9 .Nd IP firewall and traffic shaper control program
19 .Op Ar rule | first-last ...
27 .Brq Cm delete | zero | resetlog
31 .Brq Cm firewall | altq | one_pass | debug | verbose | dyn_keepalive
34 .Brq Cm firewall | altq | one_pass | debug | verbose | dyn_keepalive
37 .Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
41 .Ar number Cm to Ar number
43 .Cm set swap Ar number number
48 .Cm table Ar number Cm add Ar addr Ns Oo / Ns Ar masklen Oc Op Ar value
50 .Cm table Ar number Cm delete Ar addr Ns Op / Ns Ar masklen
68 .Brq Cm delete | list | show
90 utility is the user interface for controlling the
103 numbered from 1 to 65535.
104 Packets are passed to
106 from a number of different places in the protocol stack
107 (depending on the source and destination of the packet,
110 is invoked multiple times on the same packet).
111 The packet passed to the firewall is compared
112 against each of the rules in the firewall
114 When a match is found, the action corresponding to the
115 matching rule is performed.
117 Depending on the action and certain system settings, packets
118 can be reinjected into the firewall at some rule after the
119 matching one for further processing.
123 ruleset always includes a
125 rule (numbered 65535) which cannot be modified or deleted,
126 and matches all packets.
127 The action associated with the
133 depending on how the kernel is configured.
135 If the ruleset includes one or more rules with the
143 behaviour, i.e., upon a match it will create dynamic rules matching
144 the exact parameters (source and destination addresses and ports)
145 of the matching packet.
147 These dynamic rules, which have a limited lifetime, are checked
148 at the first occurrence of a
153 rule, and are typically used to open the firewall on-demand to
154 legitimate traffic only.
156 .Sx STATEFUL FIREWALL
159 Sections below for more information on the stateful behaviour of
162 All rules (including dynamic ones) have a few associated counters:
163 a packet count, a byte count, a log count and a timestamp
164 indicating the time of the last match.
165 Counters can be displayed or reset with
169 Each rule belongs to one of 32 different
173 commands to atomically manipulate sets, such as enable,
174 disable, swap sets, move all rules in a set to another
175 one, delete all rules in a set.
176 These can be useful to
177 install temporary configurations, or to test them.
180 for more information on
184 Rules can be added with the
186 command; deleted individually or in groups with the
188 command, and globally (except those in set 31) with the
190 command; displayed, optionally with the content of the
196 Finally, counters can be reset with the
203 The following general options are available when invoking
205 .Bl -tag -width indent
207 Show counter values when listing rules.
210 command implies this option.
212 Only show the action and the comment, not the body of a rule.
216 When entering or showing rules, print them in compact form,
217 i.e., omitting the "ip from any to any" string
218 when this does not carry any additional information.
220 When listing, show dynamic rules in addition to static ones.
224 is specified, also show expired dynamic rules.
226 Do not ask for confirmation for commands that can cause problems
229 If there is no tty associated with the process, this is implied.
231 When listing a table (see the
233 section below for more information on lookup tables), format values
234 as IP addresses. By default, values are shown as integers.
236 Only check syntax of the command strings, without actually passing
239 Try to resolve addresses and service names in output.
241 Be quiet when executing the
251 This is useful when updating rulesets by executing multiple
255 .Ql sh\ /etc/rc.firewall ) ,
256 or by processing a file with many
258 rules across a remote login session.
259 It also stops a table add or delete
260 from failing if the entry already exists or is not present.
262 The reason why this option may be important is that
263 for some of these actions,
265 may print a message; if the action results in blocking the
266 traffic to the remote client,
267 the remote login session will be closed
268 and the rest of the ruleset will not be processed.
269 Access to the console would then be required to recover.
271 When listing rules, show the
273 each rule belongs to.
274 If this flag is not specified, disabled rules will not be
277 When listing pipes, sort according to one of the four
278 counters (total or current packets or bytes).
280 When listing, show last match timestamp converted with ctime().
282 When listing, show last match timestamp as seconds from the epoch.
283 This form can be more convenient for postprocessing by scripts.
286 To ease configuration, rules can be put into a file which is
289 as shown in the last synopsis line.
293 The file will be read line by line and applied as arguments to the
297 Optionally, a preprocessor can be specified using
301 is to be piped through.
302 Useful preprocessors include
308 does not start with a slash
310 as its first character, the usual
312 name search is performed.
313 Care should be taken with this in environments where not all
314 file systems are mounted (yet) by the time
316 is being run (e.g.\& when they are mounted over NFS).
319 has been specified, any additional arguments are passed on to the preprocessor
321 This allows for flexible configuration files (like conditionalizing
322 them on the local hostname) and the use of macros to centralize
323 frequently required arguments like IP addresses.
330 commands are used to configure the traffic shaper, as shown in the
331 .Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
334 If the world and the kernel get out of sync the
336 ABI may break, preventing you from being able to add any rules.
338 adversely effect the booting process.
343 to temporarily disable the firewall to regain access to the network,
344 allowing you to fix the problem.
346 A packet is checked against the active ruleset in multiple places
347 in the protocol stack, under control of several sysctl variables.
348 These places and variables are shown below, and it is important to
349 have this picture in mind in order to design a correct ruleset.
350 .Bd -literal -offset indent
353 +----------->-----------+
355 [ip(6)_input] [ip(6)_output] net.inet(6).ip(6).fw.enable=1
358 [ether_demux] [ether_output_frame] net.link.ether.ipfw=1
360 +-->--[bdg_forward]-->--+ net.link.bridge.ipfw=1
365 As can be noted from the above picture, the number of
366 times the same packet goes through the firewall can
367 vary between 0 and 4 depending on packet source and
368 destination, and system configuration.
370 Note that as packets flow through the stack, headers can be
371 stripped or added to it, and so they may or may not be available
373 E.g., incoming packets will include the MAC header when
377 but the same packets will have the MAC header stripped off when
384 Also note that each packet is always checked against the complete ruleset,
385 irrespective of the place where the check occurs, or the source of the packet.
386 If a rule contains some match patterns or actions which are not valid
387 for the place of invocation (e.g.\& trying to match a MAC header within
391 the match pattern will not match, but a
393 operator in front of such patterns
397 match on those packets.
398 It is thus the responsibility of
399 the programmer, if necessary, to write a suitable ruleset to
400 differentiate among the possible places.
402 rules can be useful here, as an example:
403 .Bd -literal -offset indent
404 # packets from ether_demux or bdg_forward
405 ipfw add 10 skipto 1000 all from any to any layer2 in
406 # packets from ip_input
407 ipfw add 10 skipto 2000 all from any to any not layer2 in
408 # packets from ip_output
409 ipfw add 10 skipto 3000 all from any to any not layer2 out
410 # packets from ether_output_frame
411 ipfw add 10 skipto 4000 all from any to any layer2 out
414 (yes, at the moment there is no way to differentiate between
415 ether_demux and bdg_forward).
417 In general, each keyword or argument must be provided as
418 a separate command line argument, with no leading or trailing
420 Keywords are case-sensitive, whereas arguments may
421 or may not be case-sensitive depending on their nature
422 (e.g.\& uid's are, hostnames are not).
426 you can introduce spaces after commas ',' to make
427 the line more readable.
428 You can also put the entire
429 command (including flags) into a single argument.
430 E.g., the following forms are equivalent:
431 .Bd -literal -offset indent
432 ipfw -q add deny src-ip 10.0.0.0/24,127.0.0.1/8
433 ipfw -q add deny src-ip 10.0.0.0/24, 127.0.0.1/8
434 ipfw "-q add deny src-ip 10.0.0.0/24, 127.0.0.1/8"
439 rules is the following:
440 .Bd -ragged -offset indent
443 .Op Cm set Ar set_number
444 .Op Cm prob Ar match_probability
446 .Op Cm log Op Cm logamount Ar number
456 where the body of the rule specifies which information is used
457 for filtering packets, among the following:
459 .Bl -tag -width "Source and dest. addresses and ports" -offset XXX -compact
460 .It Layer-2 header fields
462 .It IPv4 and IPv6 Protocol
464 .It Source and dest. addresses and ports
468 .It Transmit and receive interface
470 .It Misc. IP header fields
471 Version, type of service, datagram length, identification,
472 fragment flag (non-zero IP offset),
475 .It IPv6 Extension headers
476 Fragmentation, Hop-by-Hop options,
477 Routing Headers, Source routing rthdr0, Mobile IPv6 rthdr2, IPSec options.
479 .It Misc. TCP header fields
480 TCP flags (SYN, FIN, ACK, RST, etc.),
481 sequence number, acknowledgment number,
489 When the packet can be associated with a local socket.
491 Whether a packet came from a divert socket (e.g.,
493 .It Fib annotation state
494 Whether a packet has been tagged for using a specific FIB (routing table)
495 in future forwarding decisions.
498 Note that some of the above information, e.g.\& source MAC or IP addresses and
499 TCP/UDP ports, could easily be spoofed, so filtering on those fields
500 alone might not guarantee the desired results.
501 .Bl -tag -width indent
503 Each rule is associated with a
505 in the range 1..65535, with the latter reserved for the
508 Rules are checked sequentially by rule number.
509 Multiple rules can have the same number, in which case they are
510 checked (and listed) according to the order in which they have
512 If a rule is entered without specifying a number, the kernel will
513 assign one in such a way that the rule becomes the last one
517 Automatic rule numbers are assigned by incrementing the last
518 non-default rule number by the value of the sysctl variable
519 .Ar net.inet.ip.fw.autoinc_step
520 which defaults to 100.
521 If this is not possible (e.g.\& because we would go beyond the
522 maximum allowed rule number), the number of the last
523 non-default value is used instead.
524 .It Cm set Ar set_number
525 Each rule is associated with a
528 Sets can be individually disabled and enabled, so this parameter
529 is of fundamental importance for atomic ruleset manipulation.
530 It can be also used to simplify deletion of groups of rules.
531 If a rule is entered without specifying a set number,
534 Set 31 is special in that it cannot be disabled,
535 and rules in set 31 are not deleted by the
537 command (but you can delete them with the
538 .Nm ipfw delete set 31
540 Set 31 is also used for the
543 .It Cm prob Ar match_probability
544 A match is only declared with the specified probability
545 (floating point number between 0 and 1).
546 This can be useful for a number of applications such as
547 random packet drop or
550 to simulate the effect of multiple paths leading to out-of-order
553 Note: this condition is checked before any other condition, including
554 ones such as keep-state or check-state which might have side effects.
555 .It Cm log Op Cm logamount Ar number
556 When a packet matches a rule with the
558 keyword, a message will be
564 The logging only occurs if the sysctl variable
565 .Va net.inet.ip.fw.verbose
567 (which is the default when the kernel is compiled with
568 .Dv IPFIREWALL_VERBOSE )
569 and the number of packets logged so far for that
570 particular rule does not exceed the
575 is specified, the limit is taken from the sysctl variable
576 .Va net.inet.ip.fw.verbose_limit .
577 In both cases, a value of 0 removes the logging limit.
579 Once the limit is reached, logging can be re-enabled by
580 clearing the logging counter or the packet counter for that entry, see the
584 Note: logging is done after all other packet matching conditions
585 have been successfully verified, and before performing the final
586 action (accept, deny, etc.) on the packet.
588 When a packet matches a rule with the
590 keyword, the numeric tag for the given
592 in the range 1..65534 will be attached to the packet.
593 The tag acts as an internal marker (it is not sent out over
594 the wire) that can be used to identify these packets later on.
595 This can be used, for example, to provide trust between interfaces
596 and to start doing policy-based filtering.
597 A packet can have multiple tags at the same time.
598 Tags are "sticky", meaning once a tag is applied to a packet by a
599 matching rule it exists until explicit removal.
600 Tags are kept with the packet everywhere within the kernel, but are
601 lost when packet leaves the kernel, for example, on transmitting
602 packet out to the network or sending packet to a
606 To check for previously applied tags, use the
609 To delete previously applied tag, use the
613 Note: since tags are kept with the packet everywhere in kernelspace,
614 they can be set and unset anywhere in the kernel network subsystem
617 facility), not only by means of the
623 For example, there can be a specialized
625 node doing traffic analyzing and tagging for later inspecting
627 .It Cm untag Ar number
628 When a packet matches a rule with the
630 keyword, the tag with the number
632 is searched among the tags attached to this packet and,
633 if found, removed from it.
634 Other tags bound to packet, if present, are left untouched.
636 When a packet matches a rule with the
638 keyword, the ALTQ identifier for the given
643 Note that this ALTQ tag is only meaningful for packets going "out" of IPFW,
644 and not being rejected or going to divert sockets.
645 Note that if there is insufficient memory at the time the packet is
646 processed, it will not be tagged, so it is wise to make your ALTQ
647 "default" queue policy account for this.
650 rules match a single packet, only the first one adds the ALTQ classification
652 In doing so, traffic may be shaped by using
653 .Cm count Cm altq Ar queue
654 rules for classification early in the ruleset, then later applying
655 the filtering decision.
660 rules may come later and provide the actual filtering decisions in
661 addition to the fallback ALTQ tag.
665 to set up the queues before IPFW will be able to look them up by name,
666 and if the ALTQ disciplines are rearranged, the rules in containing the
667 queue identifiers in the kernel will likely have gone stale and need
669 Stale queue identifiers will probably result in misclassification.
671 All system ALTQ processing can be turned on or off via
676 .Cm disable Ar altq .
678 .Va net.inet.ip.fw.one_pass
679 is irrelevant to ALTQ traffic shaping, as the actual rule action is followed
680 always after adding an ALTQ tag.
683 A rule can be associated with one of the following actions, which
684 will be executed when the packet matches the body of the rule.
685 .Bl -tag -width indent
686 .It Cm allow | accept | pass | permit
687 Allow packets that match rule.
688 The search terminates.
690 Checks the packet against the dynamic ruleset.
691 If a match is found, execute the action associated with
692 the rule which generated this dynamic rule, otherwise
693 move to the next rule.
696 rules do not have a body.
699 rule is found, the dynamic ruleset is checked at the first
705 Update counters for all packets that match rule.
706 The search continues with the next rule.
708 Discard packets that match this rule.
709 The search terminates.
710 .It Cm divert Ar port
711 Divert packets that match this rule to the
715 The search terminates.
716 .It Cm fwd | forward Ar ipaddr | tablearg Ns Op , Ns Ar port
717 Change the next-hop on matching packets to
719 which can be an IP address or a host name.
720 The next hop can also be supplied by the last table
721 looked up for the packet by using the
723 keyword instead of an explicit address.
724 The search terminates if this rule matches.
728 is a local address, then matching packets will be forwarded to
730 (or the port number in the packet if one is not specified in the rule)
731 on the local machine.
735 is not a local address, then the port number
736 (if specified) is ignored, and the packet will be
737 forwarded to the remote address, using the route as found in
738 the local routing table for that IP.
742 rule will not match layer-2 packets (those received
743 on ether_input, ether_output, or bridged).
747 action does not change the contents of the packet at all.
748 In particular, the destination address remains unmodified, so
749 packets forwarded to another system will usually be rejected by that system
750 unless there is a matching rule on that system to capture them.
751 For packets forwarded locally,
752 the local address of the socket will be
753 set to the original destination address of the packet.
756 entry look rather weird but is intended for
757 use with transparent proxy servers.
761 a custom kernel needs to be compiled with the option
762 .Cd "options IPFIREWALL_FORWARD" .
766 (for network address translation, address redirect, etc.):
768 .Sx NETWORK ADDRESS TRANSLATION (NAT)
769 Section for further information.
770 .It Cm pipe Ar pipe_nr
774 (for bandwidth limitation, delay, etc.).
776 .Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
777 Section for further information.
778 The search terminates; however, on exit from the pipe and if
782 .Va net.inet.ip.fw.one_pass
783 is not set, the packet is passed again to the firewall code
784 starting from the next rule.
785 .It Cm queue Ar queue_nr
789 (for bandwidth limitation using WF2Q+).
795 Discard packets that match this rule, and if the
796 packet is a TCP packet, try to send a TCP reset (RST) notice.
797 The search terminates.
799 Discard packets that match this rule, and if the
800 packet is a TCP packet, try to send a TCP reset (RST) notice.
801 The search terminates.
802 .It Cm skipto Ar number | tablearg
803 Skip all subsequent rules numbered less than
805 The search continues with the first rule numbered
808 It is possible to use the
810 keyword with a skipto for a
812 skipto, but care should be used, as no destination caching
813 is possible in this case so the rules are always walked to find it,
817 Send a copy of packets matching this rule to the
821 The search continues with the next rule.
822 .It Cm unreach Ar code
823 Discard packets that match this rule, and try to send an ICMP
824 unreachable notice with code
828 is a number from 0 to 255, or one of these aliases:
829 .Cm net , host , protocol , port ,
830 .Cm needfrag , srcfail , net-unknown , host-unknown ,
831 .Cm isolated , net-prohib , host-prohib , tosnet ,
832 .Cm toshost , filter-prohib , host-precedence
834 .Cm precedence-cutoff .
835 The search terminates.
836 .It Cm unreach6 Ar code
837 Discard packets that match this rule, and try to send an ICMPv6
838 unreachable notice with code
842 is a number from 0, 1, 3 or 4, or one of these aliases:
843 .Cm no-route, admin-prohib, address
846 The search terminates.
847 .It Cm netgraph Ar cookie
848 Divert packet into netgraph with given
850 The search terminates.
851 If packet is later returned from netgraph it is either
852 accepted or continues with the next rule, depending on
853 .Va net.inet.ip.fw.one_pass
855 .It Cm ngtee Ar cookie
856 A copy of packet is diverted into netgraph, original
857 packet is either accepted or continues with the next rule, depending on
858 .Va net.inet.ip.fw.one_pass
862 for more information on
867 .It Cm setfib Ar fibnum
868 The packet is tagged so as to use the FIB (routing table)
870 in any subsequent forwarding decisions.
871 Initially this is limited to the values 0 through 15, see
873 Processing continues at the next rule.
875 Queue and reassemble ip fragments.
876 If the packet is not fragmented, counters are updated and processing continues with the next rule.
877 If the packet is the last logical fragment, the packet is reassembled and, if
878 .Va net.inet.ip.fw.one_pass
879 is set to 0, processing continues with the next rule, else packet is allowed to pass and search terminates.
880 If the packet is a fragment in the middle, it is consumed and processing stops immediately.
882 Fragments handling can be tuned via
883 .Va net.inet.ip.maxfragpackets
885 .Va net.inet.ip.maxfragsperpacket
886 which limit, respectively, the maximum number of processable fragments (default: 800) and
887 the maximum number of fragments per packet (default: 16).
889 NOTA BENE: since fragments do not contain port numbers, they should be avoided with the
892 Alternatively, direction-based (like
896 ) and source-based (like
898 ) match patterns can be used to select fragments.
900 Usually a simple rule like:
901 .Bd -literal -offset indent
902 # reassemble incoming fragments
903 ipfw add reass all from any to any in
906 is all you need at the beginning of your ruleset.
909 The body of a rule contains zero or more patterns (such as
910 specific source and destination addresses or ports,
911 protocol options, incoming or outgoing interfaces, etc.)
912 that the packet must match in order to be recognised.
913 In general, the patterns are connected by (implicit)
915 operators -- i.e., all must match in order for the
917 Individual patterns can be prefixed by the
919 operator to reverse the result of the match, as in
921 .Dl "ipfw add 100 allow ip from not 1.2.3.4 to any"
923 Additionally, sets of alternative match patterns
925 can be constructed by putting the patterns in
926 lists enclosed between parentheses ( ) or braces { }, and
931 .Dl "ipfw add 100 allow ip from { x or not y or z } to any"
933 Only one level of parentheses is allowed.
934 Beware that most shells have special meanings for parentheses
935 or braces, so it is advisable to put a backslash \\ in front of them
936 to prevent such interpretations.
938 The body of a rule must in general include a source and destination
942 can be used in various places to specify that the content of
943 a required field is irrelevant.
945 The rule body has the following format:
946 .Bd -ragged -offset indent
947 .Op Ar proto Cm from Ar src Cm to Ar dst
951 The first part (proto from src to dst) is for backward
952 compatibility with earlier versions of
956 any match pattern (including MAC headers, IP protocols,
957 addresses and ports) can be specified in the
961 Rule fields have the following meaning:
962 .Bl -tag -width indent
963 .It Ar proto : protocol | Cm { Ar protocol Cm or ... }
964 .It Ar protocol : Oo Cm not Oc Ar protocol-name | protocol-number
965 An IP protocol specified by number or name
966 (for a complete list see
967 .Pa /etc/protocols ) ,
968 or one of the following keywords:
969 .Bl -tag -width indent
971 Matches IPv4 packets.
973 Matches IPv6 packets.
982 option will be treated as inner protocol.
990 .Cm { Ar protocol Cm or ... }
993 is provided for convenience only but its use is deprecated.
994 .It Ar src No and Ar dst : Bro Cm addr | Cm { Ar addr Cm or ... } Brc Op Oo Cm not Oc Ar ports
995 An address (or a list, see below)
996 optionally followed by
1002 with multiple addresses) is provided for convenience only and
1003 its use is discouraged.
1004 .It Ar addr : Oo Cm not Oc Bro
1005 .Cm any | me | me6 |
1006 .Cm table Ns Pq Ar number Ns Op , Ns Ar value
1007 .Ar | addr-list | addr-set
1010 matches any IP address.
1012 matches any IP address configured on an interface in the system.
1014 matches any IPv6 address configured on an interface in the system.
1015 The address list is evaluated at the time the packet is
1017 .It Cm table Ns Pq Ar number Ns Op , Ns Ar value
1018 Matches any IPv4 address for which an entry exists in the lookup table
1020 If an optional 32-bit unsigned
1022 is also specified, an entry will match only if it has this value.
1025 section below for more information on lookup tables.
1026 .It Ar addr-list : ip-addr Ns Op Ns , Ns Ar addr-list
1028 A host or subnet address specified in one of the following ways:
1029 .Bl -tag -width indent
1030 .It Ar numeric-ip | hostname
1031 Matches a single IPv4 address, specified as dotted-quad or a hostname.
1032 Hostnames are resolved at the time the rule is added to the firewall list.
1033 .It Ar addr Ns / Ns Ar masklen
1034 Matches all addresses with base
1036 (specified as an IP address, a network number, or a hostname)
1040 As an example, 1.2.3.4/25 or 1.2.3.0/25 will match
1041 all IP numbers from 1.2.3.0 to 1.2.3.127 .
1042 .It Ar addr Ns : Ns Ar mask
1043 Matches all addresses with base
1045 (specified as an IP address, a network number, or a hostname)
1048 specified as a dotted quad.
1049 As an example, 1.2.3.4:255.0.255.0 or 1.0.3.0:255.0.255.0 will match
1051 This form is advised only for non-contiguous
1053 It is better to resort to the
1054 .Ar addr Ns / Ns Ar masklen
1055 format for contiguous masks, which is more compact and less
1058 .It Ar addr-set : addr Ns Oo Ns / Ns Ar masklen Oc Ns Cm { Ns Ar list Ns Cm }
1059 .It Ar list : Bro Ar num | num-num Brc Ns Op Ns , Ns Ar list
1060 Matches all addresses with base address
1062 (specified as an IP address, a network number, or a hostname)
1063 and whose last byte is in the list between braces { } .
1064 Note that there must be no spaces between braces and
1065 numbers (spaces after commas are allowed).
1066 Elements of the list can be specified as single entries
1070 field is used to limit the size of the set of addresses,
1071 and can have any value between 24 and 32.
1073 it will be assumed as 24.
1075 This format is particularly useful to handle sparse address sets
1076 within a single rule.
1077 Because the matching occurs using a
1078 bitmask, it takes constant time and dramatically reduces
1079 the complexity of rulesets.
1081 As an example, an address specified as 1.2.3.4/24{128,35-55,89}
1082 or 1.2.3.0/24{128,35-55,89}
1083 will match the following IP addresses:
1085 1.2.3.128, 1.2.3.35 to 1.2.3.55, 1.2.3.89 .
1086 .It Ar addr6-list : ip6-addr Ns Op Ns , Ns Ar addr6-list
1088 A host or subnet specified one of the following ways:
1090 .Bl -tag -width indent
1091 .It Ar numeric-ip | hostname
1092 Matches a single IPv6 address as allowed by
1095 Hostnames are resolved at the time the rule is added to the firewall
1097 .It Ar addr Ns / Ns Ar masklen
1098 Matches all IPv6 addresses with base
1100 (specified as allowed by
1108 No support for sets of IPv6 addresses is provided because IPv6 addresses
1109 are typically random past the initial prefix.
1110 .It Ar ports : Bro Ar port | port Ns \&- Ns Ar port Ns Brc Ns Op , Ns Ar ports
1111 For protocols which support port numbers (such as TCP and UDP), optional
1113 may be specified as one or more ports or port ranges, separated
1114 by commas but no spaces, and an optional
1119 notation specifies a range of ports (including boundaries).
1123 may be used instead of numeric port values.
1124 The length of the port list is limited to 30 ports or ranges,
1125 though one can specify larger ranges by using an
1129 section of the rule.
1133 can be used to escape the dash
1135 character in a service name (from a shell, the backslash must be
1136 typed twice to avoid the shell itself interpreting it as an escape
1139 .Dl "ipfw add count tcp from any ftp\e\e-data-ftp to any"
1141 Fragmented packets which have a non-zero offset (i.e., not the first
1142 fragment) will never match a rule which has one or more port
1146 option for details on matching fragmented packets.
1148 .Ss RULE OPTIONS (MATCH PATTERNS)
1149 Additional match patterns can be used within
1151 Zero or more of these so-called
1153 can be present in a rule, optionally prefixed by the
1155 operand, and possibly grouped into
1158 The following match patterns can be used (listed in alphabetical order):
1159 .Bl -tag -width indent
1160 .It Cm // this is a comment.
1161 Inserts the specified text as a comment in the rule.
1162 Everything following // is considered as a comment and stored in the rule.
1163 You can have comment-only rules, which are listed as having a
1165 action followed by the comment.
1170 Matches only packets generated by a divert socket.
1171 .It Cm diverted-loopback
1172 Matches only packets coming from a divert socket back into the IP stack
1174 .It Cm diverted-output
1175 Matches only packets going from a divert socket back outward to the IP
1176 stack output for delivery.
1177 .It Cm dst-ip Ar ip-address
1178 Matches IPv4 packets whose destination IP is one of the address(es)
1179 specified as argument.
1180 .It Bro Cm dst-ip6 | dst-ipv6 Brc Ar ip6-address
1181 Matches IPv6 packets whose destination IP is one of the address(es)
1182 specified as argument.
1183 .It Cm dst-port Ar ports
1184 Matches IP packets whose destination port is one of the port(s)
1185 specified as argument.
1187 Matches TCP packets that have the RST or ACK bits set.
1188 .It Cm ext6hdr Ar header
1189 Matches IPv6 packets containing the extended header given by
1191 Supported headers are:
1197 any type of Routing Header
1199 Source routing Routing Header Type 0
1201 Mobile IPv6 Routing Header Type 2
1205 IPSec authentication headers
1207 and IPsec encapsulated security payload headers
1209 .It Cm fib Ar fibnum
1210 Matches a packet that has been tagged to use
1211 the given FIB (routing table) number.
1212 .It Cm flow-id Ar labels
1213 Matches IPv6 packets containing any of the flow labels given in
1216 is a comma separated list of numeric flow labels.
1218 Matches packets that are fragments and not the first
1219 fragment of an IP datagram.
1220 Note that these packets will not have
1221 the next protocol header (e.g.\& TCP, UDP) so options that look into
1222 these headers cannot match.
1224 Matches all TCP or UDP packets sent by or received for a
1228 may be specified by name or number.
1229 .It Cm jail Ar prisonID
1230 Matches all TCP or UDP packets sent by or received for the
1231 jail whos prison ID is
1233 .It Cm icmptypes Ar types
1234 Matches ICMP packets whose ICMP type is in the list
1236 The list may be specified as any combination of
1237 individual types (numeric) separated by commas.
1238 .Em Ranges are not allowed .
1239 The supported ICMP types are:
1243 destination unreachable
1251 router advertisement
1255 time-to-live exceeded
1267 address mask request
1269 and address mask reply
1271 .It Cm icmp6types Ar types
1272 Matches ICMP6 packets whose ICMP6 type is in the list of
1274 The list may be specified as any combination of
1275 individual types (numeric) separated by commas.
1276 .Em Ranges are not allowed .
1278 Matches incoming or outgoing packets, respectively.
1282 are mutually exclusive (in fact,
1286 .It Cm ipid Ar id-list
1287 Matches IPv4 packets whose
1289 field has value included in
1291 which is either a single value or a list of values or ranges
1292 specified in the same way as
1294 .It Cm iplen Ar len-list
1295 Matches IP packets whose total length, including header and data, is
1298 which is either a single value or a list of values or ranges
1299 specified in the same way as
1301 .It Cm ipoptions Ar spec
1302 Matches packets whose IPv4 header contains the comma separated list of
1303 options specified in
1305 The supported IP options are:
1308 (strict source route),
1310 (loose source route),
1312 (record packet route) and
1315 The absence of a particular option may be denoted
1318 .It Cm ipprecedence Ar precedence
1319 Matches IPv4 packets whose precedence field is equal to
1322 Matches packets that have IPSEC history associated with them
1323 (i.e., the packet comes encapsulated in IPSEC, the kernel
1324 has IPSEC support and IPSEC_FILTERTUNNEL option, and can correctly
1327 Note that specifying
1329 is different from specifying
1331 as the latter will only look at the specific IP protocol field,
1332 irrespective of IPSEC kernel support and the validity of the IPSEC data.
1334 Further note that this flag is silently ignored in kernels without
1336 It does not affect rule processing when given and the
1337 rules are handled as if with no
1340 .It Cm iptos Ar spec
1341 Matches IPv4 packets whose
1343 field contains the comma separated list of
1344 service types specified in
1346 The supported IP types of service are:
1349 .Pq Dv IPTOS_LOWDELAY ,
1351 .Pq Dv IPTOS_THROUGHPUT ,
1353 .Pq Dv IPTOS_RELIABILITY ,
1355 .Pq Dv IPTOS_MINCOST ,
1357 .Pq Dv IPTOS_ECN_CE .
1358 The absence of a particular type may be denoted
1361 .It Cm ipttl Ar ttl-list
1362 Matches IPv4 packets whose time to live is included in
1364 which is either a single value or a list of values or ranges
1365 specified in the same way as
1367 .It Cm ipversion Ar ver
1368 Matches IP packets whose IP version field is
1371 Upon a match, the firewall will create a dynamic rule, whose
1372 default behaviour is to match bidirectional traffic between
1373 source and destination IP/port using the same protocol.
1374 The rule has a limited lifetime (controlled by a set of
1376 variables), and the lifetime is refreshed every time a matching
1379 Matches only layer2 packets, i.e., those passed to
1381 from ether_demux() and ether_output_frame().
1382 .It Cm limit Bro Cm src-addr | src-port | dst-addr | dst-port Brc Ar N
1383 The firewall will only allow
1385 connections with the same
1386 set of parameters as specified in the rule.
1388 of source and destination addresses and ports can be
1391 only IPv4 flows are supported.
1392 .It Cm { MAC | mac } Ar dst-mac src-mac
1393 Match packets with a given
1397 addresses, specified as the
1399 keyword (matching any MAC address), or six groups of hex digits
1400 separated by colons,
1401 and optionally followed by a mask indicating the significant bits.
1402 The mask may be specified using either of the following methods:
1403 .Bl -enum -width indent
1407 followed by the number of significant bits.
1408 For example, an address with 33 significant bits could be specified as:
1410 .Dl "MAC 10:20:30:40:50:60/33 any"
1415 followed by a bitmask specified as six groups of hex digits separated
1417 For example, an address in which the last 16 bits are significant could
1420 .Dl "MAC 10:20:30:40:50:60&00:00:00:00:ff:ff any"
1422 Note that the ampersand character has a special meaning in many shells
1423 and should generally be escaped.
1426 Note that the order of MAC addresses (destination first,
1428 the same as on the wire, but the opposite of the one used for
1430 .It Cm mac-type Ar mac-type
1431 Matches packets whose Ethernet Type field
1432 corresponds to one of those specified as argument.
1434 is specified in the same way as
1436 (i.e., one or more comma-separated single values or ranges).
1437 You can use symbolic names for known values such as
1438 .Em vlan , ipv4, ipv6 .
1439 Values can be entered as decimal or hexadecimal (if prefixed by 0x),
1440 and they are always printed as hexadecimal (unless the
1442 option is used, in which case symbolic resolution will be attempted).
1443 .It Cm proto Ar protocol
1444 Matches packets with the corresponding IP protocol.
1445 .It Cm recv | xmit | via Brq Ar ifX | Ar if Ns Cm * | Ar ipno | Ar any
1446 Matches packets received, transmitted or going through,
1447 respectively, the interface specified by exact name
1448 .Ns No ( Ar ifX Ns No ),
1450 .Ns No ( Ar if Ns Ar * Ns No ),
1451 by IP address, or through some interface.
1455 keyword causes the interface to always be checked.
1462 then only the receive or transmit interface (respectively)
1464 By specifying both, it is possible to match packets based on
1465 both receive and transmit interface, e.g.:
1467 .Dl "ipfw add deny ip from any to any out recv ed0 xmit ed1"
1471 interface can be tested on either incoming or outgoing packets,
1474 interface can only be tested on outgoing packets.
1479 is invalid) whenever
1483 A packet may not have a receive or transmit interface: packets
1484 originating from the local host have no receive interface,
1485 while packets destined for the local host have no transmit
1488 Matches TCP packets that have the SYN bit set but no ACK bit.
1489 This is the short form of
1490 .Dq Li tcpflags\ syn,!ack .
1491 .It Cm src-ip Ar ip-address
1492 Matches IPv4 packets whose source IP is one of the address(es)
1493 specified as an argument.
1494 .It Cm src-ip6 Ar ip6-address
1495 Matches IPv6 packets whose source IP is one of the address(es)
1496 specified as an argument.
1497 .It Cm src-port Ar ports
1498 Matches IP packets whose source port is one of the port(s)
1499 specified as argument.
1500 .It Cm tagged Ar tag-list
1501 Matches packets whose tags are included in
1503 which is either a single value or a list of values or ranges
1504 specified in the same way as
1506 Tags can be applied to the packet using
1508 rule action parameter (see it's description for details on tags).
1509 .It Cm tcpack Ar ack
1511 Match if the TCP header acknowledgment number field is set to
1513 .It Cm tcpdatalen Ar tcpdatalen-list
1514 Matches TCP packets whose length of TCP data is
1515 .Ar tcpdatalen-list ,
1516 which is either a single value or a list of values or ranges
1517 specified in the same way as
1519 .It Cm tcpflags Ar spec
1521 Match if the TCP header contains the comma separated list of
1524 The supported TCP flags are:
1533 The absence of a particular flag may be denoted
1536 A rule which contains a
1538 specification can never match a fragmented packet which has
1542 option for details on matching fragmented packets.
1543 .It Cm tcpseq Ar seq
1545 Match if the TCP header sequence number field is set to
1547 .It Cm tcpwin Ar win
1549 Match if the TCP header window field is set to
1551 .It Cm tcpoptions Ar spec
1553 Match if the TCP header contains the comma separated list of
1554 options specified in
1556 The supported TCP options are:
1559 (maximum segment size),
1561 (tcp window advertisement),
1565 (rfc1323 timestamp) and
1567 (rfc1644 t/tcp connection count).
1568 The absence of a particular option may be denoted
1572 Match all TCP or UDP packets sent by or received for a
1576 may be matched by name or identification number.
1578 For incoming packets,
1579 a routing table lookup is done on the packet's source address.
1580 If the interface on which the packet entered the system matches the
1581 outgoing interface for the route,
1583 If the interfaces do not match up,
1584 the packet does not match.
1585 All outgoing packets or packets with no incoming interface match.
1587 The name and functionality of the option is intentionally similar to
1588 the Cisco IOS command:
1590 .Dl ip verify unicast reverse-path
1592 This option can be used to make anti-spoofing rules to reject all
1593 packets with source addresses not from this interface.
1597 For incoming packets,
1598 a routing table lookup is done on the packet's source address.
1599 If a route to the source address exists, but not the default route
1600 or a blackhole/reject route, the packet matches.
1601 Otherwise, the packet does not match.
1602 All outgoing packets match.
1604 The name and functionality of the option is intentionally similar to
1605 the Cisco IOS command:
1607 .Dl ip verify unicast source reachable-via any
1609 This option can be used to make anti-spoofing rules to reject all
1610 packets whose source address is unreachable.
1612 For incoming packets, the packet's source address is checked if it
1613 belongs to a directly connected network.
1614 If the network is directly connected, then the interface the packet
1615 came on in is compared to the interface the network is connected to.
1616 When incoming interface and directly connected interface are not the
1617 same, the packet does not match.
1618 Otherwise, the packet does match.
1619 All outgoing packets match.
1621 This option can be used to make anti-spoofing rules to reject all
1622 packets that pretend to be from a directly connected network but do
1623 not come in through that interface.
1624 This option is similar to but more restricted than
1626 because it engages only on packets with source addresses of directly
1627 connected networks instead of all source addresses.
1630 Lookup tables are useful to handle large sparse address sets,
1631 typically from a hundred to several thousands of entries.
1632 There may be up to 128 different lookup tables, numbered 0 to 127.
1634 Each entry is represented by an
1635 .Ar addr Ns Op / Ns Ar masklen
1636 and will match all addresses with base
1638 (specified as an IP address or a hostname)
1644 is not specified, it defaults to 32.
1645 When looking up an IP address in a table, the most specific
1647 Associated with each entry is a 32-bit unsigned
1649 which can optionally be checked by a rule matching code.
1650 When adding an entry, if
1652 is not specified, it defaults to 0.
1654 An entry can be added to a table
1656 removed from a table
1658 a table can be examined
1663 Internally, each table is stored in a Radix tree, the same way as
1664 the routing table (see
1667 Lookup tables currently support IPv4 addresses only.
1671 feature provides the ability to use a value, looked up in the table, as
1672 the argument for a rule action, action parameter or rule option.
1673 This can significantly reduce number of rules in some configurations.
1674 If two tables are used in a rule, the result of the second (destination)
1678 argument can be used with the following actions:
1679 .Cm nat, pipe , queue, divert, tee, netgraph, ngtee, fwd, skipto
1687 it is possible to supply table entries with values
1688 that are in the form of IP addresses or hostnames.
1691 Section for example usage of tables and the tablearg keyword.
1695 action, the user should be aware that the code will walk the ruleset
1696 up to a rule equal to, or past, the given number, and should therefore try keep the
1697 ruleset compact between the skipto and the target rules.
1699 Each rule belongs to one of 32 different
1702 Set 31 is reserved for the default rule.
1704 By default, rules are put in set 0, unless you use the
1706 attribute when entering a new rule.
1707 Sets can be individually and atomically enabled or disabled,
1708 so this mechanism permits an easy way to store multiple configurations
1709 of the firewall and quickly (and atomically) switch between them.
1710 The command to enable/disable sets is
1711 .Bd -ragged -offset indent
1713 .Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
1720 sections can be specified.
1721 Command execution is atomic on all the sets specified in the command.
1722 By default, all sets are enabled.
1724 When you disable a set, its rules behave as if they do not exist
1725 in the firewall configuration, with only one exception:
1726 .Bd -ragged -offset indent
1727 dynamic rules created from a rule before it had been disabled
1728 will still be active until they expire.
1730 dynamic rules you have to explicitly delete the parent rule
1731 which generated them.
1734 The set number of rules can be changed with the command
1735 .Bd -ragged -offset indent
1738 .Brq Cm rule Ar rule-number | old-set
1742 Also, you can atomically swap two rulesets with the command
1743 .Bd -ragged -offset indent
1745 .Cm set swap Ar first-set second-set
1750 Section on some possible uses of sets of rules.
1751 .Sh STATEFUL FIREWALL
1752 Stateful operation is a way for the firewall to dynamically
1753 create rules for specific flows when packets that
1754 match a given pattern are detected.
1755 Support for stateful
1756 operation comes through the
1757 .Cm check-state , keep-state
1763 Dynamic rules are created when a packet matches a
1767 rule, causing the creation of a
1769 rule which will match all and only packets with
1773 .Em src-ip/src-port dst-ip/dst-port
1778 are used here only to denote the initial match addresses, but they
1779 are completely equivalent afterwards).
1780 Dynamic rules will be checked at the first
1781 .Cm check-state, keep-state
1784 occurrence, and the action performed upon a match will be the same
1785 as in the parent rule.
1787 Note that no additional attributes other than protocol and IP addresses
1788 and ports are checked on dynamic rules.
1790 The typical use of dynamic rules is to keep a closed firewall configuration,
1791 but let the first TCP SYN packet from the inside network install a
1792 dynamic rule for the flow so that packets belonging to that session
1793 will be allowed through the firewall:
1795 .Dl "ipfw add check-state"
1796 .Dl "ipfw add allow tcp from my-subnet to any setup keep-state"
1797 .Dl "ipfw add deny tcp from any to any"
1799 A similar approach can be used for UDP, where an UDP packet coming
1800 from the inside will install a dynamic rule to let the response through
1803 .Dl "ipfw add check-state"
1804 .Dl "ipfw add allow udp from my-subnet to any keep-state"
1805 .Dl "ipfw add deny udp from any to any"
1807 Dynamic rules expire after some time, which depends on the status
1808 of the flow and the setting of some
1812 .Sx SYSCTL VARIABLES
1814 For TCP sessions, dynamic rules can be instructed to periodically
1815 send keepalive packets to refresh the state of the rule when it is
1820 for more examples on how to use dynamic rules.
1821 .Sh TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
1823 is also the user interface for the
1825 traffic shaper and network emulator, a subsystem that
1826 can artificially queue, delay or drop packets
1827 emulator the behaviour of certain network links
1828 or queueing systems.
1831 operates by first using the firewall to select packets
1832 using any match pattern that can be used in
1835 Matching packets are then passed to either of two
1836 different objects, which implement the traffic regulation:
1837 .Bl -hang -offset XXXX
1839 A pipe emulates a link with given bandwidth, propagation delay,
1840 queue size and packet loss rate.
1841 Packets are queued in front of the pipe as they come out from the classifier,
1842 and then transferred to the pipe according to the pipe's parameters.
1845 is an abstraction used to implement the WF2Q+
1846 (Worst-case Fair Weighted Fair Queueing) policy, which is
1847 an efficient variant of the WFQ policy.
1849 The queue associates a
1851 and a reference pipe to each flow (a flow is a set of packets
1852 with the same addresses and ports after masking).
1853 All backlogged flows (i.e., those
1854 with packets queued) linked to the same pipe share the pipe's
1855 bandwidth proportionally to their weights.
1856 Note that weights are not priorities; a flow with a lower weight
1857 is still guaranteed to get its fraction of the bandwidth even if a
1858 flow with a higher weight is permanently backlogged.
1863 can be used to set hard limits to the bandwidth that a flow can use, whereas
1865 can be used to determine how different flows share the available bandwidth.
1867 There are two modes of
1875 mode tries to emulate a real link: the
1877 scheduler ensures that the packet will not leave the pipe faster than it
1878 would on the real link with a given bandwidth.
1881 mode allows certain packets to bypass the
1883 scheduler (if packet flow does not exceed pipe's bandwidth).
1884 This is the reason why the
1886 mode requires less CPU cycles per packet (on average) and packet latency
1887 can be significantly lower in comparison to a real link with the same
1893 mode can be enabled by setting the
1894 .Va net.inet.ip.dummynet.io_fast
1896 variable to a non-zero value.
1898 .Ss PIPE AND QUEUE CONFIGURATION
1903 configuration commands are the following:
1904 .Bd -ragged -offset indent
1905 .Cm pipe Ar number Cm config Ar pipe-configuration
1907 .Cm queue Ar number Cm config Ar queue-configuration
1910 The following parameters can be configured for a pipe:
1912 .Bl -tag -width indent -compact
1913 .It Cm bw Ar bandwidth | device
1914 Bandwidth, measured in
1917 .Brq Cm bit/s | Byte/s .
1920 A value of 0 (default) means unlimited bandwidth.
1921 The unit must immediately follow the number, as in
1923 .Dl "ipfw pipe 1 config bw 300Kbit/s"
1925 If a device name is specified instead of a numeric value, as in
1927 .Dl "ipfw pipe 1 config bw tun0"
1929 then the transmit clock is supplied by the specified device.
1930 At the moment only the
1932 device supports this
1933 functionality, for use in conjunction with
1936 .It Cm delay Ar ms-delay
1937 Propagation delay, measured in milliseconds.
1938 The value is rounded to the next multiple of the clock tick
1939 (typically 10ms, but it is a good practice to run kernels
1941 .Dq "options HZ=1000"
1943 the granularity to 1ms or less).
1944 The default value is 0, meaning no delay.
1946 .It Cm burst Ar size
1947 If the data to be sent exceeds the pipe's bandwidth limit
1948 (and the pipe was previously idle), up to
1950 bytes of data are allowed to bypass the
1952 scheduler, and will be sent as fast as the physical link allows.
1953 Any additional data will be transmitted at the rate specified
1957 The burst size depends on how long the pipe has been idle;
1958 the effective burst size is calculated as follows:
1965 .It Cm profile Ar filename
1966 A file specifying the additional overhead incurred in the transmission
1967 of a packet on the link.
1969 Some link types introduce extra delays in the transmission
1970 of a packet, e.g. because of MAC level framing, contention on
1971 the use of the channel, MAC level retransmissions and so on.
1972 From our point of view, the channel is effectively unavailable
1973 for this extra time, which is constant or variable depending
1974 on the link type. Additionally, packets may be dropped after this
1975 time (e.g. on a wireless link after too many retransmissions).
1976 We can model the additional delay with an empirical curve
1977 that represents its distribution.
1978 .Bd -literal -offset indent
1979 cumulative probability
1989 +-------*------------------->
1992 The empirical curve may have both vertical and horizontal lines.
1993 Vertical lines represent constant delay for a range of
1995 Horizontal lines correspond to a discontinuity in the delay
1996 distribution: the pipe will use the largest delay for a
1999 The file format is the following, with whitespace acting as
2000 a separator and '#' indicating the beginning a comment:
2001 .Bl -tag -width indent
2002 .It Cm name Ar identifier
2003 optional name (listed by "ipfw pipe show")
2004 to identify the delay distribution;
2006 the bandwidth used for the pipe.
2007 If not specified here, it must be present
2008 explicitly as a configuration parameter for the pipe;
2009 .It Cm loss-level Ar L
2010 the probability above which packets are lost.
2011 (0.0 <= L <= 1.0, default 1.0 i.e. no loss);
2013 the number of samples used in the internal
2014 representation of the curve (2..1024; default 100);
2015 .It Cm "delay prob" | "prob delay"
2016 One of these two lines is mandatory and defines
2017 the format of the following lines with data points.
2019 2 or more lines representing points in the curve,
2020 with either delay or probability first, according
2021 to the chosen format.
2022 The unit for delay is milliseconds.
2023 Data points do not need to be sorted.
2024 Also, tne number of actual lines can be different
2025 from the value of the "samples" parameter:
2027 utility will sort and interpolate
2028 the curve as needed.
2031 Example of a profile file:
2032 .Bd -literal -offset indent
2037 0 200 # minimum overhead is 200ms
2043 #configuration file end
2047 The following parameters can be configured for a queue:
2049 .Bl -tag -width indent -compact
2050 .It Cm pipe Ar pipe_nr
2051 Connects a queue to the specified pipe.
2052 Multiple queues (with the same or different weights) can be connected to
2053 the same pipe, which specifies the aggregate rate for the set of queues.
2055 .It Cm weight Ar weight
2056 Specifies the weight to be used for flows matching this queue.
2057 The weight must be in the range 1..100, and defaults to 1.
2060 Finally, the following parameters can be configured for both
2063 .Bl -tag -width XXXX -compact
2065 .It Cm buckets Ar hash-table-size
2066 Specifies the size of the hash table used for storing the
2068 Default value is 64 controlled by the
2071 .Va net.inet.ip.dummynet.hash_size ,
2072 allowed range is 16 to 65536.
2074 .It Cm mask Ar mask-specifier
2075 Packets sent to a given pipe or queue by an
2077 rule can be further classified into multiple flows, each of which is then
2081 A flow identifier is constructed by masking the IP addresses,
2082 ports and protocol types as specified with the
2084 options in the configuration of the pipe or queue.
2085 For each different flow identifier, a new pipe or queue is created
2086 with the same parameters as the original object, and matching packets
2091 are used, each flow will get the same bandwidth as defined by the pipe,
2094 are used, each flow will share the parent's pipe bandwidth evenly
2095 with other flows generated by the same queue (note that other queues
2096 with different weights might be connected to the same pipe).
2098 Available mask specifiers are a combination of one or more of the following:
2100 .Cm dst-ip Ar mask ,
2101 .Cm dst-ip6 Ar mask ,
2102 .Cm src-ip Ar mask ,
2103 .Cm src-ip6 Ar mask ,
2104 .Cm dst-port Ar mask ,
2105 .Cm src-port Ar mask ,
2106 .Cm flow-id Ar mask ,
2111 where the latter means all bits in all fields are significant.
2114 When a packet is dropped by a
2116 queue or pipe, the error
2117 is normally reported to the caller routine in the kernel, in the
2118 same way as it happens when a device queue fills up.
2120 option reports the packet as successfully delivered, which can be
2121 needed for some experimental setups where you want to simulate
2122 loss or congestion at a remote router.
2124 .It Cm plr Ar packet-loss-rate
2127 .Ar packet-loss-rate
2128 is a floating-point number between 0 and 1, with 0 meaning no
2129 loss, 1 meaning 100% loss.
2130 The loss rate is internally represented on 31 bits.
2132 .It Cm queue Brq Ar slots | size Ns Cm Kbytes
2137 Default value is 50 slots, which
2138 is the typical queue size for Ethernet devices.
2139 Note that for slow speed links you should keep the queue
2140 size short or your traffic might be affected by a significant
2142 E.g., 50 max-sized ethernet packets (1500 bytes) mean 600Kbit
2143 or 20s of queue on a 30Kbit/s pipe.
2144 Even worse effects can result if you get packets from an
2145 interface with a much larger MTU, e.g.\& the loopback interface
2146 with its 16KB packets.
2150 .Em net.inet.ip.dummynet.pipe_byte_limit
2152 .Em net.inet.ip.dummynet.pipe_slot_limit
2153 control the maximum lengths that can be specified.
2155 .It Cm red | gred Ar w_q Ns / Ns Ar min_th Ns / Ns Ar max_th Ns / Ns Ar max_p
2156 Make use of the RED (Random Early Detection) queue management algorithm.
2161 point numbers between 0 and 1 (0 not included), while
2165 are integer numbers specifying thresholds for queue management
2166 (thresholds are computed in bytes if the queue has been defined
2167 in bytes, in slots otherwise).
2170 also supports the gentle RED variant (gred).
2173 variables can be used to control the RED behaviour:
2174 .Bl -tag -width indent
2175 .It Va net.inet.ip.dummynet.red_lookup_depth
2176 specifies the accuracy in computing the average queue
2177 when the link is idle (defaults to 256, must be greater than zero)
2178 .It Va net.inet.ip.dummynet.red_avg_pkt_size
2179 specifies the expected average packet size (defaults to 512, must be
2181 .It Va net.inet.ip.dummynet.red_max_pkt_size
2182 specifies the expected maximum packet size, only used when queue
2183 thresholds are in bytes (defaults to 1500, must be greater than zero).
2187 When used with IPv6 data,
2189 currently has several limitations.
2190 Information necessary to route link-local packets to an
2191 interface is not available after processing by
2193 so those packets are dropped in the output path.
2194 Care should be taken to insure that link-local packets are not passed to
2197 Here are some important points to consider when designing your
2201 Remember that you filter both packets going
2205 Most connections need packets going in both directions.
2207 Remember to test very carefully.
2208 It is a good idea to be near the console when doing this.
2209 If you cannot be near the console,
2210 use an auto-recovery script such as the one in
2211 .Pa /usr/share/examples/ipfw/change_rules.sh .
2213 Do not forget the loopback interface.
2218 There are circumstances where fragmented datagrams are unconditionally
2220 TCP packets are dropped if they do not contain at least 20 bytes of
2221 TCP header, UDP packets are dropped if they do not contain a full 8
2222 byte UDP header, and ICMP packets are dropped if they do not contain
2223 4 bytes of ICMP header, enough to specify the ICMP type, code, and
2225 These packets are simply logged as
2227 since there may not be enough good data in the packet to produce a
2228 meaningful log entry.
2230 Another type of packet is unconditionally dropped, a TCP packet with a
2231 fragment offset of one.
2232 This is a valid packet, but it only has one use, to try
2233 to circumvent firewalls.
2234 When logging is enabled, these packets are
2235 reported as being dropped by rule -1.
2237 If you are logged in over a network, loading the
2241 is probably not as straightforward as you would think.
2242 The following command line is recommended:
2243 .Bd -literal -offset indent
2245 ipfw add 32000 allow ip from any to any
2248 Along the same lines, doing an
2249 .Bd -literal -offset indent
2253 in similar surroundings is also a bad idea.
2257 filter list may not be modified if the system security level
2258 is set to 3 or higher
2261 for information on system security levels).
2263 .Sh PACKET DIVERSION
2266 socket bound to the specified port will receive all packets
2267 diverted to that port.
2268 If no socket is bound to the destination port, or if the divert module is
2269 not loaded, or if the kernel was not compiled with divert socket support,
2270 the packets are dropped.
2271 .Sh NETWORK ADDRESS TRANSLATION (NAT)
2274 support in-kernel NAT using the kernel version of
2277 The nat configuration command is the following:
2278 .Bd -ragged -offset indent
2283 .Ar nat-configuration
2287 The following parameters can be configured:
2288 .Bl -tag -width indent
2289 .It Cm ip Ar ip_address
2290 Define an ip address to use for aliasing.
2292 Use ip address of NIC for aliasing, dynamically changing
2293 it if NIC's ip address changes.
2295 Enable logging on this nat instance.
2297 Deny any incoming connection from outside world.
2299 Try to leave the alias port numbers unchanged from
2300 the actual local port numbers.
2302 Traffic on the local network not originating from an
2303 unregistered address spaces will be ignored.
2305 Reset table of the packet aliasing engine on address change.
2307 Reverse the way libalias handles aliasing.
2309 Obey transparent proxy rules only, packet aliasing is not performed.
2312 To let the packet continue after being (de)aliased, set the sysctl variable
2313 .Va net.inet.ip.fw.one_pass
2315 For more information about aliasing modes, refer to
2319 for some examples about nat usage.
2320 .Ss REDIRECT AND LSNAT SUPPORT IN IPFW
2321 Redirect and LSNAT support follow closely the syntax used in
2325 for some examples on how to do redirect and lsnat.
2326 .Ss SCTP NAT SUPPORT
2327 SCTP nat can be configured in a similar manner to TCP through the
2330 The main difference is that
2332 does not do port translation.
2333 Since the local and global side ports will be the same,
2334 there is no need to specify both.
2335 Ports are redirected as follows:
2336 .Bd -ragged -offset indent
2342 .Cm redirect_port sctp
2343 .Ar ip_address [,addr_list] {[port | port-port] [,ports]}
2349 configuration can be done in real-time through the
2352 All may be changed dynamically, though the hash_table size will only
2357 .Sx SYSCTL VARIABLES
2359 .Sh SYSCTL VARIABLES
2362 variables controls the behaviour of the firewall and
2364 .Pq Nm dummynet , bridge , sctp nat .
2365 These are shown below together with their default value
2366 (but always check with the
2368 command what value is actually in use) and meaning:
2369 .Bl -tag -width indent
2370 .It Va net.inet.ip.alias.sctp.accept_global_ootb_addip: No 0
2373 responds to receipt of global OOTB ASCONF-AddIP:
2374 .Bl -tag -width indent
2376 No response (unless a partially matching association exists -
2377 ports and vtags match but global address does not)
2380 will accept and process all OOTB global AddIP messages.
2383 Option 1 should never be selected as this forms a security risk.
2385 establish multiple fake associations by sending AddIP messages.
2386 .It Va net.inet.ip.alias.sctp.chunk_proc_limit: No 5
2387 Defines the maximum number of chunks in an SCTP packet that will be parsed for a
2388 packet that matches an existing association.
2389 This value is enforced to be greater or equal than
2390 .Cm net.inet.ip.alias.sctp.initialising_chunk_proc_limit .
2392 a DoS risk yet setting too low a value may result in important control chunks in
2393 the packet not being located and parsed.
2394 .It Va net.inet.ip.alias.sctp.error_on_ootb: No 1
2397 responds to any Out-of-the-Blue (OOTB) packets with ErrorM packets.
2398 An OOTB packet is a packet that arrives with no existing association
2401 and is not an INIT or ASCONF-AddIP packet:
2402 .Bl -tag -width indent
2404 ErrorM is never sent in response to OOTB packets.
2406 ErrorM is only sent to OOTB packets received on the local side.
2408 ErrorM is sent to the local side and on the global side ONLY if there is a
2409 partial match (ports and vtags match but the source global IP does not).
2410 This value is only useful if the
2412 is tracking global IP addresses.
2414 ErrorM is sent in response to all OOTB packets on both the local and global side
2418 At the moment the default is 0, since the ErrorM packet is not yet
2419 supported by most SCTP stacks.
2420 When it is supported, and if not tracking
2421 global addresses, we recommend setting this value to 1 to allow
2422 multi-homed local hosts to function with the
2424 To track global addresses, we recommend setting this value to 2 to
2425 allow global hosts to be informed when they need to (re)send an
2427 Value 3 should never be chosen (except for debugging) as the
2429 will respond to all OOTB global packets (a DoS risk).
2430 .It Va net.inet.ip.alias.sctp.hashtable_size: No 2003
2431 Size of hash tables used for
2433 lookups (100 < prime_number > 1000001).
2436 size for any future created
2438 instance and therefore must be set prior to creating a
2441 The table sizes may be changed to suit specific needs.
2442 If there will be few
2443 concurrent associations, and memory is scarce, you may make these smaller.
2444 If there will be many thousands (or millions) of concurrent associations, you
2445 should make these larger.
2446 A prime number is best for the table size.
2448 update function will adjust your input value to the next highest prime number.
2449 .It Va net.inet.ip.alias.sctp.holddown_time: No 0
2450 Hold association in table for this many seconds after receiving a
2452 This allows endpoints to correct shutdown gracefully if a
2453 shutdown_complete is lost and retransmissions are required.
2454 .It Va net.inet.ip.alias.sctp.init_timer: No 15
2455 Timeout value while waiting for (INIT-ACK|AddIP-ACK).
2456 This value cannot be 0.
2457 .It Va net.inet.ip.alias.sctp.initialising_chunk_proc_limit: No 2
2458 Defines the maximum number of chunks in an SCTP packet that will be parsed when
2459 no existing association exists that matches that packet.
2461 will only be an INIT or ASCONF-AddIP packet.
2462 A higher value may become a DoS
2463 risk as malformed packets can consume processing resources.
2464 .It Va net.inet.ip.alias.sctp.param_proc_limit: No 25
2465 Defines the maximum number of parameters within a chunk that will be parsed in a
2467 As for other similar sysctl variables, larger values pose a DoS risk.
2468 .It Va net.inet.ip.alias.sctp.log_level: No 0
2469 Level of detail in the system log messages (0 \- minimal, 1 \- event,
2470 2 \- info, 3 \- detail, 4 \- debug, 5 \- max debug). May be a good
2471 option in high loss environments.
2472 .It Va net.inet.ip.alias.sctp.shutdown_time: No 15
2473 Timeout value while waiting for SHUTDOWN-COMPLETE.
2474 This value cannot be 0.
2475 .It Va net.inet.ip.alias.sctp.track_global_addresses: No 0
2476 Enables/disables global IP address tracking within the
2479 upper limit on the number of addresses tracked for each association:
2480 .Bl -tag -width indent
2482 Global tracking is disabled
2484 Enables tracking, the maximum number of addresses tracked for each
2485 association is limited to this value
2488 This variable is fully dynamic, the new value will be adopted for all newly
2489 arriving associations, existing associations are treated as they were previously.
2490 Global tracking will decrease the number of collisions within the
2493 of increased processing load, memory usage, complexity, and possible
2496 problems in complex networks with multiple
2498 We recommend not tracking
2499 global IP addresses, this will still result in a fully functional
2501 .It Va net.inet.ip.alias.sctp.up_timer: No 300
2502 Timeout value to keep an association up with no traffic.
2503 This value cannot be 0.
2504 .It Va net.inet.ip.dummynet.expire : No 1
2505 Lazily delete dynamic pipes/queue once they have no pending traffic.
2506 You can disable this by setting the variable to 0, in which case
2507 the pipes/queues will only be deleted when the threshold is reached.
2508 .It Va net.inet.ip.dummynet.hash_size : No 64
2509 Default size of the hash table used for dynamic pipes/queues.
2510 This value is used when no
2512 option is specified when configuring a pipe/queue.
2513 .It Va net.inet.ip.dummynet.io_fast : No 0
2514 If set to a non-zero value,
2519 operation (see above) is enabled.
2520 .It Va net.inet.ip.dummynet.io_pkt
2521 Number of packets passed to
2523 .It Va net.inet.ip.dummynet.io_pkt_drop
2524 Number of packets dropped by
2526 .It Va net.inet.ip.dummynet.io_pkt_fast
2527 Number of packets bypassed by the
2530 .It Va net.inet.ip.dummynet.max_chain_len : No 16
2531 Target value for the maximum number of pipes/queues in a hash bucket.
2533 .Cm max_chain_len*hash_size
2534 is used to determine the threshold over which empty pipes/queues
2535 will be expired even when
2536 .Cm net.inet.ip.dummynet.expire=0 .
2537 .It Va net.inet.ip.dummynet.red_lookup_depth : No 256
2538 .It Va net.inet.ip.dummynet.red_avg_pkt_size : No 512
2539 .It Va net.inet.ip.dummynet.red_max_pkt_size : No 1500
2540 Parameters used in the computations of the drop probability
2541 for the RED algorithm.
2542 .It Va net.inet.ip.dummynet.pipe_byte_limit : No 1048576
2543 .It Va net.inet.ip.dummynet.pipe_slot_limit : No 100
2544 The maximum queue size that can be specified in bytes or packets.
2545 These limits prevent accidental exhaustion of resources such as mbufs.
2546 If you raise these limits,
2547 you should make sure the system is configured so that sufficient resources
2549 .It Va net.inet.ip.fw.autoinc_step : No 100
2550 Delta between rule numbers when auto-generating them.
2551 The value must be in the range 1..1000.
2552 .It Va net.inet.ip.fw.curr_dyn_buckets : Va net.inet.ip.fw.dyn_buckets
2553 The current number of buckets in the hash table for dynamic rules
2555 .It Va net.inet.ip.fw.debug : No 1
2556 Controls debugging messages produced by
2558 .It Va net.inet.ip.fw.default_rule : No 65535
2559 The default rule number (read-only).
2561 .Nm , the default rule is the last one, so its number
2562 can also serve as the highest number allowed for a rule.
2563 .It Va net.inet.ip.fw.dyn_buckets : No 256
2564 The number of buckets in the hash table for dynamic rules.
2565 Must be a power of 2, up to 65536.
2566 It only takes effect when all dynamic rules have expired, so you
2567 are advised to use a
2569 command to make sure that the hash table is resized.
2570 .It Va net.inet.ip.fw.dyn_count : No 3
2571 Current number of dynamic rules
2573 .It Va net.inet.ip.fw.dyn_keepalive : No 1
2574 Enables generation of keepalive packets for
2576 rules on TCP sessions.
2577 A keepalive is generated to both
2578 sides of the connection every 5 seconds for the last 20
2579 seconds of the lifetime of the rule.
2580 .It Va net.inet.ip.fw.dyn_max : No 8192
2581 Maximum number of dynamic rules.
2582 When you hit this limit, no more dynamic rules can be
2583 installed until old ones expire.
2584 .It Va net.inet.ip.fw.dyn_ack_lifetime : No 300
2585 .It Va net.inet.ip.fw.dyn_syn_lifetime : No 20
2586 .It Va net.inet.ip.fw.dyn_fin_lifetime : No 1
2587 .It Va net.inet.ip.fw.dyn_rst_lifetime : No 1
2588 .It Va net.inet.ip.fw.dyn_udp_lifetime : No 5
2589 .It Va net.inet.ip.fw.dyn_short_lifetime : No 30
2590 These variables control the lifetime, in seconds, of dynamic
2592 Upon the initial SYN exchange the lifetime is kept short,
2593 then increased after both SYN have been seen, then decreased
2594 again during the final FIN exchange or when a RST is received.
2596 .Em dyn_fin_lifetime
2598 .Em dyn_rst_lifetime
2599 must be strictly lower than 5 seconds, the period of
2600 repetition of keepalives.
2601 The firewall enforces that.
2602 .It Va net.inet.ip.fw.enable : No 1
2603 Enables the firewall.
2604 Setting this variable to 0 lets you run your machine without
2605 firewall even if compiled in.
2606 .It Va net.inet6.ip6.fw.enable : No 1
2607 provides the same functionality as above for the IPv6 case.
2608 .It Va net.inet.ip.fw.one_pass : No 1
2609 When set, the packet exiting from the
2613 node is not passed though the firewall again.
2614 Otherwise, after an action, the packet is
2615 reinjected into the firewall at the next rule.
2616 .It Va net.inet.ip.fw.tables_max : No 128
2617 Maximum number of tables (read-only).
2618 .It Va net.inet.ip.fw.verbose : No 1
2619 Enables verbose messages.
2620 .It Va net.inet.ip.fw.verbose_limit : No 0
2621 Limits the number of messages produced by a verbose firewall.
2622 .It Va net.inet6.ip6.fw.deny_unknown_exthdrs : No 1
2623 If enabled packets with unknown IPv6 Extension Headers will be denied.
2624 .It Va net.link.ether.ipfw : No 0
2625 Controls whether layer-2 packets are passed to
2628 .It Va net.link.bridge.ipfw : No 0
2629 Controls whether bridged packets are passed to
2635 There are far too many possible uses of
2637 so this Section will only give a small set of examples.
2639 .Ss BASIC PACKET FILTERING
2640 This command adds an entry which denies all tcp packets from
2641 .Em cracker.evil.org
2642 to the telnet port of
2644 from being forwarded by the host:
2646 .Dl "ipfw add deny tcp from cracker.evil.org to wolf.tambov.su telnet"
2648 This one disallows any connection from the entire cracker's
2651 .Dl "ipfw add deny ip from 123.45.67.0/24 to my.host.org"
2653 A first and efficient way to limit access (not using dynamic rules)
2654 is the use of the following rules:
2656 .Dl "ipfw add allow tcp from any to any established"
2657 .Dl "ipfw add allow tcp from net1 portlist1 to net2 portlist2 setup"
2658 .Dl "ipfw add allow tcp from net3 portlist3 to net3 portlist3 setup"
2660 .Dl "ipfw add deny tcp from any to any"
2662 The first rule will be a quick match for normal TCP packets,
2663 but it will not match the initial SYN packet, which will be
2666 rules only for selected source/destination pairs.
2667 All other SYN packets will be rejected by the final
2671 If you administer one or more subnets, you can take advantage
2672 of the address sets and or-blocks and write extremely
2673 compact rulesets which selectively enable services to blocks
2674 of clients, as below:
2676 .Dl "goodguys=\*q{ 10.1.2.0/24{20,35,66,18} or 10.2.3.0/28{6,3,11} }\*q"
2677 .Dl "badguys=\*q10.1.2.0/24{8,38,60}\*q"
2679 .Dl "ipfw add allow ip from ${goodguys} to any"
2680 .Dl "ipfw add deny ip from ${badguys} to any"
2681 .Dl "... normal policies ..."
2685 option could be used to do automated anti-spoofing by adding the
2686 following to the top of a ruleset:
2688 .Dl "ipfw add deny ip from any to any not verrevpath in"
2690 This rule drops all incoming packets that appear to be coming to the
2691 system on the wrong interface.
2692 For example, a packet with a source
2693 address belonging to a host on a protected internal network would be
2694 dropped if it tried to enter the system from an external interface.
2698 option could be used to do similar but more restricted anti-spoofing
2699 by adding the following to the top of a ruleset:
2701 .Dl "ipfw add deny ip from any to any not antispoof in"
2703 This rule drops all incoming packets that appear to be coming from another
2704 directly connected system but on the wrong interface.
2705 For example, a packet with a source address of
2706 .Li 192.168.0.0/24 ,
2713 In order to protect a site from flood attacks involving fake
2714 TCP packets, it is safer to use dynamic rules:
2716 .Dl "ipfw add check-state"
2717 .Dl "ipfw add deny tcp from any to any established"
2718 .Dl "ipfw add allow tcp from my-net to any setup keep-state"
2720 This will let the firewall install dynamic rules only for
2721 those connection which start with a regular SYN packet coming
2722 from the inside of our network.
2723 Dynamic rules are checked when encountering the first
2730 rule should usually be placed near the beginning of the
2731 ruleset to minimize the amount of work scanning the ruleset.
2732 Your mileage may vary.
2734 To limit the number of connections a user can open
2735 you can use the following type of rules:
2737 .Dl "ipfw add allow tcp from my-net/24 to any setup limit src-addr 10"
2738 .Dl "ipfw add allow tcp from any to me setup limit src-addr 4"
2740 The former (assuming it runs on a gateway) will allow each host
2741 on a /24 network to open at most 10 TCP connections.
2742 The latter can be placed on a server to make sure that a single
2743 client does not use more than 4 simultaneous connections.
2746 stateful rules can be subject to denial-of-service attacks
2747 by a SYN-flood which opens a huge number of dynamic rules.
2748 The effects of such attacks can be partially limited by
2751 variables which control the operation of the firewall.
2753 Here is a good usage of the
2755 command to see accounting records and timestamp information:
2759 or in short form without timestamps:
2763 which is equivalent to:
2767 Next rule diverts all incoming packets from 192.168.2.0/24
2768 to divert port 5000:
2770 .Dl ipfw divert 5000 ip from 192.168.2.0/24 to any in
2773 The following rules show some of the applications of
2777 for simulations and the like.
2779 This rule drops random incoming packets with a probability
2782 .Dl "ipfw add prob 0.05 deny ip from any to any in"
2784 A similar effect can be achieved making use of
2788 .Dl "ipfw add pipe 10 ip from any to any"
2789 .Dl "ipfw pipe 10 config plr 0.05"
2791 We can use pipes to artificially limit bandwidth, e.g.\& on a
2792 machine acting as a router, if we want to limit traffic from
2793 local clients on 192.168.2.0/24 we do:
2795 .Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
2796 .Dl "ipfw pipe 1 config bw 300Kbit/s queue 50KBytes"
2798 note that we use the
2800 modifier so that the rule is not used twice.
2801 Remember in fact that
2803 rules are checked both on incoming and outgoing packets.
2805 Should we want to simulate a bidirectional link with bandwidth
2806 limitations, the correct way is the following:
2808 .Dl "ipfw add pipe 1 ip from any to any out"
2809 .Dl "ipfw add pipe 2 ip from any to any in"
2810 .Dl "ipfw pipe 1 config bw 64Kbit/s queue 10Kbytes"
2811 .Dl "ipfw pipe 2 config bw 64Kbit/s queue 10Kbytes"
2813 The above can be very useful, e.g.\& if you want to see how
2814 your fancy Web page will look for a residential user who
2815 is connected only through a slow link.
2816 You should not use only one pipe for both directions, unless
2817 you want to simulate a half-duplex medium (e.g.\& AppleTalk,
2819 It is not necessary that both pipes have the same configuration,
2820 so we can also simulate asymmetric links.
2822 Should we want to verify network performance with the RED queue
2823 management algorithm:
2825 .Dl "ipfw add pipe 1 ip from any to any"
2826 .Dl "ipfw pipe 1 config bw 500Kbit/s queue 100 red 0.002/30/80/0.1"
2828 Another typical application of the traffic shaper is to
2829 introduce some delay in the communication.
2830 This can significantly affect applications which do a lot of Remote
2831 Procedure Calls, and where the round-trip-time of the
2832 connection often becomes a limiting factor much more than
2835 .Dl "ipfw add pipe 1 ip from any to any out"
2836 .Dl "ipfw add pipe 2 ip from any to any in"
2837 .Dl "ipfw pipe 1 config delay 250ms bw 1Mbit/s"
2838 .Dl "ipfw pipe 2 config delay 250ms bw 1Mbit/s"
2840 Per-flow queueing can be useful for a variety of purposes.
2841 A very simple one is counting traffic:
2843 .Dl "ipfw add pipe 1 tcp from any to any"
2844 .Dl "ipfw add pipe 1 udp from any to any"
2845 .Dl "ipfw add pipe 1 ip from any to any"
2846 .Dl "ipfw pipe 1 config mask all"
2848 The above set of rules will create queues (and collect
2849 statistics) for all traffic.
2850 Because the pipes have no limitations, the only effect is
2851 collecting statistics.
2852 Note that we need 3 rules, not just the last one, because
2855 tries to match IP packets it will not consider ports, so we
2856 would not see connections on separate ports as different
2859 A more sophisticated example is limiting the outbound traffic
2860 on a net with per-host limits, rather than per-network limits:
2862 .Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
2863 .Dl "ipfw add pipe 2 ip from any to 192.168.2.0/24 in"
2864 .Dl "ipfw pipe 1 config mask src-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
2865 .Dl "ipfw pipe 2 config mask dst-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
2867 In the following example, we need to create several traffic bandwidth
2868 classes and we need different hosts/networks to fall into different classes.
2869 We create one pipe for each class and configure them accordingly.
2870 Then we create a single table and fill it with IP subnets and addresses.
2871 For each subnet/host we set the argument equal to the number of the pipe
2873 Then we classify traffic using a single rule:
2875 .Dl "ipfw pipe 1 config bw 1000Kbyte/s"
2876 .Dl "ipfw pipe 4 config bw 4000Kbyte/s"
2878 .Dl "ipfw table 1 add 192.168.2.0/24 1"
2879 .Dl "ipfw table 1 add 192.168.0.0/27 4"
2880 .Dl "ipfw table 1 add 192.168.0.2 1"
2882 .Dl "ipfw add pipe tablearg ip from table(1) to any"
2886 action, the table entries may include hostnames and IP addresses.
2888 .Dl "ipfw table 1 add 192.168.2.0/24 10.23.2.1"
2889 .Dl "ipfw table 1 add 192.168.0.0/27 router1.dmz"
2891 .Dl "ipfw add 100 fwd tablearg ip from any to table(1)"
2893 To add a set of rules atomically, e.g.\& set 18:
2895 .Dl "ipfw set disable 18"
2896 .Dl "ipfw add NN set 18 ... # repeat as needed"
2897 .Dl "ipfw set enable 18"
2899 To delete a set of rules atomically the command is simply:
2901 .Dl "ipfw delete set 18"
2903 To test a ruleset and disable it and regain control if something goes wrong:
2905 .Dl "ipfw set disable 18"
2906 .Dl "ipfw add NN set 18 ... # repeat as needed"
2907 .Dl "ipfw set enable 18; echo done; sleep 30 && ipfw set disable 18"
2909 Here if everything goes well, you press control-C before the "sleep"
2910 terminates, and your ruleset will be left active.
2911 Otherwise, e.g.\& if
2912 you cannot access your box, the ruleset will be disabled after
2913 the sleep terminates thus restoring the previous situation.
2915 To show rules of the specific set:
2917 .Dl "ipfw set 18 show"
2919 To show rules of the disabled set:
2921 .Dl "ipfw -S set 18 show"
2923 To clear a specific rule counters of the specific set:
2925 .Dl "ipfw set 18 zero NN"
2927 To delete a specific rule of the specific set:
2929 .Dl "ipfw set 18 delete NN"
2930 .Ss NAT, REDIRECT AND LSNAT
2931 First redirect all the traffic to nat instance 123:
2933 .Dl "ipfw add nat 123 all from any to any"
2935 Then to configure nat instance 123 to alias all the outgoing traffic with ip
2936 192.168.0.123, blocking all incoming connections, trying to keep
2937 same ports on both sides, clearing aliasing table on address change
2938 and keeping a log of traffic/link statistics:
2940 .Dl "ipfw nat 123 config ip 192.168.0.123 log deny_in reset same_ports"
2942 Or to change address of instance 123, aliasing table will be cleared (see
2945 .Dl "ipfw nat 123 config ip 10.0.0.1"
2947 To see configuration of nat instance 123:
2949 .Dl "ipfw nat 123 show config"
2951 To show logs of all the instances in range 111-999:
2953 .Dl "ipfw nat 111-999 show"
2955 To see configurations of all instances:
2957 .Dl "ipfw nat show config"
2959 Or a redirect rule with mixed modes could looks like:
2961 .Dl "ipfw nat 123 config redirect_addr 10.0.0.1 10.0.0.66"
2962 .Dl " redirect_port tcp 192.168.0.1:80 500"
2963 .Dl " redirect_proto udp 192.168.1.43 192.168.1.1"
2964 .Dl " redirect_addr 192.168.0.10,192.168.0.11"
2965 .Dl " 10.0.0.100 # LSNAT"
2966 .Dl " redirect_port tcp 192.168.0.1:80,192.168.0.10:22"
2969 or it could be split in:
2971 .Dl "ipfw nat 1 config redirect_addr 10.0.0.1 10.0.0.66"
2972 .Dl "ipfw nat 2 config redirect_port tcp 192.168.0.1:80 500"
2973 .Dl "ipfw nat 3 config redirect_proto udp 192.168.1.43 192.168.1.1"
2974 .Dl "ipfw nat 4 config redirect_addr 192.168.0.10,192.168.0.11,192.168.0.12"
2976 .Dl "ipfw nat 5 config redirect_port tcp"
2977 .Dl " 192.168.0.1:80,192.168.0.10:22,192.168.0.20:25 500"
2999 utility first appeared in
3004 Stateful extensions were introduced in
3007 was introduced in Summer 2002.
3009 .An Ugen J. S. Antsilevich ,
3010 .An Poul-Henning Kamp ,
3016 API based upon code written by
3020 Dummynet has been introduced by Luigi Rizzo in 1997-1998.
3022 Some early work (1999-2000) on the
3024 traffic shaper supported by Akamba Corp.
3026 The ipfw core (ipfw2) has been completely redesigned and
3027 reimplemented by Luigi Rizzo in summer 2002. Further
3029 options have been added by various developer over the years.
3032 In-kernel NAT support written by
3033 .An Paolo Pisati Aq piso@FreeBSD.org
3034 as part of a Summer of Code 2005 project.
3038 support has been developed by
3039 .An The Centre for Advanced Internet Architectures (CAIA) Aq http://www.caia.swin.edu.au .
3040 The primary developers and maintainers are David Hayes and Jason But.
3041 For further information visit:
3042 .Aq http://www.caia.swin.edu.au/urp/SONATA
3044 Delay profiles have been developed by Alessandro Cerri and
3045 Luigi Rizzo, supported by the
3046 European Commission within Projects Onelab and Onelab2.
3048 The syntax has grown over the years and sometimes it might be confusing.
3049 Unfortunately, backward compatibility prevents cleaning up mistakes
3050 made in the definition of the syntax.
3054 Misconfiguring the firewall can put your computer in an unusable state,
3055 possibly shutting down network services and requiring console access to
3056 regain control of it.
3058 Incoming packet fragments diverted by
3060 are reassembled before delivery to the socket.
3061 The action used on those packet is the one from the
3062 rule which matches the first fragment of the packet.
3064 Packets diverted to userland, and then reinserted by a userland process
3065 may lose various packet attributes.
3066 The packet source interface name
3067 will be preserved if it is shorter than 8 bytes and the userland process
3068 saves and reuses the sockaddr_in
3071 otherwise, it may be lost.
3072 If a packet is reinserted in this manner, later rules may be incorrectly
3073 applied, making the order of
3075 rules in the rule sequence very important.
3077 Dummynet drops all packets with IPv6 link-local addresses.
3083 may not behave as expected.
3084 In particular, incoming SYN packets may
3085 have no uid or gid associated with them since they do not yet belong
3086 to a TCP connection, and the uid/gid associated with a packet may not
3087 be as expected if the associated process calls
3089 or similar system calls.
3091 Rule syntax is subject to the command line environment and some patterns
3092 may need to be escaped with the backslash character
3093 or quoted appropriately.
3095 Due to the architecture of
3097 ipfw nat is not compatible with the TCP segmentation offloading (TSO).
3098 Thus, to reliably nat your network traffic, please disable TSO
3102 ICMP error messages are not implicitly matched by dynamic rules
3103 for the respective conversations.
3104 To avoid failures of network error detection and path MTU discovery,
3105 ICMP error messages may need to be allowed explicitly through static