line with
1982 # a result of "allow" is used as the maximum body size for this reply.
1983 # This size is then checked twice. First when we get the reply headers,
1984 # we check the content-length value. If the content length value exists
1985 # and is larger than the allowed size, the request is denied and the
1986 # user receives an error message that says "the request or reply
1987 # is too large." If there is no content-length, and the reply
1988 # size exceeds this limit, the client's connection is just closed
1989 # and they will receive a partial reply.
1990 #
1991 # WARNING: downstream caches probably can not detect a partial reply
1992 # if there is no content-length header, so they will cache
1993 # partial responses and give them out as hits. You should NOT
1994 # use this option if you have downstream caches.
1995 #
1996 # If you set this parameter to zero (the default), there will be
1997 # no limit imposed.
1998 #
1999 #Default:
2000 # reply_body_max_size 0 allow all
2003 # ADMINISTRATIVE PARAMETERS
2004 # -----------------------------------------------------------------------------
2006 # TAG: cache_mgr
2007 # Email-address of local cache manager who will receive
2008 # mail if the cache dies. The default is "webmaster."
2009 #
2010 #Default:
2011 # cache_mgr webmaster
2013 # TAG: cache_effective_user
2014 # TAG: cache_effective_group
2015 #
2016 # If you start Squid as root, it will change its effective/real
2017 # UID/GID to the UID/GID specified below. The default is to
2018 # change to UID to nobody. If you define cache_effective_user,
2019 # but not cache_effective_group, Squid sets the GID the
2020 # effective user's default group ID (taken from the password
2021 # file).
2022 #
2023 # If Squid is not started as root, the cache_effective_user
2024 # value is ignored and the GID value is unchanged by default.
2025 # However, you can make Squid change its GID to another group
2026 # that the process owner is a member of. Note that if Squid
2027 # is not started as root then you cannot set http_port to a
2028 # value lower than 1024.
2029 #
2030 #Default:
2031 # cache_effective_user squid
2032 # cache_effective_group squid
2034 # TAG: visible_hostname
2035 # If you want to present a special hostname in error messages, etc,
2036 # then define this. Otherwise, the return value of gethostname()
2037 # will be used. If you have multiple caches in a cluster and
2038 # get errors about IP-forwarding you must set them to have individual
2039 # names with this setting.
2040 #
2041 #Default:
2042 # none
2044 # TAG: unique_hostname
2045 # If you want to have multiple machines with the same
2046 # 'visible_hostname' then you must give each machine a different
2047 # 'unique_hostname' so that forwarding loops can be detected.
2048 #
2049 #Default:
2050 # none
2052 # TAG: hostname_aliases
2053 # A list of other DNS names that your cache has.
2054 #
2055 #Default:
2056 # none
2059 # OPTIONS FOR THE CACHE REGISTRATION SERVICE
2060 # -----------------------------------------------------------------------------
2061 #
2062 # This section contains parameters for the (optional) cache
2063 # announcement service. This service is provided to help
2064 # cache administrators locate one another in order to join or
2065 # create cache hierarchies.
2066 #
2067 # An 'announcement' message is sent (via UDP) to the registration
2068 # service by Squid. By default, the announcement message is NOT
2069 # SENT unless you enable it with 'announce_period' below.
2070 #
2071 # The announcement message includes your hostname, plus the
2072 # following information from this configuration file:
2073 #
2074 # http_port
2075 # icp_port
2076 # cache_mgr
2077 #
2078 # All current information is processed regularly and made
2079 # available on the Web at http://www.ircache.net/Cache/Tracker/.
2081 # TAG: announce_period
2082 # This is how frequently to send cache announcements. The
2083 # default is `0' which disables sending the announcement
2084 # messages.
2085 #
2086 # To enable announcing your cache, just uncomment the line
2087 # below.
2088 #
2089 #Default:
2090 # announce_period 0
2091 #
2092 #To enable announcing your cache, just uncomment the line below.
2093 #announce_period 1 day
2095 # TAG: announce_host
2096 # TAG: announce_file
2097 # TAG: announce_port
2098 # announce_host and announce_port set the hostname and port
2099 # number where the registration message will be sent.
2100 #
2101 # Hostname will default to 'tracker.ircache.net' and port will
2102 # default default to 3131. If the 'filename' argument is given,
2103 # the contents of that file will be included in the announce
2104 # message.
2105 #
2106 #Default:
2107 # announce_host tracker.ircache.net
2108 # announce_port 3131
2111 # HTTPD-ACCELERATOR OPTIONS
2112 # -----------------------------------------------------------------------------
2114 # TAG: httpd_accel_host
2115 # TAG: httpd_accel_port
2116 # If you want to run Squid as an httpd accelerator, define the
2117 # host name and port number where the real HTTP server is.
2118 #
2119 # If you want IP based virtual host support then specify the
2120 # hostname as "virtual". This will make Squid use the IP address
2121 # where it accepted the request as hostname in the URL.
2122 #
2123 # If you want virtual port support then specify the port as "0".
2124 #
2125 # NOTE: enabling httpd_accel_host disables proxy-caching and
2126 # ICP. If you want these features enabled also, then set
2127 # the 'httpd_accel_with_proxy' option.
2128 #
2129 #Default:
2130 # httpd_accel_port 80
2132 # TAG: httpd_accel_single_host on|off
2133 # If you are running Squid as an accelerator and have a single backend
2134 # server then set this to on. This causes Squid to forward the request
2135 # to this server irregardles of what any redirectors or Host headers
2136 # says.
2137 #
2138 # Leave this at off if you have multiple backend servers, and use a
2139 # redirector (or host table or private DNS) to map the requests to the
2140 # appropriate backend servers. Note that the mapping needs to be a
2141 # 1-1 mapping between requested and backend (from redirector) domain
2142 # names or caching will fail, as cacing is performed using the
2143 # URL returned from the redirector.
2144 #
2145 # See also redirect_rewrites_host_header.
2146 #
2147 #Default:
2148 # httpd_accel_single_host off
2150 # TAG: httpd_accel_with_proxy on|off
2151 # If you want to use Squid as both a local httpd accelerator
2152 # and as a proxy, change this to 'on'. Note however that your
2153 # proxy users may have trouble to reach the accelerated domains
2154 # unless their browsers are configured not to use this proxy for
2155 # those domains (for example via the no_proxy browser configuration
2156 # setting)
2157 #
2158 #Default:
2159 # httpd_accel_with_proxy off
2161 # TAG: httpd_accel_uses_host_header on|off
2162 # HTTP/1.1 requests include a Host: header which is basically the
2163 # hostname from the URL. The Host: header is used for domain based
2164 # virutal hosts. If your accelerator needs to provide domain based
2165 # virtual hosts on the same IP address then you will need to turn this
2166 # on.
2167 #
2168 # Note that Squid does NOT check the value of the Host header matches
2169 # any of your accelerated server, so it may open a big security hole
2170 # unless you take care to set up access controls proper. We recommend
2171 # that this option remain disabled unless you are sure of what you
2172 # are doing.
2173 #
2174 # However, you will need to enable this option if you run Squid
2175 # as a transparent proxy. Otherwise, virtual servers which
2176 # require the Host: header will not be properly cached.
2177 #
2178 #Default:
2179 # httpd_accel_uses_host_header off
2182 # MISCELLANEOUS
2183 # -----------------------------------------------------------------------------
2185 # TAG: dns_testnames
2186 # The DNS tests exit as soon as the first site is successfully looked up
2187 #
2188 # This test can be disabled with the -D command line option.
2189 #
2190 #Default:
2191 # dns_testnames netscape.com internic.net nlanr.net microsoft.com
2193 # TAG: logfile_rotate
2194 # Specifies the number of logfile rotations to make when you
2195 # type 'squid -k rotate'. The default is 10, which will rotate
2196 # with extensions 0 through 9. Setting logfile_rotate to 0 will
2197 # disable the rotation, but the logfiles are still closed and
2198 # re-opened. This will enable you to rename the logfiles
2199 # yourself just before sending the rotate signal.
2200 #
2201 # Note, the 'squid -k rotate' command normally sends a USR1
2202 # signal to the running squid process. In certain situations
2203 # (e.g. on Linux with Async I/O), USR1 is used for other
2204 # purposes, so -k rotate uses another signal. It is best to get
2205 # in the habit of using 'squid -k rotate' instead of 'kill -USR1
2206 # <pid>'.
2207 #
2208 #Default:
2209 # logfile_rotate 0
2211 # TAG: append_domain
2212 # Appends local domain name to hostnames without any dots in
2213 # them. append_domain must begin with a period.
2214 #
2215 # Be warned that there today is Internet names with no dots in
2216 # them using only top-domain names, so setting this may
2217 # cause some Internet sites to become unavailable.
2218 #
2219 #Example:
2220 # append_domain .yourdomain.com
2221 #
2222 #Default:
2223 # none
2225 # TAG: tcp_recv_bufsize (bytes)
2226 # Size of receive buffer to set for TCP sockets. Probably just
2227 # as easy to change your kernel's default. Set to zero to use
2228 # the default buffer size.
2229 #
2230 #Default:
2231 # tcp_recv_bufsize 0 bytes
2233 # TAG: err_html_text
2234 # HTML text to include in error messages. Make this a "mailto"
2235 # URL to your admin address, or maybe just a link to your
2236 # organizations Web page.
2237 #
2238 # To include this in your error messages, you must rewrite
2239 # the error template files (found in the "errors" directory).
2240 # Wherever you want the 'err_html_text' line to appear,
2241 # insert a %L tag in the error template file.
2242 #
2243 #Default:
2244 # none
2246 # TAG: deny_info
2247 # Usage: deny_info err_page_name acl
2248 # or deny_info http://... acl
2249 # Example: deny_info ERR_CUSTOM_ACCESS_DENIED bad_guys
2250 #
2251 # This can be used to return a ERR_ page for requests which
2252 # do not pass the 'http_access' rules. A single ACL will cause
2253 # the http_access check to fail. If a 'deny_info' line exists
2254 # for that ACL then Squid returns a corresponding error page.
2255 #
2256 # You may use ERR_ pages that come with Squid or create your own pages
2257 # and put them into the configured errors/ directory.
2258 #
2259 # Alternatively you can specify an error URL. The browsers will then
2260 # get redirected (302) to the specified URL. %s in the redirection
2261 # URL will be replaced by the requested URL.
2262 #
2263 # Alternatively you can tell Squid to reset the TCP connection
2264 # by specifying TCP_RESET.
2265 #
2266 #Default:
2267 # none
2269 # TAG: memory_pools on|off
2270 # If set, Squid will keep pools of allocated (but unused) memory
2271 # available for future use. If memory is a premium on your
2272 # system and you believe your malloc library outperforms Squid
2273 # routines, disable this.
2274 #
2275 #Default:
2276 # memory_pools on
2278 # TAG: memory_pools_limit (bytes)
2279 # Used only with memory_pools on:
2280 # memory_pools_limit 50 MB
2281 #
2282 # If set to a non-zero value, Squid will keep at most the specified
2283 # limit of allocated (but unused) memory in memory pools. All free()
2284 # requests that exceed this limit will be handled by your malloc
2285 # library. Squid does not pre-allocate any memory, just safe-keeps
2286 # objects that otherwise would be free()d. Thus, it is safe to set
2287 # memory_pools_limit to a reasonably high value even if your
2288 # configuration will use less memory.
2289 #
2290 # If not set (default) or set to zero, Squid will keep all memory it
2291 # can. That is, there will be no limit on the total amount of memory
2292 # used for safe-keeping.
2293 #
2294 # To disable memory allocation optimization, do not set
2295 # memory_pools_limit to 0. Set memory_pools to "off" instead.
2296 #
2297 # An overhead for maintaining memory pools is not taken into account
2298 # when the limit is checked. This overhead is close to four bytes per
2299 # object kept. However, pools may actually _save_ memory because of
2300 # reduced memory thrashing in your malloc library.
2301 #
2302 #Default:
2303 # none
2305 # TAG: forwarded_for on|off
2306 # If set, Squid will include your system's IP address or name
2307 # in the HTTP requests it forwards. By default it looks like
2308 # this:
2309 #
2310 # X-Forwarded-For: 192.1.2.3
2311 #
2312 # If you disable this, it will appear as
2313 #
2314 # X-Forwarded-For: unknown
2315 #
2316 #Default:
2317 # forwarded_for on
2319 # TAG: log_icp_queries on|off
2320 # If set, ICP queries are logged to access.log. You may wish
2321 # do disable this if your ICP load is VERY high to speed things
2322 # up or to simplify log analysis.
2323 #
2324 #Default:
2325 # log_icp_queries on
2327 # TAG: icp_hit_stale on|off
2328 # If you want to return ICP_HIT for stale cache objects, set this
2329 # option to 'on'. If you have sibling relationships with caches
2330 # in other administrative domains, this should be 'off'. If you only
2331 # have sibling relationships with caches under your control, then
2332 # it is probably okay to set this to 'on'.
2333 # If set to 'on', then your siblings should use the option "allow-miss"
2334 # on their cache_peer lines for connecting to you.
2335 #
2336 #Default:
2337 # icp_hit_stale off
2339 # TAG: minimum_direct_hops
2340 # If using the ICMP pinging stuff, do direct fetches for sites
2341 # which are no more than this many hops away.
2342 #
2343 #Default:
2344 # minimum_direct_hops 4
2346 # TAG: minimum_direct_rtt
2347 # If using the ICMP pinging stuff, do direct fetches for sites
2348 # which are no more than this many rtt milliseconds away.
2349 #
2350 #Default:
2351 # minimum_direct_rtt 400
2353 # TAG: cachemgr_passwd
2354 # Specify passwords for cachemgr operations.
2355 #
2356 # Usage: cachemgr_passwd password action action ...
2357 #
2358 # Some valid actions are (see cache manager menu for a full list):
2359 # 5min
2360 # 60min
2361 # asndb
2362 # authenticator
2363 # cbdata
2364 # client_list
2365 # comm_incoming
2366 # config *
2367 # counters
2368 # delay
2369 # digest_stats
2370 # dns
2371 # events
2372 # filedescriptors
2373 # fqdncache
2374 # histograms
2375 # http_headers
2376 # info
2377 # io
2378 # ipcache
2379 # mem
2380 # menu
2381 # netdb
2382 # non_peers
2383 # objects
2384 # offline_toggle *
2385 # pconn
2386 # peer_select
2387 # redirector
2388 # refresh
2389 # server_list
2390 # shutdown *
2391 # store_digest
2392 # storedir
2393 # utilization
2394 # via_headers
2395 # vm_objects
2396 #
2397 # * Indicates actions which will not be performed without a
2398 # valid password, others can be performed if not listed here.
2399 #
2400 # To disable an action, set the password to "disable".
2401 # To allow performing an action without a password, set the
2402 # password to "none".
2403 #
2404 # Use the keyword "all" to set the same password for all actions.
2405 #
2406 #Example:
2407 # cachemgr_passwd secret shutdown
2408 # cachemgr_passwd lesssssssecret info stats/objects
2409 # cachemgr_passwd disable all
2410 #
2411 #Default:
2412 # none
2414 # TAG: store_avg_object_size (kbytes)
2415 # Average object size, used to estimate number of objects your
2416 # cache can hold. See doc/Release-Notes-1.1.txt. The default is
2417 # 13 KB.
2418 #
2419 #Default:
2420 # store_avg_object_size 13 KB
2422 # TAG: store_objects_per_bucket
2423 # Target number of objects per bucket in the store hash table.
2424 # Lowering this value increases the total number of buckets and
2425 # also the storage maintenance rate. The default is 50.
2426 #
2427 #Default:
2428 # store_objects_per_bucket 20
2430 # TAG: client_db on|off
2431 # If you want to disable collecting per-client statistics, then
2432 # turn off client_db here.
2433 #
2434 #Default:
2435 # client_db on
2437 # TAG: netdb_low
2438 # TAG: netdb_high
2439 # The low and high water marks for the ICMP measurement
2440 # database. These are counts, not percents. The defaults are
2441 # 900 and 1000. When the high water mark is reached, database
2442 # entries will be deleted until the low mark is reached.
2443 #
2444 #Default:
2445 # netdb_low 900
2446 # netdb_high 1000
2448 # TAG: netdb_ping_period
2449 # The minimum period for measuring a site. There will be at
2450 # least this much delay between successive pings to the same
2451 # network. The default is five minutes.
2452 #
2453 #Default:
2454 # netdb_ping_period 5 minutes
2456 # TAG: query_icmp on|off
2457 # If you want to ask your peers to include ICMP data in their ICP
2458 # replies, enable this option.
2459 #
2460 # If your peer has configured Squid (during compilation) with
2461 # '--enable-icmp' then that peer will send ICMP pings to origin server
2462 # sites of the URLs it receives. If you enable this option then the
2463 # ICP replies from that peer will include the ICMP data (if available).
2464 # Then, when choosing a parent cache, Squid will choose the parent with
2465 # the minimal RTT to the origin server. When this happens, the
2466 # hierarchy field of the access.log will be
2467 # "CLOSEST_PARENT_MISS". This option is off by default.
2468 #
2469 #Default:
2470 # query_icmp off
2472 # TAG: test_reachability on|off
2473 # When this is 'on', ICP MISS replies will be ICP_MISS_NOFETCH
2474 # instead of ICP_MISS if the target host is NOT in the ICMP
2475 # database, or has a zero RTT.
2476 #
2477 #Default:
2478 # test_reachability off
2480 # TAG: buffered_logs on|off
2481 # cache.log log file is written with stdio functions, and as such
2482 # it can be buffered or unbuffered. By default it will be unbuffered.
2483 # Buffering it can speed up the writing slightly (though you are
2484 # unlikely to need to worry unless you run with tons of debugging
2485 # enabled in which case performance will suffer badly anyway..).
2486 #
2487 #Default:
2488 # buffered_logs off
2490 # TAG: reload_into_ims on|off
2491 # When you enable this option, client no-cache or ``reload''
2492 # requests will be changed to If-Modified-Since requests.
2493 # Doing this VIOLATES the HTTP standard. Enabling this
2494 # feature could make you liable for problems which it
2495 # causes.
2496 #
2497 # see also refresh_pattern for a more selective approach.
2498 #
2499 #Default:
2500 # reload_into_ims off
2502 # TAG: always_direct
2503 # Usage: always_direct allow|deny [!]aclname ...
2504 #
2505 # Here you can use ACL elements to specify requests which should
2506 # ALWAYS be forwarded directly to origin servers. For example,
2507 # to always directly forward requests for local servers use
2508 # something like:
2509 #
2510 # acl local-servers dstdomain my.domain.net
2511 # always_direct allow local-servers
2512 #
2513 # To always forward FTP requests directly, use
2514 #
2515 # acl FTP proto FTP
2516 # always_direct allow FTP
2517 #
2518 # NOTE: There is a similar, but opposite option named
2519 # 'never_direct'. You need to be aware that "always_direct deny
2520 # foo" is NOT the same thing as "never_direct allow foo". You
2521 # may need to use a deny rule to exclude a more-specific case of
2522 # some other rule. Example:
2523 #
2524 # acl local-external dstdomain external.foo.net
2525 # acl local-servers dstdomain .foo.net
2526 # always_direct deny local-external
2527 # always_direct allow local-servers
2528 #
2529 # This option replaces some v1.1 options such as local_domain
2530 # and local_ip.
2531 #
2532 #Default:
2533 # none
2535 # TAG: never_direct
2536 # Usage: never_direct allow|deny [!]aclname ...
2537 #
2538 # never_direct is the opposite of always_direct. Please read
2539 # the description for always_direct if you have not already.
2540 #
2541 # With 'never_direct' you can use ACL elements to specify
2542 # requests which should NEVER be forwarded directly to origin
2543 # servers. For example, to force the use of a proxy for all
2544 # requests, except those in your local domain use something like:
2545 #
2546 # acl local-servers dstdomain .foo.net
2547 # acl all src 0.0.0.0/0.0.0.0
2548 # never_direct deny local-servers
2549 # never_direct allow all
2550 #
2551 # or if squid is inside a firewall and there is local intranet
2552 # servers inside the firewall then use something like:
2553 #
2554 # acl local-intranet dstdomain .foo.net
2555 # acl local-external dstdomain external.foo.net
2556 # always_direct deny local-external
2557 # always_direct allow local-intranet
2558 # never_direct allow all
2559 #
2560 # This option replaces some v1.1 options such as inside_firewall
2561 # and firewall_ip.
2562 #
2563 #Default:
2564 # none
2566 # TAG: header_access
2567 # Usage: header_access header_name allow|deny [!]aclname ...
2568 #
2569 # WARNING: Doing this VIOLATES the HTTP standard. Enabling
2570 # this feature could make you liable for problems which it
2571 # causes.
2572 #
2573 # This option replaces the old 'anonymize_headers' and the
2574 # older 'http_anonymizer' option with something that is much
2575 # more configurable. This new method creates a list of ACLs
2576 # for each header, allowing you very fine-tuned header
2577 # mangling.
2578 #
2579 # You can only specify known headers for the header name.
2580 # Other headers are reclassified as 'Other'. You can also
2581 # refer to all the headers with 'All'.
2582 #
2583 # For example, to achieve the same behaviour as the old
2584 # 'http_anonymizer standard' option, you should use:
2585 #
2586 # header_access From deny all
2587 # header_access Referer deny all
2588 # header_access Server deny all
2589 # header_access User-Agent deny all
2590 # header_access WWW-Authenticate deny all
2591 # header_access Link deny all
2592 #
2593 # Or, to reproduce the old 'http_anonymizer paranoid' feature
2594 # you should use:
2595 #
2596 # header_access Allow allow all
2597 # header_access Authorization allow all
2598 # header_access WWW-Authenticate allow all
2599 # header_access Cache-Control allow all
2600 # header_access Content-Encoding allow all
2601 # header_access Content-Length allow all
2602 # header_access Content-Type allow all
2603 # header_access Date allow all
2604 # header_access Expires allow all
2605 # header_access Host allow all
2606 # header_access If-Modified-Since allow all
2607 # header_access Last-Modified allow all
2608 # header_access Location allow all
2609 # header_access Pragma allow all
2610 # header_access Accept allow all
2611 # header_access Accept-Charset allow all
2612 # header_access Accept-Encoding allow all
2613 # header_access Accept-Language allow all
2614 # header_access Content-Language allow all
2615 # header_access Mime-Version allow all
2616 # header_access Retry-After allow all
2617 # header_access Title allow all
2618 # header_access Connection allow all
2619 # header_access Proxy-Connection allow all
2620 # header_access All deny all
2621 #
2622 # By default, all headers are allowed (no anonymizing is
2623 # performed).
2624 #
2625 #Default:
2626 # none
2628 # TAG: header_replace
2629 # Usage: header_replace header_name message
2630 # Example: header_replace User-Agent Nutscrape/1.0 (CP/M; 8-bit)
2631 #
2632 # This option allows you to change the contents of headers
2633 # denied with header_access above, by replacing them with
2634 # some fixed string. This replaces the old fake_user_agent
2635 # option.
2636 #
2637 # By default, headers are removed if denied.
2638 #
2639 #Default:
2640 # none
2642 # TAG: icon_directory
2643 # Where the icons are stored. These are normally kept in
2644 # /usr/share/squid/icons
2645 #
2646 #Default:
2647 # icon_directory /usr/share/squid/icons
2649 # TAG: short_icon_urls
2650 # If this is enabled then Squid will use short URLs for icons.
2651 #
2652 # If off then the URLs for icons will always be absolute URLs
2653 # including the proxy name and port.
2654 #
2655 #Default:
2656 # short_icon_urls off
2658 # TAG: error_directory
2659 # If you wish to create your own versions of the default
2660 # (English) error files, either to customize them to suit your
2661 # language or company copy the template English files to another
2662 # directory and point this tag at them.
2663 #
2664 #Default:
2665 # error_directory /usr/share/squid/errors/English
2667 # TAG: maximum_single_addr_tries
2668 # This sets the maximum number of connection attempts for a
2669 # host that only has one address (for multiple-address hosts,
2670 # each address is tried once).
2671 #
2672 # The default value is one attempt, the (not recommended)
2673 # maximum is 255 tries. A warning message will be generated
2674 # if it is set to a value greater than ten.
2675 #
2676 # Note: This is in addition to the request reforwarding which
2677 # takes place if Squid fails to get a satisfying response.
2678 #
2679 #Default:
2680 # maximum_single_addr_tries 1
2682 # TAG: snmp_port
2683 # Squid can now serve statistics and status information via SNMP.
2684 # By default it listens to port 3401 on the machine. If you don't
2685 # wish to use SNMP, set this to "0".
2686 #
2687 #Default:
2688 # snmp_port 3401
2690 # TAG: snmp_access
2691 # Allowing or denying access to the SNMP port.
2692 #
2693 # All access to the agent is denied by default.
2694 # usage:
2695 #
2696 # snmp_access allow|deny [!]aclname ...
2697 #
2698 #Example:
2699 # snmp_access allow snmppublic localhost
2700 # snmp_access deny all
2701 #
2702 #Default:
2703 # snmp_access deny all
2705 # TAG: snmp_incoming_address
2706 # TAG: snmp_outgoing_address
2707 # Just like 'udp_incoming_address' above, but for the SNMP port.
2708 #
2709 # snmp_incoming_address is used for the SNMP socket receiving
2710 # messages from SNMP agents.
2711 # snmp_outgoing_address is used for SNMP packets returned to SNMP
2712 # agents.
2713 #
2714 # The default snmp_incoming_address (0.0.0.0) is to listen on all
2715 # available network interfaces.
2716 #
2717 # If snmp_outgoing_address is set to 255.255.255.255 (the default)
2718 # then it will use the same socket as snmp_incoming_address. Only
2719 # change this if you want to have SNMP replies sent using another
2720 # address than where this Squid listens for SNMP queries.
2721 #
2722 # NOTE, snmp_incoming_address and snmp_outgoing_address can not have
2723 # the same value since they both use port 3401.
2724 #
2725 #Default:
2726 # snmp_incoming_address 0.0.0.0
2727 # snmp_outgoing_address 255.255.255.255
2729 # TAG: as_whois_server
2730 # WHOIS server to query for AS numbers. NOTE: AS numbers are
2731 # queried only when Squid starts up, not for every request.
2732 #
2733 #Default:
2734 # as_whois_server whois.ra.net
2735 # as_whois_server whois.ra.net
2737 # TAG: wccp_router
2738 # Use this option to define your WCCP ``home'' router for
2739 # Squid. Setting the 'wccp_router' to 0.0.0.0 (the default)
2740 # disables WCCP.
2741 #
2742 #Default:
2743 # wccp_router 0.0.0.0
2745 # TAG: wccp_version
2746 # According to some users, Cisco IOS 11.2 only supports WCCP
2747 # version 3. If you're using that version of IOS, change
2748 # this value to 3.
2749 #
2750 #Default:
2751 # wccp_version 4
2753 # TAG: wccp_incoming_address
2754 # TAG: wccp_outgoing_address
2755 # wccp_incoming_address Use this option if you require WCCP
2756 # messages to be received on only one
2757 # interface. Do NOT use this option if
2758 # you're unsure how many interfaces you
2759 # have, or if you know you have only one
2760 # interface.
2761 #
2762 # wccp_outgoing_address Use this option if you require WCCP
2763 # messages to be sent out on only one
2764 # interface. Do NOT use this option if
2765 # you're unsure how many interfaces you
2766 # have, or if you know you have only one
2767 # interface.
2768 #
2769 # The default behavior is to not bind to any specific address.
2770 #
2771 # NOTE, wccp_incoming_address and wccp_outgoing_address can not have
2772 # the same value since they both use port 2048.
2773 #
2774 #Default:
2775 # wccp_incoming_address 0.0.0.0
2776 # wccp_outgoing_address 255.255.255.255
2779 # DELAY POOL PARAMETERS (all require DELAY_POOLS compilation option)
2780 # -----------------------------------------------------------------------------
2782 # TAG: delay_pools
2783 # This represents the number of delay pools to be used. For example,
2784 # if you have one class 2 delay pool and one class 3 delays pool, you
2785 # have a total of 2 delay pools.
2786 #
2787 #Default:
2788 # delay_pools 0
2790 # TAG: delay_class
2791 # This defines the class of each delay pool. There must be exactly one
2792 # delay_class line for each delay pool. For example, to define two
2793 # delay pools, one of class 2 and one of class 3, the settings above
2794 # and here would be:
2795 #
2796 #Example:
2797 # delay_pools 2 # 2 delay pools
2798 # delay_class 1 2 # pool 1 is a class 2 pool
2799 # delay_class 2 3 # pool 2 is a class 3 pool
2800 #
2801 # The delay pool classes are:
2802 #
2803 # class 1 Everything is limited by a single aggregate
2804 # bucket.
2805 #
2806 # class 2 Everything is limited by a single aggregate
2807 # bucket as well as an "individual" bucket chosen
2808 # from bits 25 through 32 of the IP address.
2809 #
2810 # class 3 Everything is limited by a single aggregate
2811 # bucket as well as a "network" bucket chosen
2812 # from bits 17 through 24 of the IP address and a
2813 # "individual" bucket chosen from bits 17 through
2814 # 32 of the IP address.
2815 #
2816 # NOTE: If an IP address is a.b.c.d
2817 # -> bits 25 through 32 are "d"
2818 # -> bits 17 through 24 are "c"
2819 # -> bits 17 through 32 are "c * 256 + d"
2820 #
2821 #Default:
2822 # none
2824 # TAG: delay_access
2825 # This is used to determine which delay pool a request falls into.
2826 # The first matched delay pool is always used, i.e., if a request falls
2827 # into delay pool number one, no more delay are checked, otherwise the
2828 # rest are checked in order of their delay pool number until they have
2829 # all been checked. For example, if you want some_big_clients in delay
2830 # pool 1 and lotsa_little_clients in delay pool 2:
2831 #
2832 #Example:
2833 # delay_access 1 allow some_big_clients
2834 # delay_access 1 deny all
2835 # delay_access 2 allow lotsa_little_clients
2836 # delay_access 2 deny all
2837 #
2838 #Default:
2839 # none
2841 # TAG: delay_parameters
2842 # This defines the parameters for a delay pool. Each delay pool has
2843 # a number of "buckets" associated with it, as explained in the
2844 # description of delay_class. For a class 1 delay pool, the syntax is:
2845 #
2846 #delay_parameters pool aggregate
2847 #
2848 # For a class 2 delay pool:
2849 #
2850 #delay_parameters pool aggregate individual
2851 #
2852 # For a class 3 delay pool:
2853 #
2854 #delay_parameters pool aggregate network individual
2855 #
2856 # The variables here are:
2857 #
2858 # pool a pool number - ie, a number between 1 and the
2859 # number specified in delay_pools as used in
2860 # delay_class lines.
2861 #
2862 # aggregate the "delay parameters" for the aggregate bucket
2863 # (class 1, 2, 3).
2864 #
2865 # individual the "delay parameters" for the individual
2866 # buckets (class 2, 3).
2867 #
2868 # network the "delay parameters" for the network buckets
2869 # (class 3).
2870 #
2871 # A pair of delay parameters is written restore/maximum, where restore is
2872 # the number of bytes (not bits - modem and network speeds are usually
2873 # quoted in bits) per second placed into the bucket, and maximum is the
2874 # maximum number of bytes which can be in the bucket at any time.
2875 #
2876 # For example, if delay pool number 1 is a class 2 delay pool as in the
2877 # above example, and is being used to strictly limit each host to 64kbps
2878 # (plus overheads), with no overall limit, the line is:
2879 #
2880 #delay_parameters 1 -1/-1 8000/8000
2881 #
2882 # Note that the figure -1 is used to represent "unlimited".
2883 #
2884 # And, if delay pool number 2 is a class 3 delay pool as in the above
2885 # example, and you want to limit it to a total of 256kbps (strict limit)
2886 # with each 8-bit network permitted 64kbps (strict limit) and each
2887 # individual host permitted 4800bps with a bucket maximum size of 64kb
2888 # to permit a decent web page to be downloaded at a decent speed
2889 # (if the network is not being limited due to overuse) but slow down
2890 # large downloads more significantly:
2891 #
2892 #delay_parameters 2 32000/32000 8000/8000 600/8000
2893 #
2894 # There must be one delay_parameters line for each delay pool.
2895 #
2896 #Default:
2897 # none
2899 # TAG: delay_initial_bucket_level (percent, 0-100)
2900 # The initial bucket percentage is used to determine how much is put
2901 # in each bucket when squid starts, is reconfigured, or first notices
2902 # a host accessing it (in class 2 and class 3, individual hosts and
2903 # networks only have buckets associated with them once they have been
2904 # "seen" by squid).
2905 #
2906 #Default:
2907 # delay_initial_bucket_level 50
2909 # TAG: incoming_icp_average
2910 # TAG: incoming_http_average
2911 # TAG: incoming_dns_average
2912 # TAG: min_icp_poll_cnt
2913 # TAG: min_dns_poll_cnt
2914 # TAG: min_http_poll_cnt
2915 # Heavy voodoo here. I can't even believe you are reading this.
2916 # Are you crazy? Don't even think about adjusting these unless
2917 # you understand the algorithms in comm_select.c first!
2918 #
2919 #Default:
2920 # incoming_icp_average 6
2921 # incoming_http_average 4
2922 # incoming_dns_average 4
2923 # min_icp_poll_cnt 8
2924 # min_dns_poll_cnt 8
2925 # min_http_poll_cnt 8
2927 # TAG: max_open_disk_fds
2928 # To avoid having disk as the I/O bottleneck Squid can optionally
2929 # bypass the on-disk cache if more than this amount of disk file
2930 # descriptors are open.
2931 #
2932 # A value of 0 indicates no limit.
2933 #
2934 #Default:
2935 # max_open_disk_fds 0
2937 # TAG: offline_mode
2938 # Enable this option and Squid will never try to validate cached
2939 # objects.
2940 #
2941 #Default:
2942 # offline_mode off
2944 # TAG: uri_whitespace
2945 # What to do with requests that have whitespace characters in the
2946 # URI. Options:
2947 #
2948 # strip: The whitespace characters are stripped out of the URL.
2949 # This is the behavior recommended by RFC2396.
2950 # deny: The request is denied. The user receives an "Invalid
2951 # Request" message.
2952 # allow: The request is allowed and the URI is not changed. The
2953 # whitespace characters remain in the URI. Note the
2954 # whitespace is passed to redirector processes if they
2955 # are in use.
2956 # encode: The request is allowed and the whitespace characters are
2957 # encoded according to RFC1738. This could be considered
2958 # a violation of the HTTP/1.1
2959 # RFC because proxies are not allowed to rewrite URI's.
2960 # chop: The request is allowed and the URI is chopped at the
2961 # first whitespace. This might also be considered a
2962 # violation.
2963 #
2964 #Default:
2965 # uri_whitespace strip
2967 # TAG: broken_posts
2968 # A list of ACL elements which, if matched, causes Squid to send
2969 # an extra CRLF pair after the body of a PUT/POST request.
2970 #
2971 # Some HTTP servers has broken implementations of PUT/POST,
2972 # and rely on an extra CRLF pair sent by some WWW clients.
2973 #
2974 # Quote from RFC 2068 section 4.1 on this matter:
2975 #
2976 # Note: certain buggy HTTP/1.0 client implementations generate an
2977 # extra CRLF's after a POST request. To restate what is explicitly
2978 # forbidden by the BNF, an HTTP/1.1 client must not preface or follow
2979 # a request with an extra CRLF.
2980 #
2981 #Example:
2982 # acl buggy_server url_regex ^http://....
2983 # broken_posts allow buggy_server
2984 #
2985 #Default:
2986 # none
2988 # TAG: mcast_miss_addr
2989 # Note: This option is only available if Squid is rebuilt with the
2990 # -DMULTICAST_MISS_STREAM option
2991 #
2992 # If you enable this option, every "cache miss" URL will
2993 # be sent out on the specified multicast address.
2994 #
2995 # Do not enable this option unless you are are absolutely
2996 # certain you understand what you are doing.
2997 #
2998 #Default:
2999 # mcast_miss_addr 255.255.255.255
3001 # TAG: mcast_miss_ttl
3002 # Note: This option is only available if Squid is rebuilt with the
3003 # -DMULTICAST_MISS_TTL option
3004 #
3005 # This is the time-to-live value for packets multicasted
3006 # when multicasting off cache miss URLs is enabled. By
3007 # default this is set to 'site scope', i.e. 16.
3008 #
3009 #Default:
3010 # mcast_miss_ttl 16
3012 # TAG: mcast_miss_port
3013 # Note: This option is only available if Squid is rebuilt with the
3014 # -DMULTICAST_MISS_STREAM option
3015 #
3016 # This is the port number to be used in conjunction with
3017 # 'mcast_miss_addr'.
3018 #
3019 #Default:
3020 # mcast_miss_port 3135
3022 # TAG: mcast_miss_encode_key
3023 # Note: This option is only available if Squid is rebuilt with the
3024 # -DMULTICAST_MISS_STREAM option
3025 #
3026 # The URLs that are sent in the multicast miss stream are
3027 # encrypted. This is the encryption key.
3028 #
3029 #Default:
3030 # mcast_miss_encode_key XXXXXXXXXXXXXXXX
3032 # TAG: nonhierarchical_direct
3033 # By default, Squid will send any non-hierarchical requests
3034 # (matching hierarchy_stoplist or not cachable request type) direct
3035 # to origin servers.
3036 #
3037 # If you set this to off, then Squid will prefer to send these
3038 # requests to parents.
3039 #
3040 # Note that in most configurations, by turning this off you will only
3041 # add latency to these request without any improvement in global hit
3042 # ratio.
3043 #
3044 # If you are inside an firewall then see never_direct instead of
3045 # this directive.
3046 #
3047 #Default:
3048 # nonhierarchical_direct on
3050 # TAG: prefer_direct
3051 # Normally Squid tries to use parents for most requests. If you by some
3052 # reason like it to first try going direct and only use a parent if
3053 # going direct fails then set this to on.
3054 #
3055 # By combining nonhierarchical_direct off and prefer_direct on you
3056 # can set up Squid to use a parent as a backup path if going direct
3057 # fails.
3058 #
3059 # Note: If you want Squid to use parents for all requests then see
3060 # the never_direct directive. prefer_direct only modifies how Squid
3061 # acts on cachable requests.
3062 #
3063 #Default:
3064 # prefer_direct off
3066 # TAG: strip_query_terms
3067 # By default, Squid strips query terms from requested URLs before
3068 # logging. This protects your user's privacy.
3069 #
3070 #Default:
3071 # strip_query_terms on
3073 # TAG: coredump_dir
3074 # By default Squid leaves core files in the directory from where
3075 # it was started. If you set 'coredump_dir' to a directory
3076 # that exists, Squid will chdir() to that directory at startup
3077 # and coredump files will be left there.
3078 #
3079 #Default:
3080 # coredump_dir none
3081 #
3082 # Leave coredumps in the first cache dir
3083 coredump_dir /var/spool/squid
3085 # TAG: redirector_bypass
3086 # When this is 'on', a request will not go through the
3087 # redirector if all redirectors are busy. If this is 'off'
3088 # and the redirector queue grows too large, Squid will exit
3089 # with a FATAL error and ask you to increase the number of
3090 # redirectors. You should only enable this if the redirectors
3091 # are not critical to your caching system. If you use
3092 # redirectors for access control, and you enable this option,
3093 # then users may have access to pages that they should not
3094 # be allowed to request.
3095 #
3096 #Default:
3097 # redirector_bypass off
3099 # TAG: ignore_unknown_nameservers
3100 # By default Squid checks that DNS responses are received
3101 # from the same IP addresses that they are sent to. If they
3102 # don't match, Squid ignores the response and writes a warning
3103 # message to cache.log. You can allow responses from unknown
3104 # nameservers by setting this option to 'off'.
3105 #
3106 #Default:
3107 # ignore_unknown_nameservers on
3109 # TAG: digest_generation
3110 # This controls whether the server will generate a Cache Digest
3111 # of its contents. By default, Cache Digest generation is
3112 # enabled if Squid is compiled with USE_CACHE_DIGESTS defined.
3113 #
3114 #Default:
3115 # digest_generation on
3117 # TAG: digest_bits_per_entry
3118 # This is the number of bits of the server's Cache Digest which
3119 # will be associated with the Digest entry for a given HTTP
3120 # Method and URL (public key) combination. The default is 5.
3121 #
3122 #Default:
3123 # digest_bits_per_entry 5
3125 # TAG: digest_rebuild_period (seconds)
3126 # This is the number of seconds between Cache Digest rebuilds.
3127 #
3128 #Default:
3129 # digest_rebuild_period 1 hour
3131 # TAG: digest_rewrite_period (seconds)
3132 # This is the number of seconds between Cache Digest writes to
3133 # disk.
3134 #
3135 #Default:
3136 # digest_rewrite_period 1 hour
3138 # TAG: digest_swapout_chunk_size (bytes)
3139 # This is the number of bytes of the Cache Digest to write to
3140 # disk at a time. It defaults to 4096 bytes (4KB), the Squid
3141 # default swap page.
3142 #
3143 #Default:
3144 # digest_swapout_chunk_size 4096 bytes
3146 # TAG: digest_rebuild_chunk_percentage (percent, 0-100)
3147 # This is the percentage of the Cache Digest to be scanned at a
3148 # time. By default it is set to 10% of the Cache Digest.
3149 #
3150 #Default:
3151 # digest_rebuild_chunk_percentage 10
3153 # TAG: chroot
3154 # Use this to have Squid do a chroot() while initializing. This
3155 # also causes Squid to fully drop root privileges after
3156 # initializing. This means, for example, that if you use a HTTP
3157 # port less than 1024 and try to reconfigure, you will get an
3158 # error.
3159 #
3160 #Default:
3161 # none
3163 # TAG: client_persistent_connections
3164 # TAG: server_persistent_connections
3165 # Persistent connection support for clients and servers. By
3166 # default, Squid uses persistent connections (when allowed)
3167 # with its clients and servers. You can use these options to
3168 # disable persistent connections with clients and/or servers.
3169 #
3170 #Default:
3171 # client_persistent_connections on
3172 # server_persistent_connections on
3174 # TAG: detect_broken_pconn
3175 # Some servers have been found to incorrectly signal the use
3176 # of HTTP/1.0 persistent connections even on replies not
3177 # compatible, causing significant delays. This server problem
3178 # has mostly been seen on redirects.
3179 #
3180 # By enabling this directive Squid attempts to detect such
3181 # broken replies and automatically assume the reply is finished
3182 # after 10 seconds timeout.
3183 #
3184 #Default:
3185 # detect_broken_pconn off
3187 # TAG: pipeline_prefetch
3188 # To boost the performance of pipelined requests to closer
3189 # match that of a non-proxied environment Squid can try to fetch
3190 # up to two requests in parallell from a pipeline.
3191 #
3192 # Defaults to off for bandwidth management and access logging
3193 # reasons.
3194 #
3195 #Default:
3196 # pipeline_prefetch off
3198 # TAG: extension_methods
3199 # Squid only knows about standardized HTTP request methods.
3200 # You can add up to 20 additional "extension" methods here.
3201 #
3202 #Default:
3203 # none
3205 # TAG: request_entities
3206 # Squid defaults to deny GET and HEAD requests with request entities,
3207 # as the meaning of such requests are undefined in the HTTP standard
3208 # even if not explicitly forbidden.
3209 #
3210 # Set this directive to on if you have clients which insists
3211 # on sending request entities in GET or HEAD requests.
3212 #
3213 #Default:
3214 # request_entities off
3216 # TAG: high_response_time_warning (msec)
3217 # If the one-minute median response time exceeds this value,
3218 # Squid prints a WARNING with debug level 0 to get the
3219 # administrators attention. The value is in milliseconds.
3220 #
3221 #Default:
3222 # high_response_time_warning 0
3224 # TAG: high_page_fault_warning
3225 # If the one-minute average page fault rate exceeds this
3226 # value, Squid prints a WARNING with debug level 0 to get
3227 # the administrators attention. The value is in page faults
3228 # per second.
3229 #
3230 #Default:
3231 # high_page_fault_warning 0
3233 # TAG: high_memory_warning
3234 # If the memory usage (as determined by mallinfo) exceeds
3235 # value, Squid prints a WARNING with debug level 0 to get
3236 # the administrators attention.
3237 #
3238 #Default:
3239 # high_memory_warning 0
3241 # TAG: store_dir_select_algorithm
3242 # Set this to 'round-robin' as an alternative.
3243 #
3244 #Default:
3245 # store_dir_select_algorithm least-load
3247 # TAG: forward_log
3248 # Note: This option is only available if Squid is rebuilt with the
3249 # -DWIP_FWD_LOG option
3250 #
3251 # Logs the server-side requests.
3252 #
3253 # This is currently work in progress.
3254 #
3255 #Default:
3256 # none
3258 # TAG: ie_refresh on|off
3259 # Microsoft Internet Explorer up until version 5.5 Service
3260 # Pack 1 has an issue with transparent proxies, wherein it
3261 # is impossible to force a refresh. Turning this on provides
3262 # a partial fix to the problem, by causing all IMS-REFRESH
3263 # requests from older IE versions to check the origin server
3264 # for fresh content. This reduces hit ratio by some amount
3265 # (~10% in my experience), but allows users to actually get
3266 # fresh content when they want it. Note that because Squid
3267 # cannot tell if the user is using 5.5 or 5.5SP1, the behavior
3268 # of 5.5 is unchanged from old versions of Squid (i.e. a
3269 # forced refresh is impossible). Newer versions of IE will,
3270 # hopefully, continue to have the new behavior and will be
3271 # handled based on that assumption. This option defaults to
3272 # the old Squid behavior, which is better for hit ratios but
3273 # worse for clients using IE, if they need to be able to
3274 # force fresh content.
3275 #
3276 #Default:
3277 # ie_refresh off
3279 # TAG: vary_ignore_expire on|off
3280 # Many HTTP servers supporting Vary gives such objects
3281 # immediate expiry time with no cache-control header
3282 # when requested by a HTTP/1.0 client. This option
3283 # enables Squid to ignore such expiry times until
3284 # HTTP/1.1 is fully implemented.
3285 # WARNING: This may eventually cause some varying
3286 # objects not intended for caching to get cached.
3287 #
3288 #Default:
3289 # vary_ignore_expire off
3291 # TAG: sleep_after_fork (microseconds)
3292 # When this is set to a non-zero value, the main Squid process
3293 # sleeps the specified number of microseconds after a fork()
3294 # system call. This sleep may help the situation where your
3295 # system reports fork() failures due to lack of (virtual)
3296 # memory. Note, however, that if you have a lot of child
3297 # processes, then these sleep delays will add up and your
3298 # Squid will not service requests for some amount of time
3299 # until all the child processes have been started.
3300 #
3301 #Default:
3302 # sleep_after_fork 0