9 .Nd User interface for firewall, traffic shaper, packet scheduler,
12 .Ss FIREWALL CONFIGURATION
21 .Op Ar rule | first-last ...
29 .Brq Cm delete | zero | resetlog
33 .Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
37 .Ar number Cm to Ar number
39 .Cm set swap Ar number number
45 .Brq Cm firewall | altq | one_pass | debug | verbose | dyn_keepalive
48 .Brq Cm firewall | altq | one_pass | debug | verbose | dyn_keepalive
51 .Oo Cm set Ar N Oc Cm table Ar name Cm create Ar create-options
53 .Oo Cm set Ar N Oc Cm table
57 .Oo Cm set Ar N Oc Cm table Ar name Cm modify Ar modify-options
59 .Oo Cm set Ar N Oc Cm table Ar name Cm swap Ar name
61 .Oo Cm set Ar N Oc Cm table Ar name Cm add Ar table-key Op Ar value
63 .Oo Cm set Ar N Oc Cm table Ar name Cm add Op Ar table-key Ar value ...
65 .Oo Cm set Ar N Oc Cm table Ar name Cm atomic add Op Ar table-key Ar value ...
67 .Oo Cm set Ar N Oc Cm table Ar name Cm delete Op Ar table-key ...
69 .Oo Cm set Ar N Oc Cm table Ar name Cm lookup Ar addr
71 .Oo Cm set Ar N Oc Cm table Ar name Cm lock
73 .Oo Cm set Ar N Oc Cm table Ar name Cm unlock
75 .Oo Cm set Ar N Oc Cm table
79 .Oo Cm set Ar N Oc Cm table
83 .Oo Cm set Ar N Oc Cm table
87 .Oo Cm set Ar N Oc Cm table
90 .Ss DUMMYNET CONFIGURATION (TRAFFIC SHAPER AND PACKET SCHEDULER)
92 .Brq Cm pipe | queue | sched
98 .Brq Cm pipe | queue | sched
99 .Brq Cm delete | list | show
118 .Ss STATEFUL IPv6/IPv4 NETWORK ADDRESS AND PROTOCOL TRANSLATION
120 .Oo Cm set Ar N Oc Cm nat64lsn Ar name Cm create Ar create-options
122 .Oo Cm set Ar N Oc Cm nat64lsn Ar name Cm config Ar config-options
124 .Oo Cm set Ar N Oc Cm nat64lsn
129 .Oo Cm set Ar N Oc Cm nat64lsn
133 .Oo Cm set Ar N Oc Cm nat64lsn Ar name Cm stats Op Cm reset
134 .Ss STATELESS IPv6/IPv4 NETWORK ADDRESS AND PROTOCOL TRANSLATION
136 .Oo Cm set Ar N Oc Cm nat64stl Ar name Cm create Ar create-options
138 .Oo Cm set Ar N Oc Cm nat64stl Ar name Cm config Ar config-options
140 .Oo Cm set Ar N Oc Cm nat64stl
144 .Oo Cm set Ar N Oc Cm nat64stl
148 .Oo Cm set Ar N Oc Cm nat64stl Ar name Cm stats Op Cm reset
149 .Ss IPv6-to-IPv6 NETWORK PREFIX TRANSLATION
151 .Oo Cm set Ar N Oc Cm nptv6 Ar name Cm create Ar create-options
153 .Oo Cm set Ar N Oc Cm nptv6
157 .Oo Cm set Ar N Oc Cm nptv6
161 .Oo Cm set Ar N Oc Cm nptv6 Ar name Cm stats Op Cm reset
162 .Ss INTERNAL DIAGNOSTICS
172 utility is the user interface for controlling the
176 traffic shaper/packet scheduler, and the
177 in-kernel NAT services.
179 A firewall configuration, or
183 numbered from 1 to 65535.
184 Packets are passed to the firewall
185 from a number of different places in the protocol stack
186 (depending on the source and destination of the packet,
187 it is possible for the firewall to be
188 invoked multiple times on the same packet).
189 The packet passed to the firewall is compared
190 against each of the rules in the
193 (multiple rules with the same number are permitted, in which case
194 they are processed in order of insertion).
195 When a match is found, the action corresponding to the
196 matching rule is performed.
198 Depending on the action and certain system settings, packets
199 can be reinjected into the firewall at some rule after the
200 matching one for further processing.
202 A ruleset always includes a
204 rule (numbered 65535) which cannot be modified or deleted,
205 and matches all packets.
206 The action associated with the
212 depending on how the kernel is configured.
214 If the ruleset includes one or more rules with the
219 the firewall will have a
221 behaviour, i.e., upon a match it will create
223 i.e., rules that match packets with the same 5-tuple
224 (protocol, source and destination addresses and ports)
225 as the packet which caused their creation.
226 Dynamic rules, which have a limited lifetime, are checked
227 at the first occurrence of a
232 rule, and are typically used to open the firewall on-demand to
233 legitimate traffic only.
235 .Sx STATEFUL FIREWALL
238 Sections below for more information on the stateful behaviour of
241 All rules (including dynamic ones) have a few associated counters:
242 a packet count, a byte count, a log count and a timestamp
243 indicating the time of the last match.
244 Counters can be displayed or reset with
248 Each rule belongs to one of 32 different
252 commands to atomically manipulate sets, such as enable,
253 disable, swap sets, move all rules in a set to another
254 one, delete all rules in a set.
255 These can be useful to
256 install temporary configurations, or to test them.
259 for more information on
262 Rules can be added with the
264 command; deleted individually or in groups with the
266 command, and globally (except those in set 31) with the
268 command; displayed, optionally with the content of the
274 Finally, counters can be reset with the
281 The following general options are available when invoking
283 .Bl -tag -width indent
285 Show counter values when listing rules.
288 command implies this option.
290 Only show the action and the comment, not the body of a rule.
294 When entering or showing rules, print them in compact form,
295 i.e., omitting the "ip from any to any" string
296 when this does not carry any additional information.
298 When listing, show dynamic rules in addition to static ones.
302 is specified, also show expired dynamic rules.
304 Do not ask for confirmation for commands that can cause problems
307 If there is no tty associated with the process, this is implied.
309 When listing a table (see the
311 section below for more information on lookup tables), format values
313 By default, values are shown as integers.
315 Only check syntax of the command strings, without actually passing
318 Try to resolve addresses and service names in output.
320 Be quiet when executing the
330 This is useful when updating rulesets by executing multiple
334 .Ql sh\ /etc/rc.firewall ) ,
335 or by processing a file with many
337 rules across a remote login session.
338 It also stops a table add or delete
339 from failing if the entry already exists or is not present.
341 The reason why this option may be important is that
342 for some of these actions,
344 may print a message; if the action results in blocking the
345 traffic to the remote client,
346 the remote login session will be closed
347 and the rest of the ruleset will not be processed.
348 Access to the console would then be required to recover.
350 When listing rules, show the
352 each rule belongs to.
353 If this flag is not specified, disabled rules will not be
356 When listing pipes, sort according to one of the four
357 counters (total or current packets or bytes).
359 When listing, show last match timestamp converted with ctime().
361 When listing, show last match timestamp as seconds from the epoch.
362 This form can be more convenient for postprocessing by scripts.
364 .Ss LIST OF RULES AND PREPROCESSING
365 To ease configuration, rules can be put into a file which is
368 as shown in the last synopsis line.
372 The file will be read line by line and applied as arguments to the
376 Optionally, a preprocessor can be specified using
380 is to be piped through.
381 Useful preprocessors include
387 does not start with a slash
389 as its first character, the usual
391 name search is performed.
392 Care should be taken with this in environments where not all
393 file systems are mounted (yet) by the time
395 is being run (e.g.\& when they are mounted over NFS).
398 has been specified, any additional arguments are passed on to the preprocessor
400 This allows for flexible configuration files (like conditionalizing
401 them on the local hostname) and the use of macros to centralize
402 frequently required arguments like IP addresses.
403 .Ss TRAFFIC SHAPER CONFIGURATION
409 commands are used to configure the traffic shaper and packet scheduler.
411 .Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
412 Section below for details.
414 If the world and the kernel get out of sync the
416 ABI may break, preventing you from being able to add any rules.
417 This can adversely affect the booting process.
422 to temporarily disable the firewall to regain access to the network,
423 allowing you to fix the problem.
425 A packet is checked against the active ruleset in multiple places
426 in the protocol stack, under control of several sysctl variables.
427 These places and variables are shown below, and it is important to
428 have this picture in mind in order to design a correct ruleset.
429 .Bd -literal -offset indent
432 +----------->-----------+
434 [ip(6)_input] [ip(6)_output] net.inet(6).ip(6).fw.enable=1
437 [ether_demux] [ether_output_frame] net.link.ether.ipfw=1
439 +-->--[bdg_forward]-->--+ net.link.bridge.ipfw=1
445 times the same packet goes through the firewall can
446 vary between 0 and 4 depending on packet source and
447 destination, and system configuration.
449 Note that as packets flow through the stack, headers can be
450 stripped or added to it, and so they may or may not be available
452 E.g., incoming packets will include the MAC header when
456 but the same packets will have the MAC header stripped off when
463 Also note that each packet is always checked against the complete ruleset,
464 irrespective of the place where the check occurs, or the source of the packet.
465 If a rule contains some match patterns or actions which are not valid
466 for the place of invocation (e.g.\& trying to match a MAC header within
470 the match pattern will not match, but a
472 operator in front of such patterns
476 match on those packets.
477 It is thus the responsibility of
478 the programmer, if necessary, to write a suitable ruleset to
479 differentiate among the possible places.
481 rules can be useful here, as an example:
482 .Bd -literal -offset indent
483 # packets from ether_demux or bdg_forward
484 ipfw add 10 skipto 1000 all from any to any layer2 in
485 # packets from ip_input
486 ipfw add 10 skipto 2000 all from any to any not layer2 in
487 # packets from ip_output
488 ipfw add 10 skipto 3000 all from any to any not layer2 out
489 # packets from ether_output_frame
490 ipfw add 10 skipto 4000 all from any to any layer2 out
493 (yes, at the moment there is no way to differentiate between
494 ether_demux and bdg_forward).
496 In general, each keyword or argument must be provided as
497 a separate command line argument, with no leading or trailing
499 Keywords are case-sensitive, whereas arguments may
500 or may not be case-sensitive depending on their nature
501 (e.g.\& uid's are, hostnames are not).
503 Some arguments (e.g., port or address lists) are comma-separated
505 In this case, spaces after commas ',' are allowed to make
506 the line more readable.
507 You can also put the entire
508 command (including flags) into a single argument.
509 E.g., the following forms are equivalent:
510 .Bd -literal -offset indent
511 ipfw -q add deny src-ip 10.0.0.0/24,127.0.0.1/8
512 ipfw -q add deny src-ip 10.0.0.0/24, 127.0.0.1/8
513 ipfw "-q add deny src-ip 10.0.0.0/24, 127.0.0.1/8"
516 The format of firewall rules is the following:
517 .Bd -ragged -offset indent
520 .Op Cm set Ar set_number
521 .Op Cm prob Ar match_probability
523 .Op Cm log Op Cm logamount Ar number
533 where the body of the rule specifies which information is used
534 for filtering packets, among the following:
536 .Bl -tag -width "Source and dest. addresses and ports" -offset XXX -compact
537 .It Layer-2 header fields
539 .It IPv4 and IPv6 Protocol
540 SCTP, TCP, UDP, ICMP, etc.
541 .It Source and dest. addresses and ports
545 .It Transmit and receive interface
547 .It Misc. IP header fields
548 Version, type of service, datagram length, identification,
549 fragment flag (non-zero IP offset),
552 .It IPv6 Extension headers
553 Fragmentation, Hop-by-Hop options,
554 Routing Headers, Source routing rthdr0, Mobile IPv6 rthdr2, IPSec options.
556 .It Misc. TCP header fields
557 TCP flags (SYN, FIN, ACK, RST, etc.),
558 sequence number, acknowledgment number,
566 When the packet can be associated with a local socket.
568 Whether a packet came from a divert socket (e.g.,
570 .It Fib annotation state
571 Whether a packet has been tagged for using a specific FIB (routing table)
572 in future forwarding decisions.
575 Note that some of the above information, e.g.\& source MAC or IP addresses and
576 TCP/UDP ports, can be easily spoofed, so filtering on those fields
577 alone might not guarantee the desired results.
578 .Bl -tag -width indent
580 Each rule is associated with a
582 in the range 1..65535, with the latter reserved for the
585 Rules are checked sequentially by rule number.
586 Multiple rules can have the same number, in which case they are
587 checked (and listed) according to the order in which they have
589 If a rule is entered without specifying a number, the kernel will
590 assign one in such a way that the rule becomes the last one
594 Automatic rule numbers are assigned by incrementing the last
595 non-default rule number by the value of the sysctl variable
596 .Ar net.inet.ip.fw.autoinc_step
597 which defaults to 100.
598 If this is not possible (e.g.\& because we would go beyond the
599 maximum allowed rule number), the number of the last
600 non-default value is used instead.
601 .It Cm set Ar set_number
602 Each rule is associated with a
605 Sets can be individually disabled and enabled, so this parameter
606 is of fundamental importance for atomic ruleset manipulation.
607 It can be also used to simplify deletion of groups of rules.
608 If a rule is entered without specifying a set number,
611 Set 31 is special in that it cannot be disabled,
612 and rules in set 31 are not deleted by the
614 command (but you can delete them with the
615 .Nm ipfw delete set 31
617 Set 31 is also used for the
620 .It Cm prob Ar match_probability
621 A match is only declared with the specified probability
622 (floating point number between 0 and 1).
623 This can be useful for a number of applications such as
624 random packet drop or
627 to simulate the effect of multiple paths leading to out-of-order
630 Note: this condition is checked before any other condition, including
631 ones such as keep-state or check-state which might have side effects.
632 .It Cm log Op Cm logamount Ar number
633 Packets matching a rule with the
635 keyword will be made available for logging in two ways:
636 if the sysctl variable
637 .Va net.inet.ip.fw.verbose
638 is set to 0 (default), one can use
643 This pseudo interface can be created after a boot
644 manually by using the following command:
645 .Bd -literal -offset indent
646 # ifconfig ipfw0 create
649 Or, automatically at boot time by adding the following
653 .Bd -literal -offset indent
657 There is no overhead if no
659 is attached to the pseudo interface.
662 .Va net.inet.ip.fw.verbose
663 is set to 1, packets will be logged to
667 facility up to a maximum of
672 is specified, the limit is taken from the sysctl variable
673 .Va net.inet.ip.fw.verbose_limit .
674 In both cases, a value of 0 means unlimited logging.
676 Once the limit is reached, logging can be re-enabled by
677 clearing the logging counter or the packet counter for that entry, see the
681 Note: logging is done after all other packet matching conditions
682 have been successfully verified, and before performing the final
683 action (accept, deny, etc.) on the packet.
685 When a packet matches a rule with the
687 keyword, the numeric tag for the given
689 in the range 1..65534 will be attached to the packet.
690 The tag acts as an internal marker (it is not sent out over
691 the wire) that can be used to identify these packets later on.
692 This can be used, for example, to provide trust between interfaces
693 and to start doing policy-based filtering.
694 A packet can have multiple tags at the same time.
695 Tags are "sticky", meaning once a tag is applied to a packet by a
696 matching rule it exists until explicit removal.
697 Tags are kept with the packet everywhere within the kernel, but are
698 lost when packet leaves the kernel, for example, on transmitting
699 packet out to the network or sending packet to a
703 To check for previously applied tags, use the
706 To delete previously applied tag, use the
710 Note: since tags are kept with the packet everywhere in kernelspace,
711 they can be set and unset anywhere in the kernel network subsystem
714 facility), not only by means of the
720 For example, there can be a specialized
722 node doing traffic analyzing and tagging for later inspecting
724 .It Cm untag Ar number
725 When a packet matches a rule with the
727 keyword, the tag with the number
729 is searched among the tags attached to this packet and,
730 if found, removed from it.
731 Other tags bound to packet, if present, are left untouched.
733 When a packet matches a rule with the
735 keyword, the ALTQ identifier for the given
740 Note that this ALTQ tag is only meaningful for packets going "out" of IPFW,
741 and not being rejected or going to divert sockets.
742 Note that if there is insufficient memory at the time the packet is
743 processed, it will not be tagged, so it is wise to make your ALTQ
744 "default" queue policy account for this.
747 rules match a single packet, only the first one adds the ALTQ classification
749 In doing so, traffic may be shaped by using
750 .Cm count Cm altq Ar queue
751 rules for classification early in the ruleset, then later applying
752 the filtering decision.
757 rules may come later and provide the actual filtering decisions in
758 addition to the fallback ALTQ tag.
762 to set up the queues before IPFW will be able to look them up by name,
763 and if the ALTQ disciplines are rearranged, the rules in containing the
764 queue identifiers in the kernel will likely have gone stale and need
766 Stale queue identifiers will probably result in misclassification.
768 All system ALTQ processing can be turned on or off via
773 .Cm disable Ar altq .
775 .Va net.inet.ip.fw.one_pass
776 is irrelevant to ALTQ traffic shaping, as the actual rule action is followed
777 always after adding an ALTQ tag.
780 A rule can be associated with one of the following actions, which
781 will be executed when the packet matches the body of the rule.
782 .Bl -tag -width indent
783 .It Cm allow | accept | pass | permit
784 Allow packets that match rule.
785 The search terminates.
786 .It Cm check-state Op Ar :flowname | Cm :any
787 Checks the packet against the dynamic ruleset.
788 If a match is found, execute the action associated with
789 the rule which generated this dynamic rule, otherwise
790 move to the next rule.
793 rules do not have a body.
796 rule is found, the dynamic ruleset is checked at the first
803 is symbolic name assigned to dynamic rule by
808 can be used to ignore states flowname when matching.
811 keyword is special name used for compatibility with old rulesets.
813 Update counters for all packets that match rule.
814 The search continues with the next rule.
816 Discard packets that match this rule.
817 The search terminates.
818 .It Cm divert Ar port
819 Divert packets that match this rule to the
823 The search terminates.
824 .It Cm fwd | forward Ar ipaddr | tablearg Ns Op , Ns Ar port
825 Change the next-hop on matching packets to
827 which can be an IP address or a host name.
828 The next hop can also be supplied by the last table
829 looked up for the packet by using the
831 keyword instead of an explicit address.
832 The search terminates if this rule matches.
836 is a local address, then matching packets will be forwarded to
838 (or the port number in the packet if one is not specified in the rule)
839 on the local machine.
843 is not a local address, then the port number
844 (if specified) is ignored, and the packet will be
845 forwarded to the remote address, using the route as found in
846 the local routing table for that IP.
850 rule will not match layer-2 packets (those received
851 on ether_input, ether_output, or bridged).
855 action does not change the contents of the packet at all.
856 In particular, the destination address remains unmodified, so
857 packets forwarded to another system will usually be rejected by that system
858 unless there is a matching rule on that system to capture them.
859 For packets forwarded locally,
860 the local address of the socket will be
861 set to the original destination address of the packet.
864 entry look rather weird but is intended for
865 use with transparent proxy servers.
866 .It Cm nat Ar nat_nr | tablearg
869 (for network address translation, address redirect, etc.):
871 .Sx NETWORK ADDRESS TRANSLATION (NAT)
872 Section for further information.
873 .It Cm nat64lsn Ar name
874 Pass packet to a stateful NAT64 instance (for IPv6/IPv4 network address and
875 protocol translation): see the
876 .Sx IPv6/IPv4 NETWORK ADDRESS AND PROTOCOL TRANSLATION
877 Section for further information.
878 .It Cm nat64stl Ar name
879 Pass packet to a stateless NAT64 instance (for IPv6/IPv4 network address and
880 protocol translation): see the
881 .Sx IPv6/IPv4 NETWORK ADDRESS AND PROTOCOL TRANSLATION
882 Section for further information.
884 Pass packet to a NPTv6 instance (for IPv6-to-IPv6 network prefix translation):
886 .Sx IPv6-to-IPv6 NETWORK PREFIX TRANSLATION (NPTv6)
887 Section for further information.
888 .It Cm pipe Ar pipe_nr
892 (for bandwidth limitation, delay, etc.).
894 .Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
895 Section for further information.
896 The search terminates; however, on exit from the pipe and if
900 .Va net.inet.ip.fw.one_pass
901 is not set, the packet is passed again to the firewall code
902 starting from the next rule.
903 .It Cm queue Ar queue_nr
907 (for bandwidth limitation using WF2Q+).
913 Discard packets that match this rule, and if the
914 packet is a TCP packet, try to send a TCP reset (RST) notice.
915 The search terminates.
917 Discard packets that match this rule, and if the
918 packet is a TCP packet, try to send a TCP reset (RST) notice.
919 The search terminates.
920 .It Cm skipto Ar number | tablearg
921 Skip all subsequent rules numbered less than
923 The search continues with the first rule numbered
926 It is possible to use the
928 keyword with a skipto for a
930 skipto. Skipto may work either in O(log(N)) or in O(1) depending
931 on amount of memory and/or sysctl variables.
934 section for more details.
935 .It Cm call Ar number | tablearg
936 The current rule number is saved in the internal stack and
937 ruleset processing continues with the first rule numbered
940 If later a rule with the
942 action is encountered, the processing returns to the first rule
945 rule plus one or higher
946 (the same behaviour as with packets returning from
951 This could be used to make somewhat like an assembly language
953 calls to rules with common checks for different interfaces, etc.
955 Rule with any number could be called, not just forward jumps as with
957 So, to prevent endless loops in case of mistakes, both
961 actions don't do any jumps and simply go to the next rule if memory
962 cannot be allocated or stack overflowed/underflowed.
964 Internally stack for rule numbers is implemented using
966 facility and currently has size of 16 entries.
967 As mbuf tags are lost when packet leaves the kernel,
969 should not be used in subroutines to avoid endless loops
970 and other undesired effects.
972 Takes rule number saved to internal stack by the last
974 action and returns ruleset processing to the first rule
975 with number greater than number of corresponding
978 See description of the
980 action for more details.
986 and thus are unconditional, but
988 command-line utility currently requires every action except
991 While it is sometimes useful to return only on some packets,
992 usually you want to print just
995 A workaround for this is to use new syntax and
998 .Bd -literal -offset indent
999 # Add a rule without actual body
1000 ipfw add 2999 return via any
1002 # List rules without "from any to any" part
1006 This cosmetic annoyance may be fixed in future releases.
1008 Send a copy of packets matching this rule to the
1010 socket bound to port
1012 The search continues with the next rule.
1013 .It Cm unreach Ar code
1014 Discard packets that match this rule, and try to send an ICMP
1015 unreachable notice with code
1019 is a number from 0 to 255, or one of these aliases:
1020 .Cm net , host , protocol , port ,
1021 .Cm needfrag , srcfail , net-unknown , host-unknown ,
1022 .Cm isolated , net-prohib , host-prohib , tosnet ,
1023 .Cm toshost , filter-prohib , host-precedence
1025 .Cm precedence-cutoff .
1026 The search terminates.
1027 .It Cm unreach6 Ar code
1028 Discard packets that match this rule, and try to send an ICMPv6
1029 unreachable notice with code
1033 is a number from 0, 1, 3 or 4, or one of these aliases:
1034 .Cm no-route, admin-prohib, address
1037 The search terminates.
1038 .It Cm netgraph Ar cookie
1039 Divert packet into netgraph with given
1041 The search terminates.
1042 If packet is later returned from netgraph it is either
1043 accepted or continues with the next rule, depending on
1044 .Va net.inet.ip.fw.one_pass
1046 .It Cm ngtee Ar cookie
1047 A copy of packet is diverted into netgraph, original
1048 packet continues with the next rule.
1051 for more information on
1056 .It Cm setfib Ar fibnum | tablearg
1057 The packet is tagged so as to use the FIB (routing table)
1059 in any subsequent forwarding decisions.
1060 In the current implementation, this is limited to the values 0 through 15, see
1062 Processing continues at the next rule.
1063 It is possible to use the
1065 keyword with setfib.
1066 If the tablearg value is not within the compiled range of fibs,
1067 the packet's fib is set to 0.
1068 .It Cm setdscp Ar DSCP | number | tablearg
1069 Set specified DiffServ codepoint for an IPv4/IPv6 packet.
1070 Processing continues at the next rule.
1071 Supported values are:
1117 Additionally, DSCP value can be specified by number (0..64).
1118 It is also possible to use the
1120 keyword with setdscp.
1121 If the tablearg value is not within the 0..64 range, lower 6 bits of supplied
1123 .It Cm tcp-setmss Ar mss
1124 Set the Maximum Segment Size (MSS) in the TCP segment to value
1128 should be loaded or kernel should have
1129 .Cm options IPFIREWALL_PMOD
1130 to be able use this action.
1131 This command does not change a packet if original MSS value is lower than
1133 Both TCP over IPv4 and over IPv6 are supported.
1134 Regardless of matched a packet or not by the
1136 rule, the search continues with the next rule.
1138 Queue and reassemble IPv4 fragments.
1139 If the packet is not fragmented, counters are updated and
1140 processing continues with the next rule.
1141 If the packet is the last logical fragment, the packet is reassembled and, if
1142 .Va net.inet.ip.fw.one_pass
1143 is set to 0, processing continues with the next rule.
1144 Otherwise, the packet is allowed to pass and the search terminates.
1145 If the packet is a fragment in the middle of a logical group of fragments,
1147 processing stops immediately.
1149 Fragment handling can be tuned via
1150 .Va net.inet.ip.maxfragpackets
1152 .Va net.inet.ip.maxfragsperpacket
1153 which limit, respectively, the maximum number of processable
1154 fragments (default: 800) and
1155 the maximum number of fragments per packet (default: 16).
1157 NOTA BENE: since fragments do not contain port numbers,
1158 they should be avoided with the
1161 Alternatively, direction-based (like
1165 ) and source-based (like
1167 ) match patterns can be used to select fragments.
1169 Usually a simple rule like:
1170 .Bd -literal -offset indent
1171 # reassemble incoming fragments
1172 ipfw add reass all from any to any in
1175 is all you need at the beginning of your ruleset.
1177 Discard packets that match this rule, and if the packet is an SCTP packet,
1178 try to send an SCTP packet containing an ABORT chunk.
1179 The search terminates.
1181 Discard packets that match this rule, and if the packet is an SCTP packet,
1182 try to send an SCTP packet containing an ABORT chunk.
1183 The search terminates.
1186 The body of a rule contains zero or more patterns (such as
1187 specific source and destination addresses or ports,
1188 protocol options, incoming or outgoing interfaces, etc.)
1189 that the packet must match in order to be recognised.
1190 In general, the patterns are connected by (implicit)
1192 operators -- i.e., all must match in order for the
1194 Individual patterns can be prefixed by the
1196 operator to reverse the result of the match, as in
1198 .Dl "ipfw add 100 allow ip from not 1.2.3.4 to any"
1200 Additionally, sets of alternative match patterns
1202 can be constructed by putting the patterns in
1203 lists enclosed between parentheses ( ) or braces { }, and
1206 operator as follows:
1208 .Dl "ipfw add 100 allow ip from { x or not y or z } to any"
1210 Only one level of parentheses is allowed.
1211 Beware that most shells have special meanings for parentheses
1212 or braces, so it is advisable to put a backslash \\ in front of them
1213 to prevent such interpretations.
1215 The body of a rule must in general include a source and destination
1219 can be used in various places to specify that the content of
1220 a required field is irrelevant.
1222 The rule body has the following format:
1223 .Bd -ragged -offset indent
1224 .Op Ar proto Cm from Ar src Cm to Ar dst
1228 The first part (proto from src to dst) is for backward
1229 compatibility with earlier versions of
1233 any match pattern (including MAC headers, IP protocols,
1234 addresses and ports) can be specified in the
1238 Rule fields have the following meaning:
1239 .Bl -tag -width indent
1240 .It Ar proto : protocol | Cm { Ar protocol Cm or ... }
1241 .It Ar protocol : Oo Cm not Oc Ar protocol-name | protocol-number
1242 An IP protocol specified by number or name
1243 (for a complete list see
1244 .Pa /etc/protocols ) ,
1245 or one of the following keywords:
1246 .Bl -tag -width indent
1248 Matches IPv4 packets.
1250 Matches IPv6 packets.
1259 option will be treated as inner protocol.
1267 .Cm { Ar protocol Cm or ... }
1270 is provided for convenience only but its use is deprecated.
1271 .It Ar src No and Ar dst : Bro Cm addr | Cm { Ar addr Cm or ... } Brc Op Oo Cm not Oc Ar ports
1272 An address (or a list, see below)
1273 optionally followed by
1279 with multiple addresses) is provided for convenience only and
1280 its use is discouraged.
1281 .It Ar addr : Oo Cm not Oc Bro
1282 .Cm any | me | me6 |
1283 .Cm table Ns Pq Ar name Ns Op , Ns Ar value
1284 .Ar | addr-list | addr-set
1286 .Bl -tag -width indent
1288 matches any IP address.
1290 matches any IP address configured on an interface in the system.
1292 matches any IPv6 address configured on an interface in the system.
1293 The address list is evaluated at the time the packet is
1295 .It Cm table Ns Pq Ar name Ns Op , Ns Ar value
1296 Matches any IPv4 or IPv6 address for which an entry exists in the lookup table
1298 If an optional 32-bit unsigned
1300 is also specified, an entry will match only if it has this value.
1303 section below for more information on lookup tables.
1305 .It Ar addr-list : ip-addr Ns Op Ns , Ns Ar addr-list
1307 A host or subnet address specified in one of the following ways:
1308 .Bl -tag -width indent
1309 .It Ar numeric-ip | hostname
1310 Matches a single IPv4 address, specified as dotted-quad or a hostname.
1311 Hostnames are resolved at the time the rule is added to the firewall list.
1312 .It Ar addr Ns / Ns Ar masklen
1313 Matches all addresses with base
1315 (specified as an IP address, a network number, or a hostname)
1319 As an example, 1.2.3.4/25 or 1.2.3.0/25 will match
1320 all IP numbers from 1.2.3.0 to 1.2.3.127 .
1321 .It Ar addr Ns : Ns Ar mask
1322 Matches all addresses with base
1324 (specified as an IP address, a network number, or a hostname)
1327 specified as a dotted quad.
1328 As an example, 1.2.3.4:255.0.255.0 or 1.0.3.0:255.0.255.0 will match
1330 This form is advised only for non-contiguous
1332 It is better to resort to the
1333 .Ar addr Ns / Ns Ar masklen
1334 format for contiguous masks, which is more compact and less
1337 .It Ar addr-set : addr Ns Oo Ns / Ns Ar masklen Oc Ns Cm { Ns Ar list Ns Cm }
1338 .It Ar list : Bro Ar num | num-num Brc Ns Op Ns , Ns Ar list
1339 Matches all addresses with base address
1341 (specified as an IP address, a network number, or a hostname)
1342 and whose last byte is in the list between braces { } .
1343 Note that there must be no spaces between braces and
1344 numbers (spaces after commas are allowed).
1345 Elements of the list can be specified as single entries
1349 field is used to limit the size of the set of addresses,
1350 and can have any value between 24 and 32.
1352 it will be assumed as 24.
1354 This format is particularly useful to handle sparse address sets
1355 within a single rule.
1356 Because the matching occurs using a
1357 bitmask, it takes constant time and dramatically reduces
1358 the complexity of rulesets.
1360 As an example, an address specified as 1.2.3.4/24{128,35-55,89}
1361 or 1.2.3.0/24{128,35-55,89}
1362 will match the following IP addresses:
1364 1.2.3.128, 1.2.3.35 to 1.2.3.55, 1.2.3.89 .
1365 .It Ar addr6-list : ip6-addr Ns Op Ns , Ns Ar addr6-list
1367 A host or subnet specified one of the following ways:
1368 .Bl -tag -width indent
1369 .It Ar numeric-ip | hostname
1370 Matches a single IPv6 address as allowed by
1373 Hostnames are resolved at the time the rule is added to the firewall
1375 .It Ar addr Ns / Ns Ar masklen
1376 Matches all IPv6 addresses with base
1378 (specified as allowed by
1384 .It Ar addr Ns / Ns Ar mask
1385 Matches all IPv6 addresses with base
1387 (specified as allowed by
1392 specified as allowed by
1394 As an example, fe::640:0:0/ffff::ffff:ffff:0:0 will match
1396 This form is advised only for non-contiguous
1398 It is better to resort to the
1399 .Ar addr Ns / Ns Ar masklen
1400 format for contiguous masks, which is more compact and less
1404 No support for sets of IPv6 addresses is provided because IPv6 addresses
1405 are typically random past the initial prefix.
1406 .It Ar ports : Bro Ar port | port Ns \&- Ns Ar port Ns Brc Ns Op , Ns Ar ports
1407 For protocols which support port numbers (such as SCTP, TCP and UDP), optional
1409 may be specified as one or more ports or port ranges, separated
1410 by commas but no spaces, and an optional
1415 notation specifies a range of ports (including boundaries).
1419 may be used instead of numeric port values.
1420 The length of the port list is limited to 30 ports or ranges,
1421 though one can specify larger ranges by using an
1425 section of the rule.
1429 can be used to escape the dash
1431 character in a service name (from a shell, the backslash must be
1432 typed twice to avoid the shell itself interpreting it as an escape
1435 .Dl "ipfw add count tcp from any ftp\e\e-data-ftp to any"
1437 Fragmented packets which have a non-zero offset (i.e., not the first
1438 fragment) will never match a rule which has one or more port
1442 option for details on matching fragmented packets.
1444 .Ss RULE OPTIONS (MATCH PATTERNS)
1445 Additional match patterns can be used within
1447 Zero or more of these so-called
1449 can be present in a rule, optionally prefixed by the
1451 operand, and possibly grouped into
1454 The following match patterns can be used (listed in alphabetical order):
1455 .Bl -tag -width indent
1456 .It Cm // this is a comment.
1457 Inserts the specified text as a comment in the rule.
1458 Everything following // is considered as a comment and stored in the rule.
1459 You can have comment-only rules, which are listed as having a
1461 action followed by the comment.
1466 Matches only packets generated by a divert socket.
1467 .It Cm diverted-loopback
1468 Matches only packets coming from a divert socket back into the IP stack
1470 .It Cm diverted-output
1471 Matches only packets going from a divert socket back outward to the IP
1472 stack output for delivery.
1473 .It Cm dst-ip Ar ip-address
1474 Matches IPv4 packets whose destination IP is one of the address(es)
1475 specified as argument.
1476 .It Bro Cm dst-ip6 | dst-ipv6 Brc Ar ip6-address
1477 Matches IPv6 packets whose destination IP is one of the address(es)
1478 specified as argument.
1479 .It Cm dst-port Ar ports
1480 Matches IP packets whose destination port is one of the port(s)
1481 specified as argument.
1483 Matches TCP packets that have the RST or ACK bits set.
1484 .It Cm ext6hdr Ar header
1485 Matches IPv6 packets containing the extended header given by
1487 Supported headers are:
1493 any type of Routing Header
1495 Source routing Routing Header Type 0
1497 Mobile IPv6 Routing Header Type 2
1501 IPSec authentication headers
1503 and IPsec encapsulated security payload headers
1505 .It Cm fib Ar fibnum
1506 Matches a packet that has been tagged to use
1507 the given FIB (routing table) number.
1508 .It Cm flow Ar table Ns Pq Ar name Ns Op , Ns Ar value
1509 Search for the flow entry in lookup table
1511 If not found, the match fails.
1512 Otherwise, the match succeeds and
1514 is set to the value extracted from the table.
1516 This option can be useful to quickly dispatch traffic based on
1517 certain packet fields.
1520 section below for more information on lookup tables.
1521 .It Cm flow-id Ar labels
1522 Matches IPv6 packets containing any of the flow labels given in
1525 is a comma separated list of numeric flow labels.
1527 Matches packets that are fragments and not the first
1528 fragment of an IP datagram.
1529 Note that these packets will not have
1530 the next protocol header (e.g.\& TCP, UDP) so options that look into
1531 these headers cannot match.
1533 Matches all TCP or UDP packets sent by or received for a
1537 may be specified by name or number.
1538 .It Cm jail Ar prisonID
1539 Matches all TCP or UDP packets sent by or received for the
1540 jail whos prison ID is
1542 .It Cm icmptypes Ar types
1543 Matches ICMP packets whose ICMP type is in the list
1545 The list may be specified as any combination of
1546 individual types (numeric) separated by commas.
1547 .Em Ranges are not allowed .
1548 The supported ICMP types are:
1552 destination unreachable
1560 router advertisement
1564 time-to-live exceeded
1576 address mask request
1578 and address mask reply
1580 .It Cm icmp6types Ar types
1581 Matches ICMP6 packets whose ICMP6 type is in the list of
1583 The list may be specified as any combination of
1584 individual types (numeric) separated by commas.
1585 .Em Ranges are not allowed .
1587 Matches incoming or outgoing packets, respectively.
1591 are mutually exclusive (in fact,
1595 .It Cm ipid Ar id-list
1596 Matches IPv4 packets whose
1598 field has value included in
1600 which is either a single value or a list of values or ranges
1601 specified in the same way as
1603 .It Cm iplen Ar len-list
1604 Matches IP packets whose total length, including header and data, is
1607 which is either a single value or a list of values or ranges
1608 specified in the same way as
1610 .It Cm ipoptions Ar spec
1611 Matches packets whose IPv4 header contains the comma separated list of
1612 options specified in
1614 The supported IP options are:
1617 (strict source route),
1619 (loose source route),
1621 (record packet route) and
1624 The absence of a particular option may be denoted
1627 .It Cm ipprecedence Ar precedence
1628 Matches IPv4 packets whose precedence field is equal to
1631 Matches packets that have IPSEC history associated with them
1632 (i.e., the packet comes encapsulated in IPSEC, the kernel
1633 has IPSEC support, and can correctly decapsulate it).
1635 Note that specifying
1637 is different from specifying
1639 as the latter will only look at the specific IP protocol field,
1640 irrespective of IPSEC kernel support and the validity of the IPSEC data.
1642 Further note that this flag is silently ignored in kernels without
1644 It does not affect rule processing when given and the
1645 rules are handled as if with no
1648 .It Cm iptos Ar spec
1649 Matches IPv4 packets whose
1651 field contains the comma separated list of
1652 service types specified in
1654 The supported IP types of service are:
1657 .Pq Dv IPTOS_LOWDELAY ,
1659 .Pq Dv IPTOS_THROUGHPUT ,
1661 .Pq Dv IPTOS_RELIABILITY ,
1663 .Pq Dv IPTOS_MINCOST ,
1665 .Pq Dv IPTOS_ECN_CE .
1666 The absence of a particular type may be denoted
1669 .It Cm dscp spec Ns Op , Ns Ar spec
1670 Matches IPv4/IPv6 packets whose
1672 field value is contained in
1675 Multiple values can be specified via
1676 the comma separated list.
1677 Value can be one of keywords used in
1679 action or exact number.
1680 .It Cm ipttl Ar ttl-list
1681 Matches IPv4 packets whose time to live is included in
1683 which is either a single value or a list of values or ranges
1684 specified in the same way as
1686 .It Cm ipversion Ar ver
1687 Matches IP packets whose IP version field is
1689 .It Cm keep-state Op Ar :flowname
1690 Upon a match, the firewall will create a dynamic rule, whose
1691 default behaviour is to match bidirectional traffic between
1692 source and destination IP/port using the same protocol.
1693 The rule has a limited lifetime (controlled by a set of
1695 variables), and the lifetime is refreshed every time a matching
1699 is used to assign additional to addresses, ports and protocol parameter
1700 to dynamic rule. It can be used for more accurate matching by
1705 keyword is special name used for compatibility with old rulesets.
1707 Matches only layer2 packets, i.e., those passed to
1709 from ether_demux() and ether_output_frame().
1710 .It Cm limit Bro Cm src-addr | src-port | dst-addr | dst-port Brc Ar N Op Ar :flowname
1711 The firewall will only allow
1713 connections with the same
1714 set of parameters as specified in the rule.
1716 of source and destination addresses and ports can be
1718 .It Cm lookup Bro Cm dst-ip | dst-port | src-ip | src-port | uid | jail Brc Ar name
1719 Search an entry in lookup table
1721 that matches the field specified as argument.
1722 If not found, the match fails.
1723 Otherwise, the match succeeds and
1725 is set to the value extracted from the table.
1727 This option can be useful to quickly dispatch traffic based on
1728 certain packet fields.
1731 section below for more information on lookup tables.
1732 .It Cm { MAC | mac } Ar dst-mac src-mac
1733 Match packets with a given
1737 addresses, specified as the
1739 keyword (matching any MAC address), or six groups of hex digits
1740 separated by colons,
1741 and optionally followed by a mask indicating the significant bits.
1742 The mask may be specified using either of the following methods:
1743 .Bl -enum -width indent
1747 followed by the number of significant bits.
1748 For example, an address with 33 significant bits could be specified as:
1750 .Dl "MAC 10:20:30:40:50:60/33 any"
1754 followed by a bitmask specified as six groups of hex digits separated
1756 For example, an address in which the last 16 bits are significant could
1759 .Dl "MAC 10:20:30:40:50:60&00:00:00:00:ff:ff any"
1761 Note that the ampersand character has a special meaning in many shells
1762 and should generally be escaped.
1764 Note that the order of MAC addresses (destination first,
1766 the same as on the wire, but the opposite of the one used for
1768 .It Cm mac-type Ar mac-type
1769 Matches packets whose Ethernet Type field
1770 corresponds to one of those specified as argument.
1772 is specified in the same way as
1774 (i.e., one or more comma-separated single values or ranges).
1775 You can use symbolic names for known values such as
1776 .Em vlan , ipv4, ipv6 .
1777 Values can be entered as decimal or hexadecimal (if prefixed by 0x),
1778 and they are always printed as hexadecimal (unless the
1780 option is used, in which case symbolic resolution will be attempted).
1781 .It Cm proto Ar protocol
1782 Matches packets with the corresponding IP protocol.
1783 .It Cm recv | xmit | via Brq Ar ifX | Ar if Ns Cm * | Ar table Ns Po Ar name Ns Oo , Ns Ar value Oc Pc | Ar ipno | Ar any
1784 Matches packets received, transmitted or going through,
1785 respectively, the interface specified by exact name
1789 by IP address, or through some interface.
1792 may be used to match interface by its kernel ifindex.
1795 section below for more information on lookup tables.
1799 keyword causes the interface to always be checked.
1806 then only the receive or transmit interface (respectively)
1808 By specifying both, it is possible to match packets based on
1809 both receive and transmit interface, e.g.:
1811 .Dl "ipfw add deny ip from any to any out recv ed0 xmit ed1"
1815 interface can be tested on either incoming or outgoing packets,
1818 interface can only be tested on outgoing packets.
1823 is invalid) whenever
1827 A packet might not have a receive or transmit interface: packets
1828 originating from the local host have no receive interface,
1829 while packets destined for the local host have no transmit
1832 Matches TCP packets that have the SYN bit set but no ACK bit.
1833 This is the short form of
1834 .Dq Li tcpflags\ syn,!ack .
1836 Matches packets that are associated to a local socket and
1837 for which the SO_USER_COOKIE socket option has been set
1838 to a non-zero value.
1839 As a side effect, the value of the
1840 option is made available as
1842 value, which in turn can be used as
1847 .It Cm src-ip Ar ip-address
1848 Matches IPv4 packets whose source IP is one of the address(es)
1849 specified as an argument.
1850 .It Cm src-ip6 Ar ip6-address
1851 Matches IPv6 packets whose source IP is one of the address(es)
1852 specified as an argument.
1853 .It Cm src-port Ar ports
1854 Matches IP packets whose source port is one of the port(s)
1855 specified as argument.
1856 .It Cm tagged Ar tag-list
1857 Matches packets whose tags are included in
1859 which is either a single value or a list of values or ranges
1860 specified in the same way as
1862 Tags can be applied to the packet using
1864 rule action parameter (see it's description for details on tags).
1865 .It Cm tcpack Ar ack
1867 Match if the TCP header acknowledgment number field is set to
1869 .It Cm tcpdatalen Ar tcpdatalen-list
1870 Matches TCP packets whose length of TCP data is
1871 .Ar tcpdatalen-list ,
1872 which is either a single value or a list of values or ranges
1873 specified in the same way as
1875 .It Cm tcpflags Ar spec
1877 Match if the TCP header contains the comma separated list of
1880 The supported TCP flags are:
1889 The absence of a particular flag may be denoted
1892 A rule which contains a
1894 specification can never match a fragmented packet which has
1898 option for details on matching fragmented packets.
1899 .It Cm tcpseq Ar seq
1901 Match if the TCP header sequence number field is set to
1903 .It Cm tcpwin Ar tcpwin-list
1904 Matches TCP packets whose header window field is set to
1906 which is either a single value or a list of values or ranges
1907 specified in the same way as
1909 .It Cm tcpoptions Ar spec
1911 Match if the TCP header contains the comma separated list of
1912 options specified in
1914 The supported TCP options are:
1917 (maximum segment size),
1919 (tcp window advertisement),
1923 (rfc1323 timestamp) and
1925 (rfc1644 t/tcp connection count).
1926 The absence of a particular option may be denoted
1930 Match all TCP or UDP packets sent by or received for a
1934 may be matched by name or identification number.
1936 For incoming packets,
1937 a routing table lookup is done on the packet's source address.
1938 If the interface on which the packet entered the system matches the
1939 outgoing interface for the route,
1941 If the interfaces do not match up,
1942 the packet does not match.
1943 All outgoing packets or packets with no incoming interface match.
1945 The name and functionality of the option is intentionally similar to
1946 the Cisco IOS command:
1948 .Dl ip verify unicast reverse-path
1950 This option can be used to make anti-spoofing rules to reject all
1951 packets with source addresses not from this interface.
1955 For incoming packets,
1956 a routing table lookup is done on the packet's source address.
1957 If a route to the source address exists, but not the default route
1958 or a blackhole/reject route, the packet matches.
1959 Otherwise, the packet does not match.
1960 All outgoing packets match.
1962 The name and functionality of the option is intentionally similar to
1963 the Cisco IOS command:
1965 .Dl ip verify unicast source reachable-via any
1967 This option can be used to make anti-spoofing rules to reject all
1968 packets whose source address is unreachable.
1970 For incoming packets, the packet's source address is checked if it
1971 belongs to a directly connected network.
1972 If the network is directly connected, then the interface the packet
1973 came on in is compared to the interface the network is connected to.
1974 When incoming interface and directly connected interface are not the
1975 same, the packet does not match.
1976 Otherwise, the packet does match.
1977 All outgoing packets match.
1979 This option can be used to make anti-spoofing rules to reject all
1980 packets that pretend to be from a directly connected network but do
1981 not come in through that interface.
1982 This option is similar to but more restricted than
1984 because it engages only on packets with source addresses of directly
1985 connected networks instead of all source addresses.
1988 Lookup tables are useful to handle large sparse sets of
1989 addresses or other search keys (e.g., ports, jail IDs, interface names).
1990 In the rest of this section we will use the term ``key''.
1991 Table name needs to match the following spec:
1993 Tables with the same name can be created in different
1995 However, rule links to the tables in
1998 This behavior can be controlled by
1999 .Va net.inet.ip.fw.tables_sets
2003 section for more information.
2004 There may be up to 65535 different lookup tables.
2006 The following table types are supported:
2007 .Bl -tag -width indent
2008 .It Ar table-type : Ar addr | iface | number | flow
2009 .It Ar table-key : Ar addr Ns Oo / Ns Ar masklen Oc | iface-name | number | flow-spec
2010 .It Ar flow-spec : Ar flow-field Ns Op , Ns Ar flow-spec
2011 .It Ar flow-field : src-ip | proto | src-port | dst-ip | dst-port
2013 matches IPv4 or IPv6 address.
2014 Each entry is represented by an
2015 .Ar addr Ns Op / Ns Ar masklen
2016 and will match all addresses with base
2018 (specified as an IPv4/IPv6 address, or a hostname) and mask width of
2023 is not specified, it defaults to 32 for IPv4 and 128 for IPv6.
2024 When looking up an IP address in a table, the most specific
2027 matches interface names.
2028 Each entry is represented by string treated as interface name.
2029 Wildcards are not supported.
2031 maches protocol ports, uids/gids or jail IDs.
2032 Each entry is represented by 32-bit unsigned integer.
2033 Ranges are not supported.
2035 Matches packet fields specified by
2037 type suboptions with table entries.
2040 Tables require explicit creation via
2044 The following creation options are supported:
2045 .Bl -tag -width indent
2046 .It Ar create-options : Ar create-option | create-options
2047 .It Ar create-option : Cm type Ar table-type | Cm valtype Ar value-mask | Cm algo Ar algo-desc |
2048 .Cm limit Ar number | Cm locked
2054 Table algorithm to use (see below).
2056 Maximum number of items that may be inserted into table.
2058 Restrict any table modifications.
2061 Some of these options may be modified later via
2064 The following options can be changed:
2065 .Bl -tag -width indent
2066 .It Ar modify-options : Ar modify-option | modify-options
2067 .It Ar modify-option : Cm limit Ar number
2069 Alter maximum number of items that may be inserted into table.
2072 Additionally, table can be locked or unlocked using
2080 can be swapped with each other using
2083 Swap may fail if tables limits are set and data exchange
2084 would result in limits hit.
2085 Operation is performed atomically.
2087 One or more entries can be added to a table at once using
2090 Addition of all items are performed atomically.
2091 By default, error in addition of one entry does not influence
2092 addition of other entries. However, non-zero error code is returned
2096 keyword may be specified before
2098 to indicate all-or-none add request.
2100 One or more entries can be removed from a table at once using
2103 By default, error in removal of one entry does not influence
2104 removing of other entries. However, non-zero error code is returned
2107 It may be possible to check what entry will be found on particular
2113 This functionality is optional and may be unsupported in some algorithms.
2115 The following operations can be performed on
2120 .Bl -tag -width indent
2124 Removes all entries.
2126 Shows generic table information.
2128 Shows generic table information and algo-specific data.
2131 The following lookup algorithms are supported:
2132 .Bl -tag -width indent
2133 .It Ar algo-desc : algo-name | "algo-name algo-data"
2134 .It Ar algo-name: Ar addr:radix | addr:hash | iface:array | number:array | flow:hash
2136 Separate Radix trees for IPv4 and IPv6, the same way as the routing table (see
2142 Separate auto-growing hashes for IPv4 and IPv6.
2143 Accepts entries with the same mask length specified initially via
2144 .Cm "addr:hash masks=/v4,/v6"
2145 algorithm creation options.
2146 Assume /32 and /128 masks by default.
2147 Search removes host bits (according to mask) from supplied address and checks
2148 resulting key in appropriate hash.
2149 Mostly optimized for /64 and byte-ranged IPv6 masks.
2151 Array storing sorted indexes for entries which are presented in the system.
2152 Optimized for very fast lookup.
2154 Array storing sorted u32 numbers.
2156 Auto-growing hash storing flow entries.
2157 Search calculates hash on required packet fields and searches for matching
2158 entries in selected bucket.
2163 feature provides the ability to use a value, looked up in the table, as
2164 the argument for a rule action, action parameter or rule option.
2165 This can significantly reduce number of rules in some configurations.
2166 If two tables are used in a rule, the result of the second (destination)
2169 Each record may hold one or more values according to
2171 This mask is set on table creation via
2174 The following value types are supported:
2175 .Bl -tag -width indent
2176 .It Ar value-mask : Ar value-type Ns Op , Ns Ar value-mask
2177 .It Ar value-type : Ar skipto | pipe | fib | nat | dscp | tag | divert |
2178 .Ar netgraph | limit | ipv4
2180 rule number to jump to.
2184 fib number to match/set.
2186 nat number to jump to.
2188 dscp value to match/set.
2190 tag number to match/set.
2192 port number to divert traffic to.
2194 hook number to move packet to.
2196 maximum number of connections.
2198 IPv4 nexthop to fwd packets to.
2200 IPv6 nexthop to fwd packets to.
2205 argument can be used with the following actions:
2206 .Cm nat, pipe , queue, divert, tee, netgraph, ngtee, fwd, skipto, setfib,
2214 action, the user should be aware that the code will walk the ruleset
2215 up to a rule equal to, or past, the given number.
2219 Section for example usage of tables and the tablearg keyword.
2221 Each rule or table belongs to one of 32 different
2224 Set 31 is reserved for the default rule.
2226 By default, rules or tables are put in set 0, unless you use the
2228 attribute when adding a new rule or table.
2229 Sets can be individually and atomically enabled or disabled,
2230 so this mechanism permits an easy way to store multiple configurations
2231 of the firewall and quickly (and atomically) switch between them.
2233 By default, tables from set 0 are referenced when adding rule with
2234 table opcodes regardless of rule set.
2235 This behavior can be changed by setting
2236 .Va net.inet.ip.fw.tables_sets
2238 Rule's set will then be used for table references.
2240 The command to enable/disable sets is
2241 .Bd -ragged -offset indent
2243 .Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
2250 sections can be specified.
2251 Command execution is atomic on all the sets specified in the command.
2252 By default, all sets are enabled.
2254 When you disable a set, its rules behave as if they do not exist
2255 in the firewall configuration, with only one exception:
2256 .Bd -ragged -offset indent
2257 dynamic rules created from a rule before it had been disabled
2258 will still be active until they expire.
2260 dynamic rules you have to explicitly delete the parent rule
2261 which generated them.
2264 The set number of rules can be changed with the command
2265 .Bd -ragged -offset indent
2268 .Brq Cm rule Ar rule-number | old-set
2272 Also, you can atomically swap two rulesets with the command
2273 .Bd -ragged -offset indent
2275 .Cm set swap Ar first-set second-set
2280 Section on some possible uses of sets of rules.
2281 .Sh STATEFUL FIREWALL
2282 Stateful operation is a way for the firewall to dynamically
2283 create rules for specific flows when packets that
2284 match a given pattern are detected.
2285 Support for stateful
2286 operation comes through the
2287 .Cm check-state , keep-state
2293 Dynamic rules are created when a packet matches a
2297 rule, causing the creation of a
2299 rule which will match all and only packets with
2303 .Em src-ip/src-port dst-ip/dst-port
2308 are used here only to denote the initial match addresses, but they
2309 are completely equivalent afterwards).
2315 This name is used in matching together with addresses, ports and protocol.
2316 Dynamic rules will be checked at the first
2317 .Cm check-state, keep-state
2320 occurrence, and the action performed upon a match will be the same
2321 as in the parent rule.
2323 Note that no additional attributes other than protocol and IP addresses
2324 and ports and :flowname are checked on dynamic rules.
2326 The typical use of dynamic rules is to keep a closed firewall configuration,
2327 but let the first TCP SYN packet from the inside network install a
2328 dynamic rule for the flow so that packets belonging to that session
2329 will be allowed through the firewall:
2331 .Dl "ipfw add check-state :OUTBOUND"
2332 .Dl "ipfw add allow tcp from my-subnet to any setup keep-state :OUTBOUND"
2333 .Dl "ipfw add deny tcp from any to any"
2335 A similar approach can be used for UDP, where an UDP packet coming
2336 from the inside will install a dynamic rule to let the response through
2339 .Dl "ipfw add check-state :OUTBOUND"
2340 .Dl "ipfw add allow udp from my-subnet to any keep-state :OUTBOUND"
2341 .Dl "ipfw add deny udp from any to any"
2343 Dynamic rules expire after some time, which depends on the status
2344 of the flow and the setting of some
2348 .Sx SYSCTL VARIABLES
2350 For TCP sessions, dynamic rules can be instructed to periodically
2351 send keepalive packets to refresh the state of the rule when it is
2356 for more examples on how to use dynamic rules.
2357 .Sh TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
2359 is also the user interface for the
2361 traffic shaper, packet scheduler and network emulator, a subsystem that
2362 can artificially queue, delay or drop packets
2363 emulating the behaviour of certain network links
2364 or queueing systems.
2367 operates by first using the firewall to select packets
2368 using any match pattern that can be used in
2371 Matching packets are then passed to either of two
2372 different objects, which implement the traffic regulation:
2373 .Bl -hang -offset XXXX
2379 with given bandwidth and propagation delay,
2380 driven by a FIFO scheduler and a single queue with programmable
2381 queue size and packet loss rate.
2382 Packets are appended to the queue as they come out from
2384 and then transferred in FIFO order to the link at the desired rate.
2388 is an abstraction used to implement packet scheduling
2389 using one of several packet scheduling algorithms.
2392 are first grouped into flows according to a mask on the 5-tuple.
2393 Flows are then passed to the scheduler associated to the
2395 and each flow uses scheduling parameters (weight and others)
2396 as configured in the
2399 A scheduler in turn is connected to an emulated link,
2400 and arbitrates the link's bandwidth among backlogged flows according to
2401 weights and to the features of the scheduling algorithm in use.
2406 can be used to set hard limits to the bandwidth that a flow can use, whereas
2408 can be used to determine how different flows share the available bandwidth.
2410 A graphical representation of the binding of queues,
2411 flows, schedulers and links is below.
2412 .Bd -literal -offset indent
2413 (flow_mask|sched_mask) sched_mask
2414 +---------+ weight Wx +-------------+
2415 | |->-[flow]-->--| |-+
2416 -->--| QUEUE x | ... | | |
2417 | |->-[flow]-->--| SCHEDuler N | |
2419 ... | +--[LINK N]-->--
2420 +---------+ weight Wy | | +--[LINK N]-->--
2421 | |->-[flow]-->--| | |
2422 -->--| QUEUE y | ... | | |
2423 | |->-[flow]-->--| | |
2424 +---------+ +-------------+ |
2427 It is important to understand the role of the SCHED_MASK
2428 and FLOW_MASK, which are configured through the commands
2429 .Dl "ipfw sched N config mask SCHED_MASK ..."
2431 .Dl "ipfw queue X config mask FLOW_MASK ..." .
2433 The SCHED_MASK is used to assign flows to one or more
2434 scheduler instances, one for each
2435 value of the packet's 5-tuple after applying SCHED_MASK.
2436 As an example, using ``src-ip 0xffffff00'' creates one instance
2437 for each /24 destination subnet.
2439 The FLOW_MASK, together with the SCHED_MASK, is used to split
2441 As an example, using
2442 ``src-ip 0x000000ff''
2443 together with the previous SCHED_MASK makes a flow for
2444 each individual source address.
2445 In turn, flows for each /24
2446 subnet will be sent to the same scheduler instance.
2448 The above diagram holds even for the
2450 case, with the only restriction that a
2452 only supports a SCHED_MASK, and forces the use of a FIFO
2453 scheduler (these are for backward compatibility reasons;
2454 in fact, internally, a
2456 pipe is implemented exactly as above).
2458 There are two modes of
2466 mode tries to emulate a real link: the
2468 scheduler ensures that the packet will not leave the pipe faster than it
2469 would on the real link with a given bandwidth.
2472 mode allows certain packets to bypass the
2474 scheduler (if packet flow does not exceed pipe's bandwidth).
2475 This is the reason why the
2477 mode requires less CPU cycles per packet (on average) and packet latency
2478 can be significantly lower in comparison to a real link with the same
2484 mode can be enabled by setting the
2485 .Va net.inet.ip.dummynet.io_fast
2487 variable to a non-zero value.
2489 .Ss PIPE, QUEUE AND SCHEDULER CONFIGURATION
2495 configuration commands are the following:
2496 .Bd -ragged -offset indent
2497 .Cm pipe Ar number Cm config Ar pipe-configuration
2499 .Cm queue Ar number Cm config Ar queue-configuration
2501 .Cm sched Ar number Cm config Ar sched-configuration
2504 The following parameters can be configured for a pipe:
2506 .Bl -tag -width indent -compact
2507 .It Cm bw Ar bandwidth | device
2508 Bandwidth, measured in
2511 .Brq Cm bit/s | Byte/s .
2514 A value of 0 (default) means unlimited bandwidth.
2515 The unit must immediately follow the number, as in
2517 .Dl "ipfw pipe 1 config bw 300Kbit/s"
2519 If a device name is specified instead of a numeric value, as in
2521 .Dl "ipfw pipe 1 config bw tun0"
2523 then the transmit clock is supplied by the specified device.
2524 At the moment only the
2526 device supports this
2527 functionality, for use in conjunction with
2530 .It Cm delay Ar ms-delay
2531 Propagation delay, measured in milliseconds.
2532 The value is rounded to the next multiple of the clock tick
2533 (typically 10ms, but it is a good practice to run kernels
2535 .Dq "options HZ=1000"
2537 the granularity to 1ms or less).
2538 The default value is 0, meaning no delay.
2540 .It Cm burst Ar size
2541 If the data to be sent exceeds the pipe's bandwidth limit
2542 (and the pipe was previously idle), up to
2544 bytes of data are allowed to bypass the
2546 scheduler, and will be sent as fast as the physical link allows.
2547 Any additional data will be transmitted at the rate specified
2551 The burst size depends on how long the pipe has been idle;
2552 the effective burst size is calculated as follows:
2559 .It Cm profile Ar filename
2560 A file specifying the additional overhead incurred in the transmission
2561 of a packet on the link.
2563 Some link types introduce extra delays in the transmission
2564 of a packet, e.g., because of MAC level framing, contention on
2565 the use of the channel, MAC level retransmissions and so on.
2566 From our point of view, the channel is effectively unavailable
2567 for this extra time, which is constant or variable depending
2569 Additionally, packets may be dropped after this
2570 time (e.g., on a wireless link after too many retransmissions).
2571 We can model the additional delay with an empirical curve
2572 that represents its distribution.
2573 .Bd -literal -offset indent
2574 cumulative probability
2584 +-------*------------------->
2587 The empirical curve may have both vertical and horizontal lines.
2588 Vertical lines represent constant delay for a range of
2590 Horizontal lines correspond to a discontinuity in the delay
2591 distribution: the pipe will use the largest delay for a
2594 The file format is the following, with whitespace acting as
2595 a separator and '#' indicating the beginning a comment:
2596 .Bl -tag -width indent
2597 .It Cm name Ar identifier
2598 optional name (listed by "ipfw pipe show")
2599 to identify the delay distribution;
2601 the bandwidth used for the pipe.
2602 If not specified here, it must be present
2603 explicitly as a configuration parameter for the pipe;
2604 .It Cm loss-level Ar L
2605 the probability above which packets are lost.
2606 (0.0 <= L <= 1.0, default 1.0 i.e., no loss);
2608 the number of samples used in the internal
2609 representation of the curve (2..1024; default 100);
2610 .It Cm "delay prob" | "prob delay"
2611 One of these two lines is mandatory and defines
2612 the format of the following lines with data points.
2614 2 or more lines representing points in the curve,
2615 with either delay or probability first, according
2616 to the chosen format.
2617 The unit for delay is milliseconds.
2618 Data points do not need to be sorted.
2619 Also, the number of actual lines can be different
2620 from the value of the "samples" parameter:
2622 utility will sort and interpolate
2623 the curve as needed.
2626 Example of a profile file:
2627 .Bd -literal -offset indent
2632 0 200 # minimum overhead is 200ms
2638 #configuration file end
2642 The following parameters can be configured for a queue:
2644 .Bl -tag -width indent -compact
2645 .It Cm pipe Ar pipe_nr
2646 Connects a queue to the specified pipe.
2647 Multiple queues (with the same or different weights) can be connected to
2648 the same pipe, which specifies the aggregate rate for the set of queues.
2650 .It Cm weight Ar weight
2651 Specifies the weight to be used for flows matching this queue.
2652 The weight must be in the range 1..100, and defaults to 1.
2655 The following case-insensitive parameters can be configured for a
2658 .Bl -tag -width indent -compact
2659 .It Cm type Ar {fifo | wf2q+ | rr | qfq}
2660 specifies the scheduling algorithm to use.
2661 .Bl -tag -width indent -compact
2663 is just a FIFO scheduler (which means that all packets
2664 are stored in the same queue as they arrive to the scheduler).
2665 FIFO has O(1) per-packet time complexity, with very low
2666 constants (estimate 60-80ns on a 2GHz desktop machine)
2667 but gives no service guarantees.
2669 implements the WF2Q+ algorithm, which is a Weighted Fair Queueing
2670 algorithm which permits flows to share bandwidth according to
2672 Note that weights are not priorities; even a flow
2673 with a minuscule weight will never starve.
2674 WF2Q+ has O(log N) per-packet processing cost, where N is the number
2675 of flows, and is the default algorithm used by previous versions
2678 implements the Deficit Round Robin algorithm, which has O(1) processing
2679 costs (roughly, 100-150ns per packet)
2680 and permits bandwidth allocation according to weights, but
2681 with poor service guarantees.
2683 implements the QFQ algorithm, which is a very fast variant of
2684 WF2Q+, with similar service guarantees and O(1) processing
2685 costs (roughly, 200-250ns per packet).
2689 In addition to the type, all parameters allowed for a pipe can also
2690 be specified for a scheduler.
2692 Finally, the following parameters can be configured for both
2695 .Bl -tag -width XXXX -compact
2696 .It Cm buckets Ar hash-table-size
2697 Specifies the size of the hash table used for storing the
2699 Default value is 64 controlled by the
2702 .Va net.inet.ip.dummynet.hash_size ,
2703 allowed range is 16 to 65536.
2705 .It Cm mask Ar mask-specifier
2706 Packets sent to a given pipe or queue by an
2708 rule can be further classified into multiple flows, each of which is then
2712 A flow identifier is constructed by masking the IP addresses,
2713 ports and protocol types as specified with the
2715 options in the configuration of the pipe or queue.
2716 For each different flow identifier, a new pipe or queue is created
2717 with the same parameters as the original object, and matching packets
2722 are used, each flow will get the same bandwidth as defined by the pipe,
2725 are used, each flow will share the parent's pipe bandwidth evenly
2726 with other flows generated by the same queue (note that other queues
2727 with different weights might be connected to the same pipe).
2729 Available mask specifiers are a combination of one or more of the following:
2731 .Cm dst-ip Ar mask ,
2732 .Cm dst-ip6 Ar mask ,
2733 .Cm src-ip Ar mask ,
2734 .Cm src-ip6 Ar mask ,
2735 .Cm dst-port Ar mask ,
2736 .Cm src-port Ar mask ,
2737 .Cm flow-id Ar mask ,
2742 where the latter means all bits in all fields are significant.
2745 When a packet is dropped by a
2747 queue or pipe, the error
2748 is normally reported to the caller routine in the kernel, in the
2749 same way as it happens when a device queue fills up.
2751 option reports the packet as successfully delivered, which can be
2752 needed for some experimental setups where you want to simulate
2753 loss or congestion at a remote router.
2755 .It Cm plr Ar packet-loss-rate
2758 .Ar packet-loss-rate
2759 is a floating-point number between 0 and 1, with 0 meaning no
2760 loss, 1 meaning 100% loss.
2761 The loss rate is internally represented on 31 bits.
2763 .It Cm queue Brq Ar slots | size Ns Cm Kbytes
2768 Default value is 50 slots, which
2769 is the typical queue size for Ethernet devices.
2770 Note that for slow speed links you should keep the queue
2771 size short or your traffic might be affected by a significant
2773 E.g., 50 max-sized ethernet packets (1500 bytes) mean 600Kbit
2774 or 20s of queue on a 30Kbit/s pipe.
2775 Even worse effects can result if you get packets from an
2776 interface with a much larger MTU, e.g.\& the loopback interface
2777 with its 16KB packets.
2781 .Em net.inet.ip.dummynet.pipe_byte_limit
2783 .Em net.inet.ip.dummynet.pipe_slot_limit
2784 control the maximum lengths that can be specified.
2786 .It Cm red | gred Ar w_q Ns / Ns Ar min_th Ns / Ns Ar max_th Ns / Ns Ar max_p
2788 Make use of the RED (Random Early Detection) queue management algorithm.
2793 point numbers between 0 and 1 (inclusive), while
2797 are integer numbers specifying thresholds for queue management
2798 (thresholds are computed in bytes if the queue has been defined
2799 in bytes, in slots otherwise).
2800 The two parameters can also be of the same value if needed. The
2802 also supports the gentle RED variant (gred) and ECN (Explicit Congestion
2803 Notification) as optional. Three
2805 variables can be used to control the RED behaviour:
2806 .Bl -tag -width indent
2807 .It Va net.inet.ip.dummynet.red_lookup_depth
2808 specifies the accuracy in computing the average queue
2809 when the link is idle (defaults to 256, must be greater than zero)
2810 .It Va net.inet.ip.dummynet.red_avg_pkt_size
2811 specifies the expected average packet size (defaults to 512, must be
2813 .It Va net.inet.ip.dummynet.red_max_pkt_size
2814 specifies the expected maximum packet size, only used when queue
2815 thresholds are in bytes (defaults to 1500, must be greater than zero).
2819 When used with IPv6 data,
2821 currently has several limitations.
2822 Information necessary to route link-local packets to an
2823 interface is not available after processing by
2825 so those packets are dropped in the output path.
2826 Care should be taken to ensure that link-local packets are not passed to
2829 Here are some important points to consider when designing your
2833 Remember that you filter both packets going
2837 Most connections need packets going in both directions.
2839 Remember to test very carefully.
2840 It is a good idea to be near the console when doing this.
2841 If you cannot be near the console,
2842 use an auto-recovery script such as the one in
2843 .Pa /usr/share/examples/ipfw/change_rules.sh .
2845 Do not forget the loopback interface.
2850 There are circumstances where fragmented datagrams are unconditionally
2852 TCP packets are dropped if they do not contain at least 20 bytes of
2853 TCP header, UDP packets are dropped if they do not contain a full 8
2854 byte UDP header, and ICMP packets are dropped if they do not contain
2855 4 bytes of ICMP header, enough to specify the ICMP type, code, and
2857 These packets are simply logged as
2859 since there may not be enough good data in the packet to produce a
2860 meaningful log entry.
2862 Another type of packet is unconditionally dropped, a TCP packet with a
2863 fragment offset of one.
2864 This is a valid packet, but it only has one use, to try
2865 to circumvent firewalls.
2866 When logging is enabled, these packets are
2867 reported as being dropped by rule -1.
2869 If you are logged in over a network, loading the
2873 is probably not as straightforward as you would think.
2874 The following command line is recommended:
2875 .Bd -literal -offset indent
2877 ipfw add 32000 allow ip from any to any
2880 Along the same lines, doing an
2881 .Bd -literal -offset indent
2885 in similar surroundings is also a bad idea.
2889 filter list may not be modified if the system security level
2890 is set to 3 or higher
2893 for information on system security levels).
2895 .Sh PACKET DIVERSION
2898 socket bound to the specified port will receive all packets
2899 diverted to that port.
2900 If no socket is bound to the destination port, or if the divert module is
2901 not loaded, or if the kernel was not compiled with divert socket support,
2902 the packets are dropped.
2903 .Sh NETWORK ADDRESS TRANSLATION (NAT)
2905 support in-kernel NAT using the kernel version of
2909 should be loaded or kernel should have
2910 .Cm options IPFIREWALL_NAT
2913 The nat configuration command is the following:
2914 .Bd -ragged -offset indent
2919 .Ar nat-configuration
2923 The following parameters can be configured:
2924 .Bl -tag -width indent
2925 .It Cm ip Ar ip_address
2926 Define an ip address to use for aliasing.
2928 Use ip address of NIC for aliasing, dynamically changing
2929 it if NIC's ip address changes.
2931 Enable logging on this nat instance.
2933 Deny any incoming connection from outside world.
2935 Try to leave the alias port numbers unchanged from
2936 the actual local port numbers.
2938 Traffic on the local network not originating from an
2939 unregistered address spaces will be ignored.
2941 Reset table of the packet aliasing engine on address change.
2943 Reverse the way libalias handles aliasing.
2945 Obey transparent proxy rules only, packet aliasing is not performed.
2947 Skip instance in case of global state lookup (see below).
2950 Some specials value can be supplied instead of
2952 .Bl -tag -width indent
2954 Looks up translation state in all configured nat instances.
2955 If an entry is found, packet is aliased according to that entry.
2956 If no entry was found in any of the instances, packet is passed unchanged,
2957 and no new entry will be created.
2959 .Sx MULTIPLE INSTANCES
2962 for more information.
2964 Uses argument supplied in lookup table.
2967 section below for more information on lookup tables.
2970 To let the packet continue after being (de)aliased, set the sysctl variable
2971 .Va net.inet.ip.fw.one_pass
2973 For more information about aliasing modes, refer to
2977 for some examples about nat usage.
2978 .Ss REDIRECT AND LSNAT SUPPORT IN IPFW
2979 Redirect and LSNAT support follow closely the syntax used in
2983 for some examples on how to do redirect and lsnat.
2984 .Ss SCTP NAT SUPPORT
2985 SCTP nat can be configured in a similar manner to TCP through the
2988 The main difference is that
2990 does not do port translation.
2991 Since the local and global side ports will be the same,
2992 there is no need to specify both.
2993 Ports are redirected as follows:
2994 .Bd -ragged -offset indent
3000 .Cm redirect_port sctp
3001 .Ar ip_address [,addr_list] {[port | port-port] [,ports]}
3007 configuration can be done in real-time through the
3010 All may be changed dynamically, though the hash_table size will only
3015 .Sx SYSCTL VARIABLES
3017 .Sh IPv6/IPv4 NETWORK ADDRESS AND PROTOCOL TRANSLATION
3019 supports in-kernel IPv6/IPv4 network address and protocol translation.
3020 Stateful NAT64 translation allows IPv6-only clients to contact IPv4 servers
3021 using unicast TCP, UDP or ICMP protocols.
3022 One or more IPv4 addresses assigned to a stateful NAT64 translator are shared
3023 among several IPv6-only clients.
3024 When stateful NAT64 is used in conjunction with DNS64, no changes are usually
3025 required in the IPv6 client or the IPv4 server.
3028 should be loaded or kernel should have
3029 .Cm options IPFIREWALL_NAT64
3030 to be able use stateful NAT64 translator.
3032 Stateful NAT64 uses a bunch of memory for several types of objects.
3033 When IPv6 client initiates connection, NAT64 translator creates a host entry
3034 in the states table.
3035 Each host entry has a number of ports group entries allocated on demand.
3036 Ports group entries contains connection state entries.
3037 There are several options to control limits and lifetime for these objects.
3039 NAT64 translator follows RFC7915 when does ICMPv6/ICMP translation,
3040 unsupported message types will be silently dropped.
3041 IPv6 needs several ICMPv6 message types to be explicitly allowed for correct
3043 Make sure that ND6 neighbor solicitation (ICMPv6 type 135) and neighbor
3044 advertisement (ICMPv6 type 136) messages will not be handled by translation
3047 After translation NAT64 translator sends packets through corresponding netisr
3049 Thus translator host should be configured as IPv4 and IPv6 router.
3051 The stateful NAT64 configuration command is the following:
3052 .Bd -ragged -offset indent
3061 The following parameters can be configured:
3062 .Bl -tag -width indent
3063 .It Cm prefix4 Ar ipv4_prefix/plen
3064 The IPv4 prefix with mask defines the pool of IPv4 addresses used as
3065 source address after translation.
3066 Stateful NAT64 module translates IPv6 source address of client to one
3067 IPv4 address from this pool.
3068 Note that incoming IPv4 packets that don't have corresponding state entry
3069 in the states table will be dropped by translator.
3070 Make sure that translation rules handle packets, destined to configured prefix.
3071 .It Cm prefix6 Ar ipv6_prefix/length
3072 The IPv6 prefix defines IPv4-embedded IPv6 addresses used by translator
3073 to represent IPv4 addresses. This IPv6 prefix should be configured in DNS64.
3074 The translator implementation follows RFC6052, that restricts the length of
3075 prefixes to one of following: 32, 40, 48, 56, 64, or 96.
3076 The Well-Known IPv6 Prefix 64:ff9b:: must be 96 bits long.
3077 .It Cm max_ports Ar number
3078 Maximum number of ports reserved for upper level protocols to one IPv6 client.
3079 All reserved ports are divided into chunks between supported protocols.
3080 The number of connections from one IPv6 client is limited by this option.
3081 Note that closed TCP connections still remain in the list of connections until
3083 interval will not expire.
3086 .It Cm host_del_age Ar seconds
3087 The number of seconds until the host entry for a IPv6 client will be deleted
3088 and all its resources will be released due to inactivity.
3091 .It Cm pg_del_age Ar seconds
3092 The number of seconds until a ports group with unused state entries will
3096 .It Cm tcp_syn_age Ar seconds
3097 The number of seconds while a state entry for TCP connection with only SYN
3099 If TCP connection establishing will not be finished,
3100 state entry will be deleted.
3103 .It Cm tcp_est_age Ar seconds
3104 The number of seconds while a state entry for established TCP connection
3108 .It Cm tcp_close_age Ar seconds
3109 The number of seconds while a state entry for closed TCP connection
3111 Keeping state entries for closed connections is needed, because IPv4 servers
3112 typically keep closed connections in a TIME_WAIT state for a several minutes.
3113 Since translator's IPv4 addresses are shared among all IPv6 clients,
3114 new connections from the same addresses and ports may be rejected by server,
3115 because these connections are still in a TIME_WAIT state.
3116 Keeping them in translator's state table protects from such rejects.
3119 .It Cm udp_age Ar seconds
3120 The number of seconds while translator keeps state entry in a waiting for
3121 reply to the sent UDP datagram.
3124 .It Cm icmp_age Ar seconds
3125 The number of seconds while translator keeps state entry in a waiting for
3126 reply to the sent ICMP message.
3130 Turn on logging of all handled packets via BPF through
3134 is a pseudo interface and can be created after a boot manually with
3137 Note that it has different purpose than
3140 Translators sends to BPF an additional information with each packet.
3143 you are able to see each handled packet before and after translation.
3145 Turn off logging of all handled packets via BPF.
3148 To inspect a states table of stateful NAT64 the following command can be used:
3149 .Bd -ragged -offset indent
3158 Stateless NAT64 translator doesn't use a states table for translation
3159 and converts IPv4 addresses to IPv6 and vice versa solely based on the
3160 mappings taken from configured lookup tables.
3161 Since a states table doesn't used by stateless translator,
3162 it can be configured to pass IPv4 clients to IPv6-only servers.
3164 The stateless NAT64 configuration command is the following:
3165 .Bd -ragged -offset indent
3174 The following parameters can be configured:
3175 .Bl -tag -width indent
3176 .It Cm prefix6 Ar ipv6_prefix/length
3177 The IPv6 prefix defines IPv4-embedded IPv6 addresses used by translator
3178 to represent IPv4 addresses. This IPv6 prefix should be configured in DNS64.
3179 .It Cm table4 Ar table46
3182 contains mapping how IPv4 addresses should be translated to IPv6 addresses.
3183 .It Cm table6 Ar table64
3186 contains mapping how IPv6 addresses should be translated to IPv4 addresses.
3188 Turn on logging of all handled packets via BPF through
3192 Turn off logging of all handled packets via BPF.
3195 Note that the behavior of stateless translator with respect to not matched
3196 packets differs from stateful translator.
3197 If corresponding addresses was not found in the lookup tables, the packet
3198 will not be dropped and the search continues.
3199 .Sh IPv6-to-IPv6 NETWORK PREFIX TRANSLATION (NPTv6)
3201 supports in-kernel IPv6-to-IPv6 network prefix translation as described
3205 should be loaded or kernel should has
3206 .Cm options IPFIREWALL_NPTV6
3207 to be able use NPTv6 translator.
3209 The NPTv6 configuration command is the following:
3210 .Bd -ragged -offset indent
3219 The following parameters can be configured:
3220 .Bl -tag -width indent
3221 .It Cm int_prefix Ar ipv6_prefix
3222 IPv6 prefix used in internal network.
3223 NPTv6 module translates source address when it matches this prefix.
3224 .It Cm ext_prefix Ar ipv6_prefix
3225 IPv6 prefix used in external network.
3226 NPTv6 module translates destination address when it matches this prefix.
3227 .It Cm prefixlen Ar length
3228 The length of specified IPv6 prefixes. It must be in range from 8 to 64.
3231 Note that the prefix translation rules are silently ignored when IPv6 packet
3232 forwarding is disabled.
3233 To enable the packet forwarding, set the sysctl variable
3234 .Va net.inet6.ip6.forwarding
3237 To let the packet continue after being translated, set the sysctl variable
3238 .Va net.inet.ip.fw.one_pass
3241 Tunables can be set in
3247 before ipfw module gets loaded.
3248 .Bl -tag -width indent
3249 .It Va net.inet.ip.fw.default_to_accept: No 0
3250 Defines ipfw last rule behavior.
3251 This value overrides
3252 .Cd "options IPFW_DEFAULT_TO_(ACCEPT|DENY)"
3253 from kernel configuration file.
3254 .It Va net.inet.ip.fw.tables_max: No 128
3255 Defines number of tables available in ipfw.
3256 Number cannot exceed 65534.
3258 .Sh SYSCTL VARIABLES
3261 variables controls the behaviour of the firewall and
3263 .Pq Nm dummynet , bridge , sctp nat .
3264 These are shown below together with their default value
3265 (but always check with the
3267 command what value is actually in use) and meaning:
3268 .Bl -tag -width indent
3269 .It Va net.inet.ip.alias.sctp.accept_global_ootb_addip: No 0
3272 responds to receipt of global OOTB ASCONF-AddIP:
3273 .Bl -tag -width indent
3275 No response (unless a partially matching association exists -
3276 ports and vtags match but global address does not)
3279 will accept and process all OOTB global AddIP messages.
3282 Option 1 should never be selected as this forms a security risk.
3284 establish multiple fake associations by sending AddIP messages.
3285 .It Va net.inet.ip.alias.sctp.chunk_proc_limit: No 5
3286 Defines the maximum number of chunks in an SCTP packet that will be
3288 packet that matches an existing association.
3289 This value is enforced to be greater or equal than
3290 .Cm net.inet.ip.alias.sctp.initialising_chunk_proc_limit .
3292 a DoS risk yet setting too low a value may result in
3293 important control chunks in
3294 the packet not being located and parsed.
3295 .It Va net.inet.ip.alias.sctp.error_on_ootb: No 1
3298 responds to any Out-of-the-Blue (OOTB) packets with ErrorM packets.
3299 An OOTB packet is a packet that arrives with no existing association
3302 and is not an INIT or ASCONF-AddIP packet:
3303 .Bl -tag -width indent
3305 ErrorM is never sent in response to OOTB packets.
3307 ErrorM is only sent to OOTB packets received on the local side.
3309 ErrorM is sent to the local side and on the global side ONLY if there is a
3310 partial match (ports and vtags match but the source global IP does not).
3311 This value is only useful if the
3313 is tracking global IP addresses.
3315 ErrorM is sent in response to all OOTB packets on both
3316 the local and global side
3320 At the moment the default is 0, since the ErrorM packet is not yet
3321 supported by most SCTP stacks.
3322 When it is supported, and if not tracking
3323 global addresses, we recommend setting this value to 1 to allow
3324 multi-homed local hosts to function with the
3326 To track global addresses, we recommend setting this value to 2 to
3327 allow global hosts to be informed when they need to (re)send an
3329 Value 3 should never be chosen (except for debugging) as the
3331 will respond to all OOTB global packets (a DoS risk).
3332 .It Va net.inet.ip.alias.sctp.hashtable_size: No 2003
3333 Size of hash tables used for
3335 lookups (100 < prime_number > 1000001).
3338 size for any future created
3340 instance and therefore must be set prior to creating a
3343 The table sizes may be changed to suit specific needs.
3344 If there will be few
3345 concurrent associations, and memory is scarce, you may make these smaller.
3346 If there will be many thousands (or millions) of concurrent associations, you
3347 should make these larger.
3348 A prime number is best for the table size.
3350 update function will adjust your input value to the next highest prime number.
3351 .It Va net.inet.ip.alias.sctp.holddown_time: No 0
3352 Hold association in table for this many seconds after receiving a
3354 This allows endpoints to correct shutdown gracefully if a
3355 shutdown_complete is lost and retransmissions are required.
3356 .It Va net.inet.ip.alias.sctp.init_timer: No 15
3357 Timeout value while waiting for (INIT-ACK|AddIP-ACK).
3358 This value cannot be 0.
3359 .It Va net.inet.ip.alias.sctp.initialising_chunk_proc_limit: No 2
3360 Defines the maximum number of chunks in an SCTP packet that will be parsed when
3361 no existing association exists that matches that packet.
3363 will only be an INIT or ASCONF-AddIP packet.
3364 A higher value may become a DoS
3365 risk as malformed packets can consume processing resources.
3366 .It Va net.inet.ip.alias.sctp.param_proc_limit: No 25
3367 Defines the maximum number of parameters within a chunk that will be
3370 As for other similar sysctl variables, larger values pose a DoS risk.
3371 .It Va net.inet.ip.alias.sctp.log_level: No 0
3372 Level of detail in the system log messages (0 \- minimal, 1 \- event,
3373 2 \- info, 3 \- detail, 4 \- debug, 5 \- max debug).
3375 option in high loss environments.
3376 .It Va net.inet.ip.alias.sctp.shutdown_time: No 15
3377 Timeout value while waiting for SHUTDOWN-COMPLETE.
3378 This value cannot be 0.
3379 .It Va net.inet.ip.alias.sctp.track_global_addresses: No 0
3380 Enables/disables global IP address tracking within the
3383 upper limit on the number of addresses tracked for each association:
3384 .Bl -tag -width indent
3386 Global tracking is disabled
3388 Enables tracking, the maximum number of addresses tracked for each
3389 association is limited to this value
3392 This variable is fully dynamic, the new value will be adopted for all newly
3393 arriving associations, existing associations are treated
3394 as they were previously.
3395 Global tracking will decrease the number of collisions within the
3398 of increased processing load, memory usage, complexity, and possible
3401 problems in complex networks with multiple
3403 We recommend not tracking
3404 global IP addresses, this will still result in a fully functional
3406 .It Va net.inet.ip.alias.sctp.up_timer: No 300
3407 Timeout value to keep an association up with no traffic.
3408 This value cannot be 0.
3409 .It Va net.inet.ip.dummynet.expire : No 1
3410 Lazily delete dynamic pipes/queue once they have no pending traffic.
3411 You can disable this by setting the variable to 0, in which case
3412 the pipes/queues will only be deleted when the threshold is reached.
3413 .It Va net.inet.ip.dummynet.hash_size : No 64
3414 Default size of the hash table used for dynamic pipes/queues.
3415 This value is used when no
3417 option is specified when configuring a pipe/queue.
3418 .It Va net.inet.ip.dummynet.io_fast : No 0
3419 If set to a non-zero value,
3424 operation (see above) is enabled.
3425 .It Va net.inet.ip.dummynet.io_pkt
3426 Number of packets passed to
3428 .It Va net.inet.ip.dummynet.io_pkt_drop
3429 Number of packets dropped by
3431 .It Va net.inet.ip.dummynet.io_pkt_fast
3432 Number of packets bypassed by the
3435 .It Va net.inet.ip.dummynet.max_chain_len : No 16
3436 Target value for the maximum number of pipes/queues in a hash bucket.
3438 .Cm max_chain_len*hash_size
3439 is used to determine the threshold over which empty pipes/queues
3440 will be expired even when
3441 .Cm net.inet.ip.dummynet.expire=0 .
3442 .It Va net.inet.ip.dummynet.red_lookup_depth : No 256
3443 .It Va net.inet.ip.dummynet.red_avg_pkt_size : No 512
3444 .It Va net.inet.ip.dummynet.red_max_pkt_size : No 1500
3445 Parameters used in the computations of the drop probability
3446 for the RED algorithm.
3447 .It Va net.inet.ip.dummynet.pipe_byte_limit : No 1048576
3448 .It Va net.inet.ip.dummynet.pipe_slot_limit : No 100
3449 The maximum queue size that can be specified in bytes or packets.
3450 These limits prevent accidental exhaustion of resources such as mbufs.
3451 If you raise these limits,
3452 you should make sure the system is configured so that sufficient resources
3454 .It Va net.inet.ip.fw.autoinc_step : No 100
3455 Delta between rule numbers when auto-generating them.
3456 The value must be in the range 1..1000.
3457 .It Va net.inet.ip.fw.curr_dyn_buckets : Va net.inet.ip.fw.dyn_buckets
3458 The current number of buckets in the hash table for dynamic rules
3460 .It Va net.inet.ip.fw.debug : No 1
3461 Controls debugging messages produced by
3463 .It Va net.inet.ip.fw.default_rule : No 65535
3464 The default rule number (read-only).
3466 .Nm , the default rule is the last one, so its number
3467 can also serve as the highest number allowed for a rule.
3468 .It Va net.inet.ip.fw.dyn_buckets : No 256
3469 The number of buckets in the hash table for dynamic rules.
3470 Must be a power of 2, up to 65536.
3471 It only takes effect when all dynamic rules have expired, so you
3472 are advised to use a
3474 command to make sure that the hash table is resized.
3475 .It Va net.inet.ip.fw.dyn_count : No 3
3476 Current number of dynamic rules
3478 .It Va net.inet.ip.fw.dyn_keepalive : No 1
3479 Enables generation of keepalive packets for
3481 rules on TCP sessions.
3482 A keepalive is generated to both
3483 sides of the connection every 5 seconds for the last 20
3484 seconds of the lifetime of the rule.
3485 .It Va net.inet.ip.fw.dyn_max : No 8192
3486 Maximum number of dynamic rules.
3487 When you hit this limit, no more dynamic rules can be
3488 installed until old ones expire.
3489 .It Va net.inet.ip.fw.dyn_ack_lifetime : No 300
3490 .It Va net.inet.ip.fw.dyn_syn_lifetime : No 20
3491 .It Va net.inet.ip.fw.dyn_fin_lifetime : No 1
3492 .It Va net.inet.ip.fw.dyn_rst_lifetime : No 1
3493 .It Va net.inet.ip.fw.dyn_udp_lifetime : No 5
3494 .It Va net.inet.ip.fw.dyn_short_lifetime : No 30
3495 These variables control the lifetime, in seconds, of dynamic
3497 Upon the initial SYN exchange the lifetime is kept short,
3498 then increased after both SYN have been seen, then decreased
3499 again during the final FIN exchange or when a RST is received.
3501 .Em dyn_fin_lifetime
3503 .Em dyn_rst_lifetime
3504 must be strictly lower than 5 seconds, the period of
3505 repetition of keepalives.
3506 The firewall enforces that.
3507 .It Va net.inet.ip.fw.dyn_keep_states: No 0
3508 Keep dynamic states on rule/set deletion.
3509 States are relinked to default rule (65535).
3510 This can be handly for ruleset reload.
3511 Turned off by default.
3512 .It Va net.inet.ip.fw.enable : No 1
3513 Enables the firewall.
3514 Setting this variable to 0 lets you run your machine without
3515 firewall even if compiled in.
3516 .It Va net.inet6.ip6.fw.enable : No 1
3517 provides the same functionality as above for the IPv6 case.
3518 .It Va net.inet.ip.fw.one_pass : No 1
3519 When set, the packet exiting from the
3523 node is not passed though the firewall again.
3524 Otherwise, after an action, the packet is
3525 reinjected into the firewall at the next rule.
3526 .It Va net.inet.ip.fw.tables_max : No 128
3527 Maximum number of tables.
3528 .It Va net.inet.ip.fw.verbose : No 1
3529 Enables verbose messages.
3530 .It Va net.inet.ip.fw.verbose_limit : No 0
3531 Limits the number of messages produced by a verbose firewall.
3532 .It Va net.inet6.ip6.fw.deny_unknown_exthdrs : No 1
3533 If enabled packets with unknown IPv6 Extension Headers will be denied.
3534 .It Va net.link.ether.ipfw : No 0
3535 Controls whether layer-2 packets are passed to
3538 .It Va net.link.bridge.ipfw : No 0
3539 Controls whether bridged packets are passed to
3543 .Sh INTERNAL DIAGNOSTICS
3544 There are some commands that may be useful to understand current state
3545 of certain subsystems inside kernel module.
3546 These commands provide debugging output which may change without notice.
3548 Currently the following commands are available as
3551 .Bl -tag -width indent
3553 Lists all interface which are currently tracked by
3555 with their in-kernel status.
3557 List all table lookup algorithms currently available.
3560 There are far too many possible uses of
3562 so this Section will only give a small set of examples.
3564 .Ss BASIC PACKET FILTERING
3565 This command adds an entry which denies all tcp packets from
3566 .Em cracker.evil.org
3567 to the telnet port of
3569 from being forwarded by the host:
3571 .Dl "ipfw add deny tcp from cracker.evil.org to wolf.tambov.su telnet"
3573 This one disallows any connection from the entire cracker's
3576 .Dl "ipfw add deny ip from 123.45.67.0/24 to my.host.org"
3578 A first and efficient way to limit access (not using dynamic rules)
3579 is the use of the following rules:
3581 .Dl "ipfw add allow tcp from any to any established"
3582 .Dl "ipfw add allow tcp from net1 portlist1 to net2 portlist2 setup"
3583 .Dl "ipfw add allow tcp from net3 portlist3 to net3 portlist3 setup"
3585 .Dl "ipfw add deny tcp from any to any"
3587 The first rule will be a quick match for normal TCP packets,
3588 but it will not match the initial SYN packet, which will be
3591 rules only for selected source/destination pairs.
3592 All other SYN packets will be rejected by the final
3596 If you administer one or more subnets, you can take advantage
3597 of the address sets and or-blocks and write extremely
3598 compact rulesets which selectively enable services to blocks
3599 of clients, as below:
3601 .Dl "goodguys=\*q{ 10.1.2.0/24{20,35,66,18} or 10.2.3.0/28{6,3,11} }\*q"
3602 .Dl "badguys=\*q10.1.2.0/24{8,38,60}\*q"
3604 .Dl "ipfw add allow ip from ${goodguys} to any"
3605 .Dl "ipfw add deny ip from ${badguys} to any"
3606 .Dl "... normal policies ..."
3610 option could be used to do automated anti-spoofing by adding the
3611 following to the top of a ruleset:
3613 .Dl "ipfw add deny ip from any to any not verrevpath in"
3615 This rule drops all incoming packets that appear to be coming to the
3616 system on the wrong interface.
3617 For example, a packet with a source
3618 address belonging to a host on a protected internal network would be
3619 dropped if it tried to enter the system from an external interface.
3623 option could be used to do similar but more restricted anti-spoofing
3624 by adding the following to the top of a ruleset:
3626 .Dl "ipfw add deny ip from any to any not antispoof in"
3628 This rule drops all incoming packets that appear to be coming from another
3629 directly connected system but on the wrong interface.
3630 For example, a packet with a source address of
3631 .Li 192.168.0.0/24 ,
3640 option could be used to (re)mark user traffic,
3641 by adding the following to the appropriate place in ruleset:
3643 .Dl "ipfw add setdscp be ip from any to any dscp af11,af21"
3645 In order to protect a site from flood attacks involving fake
3646 TCP packets, it is safer to use dynamic rules:
3648 .Dl "ipfw add check-state"
3649 .Dl "ipfw add deny tcp from any to any established"
3650 .Dl "ipfw add allow tcp from my-net to any setup keep-state"
3652 This will let the firewall install dynamic rules only for
3653 those connection which start with a regular SYN packet coming
3654 from the inside of our network.
3655 Dynamic rules are checked when encountering the first
3664 rule should usually be placed near the beginning of the
3665 ruleset to minimize the amount of work scanning the ruleset.
3666 Your mileage may vary.
3668 To limit the number of connections a user can open
3669 you can use the following type of rules:
3671 .Dl "ipfw add allow tcp from my-net/24 to any setup limit src-addr 10"
3672 .Dl "ipfw add allow tcp from any to me setup limit src-addr 4"
3674 The former (assuming it runs on a gateway) will allow each host
3675 on a /24 network to open at most 10 TCP connections.
3676 The latter can be placed on a server to make sure that a single
3677 client does not use more than 4 simultaneous connections.
3680 stateful rules can be subject to denial-of-service attacks
3681 by a SYN-flood which opens a huge number of dynamic rules.
3682 The effects of such attacks can be partially limited by
3685 variables which control the operation of the firewall.
3687 Here is a good usage of the
3689 command to see accounting records and timestamp information:
3693 or in short form without timestamps:
3697 which is equivalent to:
3701 Next rule diverts all incoming packets from 192.168.2.0/24
3702 to divert port 5000:
3704 .Dl ipfw divert 5000 ip from 192.168.2.0/24 to any in
3706 The following rules show some of the applications of
3710 for simulations and the like.
3712 This rule drops random incoming packets with a probability
3715 .Dl "ipfw add prob 0.05 deny ip from any to any in"
3717 A similar effect can be achieved making use of
3721 .Dl "ipfw add pipe 10 ip from any to any"
3722 .Dl "ipfw pipe 10 config plr 0.05"
3724 We can use pipes to artificially limit bandwidth, e.g.\& on a
3725 machine acting as a router, if we want to limit traffic from
3726 local clients on 192.168.2.0/24 we do:
3728 .Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
3729 .Dl "ipfw pipe 1 config bw 300Kbit/s queue 50KBytes"
3731 note that we use the
3733 modifier so that the rule is not used twice.
3734 Remember in fact that
3736 rules are checked both on incoming and outgoing packets.
3738 Should we want to simulate a bidirectional link with bandwidth
3739 limitations, the correct way is the following:
3741 .Dl "ipfw add pipe 1 ip from any to any out"
3742 .Dl "ipfw add pipe 2 ip from any to any in"
3743 .Dl "ipfw pipe 1 config bw 64Kbit/s queue 10Kbytes"
3744 .Dl "ipfw pipe 2 config bw 64Kbit/s queue 10Kbytes"
3746 The above can be very useful, e.g.\& if you want to see how
3747 your fancy Web page will look for a residential user who
3748 is connected only through a slow link.
3749 You should not use only one pipe for both directions, unless
3750 you want to simulate a half-duplex medium (e.g.\& AppleTalk,
3752 It is not necessary that both pipes have the same configuration,
3753 so we can also simulate asymmetric links.
3755 Should we want to verify network performance with the RED queue
3756 management algorithm:
3758 .Dl "ipfw add pipe 1 ip from any to any"
3759 .Dl "ipfw pipe 1 config bw 500Kbit/s queue 100 red 0.002/30/80/0.1"
3761 Another typical application of the traffic shaper is to
3762 introduce some delay in the communication.
3763 This can significantly affect applications which do a lot of Remote
3764 Procedure Calls, and where the round-trip-time of the
3765 connection often becomes a limiting factor much more than
3768 .Dl "ipfw add pipe 1 ip from any to any out"
3769 .Dl "ipfw add pipe 2 ip from any to any in"
3770 .Dl "ipfw pipe 1 config delay 250ms bw 1Mbit/s"
3771 .Dl "ipfw pipe 2 config delay 250ms bw 1Mbit/s"
3773 Per-flow queueing can be useful for a variety of purposes.
3774 A very simple one is counting traffic:
3776 .Dl "ipfw add pipe 1 tcp from any to any"
3777 .Dl "ipfw add pipe 1 udp from any to any"
3778 .Dl "ipfw add pipe 1 ip from any to any"
3779 .Dl "ipfw pipe 1 config mask all"
3781 The above set of rules will create queues (and collect
3782 statistics) for all traffic.
3783 Because the pipes have no limitations, the only effect is
3784 collecting statistics.
3785 Note that we need 3 rules, not just the last one, because
3788 tries to match IP packets it will not consider ports, so we
3789 would not see connections on separate ports as different
3792 A more sophisticated example is limiting the outbound traffic
3793 on a net with per-host limits, rather than per-network limits:
3795 .Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
3796 .Dl "ipfw add pipe 2 ip from any to 192.168.2.0/24 in"
3797 .Dl "ipfw pipe 1 config mask src-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
3798 .Dl "ipfw pipe 2 config mask dst-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
3800 In the following example, we need to create several traffic bandwidth
3801 classes and we need different hosts/networks to fall into different classes.
3802 We create one pipe for each class and configure them accordingly.
3803 Then we create a single table and fill it with IP subnets and addresses.
3804 For each subnet/host we set the argument equal to the number of the pipe
3806 Then we classify traffic using a single rule:
3808 .Dl "ipfw pipe 1 config bw 1000Kbyte/s"
3809 .Dl "ipfw pipe 4 config bw 4000Kbyte/s"
3811 .Dl "ipfw table T1 create type addr"
3812 .Dl "ipfw table T1 add 192.168.2.0/24 1"
3813 .Dl "ipfw table T1 add 192.168.0.0/27 4"
3814 .Dl "ipfw table T1 add 192.168.0.2 1"
3816 .Dl "ipfw add pipe tablearg ip from 'table(T1)' to any"
3820 action, the table entries may include hostnames and IP addresses.
3822 .Dl "ipfw table T2 create type addr ftype ip"
3823 .Dl "ipfw table T2 add 192.168.2.0/24 10.23.2.1"
3824 .Dl "ipfw table T21 add 192.168.0.0/27 router1.dmz"
3826 .Dl "ipfw add 100 fwd tablearg ip from any to table(1)"
3828 In the following example per-interface firewall is created:
3830 .Dl "ipfw table IN create type iface valtype skipto,fib"
3831 .Dl "ipfw table IN add vlan20 12000,12"
3832 .Dl "ipfw table IN add vlan30 13000,13"
3833 .Dl "ipfw table OUT create type iface valtype skipto"
3834 .Dl "ipfw table OUT add vlan20 22000"
3835 .Dl "ipfw table OUT add vlan30 23000"
3837 .Dl "ipfw add 100 ipfw setfib tablearg ip from any to any recv 'table(IN)' in"
3838 .Dl "ipfw add 200 ipfw skipto tablearg ip from any to any recv 'table(IN)' in"
3839 .Dl "ipfw add 300 ipfw skipto tablearg ip from any to any xmit 'table(OUT)' out"
3841 The following example illustrate usage of flow tables:
3843 .Dl "ipfw table fl create type flow:flow:src-ip,proto,dst-ip,dst-port"
3844 .Dl "ipfw table fl add 2a02:6b8:77::88,tcp,2a02:6b8:77::99,80 11"
3845 .Dl "ipfw table fl add 10.0.0.1,udp,10.0.0.2,53 12"
3847 .Dl "ipfw add 100 allow ip from any to any flow 'table(fl,11)' recv ix0"
3849 To add a set of rules atomically, e.g.\& set 18:
3851 .Dl "ipfw set disable 18"
3852 .Dl "ipfw add NN set 18 ... # repeat as needed"
3853 .Dl "ipfw set enable 18"
3855 To delete a set of rules atomically the command is simply:
3857 .Dl "ipfw delete set 18"
3859 To test a ruleset and disable it and regain control if something goes wrong:
3861 .Dl "ipfw set disable 18"
3862 .Dl "ipfw add NN set 18 ... # repeat as needed"
3863 .Dl "ipfw set enable 18; echo done; sleep 30 && ipfw set disable 18"
3865 Here if everything goes well, you press control-C before the "sleep"
3866 terminates, and your ruleset will be left active.
3867 Otherwise, e.g.\& if
3868 you cannot access your box, the ruleset will be disabled after
3869 the sleep terminates thus restoring the previous situation.
3871 To show rules of the specific set:
3873 .Dl "ipfw set 18 show"
3875 To show rules of the disabled set:
3877 .Dl "ipfw -S set 18 show"
3879 To clear a specific rule counters of the specific set:
3881 .Dl "ipfw set 18 zero NN"
3883 To delete a specific rule of the specific set:
3885 .Dl "ipfw set 18 delete NN"
3886 .Ss NAT, REDIRECT AND LSNAT
3887 First redirect all the traffic to nat instance 123:
3889 .Dl "ipfw add nat 123 all from any to any"
3891 Then to configure nat instance 123 to alias all the outgoing traffic with ip
3892 192.168.0.123, blocking all incoming connections, trying to keep
3893 same ports on both sides, clearing aliasing table on address change
3894 and keeping a log of traffic/link statistics:
3896 .Dl "ipfw nat 123 config ip 192.168.0.123 log deny_in reset same_ports"
3898 Or to change address of instance 123, aliasing table will be cleared (see
3901 .Dl "ipfw nat 123 config ip 10.0.0.1"
3903 To see configuration of nat instance 123:
3905 .Dl "ipfw nat 123 show config"
3907 To show logs of all the instances in range 111-999:
3909 .Dl "ipfw nat 111-999 show"
3911 To see configurations of all instances:
3913 .Dl "ipfw nat show config"
3915 Or a redirect rule with mixed modes could looks like:
3917 .Dl "ipfw nat 123 config redirect_addr 10.0.0.1 10.0.0.66"
3918 .Dl " redirect_port tcp 192.168.0.1:80 500"
3919 .Dl " redirect_proto udp 192.168.1.43 192.168.1.1"
3920 .Dl " redirect_addr 192.168.0.10,192.168.0.11"
3921 .Dl " 10.0.0.100 # LSNAT"
3922 .Dl " redirect_port tcp 192.168.0.1:80,192.168.0.10:22"
3925 or it could be split in:
3927 .Dl "ipfw nat 1 config redirect_addr 10.0.0.1 10.0.0.66"
3928 .Dl "ipfw nat 2 config redirect_port tcp 192.168.0.1:80 500"
3929 .Dl "ipfw nat 3 config redirect_proto udp 192.168.1.43 192.168.1.1"
3930 .Dl "ipfw nat 4 config redirect_addr 192.168.0.10,192.168.0.11,192.168.0.12"
3932 .Dl "ipfw nat 5 config redirect_port tcp"
3933 .Dl " 192.168.0.1:80,192.168.0.10:22,192.168.0.20:25 500"
3954 utility first appeared in
3959 Stateful extensions were introduced in
3962 was introduced in Summer 2002.
3964 .An Ugen J. S. Antsilevich ,
3965 .An Poul-Henning Kamp ,
3971 API based upon code written by
3975 Dummynet has been introduced by Luigi Rizzo in 1997-1998.
3977 Some early work (1999-2000) on the
3979 traffic shaper supported by Akamba Corp.
3981 The ipfw core (ipfw2) has been completely redesigned and
3982 reimplemented by Luigi Rizzo in summer 2002.
3985 options have been added by various developer over the years.
3988 In-kernel NAT support written by
3989 .An Paolo Pisati Aq Mt piso@FreeBSD.org
3990 as part of a Summer of Code 2005 project.
3994 support has been developed by
3995 .An The Centre for Advanced Internet Architectures (CAIA) Aq http://www.caia.swin.edu.au .
3996 The primary developers and maintainers are David Hayes and Jason But.
3997 For further information visit:
3998 .Aq http://www.caia.swin.edu.au/urp/SONATA
4000 Delay profiles have been developed by Alessandro Cerri and
4001 Luigi Rizzo, supported by the
4002 European Commission within Projects Onelab and Onelab2.
4004 The syntax has grown over the years and sometimes it might be confusing.
4005 Unfortunately, backward compatibility prevents cleaning up mistakes
4006 made in the definition of the syntax.
4010 Misconfiguring the firewall can put your computer in an unusable state,
4011 possibly shutting down network services and requiring console access to
4012 regain control of it.
4014 Incoming packet fragments diverted by
4016 are reassembled before delivery to the socket.
4017 The action used on those packet is the one from the
4018 rule which matches the first fragment of the packet.
4020 Packets diverted to userland, and then reinserted by a userland process
4021 may lose various packet attributes.
4022 The packet source interface name
4023 will be preserved if it is shorter than 8 bytes and the userland process
4024 saves and reuses the sockaddr_in
4027 otherwise, it may be lost.
4028 If a packet is reinserted in this manner, later rules may be incorrectly
4029 applied, making the order of
4031 rules in the rule sequence very important.
4033 Dummynet drops all packets with IPv6 link-local addresses.
4039 may not behave as expected.
4040 In particular, incoming SYN packets may
4041 have no uid or gid associated with them since they do not yet belong
4042 to a TCP connection, and the uid/gid associated with a packet may not
4043 be as expected if the associated process calls
4045 or similar system calls.
4047 Rule syntax is subject to the command line environment and some patterns
4048 may need to be escaped with the backslash character
4049 or quoted appropriately.
4051 Due to the architecture of
4053 ipfw nat is not compatible with the TCP segmentation offloading (TSO).
4054 Thus, to reliably nat your network traffic, please disable TSO
4058 ICMP error messages are not implicitly matched by dynamic rules
4059 for the respective conversations.
4060 To avoid failures of network error detection and path MTU discovery,
4061 ICMP error messages may need to be allowed explicitly through static
4068 actions may lead to confusing behaviour if ruleset has mistakes,
4069 and/or interaction with other subsystems (netgraph, dummynet, etc.) is used.
4070 One possible case for this is packet leaving
4072 in subroutine on the input pass, while later on output encountering unpaired
4075 As the call stack is kept intact after input pass, packet will suddenly
4076 return to the rule number used on input pass, not on output one.
4077 Order of processing should be checked carefully to avoid such mistakes.