2 .\" $OpenBSD: pf.conf.5,v 1.406 2009/01/31 19:37:12 sobrado Exp $
4 .\" Copyright (c) 2002, Daniel Hartmeier
5 .\" All rights reserved.
7 .\" Redistribution and use in source and binary forms, with or without
8 .\" modification, are permitted provided that the following conditions
11 .\" - Redistributions of source code must retain the above copyright
12 .\" notice, this list of conditions and the following disclaimer.
13 .\" - Redistributions in binary form must reproduce the above
14 .\" copyright notice, this list of conditions and the following
15 .\" disclaimer in the documentation and/or other materials provided
16 .\" with the distribution.
18 .\" THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
19 .\" "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
20 .\" LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
21 .\" FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
22 .\" COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
23 .\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
24 .\" BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
25 .\" LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
26 .\" CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN
28 .\" ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
29 .\" POSSIBILITY OF SUCH DAMAGE.
36 .Nd packet filter configuration file
40 packet filter modifies, drops or passes packets according to rules or
41 definitions specified in
44 There are seven types of statements in
48 User-defined variables may be defined and used later, simplifying
49 the configuration file.
50 Macros must be defined before they are referenced in
53 Tables provide a mechanism for increasing the performance and flexibility of
54 rules with large numbers of source or destination addresses.
56 Options tune the behaviour of the packet filtering engine.
57 .It Cm Traffic Normalization Li (e.g. Em scrub )
58 Traffic normalization protects internal machines against inconsistencies
59 in Internet protocols and implementations.
61 Queueing provides rule-based bandwidth control.
62 .It Cm Translation Li (Various forms of NAT)
63 Translation rules specify how addresses are to be mapped or redirected to
65 .It Cm Packet Filtering
66 Packet filtering provides rule-based blocking or passing of packets.
73 the types of statements should be grouped and appear in
75 in the order shown above, as this matches the operation of the underlying
76 packet filtering engine.
79 enforces this order (see
83 Comments can be put anywhere in the file using a hash mark
85 and extend to the end of the current line.
87 Additional configuration files can be included with the
90 .Bd -literal -offset indent
91 include "/etc/pf/sub.filter.conf"
94 Macros can be defined that will later be expanded in context.
95 Macro names must start with a letter, and may contain letters, digits
97 Macro names may not be reserved words (for example
101 Macros are not expanded inside quotes.
104 .Bd -literal -offset indent
106 all_ifs = \&"{\&" $ext_if lo0 \&"}\&"
107 pass out on $ext_if from any to any
108 pass in on $ext_if proto tcp from any to any port 25
111 Tables are named structures which can hold a collection of addresses and
113 Lookups against tables in
115 are relatively fast, making a single rule with tables much more efficient,
117 processor usage and memory consumption, than a large number of rules which
118 differ only in IP address (either created explicitly or automatically by rule
121 Tables can be used as the source or destination of filter rules,
125 translation rules such as
129 (see below for details on the various rule types).
130 Tables can also be used for the redirect address of
134 rules and in the routing options of filter rules, but only for
138 Tables can be defined with any of the following
141 As with macros, reserved words may not be used as table names.
142 .Bl -tag -width "manually"
144 Persistent tables can be manually created with the
150 before or after the ruleset has been loaded.
152 Table definitions can be placed directly in this file, and loaded at the
153 same time as other rules are loaded, atomically.
154 Table definitions inside
158 statement, and are especially useful to define non-persistent tables.
159 The contents of a pre-existing table defined without a list of addresses
160 to initialize it is not altered when
163 A table initialized with the empty list,
165 will be cleared on load.
168 Tables may be defined with the following attributes:
169 .Bl -tag -width persist
173 flag forces the kernel to keep the table even when no rules refer to it.
174 If the flag is not set, the kernel will automatically remove the table
175 when the last rule referring to it is flushed.
179 flag prevents the user from altering the contents of the table once it
183 can be used to add or remove addresses from the table at any time, even
190 flag enables per-address packet and byte counters which can be displayed with
195 .Bd -literal -offset indent
196 table \*(Ltprivate\*(Gt const { 10/8, 172.16/12, 192.168/16 }
197 table \*(Ltbadhosts\*(Gt persist
198 block on fxp0 from { \*(Ltprivate\*(Gt, \*(Ltbadhosts\*(Gt } to any
201 creates a table called private, to hold RFC 1918 private network
202 blocks, and a table called badhosts, which is initially empty.
203 A filter rule is set up to block all traffic coming from addresses listed in
205 The private table cannot have its contents changed and the badhosts table
206 will exist even when no active filter rules reference it.
207 Addresses may later be added to the badhosts table, so that traffic from
208 these hosts can be blocked by using
209 .Bd -literal -offset indent
210 # pfctl -t badhosts -Tadd 204.92.77.111
213 A table can also be initialized with an address list specified in one or more
214 external files, using the following syntax:
215 .Bd -literal -offset indent
216 table \*(Ltspam\*(Gt persist file \&"/etc/spammers\&" file \&"/etc/openrelays\&"
217 block on fxp0 from \*(Ltspam\*(Gt to any
224 list IP addresses, one per line.
225 Any lines beginning with a # are treated as comments and ignored.
226 In addition to being specified by IP address, hosts may also be
227 specified by their hostname.
228 When the resolver is called to add a hostname to a table,
230 resulting IPv4 and IPv6 addresses are placed into the table.
231 IP addresses can also be entered in a table by specifying a valid interface
232 name, a valid interface group or the
234 keyword, in which case all addresses assigned to the interface(s) will be
238 may be tuned for various situations using the
244 .Bl -tag -width "src.track" -compact
246 Interval between purging expired states and fragments.
248 Seconds before an unassembled fragment is expired.
250 Length of time to retain a source tracking entry after the last state
254 When a packet matches a stateful connection, the seconds to live for the
255 connection will be updated to that of the
257 which corresponds to the connection state.
258 Each packet which matches this state will reset the TTL.
259 Tuning these values may improve the performance of the
260 firewall at the risk of dropping valid idle connections.
262 .Bl -tag -width xxxx -compact
264 The state after the first packet.
266 The state before the destination host ever sends a packet.
267 .It Ar tcp.established
268 The fully established state.
270 The state after the first FIN has been sent.
272 The state after both FINs have been exchanged and the connection is closed.
273 Some hosts (notably web servers on Solaris) send TCP packets even after closing
279 can prevent blocking of such packets.
281 The state after one endpoint sends an RST.
284 ICMP and UDP are handled in a fashion similar to TCP, but with a much more
285 limited set of states:
287 .Bl -tag -width xxxx -compact
289 The state after the first packet.
291 The state if the source host sends more than one packet but the destination
292 host has never sent one back.
294 The state if both hosts have sent packets.
296 The state after the first packet.
298 The state after an ICMP error came back in response to an ICMP packet.
301 Other protocols are handled similarly to UDP:
303 .Bl -tag -width xxxx -compact
306 .It Ar other.multiple
309 Timeout values can be reduced adaptively as the number of state table
312 .Bl -tag -width xxxx -compact
313 .It Ar adaptive.start
314 When the number of state entries exceeds this value, adaptive scaling
316 All timeout values are scaled linearly with factor
317 (adaptive.end - number of states) / (adaptive.end - adaptive.start).
319 When reaching this number of state entries, all timeout values become
320 zero, effectively purging all state entries immediately.
321 This value is used to define the scale factor, it should not actually
322 be reached (set a lower state limit, see below).
325 Adaptive timeouts are enabled by default, with an adaptive.start value
326 equal to 60% of the state limit, and an adaptive.end value equal to
327 120% of the state limit.
328 They can be disabled by setting both adaptive.start and adaptive.end to 0.
330 The adaptive timeout values can be defined both globally and for each rule.
331 When used on a per-rule basis, the values relate to the number of
332 states created by the rule, otherwise to the total number of
336 .Bd -literal -offset indent
337 set timeout tcp.first 120
338 set timeout tcp.established 86400
339 set timeout { adaptive.start 6000, adaptive.end 12000 }
340 set limit states 10000
343 With 9000 state table entries, the timeout values are scaled to 50%
344 (tcp.first 60, tcp.established 43200).
346 .It Ar set loginterface
347 Enable collection of packet and byte count statistics for the given
348 interface or interface group.
349 These statistics can be viewed using
350 .Bd -literal -offset indent
356 collects statistics on the interface named dc0:
357 .Bd -literal -offset indent
361 One can disable the loginterface using:
362 .Bd -literal -offset indent
363 set loginterface none
367 Sets hard limits on the memory pools used by the packet filter.
370 for an explanation of memory pools.
373 .Bd -literal -offset indent
374 set limit states 20000
377 sets the maximum number of entries in the memory pool used by state table
378 entries (generated by
380 rules which do not specify
384 .Bd -literal -offset indent
385 set limit frags 20000
388 sets the maximum number of entries in the memory pool used for fragment
389 reassembly (generated by
393 .Bd -literal -offset indent
394 set limit src-nodes 2000
397 sets the maximum number of entries in the memory pool used for tracking
398 source IP addresses (generated by the
404 .Bd -literal -offset indent
405 set limit tables 1000
406 set limit table-entries 100000
409 sets limits on the memory pools used by tables.
410 The first limits the number of tables that can exist to 1000.
411 The second limits the overall number of addresses that can be stored
414 Various limits can be combined on a single line:
415 .Bd -literal -offset indent
416 set limit { states 20000, frags 20000, src-nodes 2000 }
419 .It Ar set ruleset-optimization
420 .Bl -tag -width xxxxxxxx -compact
422 Disable the ruleset optimizer.
424 Enable basic ruleset optimization.
425 This is the default behaviour.
426 Basic ruleset optimization does four things to improve the
427 performance of ruleset evaluations:
431 remove duplicate rules
433 remove rules that are a subset of another rule
435 combine multiple rules into a table when advantageous
437 re-order the rules to improve evaluation performance
441 Uses the currently loaded ruleset as a feedback profile to tailor the
442 ordering of quick rules to actual network traffic.
445 It is important to note that the ruleset optimizer will modify the ruleset
446 to improve performance.
447 A side effect of the ruleset modification is that per-rule accounting
448 statistics will have different meanings than before.
449 If per-rule accounting is important for billing purposes or whatnot,
450 either the ruleset optimizer should not be used or a label field should
451 be added to all of the accounting rules to act as optimization barriers.
453 Optimization can also be set as a command-line argument to
455 overriding the settings in
457 .It Ar set optimization
458 Optimize state timeouts for one of the following network environments:
460 .Bl -tag -width xxxx -compact
462 A normal network environment.
463 Suitable for almost all networks.
465 A high-latency environment (such as a satellite connection).
470 Aggressively expire connections.
471 This can greatly reduce the memory usage of the firewall at the cost of
472 dropping idle connections early.
474 Extremely conservative settings.
475 Avoid dropping legitimate connections at the
476 expense of greater memory utilization (possibly much greater on a busy
477 network) and slightly increased processor utilization.
481 .Bd -literal -offset indent
482 set optimization aggressive
485 .It Ar set block-policy
488 option sets the default behaviour for the packet
492 .Bl -tag -width xxxxxxxx -compact
494 Packet is silently dropped.
496 A TCP RST is returned for blocked TCP packets,
497 an ICMP UNREACHABLE is returned for blocked UDP packets,
498 and all other packets are silently dropped.
502 .Bd -literal -offset indent
503 set block-policy return
505 .It Ar set state-policy
508 option sets the default behaviour for states:
510 .Bl -tag -width group-bound -compact
512 States are bound to interface.
514 States can match packets on any interfaces (the default).
518 .Bd -literal -offset indent
519 set state-policy if-bound
521 .It Ar set state-defaults
524 option sets the state options for states created from rules
528 .Bd -literal -offset indent
529 set state-defaults pflow, no-sync
534 identifies this firewall's state table entries to other firewalls
538 By default the hostid is set to a pseudo-random value, however it may be
539 desirable to manually configure it, for example to more easily identify the
540 source of state table entries.
541 .Bd -literal -offset indent
545 The hostid may be specified in either decimal or hexadecimal.
546 .It Ar set require-order
549 enforces an ordering of the statement types in the ruleset to:
555 Setting this option to
557 disables this enforcement.
558 There may be non-trivial and non-obvious implications to an out of
560 Consider carefully before disabling the order enforcement.
561 .It Ar set fingerprints
562 Load fingerprints of known operating systems from the given filename.
563 By default fingerprints of known operating systems are automatically
568 but can be overridden via this option.
569 Setting this option may leave a small period of time where the fingerprints
570 referenced by the currently active ruleset are inconsistent until the new
571 ruleset finishes loading.
575 .Dl set fingerprints \&"/etc/pf.os.devel\&"
577 .It Ar set skip on Aq Ar ifspec
578 List interfaces for which packets should not be filtered.
579 Packets passing in or out on such interfaces are passed as if pf was
580 disabled, i.e. pf does not process them in any way.
581 This can be useful on loopback and other virtual interfaces, when
582 packet filtering is not desired and can have unexpected effects.
590 to one of the following:
592 .Bl -tag -width xxxxxxxxxxxx -compact
594 Don't generate debug messages.
596 Generate debug messages only for serious errors.
598 Generate debug messages for various errors.
600 Generate debug messages for common conditions.
603 .Sh TRAFFIC NORMALIZATION
604 Traffic normalization is used to sanitize packet content in such
605 a way that there are no ambiguities in packet interpretation on
607 The normalizer does IP fragment reassembly to prevent attacks
608 that confuse intrusion detection systems by sending overlapping
610 Packet normalization is invoked with the
615 has the following options:
620 bit from a matching IP packet.
621 Some operating systems are known to generate fragmented packets with the
624 This is particularly true with NFS.
626 will drop such fragmented
632 Unfortunately some operating systems also generate their
634 packets with a zero IP identification field.
637 bit on packets with a zero IP ID may cause deleterious results if an
638 upstream router later fragments the packet.
641 modifier (see below) is recommended in combination with the
643 modifier to ensure unique IP identifiers.
644 .It Ar min-ttl Aq Ar number
645 Enforces a minimum TTL for matching IP packets.
646 .It Ar max-mss Aq Ar number
647 Enforces a maximum MSS for matching TCP packets.
648 .It Xo Ar set-tos Aq Ar string
649 .No \*(Ba Aq Ar number
653 for matching IP packets.
660 or as either hex or decimal.
662 Replaces the IP identification field with random values to compensate
663 for predictable values generated by many hosts.
664 This option only applies to packets that are not fragmented
665 after the optional fragment reassembly.
666 .It Ar fragment reassemble
669 rules, fragments can be reassembled by normalization.
670 In this case, fragments are buffered until they form a complete
671 packet, and only the completed packet is passed on to the filter.
672 The advantage is that filter rules have to deal only with complete
673 packets, and can ignore fragments.
674 The drawback of caching fragments is the additional memory cost.
675 But the full reassembly method is the only method that currently works
677 This is the default behavior of a
679 rule if no fragmentation modifier is supplied.
681 The default fragment reassembly method is expensive, hence the option
685 will track the fragments and cache a small range descriptor.
686 Duplicate fragments are dropped and overlaps are cropped.
687 Thus data will only occur once on the wire with ambiguities resolving to
688 the first occurrence.
690 .Ar fragment reassemble
691 modifier, fragments are not buffered, they are passed as soon as they
695 reassembly mechanism does not yet work with NAT.
697 .It Ar fragment drop-ovl
698 This option is similar to the
700 modifier except that all overlapping or duplicate fragments will be
701 dropped, and all further corresponding fragments will be
703 .It Ar reassemble tcp
704 Statefully normalizes TCP connections.
705 .Ar scrub reassemble tcp
706 rules may not have the direction (in/out) specified.
708 performs the following normalizations:
710 .Bl -tag -width timeout -compact
712 Neither side of the connection is allowed to reduce their IP TTL.
713 An attacker may send a packet such that it reaches the firewall, affects
714 the firewall state, and expires before reaching the destination host.
716 will raise the TTL of all packets back up to the highest value seen on
718 .It timestamp modulation
719 Modern TCP stacks will send a timestamp on every TCP packet and echo
720 the other endpoint's timestamp back to them.
721 Many operating systems will merely start the timestamp at zero when
722 first booted, and increment it several times a second.
723 The uptime of the host can be deduced by reading the timestamp and multiplying
725 Also observing several different timestamps can be used to count hosts
727 And spoofing TCP packets into a connection requires knowing or guessing
729 Timestamps merely need to be monotonically increasing and not derived off a
734 to modulate the TCP timestamps with a random number.
735 .It extended PAWS checks
736 There is a problem with TCP on long fat pipes, in that a packet might get
737 delayed for longer than it takes the connection to wrap its 32-bit sequence
739 In such an occurrence, the old packet would be indistinguishable from a
740 new packet and would be accepted as such.
741 The solution to this is called PAWS: Protection Against Wrapped Sequence
743 It protects against it by making sure the timestamp on each packet does
746 also makes sure the timestamp on the packet does not go forward more
750 artificially extends the security of TCP sequence numbers by 10 to 18
751 bits when the host uses appropriately randomized timestamps, since a
752 blind attacker would have to guess the timestamp as well.
757 .Bd -literal -offset indent
758 scrub in on $ext_if all fragment reassemble
763 option prefixed to a scrub rule causes matching packets to remain unscrubbed,
764 much in the same way as
766 works in the packet filter (see below).
767 This mechanism should be used when it is necessary to exclude specific packets
768 from broader scrub rules.
770 The ALTQ system is currently not available in the GENERIC kernel nor as
772 In order to use the herein after called queueing options one has to use a
776 to learn about the related kernel options.
778 Packets can be assigned to queues for the purpose of bandwidth
780 At least two declarations are required to configure queues, and later
781 any packet filtering rule can reference the defined queues by name.
782 During the filtering component of
786 name is where any packets from
788 rules will be queued, while for
790 rules it specifies where any resulting ICMP or TCP RST
791 packets should be queued.
794 defines the algorithm used to decide which packets get delayed, dropped, or
795 sent out immediately.
801 Class Based Queueing.
803 attached to an interface build a tree, thus each
805 can have further child
807 Each queue can have a
813 mainly controls the time packets take to get sent out, while
815 has primarily effects on throughput.
817 achieves both partitioning and sharing of link bandwidth
818 by hierarchically structured classes.
819 Each class has its own
821 and is assigned its share of
823 A child class can borrow bandwidth from its parent class
824 as long as excess bandwidth is available
831 are flat attached to the interface, thus,
833 cannot have further child
839 assigned, ranging from 0 to 15.
846 Hierarchical Fair Service Curve.
848 attached to an interface build a tree, thus each
850 can have further child
852 Each queue can have a
858 mainly controls the time packets take to get sent out, while
860 primarily affects throughput.
862 supports both link-sharing and guaranteed real-time services.
863 It employs a service curve based QoS model,
864 and its unique feature is an ability to decouple
871 The interfaces on which queueing should be activated are declared using
876 has the following keywords:
879 Queueing is enabled on the named interface.
881 Specifies which queueing scheduler to use.
882 Currently supported values
885 for Class Based Queueing,
887 for Priority Queueing and
889 for the Hierarchical Fair Service Curve scheduler.
890 .It Ar bandwidth Aq Ar bw
891 The maximum bitrate for all queues on an
892 interface may be specified using the
895 The value can be specified as an absolute value or as a
896 percentage of the interface bandwidth.
897 When using an absolute value, the suffixes
903 are used to represent bits, kilobits, megabits, and
904 gigabits per second, respectively.
905 The value must not exceed the interface bandwidth.
908 is not specified, the interface bandwidth is used
909 (but take note that some interfaces do not know their bandwidth,
910 or can adapt their bandwidth rates).
911 .It Ar qlimit Aq Ar limit
912 The maximum number of packets held in the queue.
914 .It Ar tbrsize Aq Ar size
915 Adjusts the size, in bytes, of the token bucket regulator.
916 If not specified, heuristics based on the
917 interface bandwidth are used to determine the size.
918 .It Ar queue Aq Ar list
919 Defines a list of subqueues to create on an interface.
922 In the following example, the interface dc0
923 should queue up to 5Mbps in four second-level queues using
924 Class Based Queueing.
925 Those four queues will be shown in a later example.
926 .Bd -literal -offset indent
927 altq on dc0 cbq bandwidth 5Mb queue { std, http, mail, ssh }
930 Once interfaces are activated for queueing using the
932 directive, a sequence of
934 directives may be defined.
935 The name associated with a
937 must match a queue defined in the
939 directive (e.g. mail), or, except for the
945 The following keywords can be used:
947 .It Ar on Aq Ar interface
948 Specifies the interface the queue operates on.
949 If not given, it operates on all matching interfaces.
950 .It Ar bandwidth Aq Ar bw
951 Specifies the maximum bitrate to be processed by the queue.
952 This value must not exceed the value of the parent
954 and can be specified as an absolute value or a percentage of the parent
956 If not specified, defaults to 100% of the parent queue's bandwidth.
959 scheduler does not support bandwidth specification.
960 .It Ar priority Aq Ar level
961 Between queues a priority level can be set.
966 the range is 0 to 7 and for
968 the range is 0 to 15.
969 The default for all is 1.
971 queues with a higher priority are always served first.
975 queues with a higher priority are preferred in the case of overload.
976 .It Ar qlimit Aq Ar limit
977 The maximum number of packets held in the queue.
983 can get additional parameters with
985 .Pf ( Aq Ar parameters ) .
987 Parameters are as follows:
990 Packets not matched by another queue are assigned to this one.
991 Exactly one default queue is required.
993 Enable RED (Random Early Detection) on this queue.
994 RED drops packets with a probability proportional to the average
997 Enables RIO on this queue.
998 RIO is RED with IN/OUT, thus running
999 RED two times more than RIO would achieve the same effect.
1000 RIO is currently not supported in the GENERIC kernel.
1002 Enables ECN (Explicit Congestion Notification) on this queue.
1009 supports an additional option:
1012 The queue can borrow bandwidth from the parent.
1018 supports some additional options:
1020 .It Ar realtime Aq Ar sc
1021 The minimum required bandwidth for the queue.
1022 .It Ar upperlimit Aq Ar sc
1023 The maximum allowed bandwidth for the queue.
1024 .It Ar linkshare Aq Ar sc
1025 The bandwidth share of a backlogged queue.
1032 The format for service curve specifications is
1033 .Ar ( m1 , d , m2 ) .
1035 controls the bandwidth assigned to the queue.
1039 are optional and can be used to control the initial bandwidth assignment.
1042 milliseconds the queue gets the bandwidth given as
1044 afterwards the value given in
1051 child queues can be specified as in an
1053 declaration, thus building a tree of queues using a part of
1054 their parent's bandwidth.
1056 Packets can be assigned to queues based on filter rules by using the
1061 is specified; when a second one is specified it will instead be used for
1062 packets which have a
1066 and for TCP ACKs with no data payload.
1068 To continue the previous example, the examples below would specify the
1070 queues, plus a few child queues.
1073 sessions get priority over bulk transfers like
1077 The queues may then be referenced by filtering rules (see
1078 .Sx PACKET FILTERING
1081 queue std bandwidth 10% cbq(default)
1082 queue http bandwidth 60% priority 2 cbq(borrow red) \e
1083 { employees, developers }
1084 queue developers bandwidth 75% cbq(borrow)
1085 queue employees bandwidth 15%
1086 queue mail bandwidth 10% priority 0 cbq(borrow ecn)
1087 queue ssh bandwidth 20% cbq(borrow) { ssh_interactive, ssh_bulk }
1088 queue ssh_interactive bandwidth 50% priority 7 cbq(borrow)
1089 queue ssh_bulk bandwidth 50% priority 0 cbq(borrow)
1091 block return out on dc0 inet all queue std
1092 pass out on dc0 inet proto tcp from $developerhosts to any port 80 \e
1094 pass out on dc0 inet proto tcp from $employeehosts to any port 80 \e
1096 pass out on dc0 inet proto tcp from any to any port 22 \e
1097 queue(ssh_bulk, ssh_interactive)
1098 pass out on dc0 inet proto tcp from any to any port 25 \e
1102 Translation rules modify either the source or destination address of the
1103 packets associated with a stateful connection.
1104 A stateful connection is automatically created to track packets matching
1105 such a rule as long as they are not blocked by the filtering section of
1107 The translation engine modifies the specified address and/or port in the
1108 packet, recalculates IP, TCP and UDP checksums as necessary, and passes it to
1109 the packet filter for evaluation.
1111 Since translation occurs before filtering the filter
1112 engine will see packets as they look after any
1113 addresses and ports have been translated.
1114 Filter rules will therefore have to filter based on the translated
1115 address and port number.
1116 Packets that match a translation rule are only automatically passed if
1119 modifier is given, otherwise they are
1126 The state entry created permits
1128 to keep track of the original address for traffic associated with that state
1129 and correctly direct return traffic for that connection.
1131 Various types of translation are possible with pf:
1132 .Bl -tag -width xxxx
1136 rule specifies a bidirectional mapping between an external IP netblock
1137 and an internal IP netblock.
1141 rule specifies that IP addresses are to be changed as the packet
1142 traverses the given interface.
1143 This technique allows one or more IP addresses
1144 on the translating host to support network traffic for a larger range of
1145 machines on an "inside" network.
1146 Although in theory any IP address can be used on the inside, it is strongly
1147 recommended that one of the address ranges defined by RFC 1918 be used.
1148 These netblocks are:
1150 10.0.0.0 - 10.255.255.255 (all of net 10, i.e., 10/8)
1151 172.16.0.0 - 172.31.255.255 (i.e., 172.16/12)
1152 192.168.0.0 - 192.168.255.255 (i.e., 192.168/16)
1155 The packet is redirected to another destination and possibly a
1158 rules can optionally specify port ranges instead of single ports.
1159 rdr ... port 2000:2999 -\*(Gt ... port 4000
1160 redirects ports 2000 to 2999 (inclusive) to port 4000.
1161 rdr ... port 2000:2999 -\*(Gt ... port 4000:*
1162 redirects port 2000 to 4000, 2001 to 4001, ..., 2999 to 4999.
1165 In addition to modifying the address, some translation rules may modify
1166 source or destination ports for
1170 connections; implicitly in the case of
1172 rules and explicitly in the case of
1175 Port numbers are never translated with a
1179 Evaluation order of the translation rules is dependent on the type
1180 of the translation rules and of the direction of a packet.
1182 rules are always evaluated first.
1185 rules are evaluated on an inbound packet or the
1187 rules on an outbound packet.
1188 Rules of the same type are evaluated in the same order in which they
1189 appear in the ruleset.
1190 The first matching rule decides what action is taken.
1194 option prefixed to a translation rule causes packets to remain untranslated,
1195 much in the same way as
1197 works in the packet filter (see below).
1198 If no rule matches the packet it is passed to the filter engine unmodified.
1200 Translation rules apply only to packets that pass through
1201 the specified interface, and if no interface is specified,
1202 translation is applied to packets on all interfaces.
1203 For instance, redirecting port 80 on an external interface to an internal
1204 web server will only work for connections originating from the outside.
1205 Connections to the address of the external interface from local hosts will
1206 not be redirected, since such packets do not actually pass through the
1208 Redirections cannot reflect packets back through the interface they arrive
1209 on, they can only be redirected to hosts connected to different interfaces
1210 or to the firewall itself.
1212 Note that redirecting external incoming connections to the loopback
1214 .Bd -literal -offset indent
1215 rdr on ne3 inet proto tcp to port smtp -\*(Gt 127.0.0.1 port spamd
1218 will effectively allow an external host to connect to daemons
1219 bound solely to the loopback address, circumventing the traditional
1220 blocking of such connections on a real interface.
1221 Unless this effect is desired, any of the local non-loopback addresses
1222 should be used as redirection target instead, which allows external
1223 connections only to daemons bound to this address or not bound to
1227 .Sx TRANSLATION EXAMPLES
1229 .Sh PACKET FILTERING
1235 packets based on attributes of their layer 3 (see
1245 In addition, packets may also be
1246 assigned to queues for the purpose of bandwidth control.
1248 For each packet processed by the packet filter, the filter rules are
1249 evaluated in sequential order, from first to last.
1250 The last matching rule decides what action is taken.
1251 If no rule matches the packet, the default action is to pass
1254 The following actions can be used in the filter:
1255 .Bl -tag -width xxxx
1257 The packet is blocked.
1258 There are a number of ways in which a
1260 rule can behave when blocking a packet.
1261 The default behaviour is to
1263 packets silently, however this can be overridden or made
1264 explicit either globally, by setting the
1266 option, or on a per-rule basis with one of the following options:
1268 .Bl -tag -width xxxx -compact
1270 The packet is silently dropped.
1272 This applies only to
1274 packets, and issues a TCP RST which closes the
1278 This causes ICMP messages to be returned for packets which match the rule.
1279 By default this is an ICMP UNREACHABLE message, however this
1280 can be overridden by specifying a message as a code or number.
1282 This causes a TCP RST to be returned for
1284 packets and an ICMP UNREACHABLE for UDP and other packets.
1287 Options returning ICMP packets currently have no effect if
1291 as the code to support this feature has not yet been implemented.
1293 The simplest mechanism to block everything by default and only pass
1294 packets that match explicit rules is specify a first filter rule of:
1295 .Bd -literal -offset indent
1299 The packet is passed;
1300 state is created unless the
1302 option is specified.
1307 filters packets statefully; the first time a packet matches a
1309 rule, a state entry is created; for subsequent packets the filter checks
1310 whether the packet matches any state.
1311 If it does, the packet is passed without evaluation of any rules.
1312 After the connection is closed or times out, the state entry is automatically
1315 This has several advantages.
1316 For TCP connections, comparing a packet to a state involves checking
1317 its sequence numbers, as well as TCP timestamps if a
1318 .Ar scrub reassemble tcp
1319 rule applies to the connection.
1320 If these values are outside the narrow windows of expected
1321 values, the packet is dropped.
1322 This prevents spoofing attacks, such as when an attacker sends packets with
1323 a fake source address/port but does not know the connection's sequence
1327 knows how to match ICMP replies to states.
1329 .Bd -literal -offset indent
1330 pass out inet proto icmp all icmp-type echoreq
1333 allows echo requests (such as those created by
1335 out statefully, and matches incoming echo replies correctly to states.
1337 Also, looking up states is usually faster than evaluating rules.
1338 If there are 50 rules, all of them are evaluated sequentially in O(n).
1339 Even with 50000 states, only 16 comparisons are needed to match a
1340 state, since states are stored in a binary search tree that allows
1341 searches in O(log2 n).
1343 Furthermore, correct handling of ICMP error messages is critical to
1344 many protocols, particularly TCP.
1346 matches ICMP error messages to the correct connection, checks them against
1347 connection parameters, and passes them if appropriate.
1348 For example if an ICMP source quench message referring to a stateful TCP
1349 connection arrives, it will be matched to the state and get passed.
1351 Finally, state tracking is required for
1352 .Ar nat , binat No and Ar rdr
1353 rules, in order to track address and port translations and reverse the
1354 translation on returning packets.
1357 will also create state for other protocols which are effectively stateless by
1359 UDP packets are matched to states using only host addresses and ports,
1360 and other protocols are matched to states using only the host addresses.
1362 If stateless filtering of individual packets is desired,
1365 keyword can be used to specify that state will not be created
1366 if this is the last matching rule.
1367 A number of parameters can also be set to affect how
1369 handles state tracking.
1371 .Sx STATEFUL TRACKING OPTIONS
1372 below for further details.
1374 The rule parameters specify the packets to which a rule applies.
1375 A packet always comes in on, or goes out through, one interface.
1376 Most parameters are optional.
1377 If a parameter is specified, the rule only applies to packets with
1378 matching attributes.
1379 Certain parameters can be expressed as lists, in which case
1381 generates all needed rule combinations.
1382 .Bl -tag -width xxxx
1383 .It Ar in No or Ar out
1384 This rule applies to incoming or outgoing packets.
1389 are specified, the rule will match packets in both directions.
1391 In addition to the action specified, a log message is generated.
1392 Only the packet that establishes the state is logged,
1395 option is specified.
1396 The logged packets are sent to a
1398 interface, by default
1400 This interface is monitored by the
1402 logging daemon, which dumps the logged packets to the file
1408 Used to force logging of all packets for a connection.
1409 This is not necessary when
1411 is explicitly specified.
1414 packets are logged to
1419 user ID of the user that owns the socket and the PID of the process that
1420 has the socket open where the packet is sourced from or destined to
1421 (depending on which socket is local).
1422 This is in addition to the normal information logged.
1424 Due to the problems described in the BUGS section only the first packet
1427 will have the user credentials logged when using stateful matching.
1428 .It Ar log (to Aq Ar interface )
1429 Send logs to the specified
1431 interface instead of
1434 If a packet matches a rule which has the
1436 option set, this rule
1437 is considered the last matching rule, and evaluation of subsequent rules
1439 .It Ar on Aq Ar interface
1440 This rule applies only to packets coming in on, or going out through, this
1441 particular interface or interface group.
1442 For more information on interface groups,
1448 This rule applies only to packets of this address family.
1449 Supported values are
1453 .It Ar proto Aq Ar protocol
1454 This rule applies only to packets of this protocol.
1455 Common protocols are
1461 For a list of all the protocol name to number mappings used by
1464 .Em /etc/protocols .
1466 .Ar from Aq Ar source
1467 .Ar port Aq Ar source
1472 This rule applies only to packets with the specified source and destination
1473 addresses and ports.
1475 Addresses can be specified in CIDR notation (matching netblocks), as
1476 symbolic host names, interface names or interface group names, or as any
1477 of the following keywords:
1479 .Bl -tag -width xxxxxxxxxxxxxx -compact
1482 .It Ar route Aq Ar label
1483 Any address whose associated route has label
1490 Any address which is not currently routable.
1492 Any source address that fails a unicast reverse path forwarding (URPF)
1493 check, i.e. packets coming in on an interface other than that which holds
1494 the route back to the packet's source address.
1496 Any address that matches the given table.
1499 Ranges of addresses are specified by using the
1503 .Dq 10.1.1.10 - 10.1.1.12
1504 means all addresses from 10.1.1.10 to 10.1.1.12,
1505 hence addresses 10.1.1.10, 10.1.1.11, and 10.1.1.12.
1507 Interface names and interface group names can have modifiers appended:
1509 .Bl -tag -width xxxxxxxxxxxx -compact
1511 Translates to the network(s) attached to the interface.
1513 Translates to the interface's broadcast address(es).
1515 Translates to the point-to-point interface's peer address(es).
1517 Do not include interface aliases.
1520 Host names may also have the
1522 option appended to restrict the name resolution to the first of each
1523 v4 and v6 address found.
1525 Host name resolution and interface to address translation are done at
1527 When the address of an interface (or host name) changes (under DHCP or PPP,
1528 for instance), the ruleset must be reloaded for the change to be reflected
1530 Surrounding the interface name (and optional modifiers) in parentheses
1531 changes this behaviour.
1532 When the interface name is surrounded by parentheses, the rule is
1533 automatically updated whenever the interface changes its address.
1534 The ruleset does not need to be reloaded.
1535 This is especially useful with
1538 Ports can be specified either by number or by name.
1539 For example, port 80 can be specified as
1541 For a list of all port name to number mappings used by
1546 Ports and ranges of ports are specified by using these operators:
1547 .Bd -literal -offset indent
1551 \*(Le (less than or equal)
1552 \*(Gt (greater than)
1553 \*(Ge (greater than or equal)
1554 : (range including boundaries)
1555 \*(Gt\*(Lt (range excluding boundaries)
1556 \*(Lt\*(Gt (except range)
1563 are binary operators (they take two arguments).
1566 .It Ar port 2000:2004
1568 .Sq all ports \*(Ge 2000 and \*(Le 2004 ,
1569 hence ports 2000, 2001, 2002, 2003 and 2004.
1570 .It Ar port 2000 \*(Gt\*(Lt 2004
1572 .Sq all ports \*(Gt 2000 and \*(Lt 2004 ,
1573 hence ports 2001, 2002 and 2003.
1574 .It Ar port 2000 \*(Lt\*(Gt 2004
1576 .Sq all ports \*(Lt 2000 or \*(Gt 2004 ,
1577 hence ports 1-1999 and 2005-65535.
1580 The operating system of the source host can be specified in the case of TCP
1585 .Sx OPERATING SYSTEM FINGERPRINTING
1586 section for more information.
1588 The host, port and OS specifications are optional, as in the following examples:
1589 .Bd -literal -offset indent
1591 pass in from any to any
1592 pass in proto tcp from any port \*(Le 1024 to any
1593 pass in proto tcp from any to any port 25
1594 pass in proto tcp from 10.0.0.0/8 port \*(Gt 1024 \e
1595 to ! 10.1.2.3 port != ssh
1596 pass in proto tcp from any os "OpenBSD"
1597 pass in proto tcp from route "DTAG"
1600 This is equivalent to "from any to any".
1601 .It Ar group Aq Ar group
1604 this rule only applies to packets of sockets owned by the specified group.
1605 .It Ar user Aq Ar user
1606 This rule only applies to packets of sockets owned by the specified user.
1607 For outgoing connections initiated from the firewall, this is the user
1608 that opened the connection.
1609 For incoming connections to the firewall itself, this is the user that
1610 listens on the destination port.
1611 For forwarded connections, where the firewall is not a connection endpoint,
1612 the user and group are
1615 All packets, both outgoing and incoming, of one connection are associated
1616 with the same user and group.
1617 Only TCP and UDP packets can be associated with users; for other protocols
1618 these parameters are ignored.
1620 User and group refer to the effective (as opposed to the real) IDs, in
1621 case the socket is created by a setuid/setgid process.
1622 User and group IDs are stored when a socket is created;
1623 when a process creates a listening socket as root (for instance, by
1624 binding to a privileged port) and subsequently changes to another
1625 user ID (to drop privileges), the credentials will remain root.
1627 User and group IDs can be specified as either numbers or names.
1628 The syntax is similar to the one for ports.
1631 matches packets of forwarded connections.
1633 can only be used with the operators
1637 Other constructs like
1638 .Cm user \*(Ge unknown
1640 Forwarded packets with unknown user and group ID match only rules
1641 that explicitly compare against
1649 does not match forwarded packets.
1650 The following example allows only selected users to open outgoing
1652 .Bd -literal -offset indent
1653 block out proto { tcp, udp } all
1654 pass out proto { tcp, udp } all user { \*(Lt 1000, dhartmei }
1656 .It Xo Ar flags Aq Ar a
1658 .No \*(Ba / Ns Aq Ar b
1661 This rule only applies to TCP packets that have the flags
1665 Flags not specified in
1668 For stateful connections, the default is
1670 To indicate that flags should not be checked at all, specify
1672 The flags are: (F)IN, (S)YN, (R)ST, (P)USH, (A)CK, (U)RG, (E)CE, and C(W)R.
1676 The other flags are ignored.
1678 This is the default setting for stateful connections.
1679 Out of SYN and ACK, exactly SYN may be set.
1680 SYN, SYN+PSH and SYN+RST match, but SYN+ACK, ACK and ACK+RST do not.
1681 This is more restrictive than the previous example.
1683 If the first set is not specified, it defaults to none.
1684 All of SYN, FIN, RST and ACK must be unset.
1689 is applied by default (unless
1691 is specified), only the initial SYN packet of a TCP handshake will create
1692 a state for a TCP connection.
1693 It is possible to be less restrictive, and allow state creation from
1696 packets, by specifying
1700 to synchronize to existing connections, for instance
1701 if one flushes the state table.
1702 However, states created from such intermediate packets may be missing
1703 connection details such as the TCP window scaling factor.
1704 States which modify the packet flow, such as those affected by
1705 .Ar nat , binat No or Ar rdr
1707 .Ar modulate No or Ar synproxy state
1708 options, or scrubbed with
1710 will also not be recoverable from intermediate packets.
1711 Such connections will stall and time out.
1712 .It Xo Ar icmp-type Aq Ar type
1715 .It Xo Ar icmp6-type Aq Ar type
1718 This rule only applies to ICMP or ICMPv6 packets with the specified type
1720 Text names for ICMP types and codes are listed in
1724 This parameter is only valid for rules that cover protocols ICMP or
1726 The protocol and the ICMP type indicator
1733 .It Xo Ar tos Aq Ar string
1734 .No \*(Ba Aq Ar number
1736 This rule applies to packets with the specified
1745 or as either hex or decimal.
1747 For example, the following rules are identical:
1748 .Bd -literal -offset indent
1749 pass all tos lowdelay
1754 By default, IPv4 packets with IP options or IPv6 packets with routing
1755 extension headers are blocked.
1760 rule, packets that pass the filter based on that rule (last matching)
1761 do so even if they contain IP options or routing extension headers.
1762 For packets that match state, the rule that initially created the
1766 rule that is used when a packet does not match any rules does not
1768 .It Ar label Aq Ar string
1769 Adds a label (name) to the rule, which can be used to identify the rule.
1772 shows per-rule statistics for rules that have labels.
1774 The following macros can be used in labels:
1776 .Bl -tag -width $srcaddr -compact -offset indent
1780 The source IP address.
1782 The destination IP address.
1784 The source port specification.
1786 The destination port specification.
1794 .Bd -literal -offset indent
1795 ips = \&"{ 1.2.3.4, 1.2.3.5 }\&"
1796 pass in proto tcp from any to $ips \e
1797 port \*(Gt 1023 label \&"$dstaddr:$dstport\&"
1801 .Bd -literal -offset indent
1802 pass in inet proto tcp from any to 1.2.3.4 \e
1803 port \*(Gt 1023 label \&"1.2.3.4:\*(Gt1023\&"
1804 pass in inet proto tcp from any to 1.2.3.5 \e
1805 port \*(Gt 1023 label \&"1.2.3.5:\*(Gt1023\&"
1808 The macro expansion for the
1810 directive occurs only at configuration file parse time, not during runtime.
1811 .It Xo Ar queue Aq Ar queue
1812 .No \*(Ba ( Aq Ar queue ,
1815 Packets matching this rule will be assigned to the specified queue.
1816 If two queues are given, packets which have a
1820 and TCP ACKs with no data payload will be assigned to the second one.
1826 .Bd -literal -offset indent
1827 pass in proto tcp to port 25 queue mail
1828 pass in proto tcp to port 22 queue(ssh_bulk, ssh_prio)
1830 .It Ar tag Aq Ar string
1831 Packets matching this rule will be tagged with the
1833 The tag acts as an internal marker that can be used to
1834 identify these packets later on.
1835 This can be used, for example, to provide trust between
1836 interfaces and to determine if packets have been
1837 processed by translation rules.
1840 meaning that the packet will be tagged even if the rule
1841 is not the last matching rule.
1842 Further matching rules can replace the tag with a
1843 new one but will not remove a previously applied tag.
1844 A packet is only ever assigned one tag at a time.
1845 Packet tagging can be done during
1850 rules in addition to filter rules.
1851 Tags take the same macros as labels (see above).
1852 .It Ar tagged Aq Ar string
1853 Used with filter, translation or scrub rules
1854 to specify that packets must already
1855 be tagged with the given tag in order to match the rule.
1856 Inverse tag matching can also be done
1862 .It Ar rtable Aq Ar number
1863 Used to select an alternate routing table for the routing lookup.
1864 Only effective before the route lookup happened, i.e. when filtering inbound.
1865 .It Xo Ar divert-to Aq Ar host
1868 Used to redirect packets to a local socket bound to
1872 The packets will not be modified, so
1874 on the socket will return the original destination address of the packet.
1876 Used to receive replies for sockets that are bound to addresses
1877 which are not local to the machine.
1880 for information on how to bind these sockets.
1881 .It Ar probability Aq Ar number
1882 A probability attribute can be attached to a rule, with a value set between
1883 0 and 1, bounds not included.
1884 In that case, the rule will be honoured using the given probability value
1886 For example, the following rule will drop 20% of incoming ICMP packets:
1887 .Bd -literal -offset indent
1888 block in proto icmp probability 20%
1892 If a packet matches a rule with a route option set, the packet filter will
1893 route the packet according to the type of route option.
1894 When such a rule creates state, the route option is also applied to all
1895 packets matching the same connection.
1896 .Bl -tag -width xxxx
1900 option does a normal route lookup to find the next hop for the packet.
1904 option routes the packet to the specified interface with an optional address
1908 rule creates state, only packets that pass in the same direction as the
1909 filter rule specifies will be routed in this way.
1910 Packets passing in the opposite direction (replies) are not affected
1911 and are routed normally.
1915 option is similar to
1917 but routes packets that pass in the opposite direction (replies) to the
1918 specified interface.
1919 Opposite direction is only defined in the context of a state entry, and
1921 is useful only in rules that create state.
1922 It can be used on systems with multiple external connections to
1923 route all outgoing packets of a connection through the interface
1924 the incoming connection arrived through (symmetric routing enforcement).
1928 option creates a duplicate of the packet and routes it like
1930 The original packet gets routed as it normally would.
1937 rules, (as well as for the
1942 rule options) for which there is a single redirection address which has a
1943 subnet mask smaller than 32 for IPv4 or 128 for IPv6 (more than one IP
1944 address), a variety of different methods for assigning this address can be
1946 .Bl -tag -width xxxx
1950 option applies the network portion of the redirection address to the address
1951 to be modified (source with
1958 option selects an address at random within the defined block of addresses.
1962 option uses a hash of the source address to determine the redirection address,
1963 ensuring that the redirection address is always the same for a given source.
1964 An optional key can be specified after this keyword either in hex or as a
1967 randomly generates a key for source-hash every time the
1968 ruleset is reloaded.
1972 option loops through the redirection address(es).
1974 When more than one redirection address is specified,
1976 is the only permitted pool type.
1984 from modifying the source port on TCP and UDP packets.
1989 option can be specified to help ensure that multiple connections from the
1990 same source are mapped to the same redirection address.
1991 This option can be used with the
1996 Note that by default these associations are destroyed as soon as there are
1997 no longer states which refer to them; in order to make the mappings last
1998 beyond the lifetime of the states, increase the global options with
1999 .Ar set timeout src.track .
2001 .Sx STATEFUL TRACKING OPTIONS
2002 for more ways to control the source tracking.
2003 .Sh STATE MODULATION
2004 Much of the security derived from TCP is attributable to how well the
2005 initial sequence numbers (ISNs) are chosen.
2006 Some popular stack implementations choose
2008 poor ISNs and thus are normally susceptible to ISN prediction exploits.
2011 rule to a TCP connection,
2013 will create a high quality random sequence number for each connection
2018 directive implicitly keeps state on the rule and is
2019 only applicable to TCP connections.
2022 .Bd -literal -offset indent
2024 pass out proto tcp from any to any modulate state
2025 pass in proto tcp from any to any port 25 flags S/SFRA modulate state
2028 Note that modulated connections will not recover when the state table
2029 is lost (firewall reboot, flushing the state table, etc...).
2031 will not be able to infer a connection again after the state table flushes
2032 the connection's modulator.
2033 When the state is lost, the connection may be left dangling until the
2034 respective endpoints time out the connection.
2035 It is possible on a fast local network for the endpoints to start an ACK
2036 storm while trying to resynchronize after the loss of the modulator.
2039 settings (or a more strict equivalent) should be used on
2041 rules to prevent ACK storms.
2043 Note that alternative methods are available
2044 to prevent loss of the state table
2045 and allow for firewall failover.
2050 for further information.
2054 passes packets that are part of a
2056 handshake between the endpoints.
2059 option can be used to cause
2061 itself to complete the handshake with the active endpoint, perform a handshake
2062 with the passive endpoint, and then forward packets between the endpoints.
2064 No packets are sent to the passive endpoint before the active endpoint has
2065 completed the handshake, hence so-called SYN floods with spoofed source
2066 addresses will not reach the passive endpoint, as the sender can't complete the
2069 The proxy is transparent to both endpoints, they each see a single
2070 connection from/to the other endpoint.
2072 chooses random initial sequence numbers for both handshakes.
2073 Once the handshakes are completed, the sequence number modulators
2074 (see previous section) are used to translate further packets of the
2078 .Ar modulate state .
2088 .Bd -literal -offset indent
2089 pass in proto tcp from any to any port www synproxy state
2091 .Sh STATEFUL TRACKING OPTIONS
2092 A number of options related to stateful tracking can be applied on a
2098 support these options, and
2100 must be specified explicitly to apply options to a rule.
2102 .Bl -tag -width xxxx -compact
2103 .It Ar max Aq Ar number
2104 Limits the number of concurrent states the rule may create.
2105 When this limit is reached, further packets that would create
2106 state will not match this rule until existing states time out.
2108 Prevent state changes for states created by this rule from appearing on the
2111 .It Xo Aq Ar timeout
2114 Changes the timeout values used for states created by this rule.
2115 For a list of all valid timeout names, see
2119 Uses a sloppy TCP connection tracker that does not check sequence
2120 numbers at all, which makes insertion and ICMP teardown attacks way
2122 This is intended to be used in situations where one does not see all
2123 packets of a connection, e.g. in asymmetric routing situations.
2124 Cannot be used with modulate or synproxy state.
2126 States created by this rule are exported on the
2131 Multiple options can be specified, separated by commas:
2132 .Bd -literal -offset indent
2133 pass in proto tcp from any to any \e
2134 port www keep state \e
2135 (max 100, source-track rule, max-src-nodes 75, \e
2136 max-src-states 3, tcp.established 60, tcp.closing 5)
2141 keyword is specified, the number of states per source IP is tracked.
2143 .Bl -tag -width xxxx -compact
2144 .It Ar source-track rule
2145 The maximum number of states created by this rule is limited by the rule's
2150 Only state entries created by this particular rule count toward the rule's
2152 .It Ar source-track global
2153 The number of states created by all rules that use this option is limited.
2154 Each rule can specify different
2158 options, however state entries created by any participating rule count towards
2159 each individual rule's limits.
2162 The following limits can be set:
2164 .Bl -tag -width xxxx -compact
2165 .It Ar max-src-nodes Aq Ar number
2166 Limits the maximum number of source addresses which can simultaneously
2167 have state table entries.
2168 .It Ar max-src-states Aq Ar number
2169 Limits the maximum number of simultaneous state entries that a single
2170 source address can create with this rule.
2173 For stateful TCP connections, limits on established connections (connections
2174 which have completed the TCP 3-way handshake) can also be enforced
2177 .Bl -tag -width xxxx -compact
2178 .It Ar max-src-conn Aq Ar number
2179 Limits the maximum number of simultaneous TCP connections which have
2180 completed the 3-way handshake that a single host can make.
2181 .It Xo Ar max-src-conn-rate Aq Ar number
2184 Limit the rate of new connections over a time interval.
2185 The connection rate is an approximation calculated as a moving average.
2188 Because the 3-way handshake ensures that the source address is not being
2189 spoofed, more aggressive action can be taken based on these limits.
2191 .Ar overload Aq Ar table
2192 state option, source IP addresses which hit either of the limits on
2193 established connections will be added to the named table.
2194 This table can be used in the ruleset to block further activity from
2195 the offending host, redirect it to a tarpit process, or restrict its
2200 keyword kills all states created by the matching rule which originate
2201 from the host which exceeds these limits.
2204 modifier to the flush command kills all states originating from the
2205 offending host, regardless of which rule created the state.
2207 For example, the following rules will protect the webserver against
2208 hosts making more than 100 connections in 10 seconds.
2209 Any host which connects faster than this rate will have its address added
2212 table and have all states originating from it flushed.
2213 Any new packets arriving from this host will be dropped unconditionally
2215 .Bd -literal -offset indent
2216 block quick from \*(Ltbad_hosts\*(Gt
2217 pass in on $ext_if proto tcp to $webserver port www keep state \e
2218 (max-src-conn-rate 100/10, overload \*(Ltbad_hosts\*(Gt flush global)
2220 .Sh OPERATING SYSTEM FINGERPRINTING
2221 Passive OS Fingerprinting is a mechanism to inspect nuances of a TCP
2222 connection's initial SYN packet and guess at the host's operating system.
2223 Unfortunately these nuances are easily spoofed by an attacker so the
2224 fingerprint is not useful in making security decisions.
2225 But the fingerprint is typically accurate enough to make policy decisions
2228 The fingerprints may be specified by operating system class, by
2229 version, or by subtype/patchlevel.
2230 The class of an operating system is typically the vendor or genre
2236 The version of the oldest available
2238 release on the main FTP site
2239 would be 2.6 and the fingerprint would be written
2241 .Dl \&"OpenBSD 2.6\&"
2243 The subtype of an operating system is typically used to describe the
2244 patchlevel if that patch led to changes in the TCP stack behavior.
2247 the only subtype is for a fingerprint that was
2250 scrub option and would be specified as
2252 .Dl \&"OpenBSD 3.3 no-df\&"
2254 Fingerprints for most popular operating systems are provided by
2258 is running, a complete list of known operating system fingerprints may
2259 be listed by running:
2263 Filter rules can enforce policy at any level of operating system specification
2264 assuming a fingerprint is present.
2265 Policy could limit traffic to approved operating systems or even ban traffic
2266 from hosts that aren't at the latest service pack.
2270 class can also be used as the fingerprint which will match packets for
2271 which no operating system fingerprint is known.
2274 .Bd -literal -offset indent
2275 pass out proto tcp from any os OpenBSD
2276 block out proto tcp from any os Doors
2277 block out proto tcp from any os "Doors PT"
2278 block out proto tcp from any os "Doors PT SP3"
2279 block out from any os "unknown"
2280 pass on lo0 proto tcp from any os "OpenBSD 3.3 lo0"
2283 Operating system fingerprinting is limited only to the TCP SYN packet.
2284 This means that it will not work on other protocols and will not match
2285 a currently established connection.
2287 Caveat: operating system fingerprints are occasionally wrong.
2288 There are three problems: an attacker can trivially craft his packets to
2289 appear as any operating system he chooses;
2290 an operating system patch could change the stack behavior and no fingerprints
2291 will match it until the database is updated;
2292 and multiple operating systems may have the same fingerprint.
2293 .Sh BLOCKING SPOOFED TRAFFIC
2294 "Spoofing" is the faking of IP addresses, typically for malicious
2298 directive expands to a set of filter rules which will block all
2299 traffic with a source IP from the network(s) directly connected
2300 to the specified interface(s) from entering the system through
2301 any other interface.
2303 For example, the line
2304 .Bd -literal -offset indent
2309 .Bd -literal -offset indent
2310 block drop in on ! lo0 inet from 127.0.0.1/8 to any
2311 block drop in on ! lo0 inet6 from ::1 to any
2314 For non-loopback interfaces, there are additional rules to block incoming
2315 packets with a source IP address identical to the interface's IP(s).
2316 For example, assuming the interface wi0 had an IP address of 10.0.0.1 and a
2317 netmask of 255.255.255.0,
2319 .Bd -literal -offset indent
2320 antispoof for wi0 inet
2324 .Bd -literal -offset indent
2325 block drop in on ! wi0 inet from 10.0.0.0/24 to any
2326 block drop in inet from 10.0.0.1 to any
2329 Caveat: Rules created by the
2331 directive interfere with packets sent over loopback interfaces
2333 One should pass these explicitly.
2334 .Sh FRAGMENT HANDLING
2335 The size of IP datagrams (packets) can be significantly larger than the
2336 maximum transmission unit (MTU) of the network.
2337 In cases when it is necessary or more efficient to send such large packets,
2338 the large packet will be fragmented into many smaller packets that will each
2340 Unfortunately for a firewalling device, only the first logical fragment will
2341 contain the necessary header information for the subprotocol that allows
2343 to filter on things such as TCP ports or to perform NAT.
2347 rules as described in
2348 .Sx TRAFFIC NORMALIZATION
2349 above, there are three options for handling fragments in the packet filter.
2351 One alternative is to filter individual fragments with filter rules.
2354 rule applies to a fragment, it is passed to the filter.
2355 Filter rules with matching IP header parameters decide whether the
2356 fragment is passed or blocked, in the same way as complete packets
2358 Without reassembly, fragments can only be filtered based on IP header
2359 fields (source/destination address, protocol), since subprotocol header
2360 fields are not available (TCP/UDP port numbers, ICMP code/type).
2363 option can be used to restrict filter rules to apply only to
2364 fragments, but not complete packets.
2365 Filter rules without the
2367 option still apply to fragments, if they only specify IP header fields.
2368 For instance, the rule
2369 .Bd -literal -offset indent
2370 pass in proto tcp from any to any port 80
2373 never applies to a fragment, even if the fragment is part of a TCP
2374 packet with destination port 80, because without reassembly this information
2375 is not available for each fragment.
2376 This also means that fragments cannot create new or match existing
2377 state table entries, which makes stateful filtering and address
2378 translation (NAT, redirection) for fragments impossible.
2380 It's also possible to reassemble only certain fragments by specifying
2381 source or destination addresses or protocols as parameters in
2385 In most cases, the benefits of reassembly outweigh the additional
2386 memory cost, and it's recommended to use
2389 all fragments via the
2390 .Ar fragment reassemble
2393 The memory allocated for fragment caching can be limited using
2395 Once this limit is reached, fragments that would have to be cached
2396 are dropped until other entries time out.
2397 The timeout value can also be adjusted.
2399 Currently, only IPv4 fragments are supported and IPv6 fragments
2400 are blocked unconditionally.
2402 Besides the main ruleset,
2404 can load rulesets into
2409 is a container that can hold rules, address tables, and other anchors.
2413 has a name which specifies the path where
2415 can be used to access the anchor to perform operations on it, such as
2416 attaching child anchors to it or loading rules into it.
2417 Anchors may be nested, with components separated by
2419 characters, similar to how file system hierarchies are laid out.
2420 The main ruleset is actually the default anchor, so filter and
2421 translation rules, for example, may also be contained in any anchor.
2423 An anchor can reference another
2426 using the following kinds
2428 .Bl -tag -width xxxx
2429 .It Ar nat-anchor Aq Ar name
2432 rules in the specified
2434 .It Ar rdr-anchor Aq Ar name
2437 rules in the specified
2439 .It Ar binat-anchor Aq Ar name
2442 rules in the specified
2444 .It Ar anchor Aq Ar name
2445 Evaluates the filter rules in the specified
2447 .It Xo Ar load anchor
2451 Loads the rules from the specified file into the
2456 When evaluation of the main ruleset reaches an
2460 will proceed to evaluate all rules specified in that anchor.
2462 Matching filter and translation rules marked with the
2464 option are final and abort the evaluation of the rules in other
2465 anchors and the main ruleset.
2468 itself is marked with the
2471 ruleset evaluation will terminate when the anchor is exited if the packet is
2472 matched by any rule within the anchor.
2475 rules are evaluated relative to the anchor in which they are contained.
2478 rules specified in the main ruleset will reference anchor
2479 attachment points underneath the main ruleset, and
2481 rules specified in a file loaded from a
2483 rule will be attached under that anchor point.
2485 Rules may be contained in
2487 attachment points which do not contain any rules when the main ruleset
2488 is loaded, and later such anchors can be manipulated through
2490 without reloading the main ruleset or other anchors.
2492 .Bd -literal -offset indent
2494 block on $ext_if all
2496 pass out on $ext_if all
2497 pass in on $ext_if proto tcp from any \e
2498 to $ext_if port smtp
2501 blocks all packets on the external interface by default, then evaluates
2504 named "spam", and finally passes all outgoing connections and
2505 incoming connections to port 25.
2506 .Bd -literal -offset indent
2507 # echo \&"block in quick from 1.2.3.4 to any\&" \&| \e
2511 This loads a single rule into the
2513 which blocks all packets from a specific address.
2515 The anchor can also be populated by adding a
2520 .Bd -literal -offset indent
2522 load anchor spam from "/etc/pf-spam.conf"
2529 it will also load all the rules from the file
2530 .Pa /etc/pf-spam.conf
2535 rules can specify packet filtering parameters using the same syntax as
2537 When parameters are used, the
2539 rule is only evaluated for matching packets.
2540 This allows conditional evaluation of anchors, like:
2541 .Bd -literal -offset indent
2542 block on $ext_if all
2543 anchor spam proto tcp from any to any port smtp
2544 pass out on $ext_if all
2545 pass in on $ext_if proto tcp from any to $ext_if port smtp
2550 spam are only evaluated for
2552 packets with destination port 25.
2554 .Bd -literal -offset indent
2555 # echo \&"block in quick from 1.2.3.4 to any" \&| \e
2559 will only block connections from 1.2.3.4 to port 25.
2561 Anchors may end with the asterisk
2563 character, which signifies that all anchors attached at that point
2564 should be evaluated in the alphabetical ordering of their anchor name.
2566 .Bd -literal -offset indent
2570 will evaluate each rule in each anchor attached to the
2573 Note that it will only evaluate anchors that are directly attached to the
2575 anchor, and will not descend to evaluate anchors recursively.
2577 Since anchors are evaluated relative to the anchor in which they are
2578 contained, there is a mechanism for accessing the parent and ancestor
2579 anchors of a given anchor.
2580 Similar to file system path name resolution, if the sequence
2582 appears as an anchor path component, the parent anchor of the current
2583 anchor in the path evaluation at that point will become the new current
2585 As an example, consider the following:
2586 .Bd -literal -offset indent
2587 # echo ' anchor "spam/allowed" ' | pfctl -f -
2588 # echo -e ' anchor "../banned" \en pass' | \e
2589 pfctl -a spam/allowed -f -
2592 Evaluation of the main ruleset will lead into the
2594 anchor, which will evaluate the rules in the
2596 anchor, if any, before finally evaluating the
2602 can also be loaded inline in the ruleset within a brace ('{' '}') delimited
2604 Brace delimited blocks may contain rules or other brace-delimited blocks.
2605 When anchors are loaded this way the anchor name becomes optional.
2606 .Bd -literal -offset indent
2607 anchor "external" on egress {
2610 pass proto tcp from any to port { 25, 80, 443 }
2612 pass in proto tcp to any port 22
2616 Since the parser specification for anchor names is a string, any
2617 reference to an anchor name containing
2619 characters will require double quote
2621 characters around the anchor name.
2622 .Sh TRANSLATION EXAMPLES
2623 This example maps incoming requests on port 80 to port 8080, on
2624 which a daemon is running (because, for example, it is not run as root,
2625 and therefore lacks permission to bind to port 80).
2627 # use a macro for the interface name, so it can be changed easily
2630 # map daemon on 8080 to appear to be on 80
2631 rdr on $ext_if proto tcp from any to any port 80 -\*(Gt 127.0.0.1 port 8080
2636 modifier is given, packets matching the translation rule are passed without
2637 inspecting the filter rules:
2639 rdr pass on $ext_if proto tcp from any to any port 80 -\*(Gt 127.0.0.1 \e
2643 In the example below, vlan12 is configured as 192.168.168.1;
2644 the machine translates all packets coming from 192.168.168.0/24 to 204.92.77.111
2645 when they are going out any interface except vlan12.
2646 This has the net effect of making traffic from the 192.168.168.0/24
2647 network appear as though it is the Internet routable address
2648 204.92.77.111 to nodes behind any interface on the router except
2649 for the nodes on vlan12.
2650 (Thus, 192.168.168.1 can talk to the 192.168.168.0/24 nodes.)
2652 nat on ! vlan12 from 192.168.168.0/24 to any -\*(Gt 204.92.77.111
2655 In the example below, the machine sits between a fake internal 144.19.74.*
2656 network, and a routable external IP of 204.92.77.100.
2659 rule excludes protocol AH from being translated.
2662 no nat on $ext_if proto ah from 144.19.74.0/24 to any
2663 nat on $ext_if from 144.19.74.0/24 to any -\*(Gt 204.92.77.100
2666 In the example below, packets bound for one specific server, as well as those
2667 generated by the sysadmins are not proxied; all other connections are.
2670 no rdr on $int_if proto { tcp, udp } from any to $server port 80
2671 no rdr on $int_if proto { tcp, udp } from $sysadmins to any port 80
2672 rdr on $int_if proto { tcp, udp } from any to any port 80 -\*(Gt 127.0.0.1 \e
2676 This longer example uses both a NAT and a redirection.
2677 The external interface has the address 157.161.48.183.
2678 On localhost, we are running
2680 waiting for FTP sessions to be redirected to it.
2681 The three mandatory anchors for
2683 are omitted from this example; see the
2688 # Translate outgoing packets' source addresses (any protocol).
2689 # In this case, any address but the gateway's external address is mapped.
2690 nat on $ext_if inet from ! ($ext_if) to any -\*(Gt ($ext_if)
2693 # Map outgoing packets' source port to an assigned proxy port instead of
2694 # an arbitrary port.
2695 # In this case, proxy outgoing isakmp with port 500 on the gateway.
2696 nat on $ext_if inet proto udp from any port = isakmp to any -\*(Gt ($ext_if) \e
2700 # Translate outgoing packets' source address (any protocol).
2701 # Translate incoming packets' destination address to an internal machine
2703 binat on $ext_if from 10.1.2.150 to any -\*(Gt $ext_if
2706 # Translate incoming packets' destination addresses.
2707 # As an example, redirect a TCP and UDP port to an internal machine.
2708 rdr on $ext_if inet proto tcp from any to ($ext_if) port 8080 \e
2709 -\*(Gt 10.1.2.151 port 22
2710 rdr on $ext_if inet proto udp from any to ($ext_if) port 8080 \e
2711 -\*(Gt 10.1.2.151 port 53
2714 # Translate outgoing ftp control connections to send them to localhost
2715 # for proxying with ftp-proxy(8) running on port 8021.
2716 rdr on $int_if proto tcp from any to any port 21 -\*(Gt 127.0.0.1 port 8021
2719 In this example, a NAT gateway is set up to translate internal addresses
2720 using a pool of public addresses (192.0.2.16/28) and to redirect
2721 incoming web server connections to a group of web servers on the internal
2725 # Translate outgoing packets' source addresses using an address pool.
2726 # A given source address is always translated to the same pool address by
2727 # using the source-hash keyword.
2728 nat on $ext_if inet from any to any -\*(Gt 192.0.2.16/28 source-hash
2731 # Translate incoming web server connections to a group of web servers on
2732 # the internal network.
2733 rdr on $ext_if proto tcp from any to any port 80 \e
2734 -\*(Gt { 10.1.2.155, 10.1.2.160, 10.1.2.161 } round-robin
2738 # The external interface is kue0
2739 # (157.161.48.183, the only routable address)
2740 # and the private network is 10.0.0.0/8, for which we are doing NAT.
2742 # use a macro for the interface name, so it can be changed easily
2745 # normalize all incoming traffic
2746 scrub in on $ext_if all fragment reassemble
2748 # block and log everything by default
2749 block return log on $ext_if all
2751 # block anything coming from source we have no back routes for
2752 block in from no-route to any
2754 # block packets whose ingress interface does not match the one in
2755 # the route back to their source address
2756 block in from urpf-failed to any
2758 # block and log outgoing packets that do not have our address as source,
2759 # they are either spoofed or something is misconfigured (NAT disabled,
2760 # for instance), we want to be nice and do not send out garbage.
2761 block out log quick on $ext_if from ! 157.161.48.183 to any
2763 # silently drop broadcasts (cable modem noise)
2764 block in quick on $ext_if from any to 255.255.255.255
2766 # block and log incoming packets from reserved address space and invalid
2767 # addresses, they are either spoofed or misconfigured, we cannot reply to
2768 # them anyway (hence, no return-rst).
2769 block in log quick on $ext_if from { 10.0.0.0/8, 172.16.0.0/12, \e
2770 192.168.0.0/16, 255.255.255.255/32 } to any
2774 # pass out/in certain ICMP queries and keep state (ping)
2775 # state matching is done on host addresses and ICMP id (not type/code),
2776 # so replies (like 0/0 for 8/0) will match queries
2777 # ICMP error messages (which always refer to a TCP/UDP packet) are
2778 # handled by the TCP/UDP states
2779 pass on $ext_if inet proto icmp all icmp-type 8 code 0
2783 # pass out all UDP connections and keep state
2784 pass out on $ext_if proto udp all
2786 # pass in certain UDP connections and keep state (DNS)
2787 pass in on $ext_if proto udp from any to any port domain
2791 # pass out all TCP connections and modulate state
2792 pass out on $ext_if proto tcp all modulate state
2794 # pass in certain TCP connections and keep state (SSH, SMTP, DNS, IDENT)
2795 pass in on $ext_if proto tcp from any to any port { ssh, smtp, domain, \e
2798 # Do not allow Windows 9x SMTP connections since they are typically
2799 # a viral worm. Alternately we could limit these OSes to 1 connection each.
2800 block in on $ext_if proto tcp from any os {"Windows 95", "Windows 98"} \e
2804 # pass in/out all IPv6 traffic: note that we have to enable this in two
2805 # different ways, on both our physical interface and our tunnel
2806 pass quick on gif0 inet6
2807 pass quick on $ext_if proto ipv6
2811 # three interfaces: $int_if, $ext_if, and $wifi_if (wireless). NAT is
2812 # being done on $ext_if for all outgoing packets. tag packets in on
2813 # $int_if and pass those tagged packets out on $ext_if. all other
2814 # outgoing packets (i.e., packets from the wireless network) are only
2815 # permitted to access port 80.
2817 pass in on $int_if from any to any tag INTNET
2818 pass in on $wifi_if from any to any
2820 block out on $ext_if from any to any
2821 pass out quick on $ext_if tagged INTNET
2822 pass out on $ext_if proto tcp from any to any port 80
2824 # tag incoming packets as they are redirected to spamd(8). use the tag
2825 # to pass those packets through the packet filter.
2827 rdr on $ext_if inet proto tcp from \*(Ltspammers\*(Gt to port smtp \e
2828 tag SPAMD -\*(Gt 127.0.0.1 port spamd
2831 pass in on $ext_if inet proto tcp tagged SPAMD
2838 line = ( option | pf-rule | nat-rule | binat-rule | rdr-rule |
2839 antispoof-rule | altq-rule | queue-rule | trans-anchors |
2840 anchor-rule | anchor-close | load-anchor | table-rule |
2843 option = "set" ( [ "timeout" ( timeout | "{" timeout-list "}" ) ] |
2844 [ "ruleset-optimization" [ "none" | "basic" | "profile" ]] |
2845 [ "optimization" [ "default" | "normal" |
2846 "high-latency" | "satellite" |
2847 "aggressive" | "conservative" ] ]
2848 [ "limit" ( limit-item | "{" limit-list "}" ) ] |
2849 [ "loginterface" ( interface-name | "none" ) ] |
2850 [ "block-policy" ( "drop" | "return" ) ] |
2851 [ "state-policy" ( "if-bound" | "floating" ) ]
2852 [ "state-defaults" state-opts ]
2853 [ "require-order" ( "yes" | "no" ) ]
2854 [ "fingerprints" filename ] |
2855 [ "skip on" ifspec ] |
2856 [ "debug" ( "none" | "urgent" | "misc" | "loud" ) ] )
2858 pf-rule = action [ ( "in" | "out" ) ]
2859 [ "log" [ "(" logopts ")"] ] [ "quick" ]
2860 [ "on" ifspec ] [ "fastroute" | route ] [ af ] [ protospec ]
2861 hosts [ filteropt-list ]
2863 logopts = logopt [ "," logopts ]
2864 logopt = "all" | "user" | "to" interface-name
2866 filteropt-list = filteropt-list filteropt | filteropt
2867 filteropt = user | group | flags | icmp-type | icmp6-type | "tos" tos |
2868 ( "no" | "keep" | "modulate" | "synproxy" ) "state"
2869 [ "(" state-opts ")" ] |
2870 "fragment" | "no-df" | "min-ttl" number | "set-tos" tos |
2871 "max-mss" number | "random-id" | "reassemble tcp" |
2872 fragmentation | "allow-opts" |
2873 "label" string | "tag" string | [ ! ] "tagged" string |
2874 "queue" ( string | "(" string [ [ "," ] string ] ")" ) |
2875 "rtable" number | "probability" number"%"
2877 nat-rule = [ "no" ] "nat" [ "pass" [ "log" [ "(" logopts ")" ] ] ]
2878 [ "on" ifspec ] [ af ]
2879 [ protospec ] hosts [ "tag" string ] [ "tagged" string ]
2880 [ "-\*(Gt" ( redirhost | "{" redirhost-list "}" )
2881 [ portspec ] [ pooltype ] [ "static-port" ] ]
2883 binat-rule = [ "no" ] "binat" [ "pass" [ "log" [ "(" logopts ")" ] ] ]
2884 [ "on" interface-name ] [ af ]
2885 [ "proto" ( proto-name | proto-number ) ]
2886 "from" address [ "/" mask-bits ] "to" ipspec
2887 [ "tag" string ] [ "tagged" string ]
2888 [ "-\*(Gt" address [ "/" mask-bits ] ]
2890 rdr-rule = [ "no" ] "rdr" [ "pass" [ "log" [ "(" logopts ")" ] ] ]
2891 [ "on" ifspec ] [ af ]
2892 [ protospec ] hosts [ "tag" string ] [ "tagged" string ]
2893 [ "-\*(Gt" ( redirhost | "{" redirhost-list "}" )
2894 [ portspec ] [ pooltype ] ]
2896 antispoof-rule = "antispoof" [ "log" ] [ "quick" ]
2897 "for" ifspec [ af ] [ "label" string ]
2899 table-rule = "table" "\*(Lt" string "\*(Gt" [ tableopts-list ]
2900 tableopts-list = tableopts-list tableopts | tableopts
2901 tableopts = "persist" | "const" | "counters" | "file" string |
2902 "{" [ tableaddr-list ] "}"
2903 tableaddr-list = tableaddr-list [ "," ] tableaddr-spec | tableaddr-spec
2904 tableaddr-spec = [ "!" ] tableaddr [ "/" mask-bits ]
2905 tableaddr = hostname | ifspec | "self" |
2906 ipv4-dotted-quad | ipv6-coloned-hex
2908 altq-rule = "altq on" interface-name queueopts-list
2910 queue-rule = "queue" string [ "on" interface-name ] queueopts-list
2913 anchor-rule = "anchor" [ string ] [ ( "in" | "out" ) ] [ "on" ifspec ]
2914 [ af ] [ protospec ] [ hosts ] [ filteropt-list ] [ "{" ]
2918 trans-anchors = ( "nat-anchor" | "rdr-anchor" | "binat-anchor" ) string
2919 [ "on" ifspec ] [ af ] [ "proto" ] [ protospec ] [ hosts ]
2921 load-anchor = "load anchor" string "from" filename
2923 queueopts-list = queueopts-list queueopts | queueopts
2924 queueopts = [ "bandwidth" bandwidth-spec ] |
2925 [ "qlimit" number ] | [ "tbrsize" number ] |
2926 [ "priority" number ] | [ schedulers ]
2927 schedulers = ( cbq-def | priq-def | hfsc-def )
2928 bandwidth-spec = "number" ( "b" | "Kb" | "Mb" | "Gb" | "%" )
2930 action = "pass" | "block" [ return ] | [ "no" ] "scrub"
2931 return = "drop" | "return" | "return-rst" [ "( ttl" number ")" ] |
2932 "return-icmp" [ "(" icmpcode [ [ "," ] icmp6code ] ")" ] |
2933 "return-icmp6" [ "(" icmp6code ")" ]
2934 icmpcode = ( icmp-code-name | icmp-code-number )
2935 icmp6code = ( icmp6-code-name | icmp6-code-number )
2937 ifspec = ( [ "!" ] ( interface-name | interface-group ) ) |
2938 "{" interface-list "}"
2939 interface-list = [ "!" ] ( interface-name | interface-group )
2940 [ [ "," ] interface-list ]
2941 route = ( "route-to" | "reply-to" | "dup-to" )
2942 ( routehost | "{" routehost-list "}" )
2944 af = "inet" | "inet6"
2946 protospec = "proto" ( proto-name | proto-number |
2947 "{" proto-list "}" )
2948 proto-list = ( proto-name | proto-number ) [ [ "," ] proto-list ]
2951 "from" ( "any" | "no-route" | "urpf-failed" | "self" | host |
2952 "{" host-list "}" | "route" string ) [ port ] [ os ]
2953 "to" ( "any" | "no-route" | "self" | host |
2954 "{" host-list "}" | "route" string ) [ port ]
2956 ipspec = "any" | host | "{" host-list "}"
2957 host = [ "!" ] ( address [ "/" mask-bits ] | "\*(Lt" string "\*(Gt" )
2958 redirhost = address [ "/" mask-bits ]
2959 routehost = "(" interface-name [ address [ "/" mask-bits ] ] ")"
2960 address = ( interface-name | interface-group |
2961 "(" ( interface-name | interface-group ) ")" |
2962 hostname | ipv4-dotted-quad | ipv6-coloned-hex )
2963 host-list = host [ [ "," ] host-list ]
2964 redirhost-list = redirhost [ [ "," ] redirhost-list ]
2965 routehost-list = routehost [ [ "," ] routehost-list ]
2967 port = "port" ( unary-op | binary-op | "{" op-list "}" )
2968 portspec = "port" ( number | name ) [ ":" ( "*" | number | name ) ]
2969 os = "os" ( os-name | "{" os-list "}" )
2970 user = "user" ( unary-op | binary-op | "{" op-list "}" )
2971 group = "group" ( unary-op | binary-op | "{" op-list "}" )
2973 unary-op = [ "=" | "!=" | "\*(Lt" | "\*(Le" | "\*(Gt" | "\*(Ge" ]
2975 binary-op = number ( "\*(Lt\*(Gt" | "\*(Gt\*(Lt" | ":" ) number
2976 op-list = ( unary-op | binary-op ) [ [ "," ] op-list ]
2978 os-name = operating-system-name
2979 os-list = os-name [ [ "," ] os-list ]
2981 flags = "flags" ( [ flag-set ] "/" flag-set | "any" )
2982 flag-set = [ "F" ] [ "S" ] [ "R" ] [ "P" ] [ "A" ] [ "U" ] [ "E" ]
2985 icmp-type = "icmp-type" ( icmp-type-code | "{" icmp-list "}" )
2986 icmp6-type = "icmp6-type" ( icmp-type-code | "{" icmp-list "}" )
2987 icmp-type-code = ( icmp-type-name | icmp-type-number )
2988 [ "code" ( icmp-code-name | icmp-code-number ) ]
2989 icmp-list = icmp-type-code [ [ "," ] icmp-list ]
2991 tos = ( "lowdelay" | "throughput" | "reliability" |
2994 state-opts = state-opt [ [ "," ] state-opts ]
2995 state-opt = ( "max" number | "no-sync" | timeout | "sloppy" | "pflow" |
2996 "source-track" [ ( "rule" | "global" ) ] |
2997 "max-src-nodes" number | "max-src-states" number |
2998 "max-src-conn" number |
2999 "max-src-conn-rate" number "/" number |
3000 "overload" "\*(Lt" string "\*(Gt" [ "flush" ] |
3001 "if-bound" | "floating" )
3003 fragmentation = [ "fragment reassemble" | "fragment crop" |
3004 "fragment drop-ovl" ]
3006 timeout-list = timeout [ [ "," ] timeout-list ]
3007 timeout = ( "tcp.first" | "tcp.opening" | "tcp.established" |
3008 "tcp.closing" | "tcp.finwait" | "tcp.closed" |
3009 "udp.first" | "udp.single" | "udp.multiple" |
3010 "icmp.first" | "icmp.error" |
3011 "other.first" | "other.single" | "other.multiple" |
3012 "frag" | "interval" | "src.track" |
3013 "adaptive.start" | "adaptive.end" ) number
3015 limit-list = limit-item [ [ "," ] limit-list ]
3016 limit-item = ( "states" | "frags" | "src-nodes" ) number
3018 pooltype = ( "bitmask" | "random" |
3019 "source-hash" [ ( hex-key | string-key ) ] |
3020 "round-robin" ) [ sticky-address ]
3022 subqueue = string | "{" queue-list "}"
3023 queue-list = string [ [ "," ] string ]
3024 cbq-def = "cbq" [ "(" cbq-opt [ [ "," ] cbq-opt ] ")" ]
3025 priq-def = "priq" [ "(" priq-opt [ [ "," ] priq-opt ] ")" ]
3026 hfsc-def = "hfsc" [ "(" hfsc-opt [ [ "," ] hfsc-opt ] ")" ]
3027 cbq-opt = ( "default" | "borrow" | "red" | "ecn" | "rio" )
3028 priq-opt = ( "default" | "red" | "ecn" | "rio" )
3029 hfsc-opt = ( "default" | "red" | "ecn" | "rio" |
3030 linkshare-sc | realtime-sc | upperlimit-sc )
3031 linkshare-sc = "linkshare" sc-spec
3032 realtime-sc = "realtime" sc-spec
3033 upperlimit-sc = "upperlimit" sc-spec
3034 sc-spec = ( bandwidth-spec |
3035 "(" bandwidth-spec number bandwidth-spec ")" )
3036 include = "include" filename
3039 .Bl -tag -width "/etc/protocols" -compact
3043 Default location of the ruleset file.
3045 Default location of OS fingerprints.
3046 .It Pa /etc/protocols
3047 Protocol name database.
3048 .It Pa /etc/services
3049 Service name database.
3052 Due to a lock order reversal (LOR) with the socket layer, the use of the
3056 filter parameter in conjuction with a Giant-free netstack
3057 can result in a deadlock.
3058 A workaround is available under the
3059 .Va debug.pfugidhack
3060 sysctl which is automatically enabled when a
3068 Route labels are not supported by the
3072 Rules with a route label do not match any traffic.
3097 file format first appeared in