xref: /freebsd/sbin/ipfw/ipfw.8 (revision 2b743a9e9ddc6736208dc8ca1ce06ce64ad20a19)
1.\"
2.\" $FreeBSD$
3.\"
4.Dd July 25, 2006
5.Dt IPFW 8
6.Os
7.Sh NAME
8.Nm ipfw
9.Nd IP firewall and traffic shaper control program
10.Sh SYNOPSIS
11.Nm
12.Op Fl cq
13.Cm add
14.Ar rule
15.Nm
16.Op Fl acdefnNStT
17.Brq Cm list | show
18.Op Ar rule | first-last ...
19.Nm
20.Op Fl f | q
21.Cm flush
22.Nm
23.Op Fl q
24.Brq Cm delete | zero | resetlog
25.Op Cm set
26.Op Ar number ...
27.Nm
28.Cm enable
29.Brq Cm firewall | altq | one_pass | debug | verbose | dyn_keepalive
30.Nm
31.Cm disable
32.Brq Cm firewall | altq | one_pass | debug | verbose | dyn_keepalive
33.Pp
34.Nm
35.Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
36.Nm
37.Cm set move
38.Op Cm rule
39.Ar number Cm to Ar number
40.Nm
41.Cm set swap Ar number number
42.Nm
43.Cm set show
44.Pp
45.Nm
46.Cm table Ar number Cm add Ar addr Ns Oo / Ns Ar masklen Oc Op Ar value
47.Nm
48.Cm table Ar number Cm delete Ar addr Ns Op / Ns Ar masklen
49.Nm
50.Cm table Ar number Cm flush
51.Nm
52.Cm table Ar number Cm list
53.Pp
54.Nm
55.Brq Cm pipe | queue
56.Ar number
57.Cm config
58.Ar config-options
59.Nm
60.Op Fl s Op Ar field
61.Brq Cm pipe | queue
62.Brq Cm delete | list | show
63.Op Ar number ...
64.Pp
65.Nm
66.Cm nat
67.Ar number
68.Cm config
69.Ar config-options
70.Pp
71.Nm
72.Op Fl cfnNqS
73.Oo
74.Fl p Ar preproc
75.Oo
76.Ar preproc-flags
77.Oc
78.Oc
79.Ar pathname
80.Sh DESCRIPTION
81The
82.Nm
83utility is the user interface for controlling the
84.Xr ipfw 4
85firewall and the
86.Xr dummynet 4
87traffic shaper in
88.Fx .
89.Pp
90An
91.Nm
92configuration, or
93.Em ruleset ,
94is made of a list of
95.Em rules
96numbered from 1 to 65535.
97Packets are passed to
98.Nm
99from a number of different places in the protocol stack
100(depending on the source and destination of the packet,
101it is possible that
102.Nm
103is invoked multiple times on the same packet).
104The packet passed to the firewall is compared
105against each of the rules in the firewall
106.Em ruleset .
107When a match is found, the action corresponding to the
108matching rule is performed.
109.Pp
110Depending on the action and certain system settings, packets
111can be reinjected into the firewall at some rule after the
112matching one for further processing.
113.Pp
114An
115.Nm
116ruleset always includes a
117.Em default
118rule (numbered 65535) which cannot be modified or deleted,
119and matches all packets.
120The action associated with the
121.Em default
122rule can be either
123.Cm deny
124or
125.Cm allow
126depending on how the kernel is configured.
127.Pp
128If the ruleset includes one or more rules with the
129.Cm keep-state
130or
131.Cm limit
132option, then
133.Nm
134assumes a
135.Em stateful
136behaviour, i.e., upon a match it will create dynamic rules matching
137the exact parameters (addresses and ports) of the matching packet.
138.Pp
139These dynamic rules, which have a limited lifetime, are checked
140at the first occurrence of a
141.Cm check-state ,
142.Cm keep-state
143or
144.Cm limit
145rule, and are typically used to open the firewall on-demand to
146legitimate traffic only.
147See the
148.Sx STATEFUL FIREWALL
149and
150.Sx EXAMPLES
151Sections below for more information on the stateful behaviour of
152.Nm .
153.Pp
154All rules (including dynamic ones) have a few associated counters:
155a packet count, a byte count, a log count and a timestamp
156indicating the time of the last match.
157Counters can be displayed or reset with
158.Nm
159commands.
160.Pp
161Rules can be added with the
162.Cm add
163command; deleted individually or in groups with the
164.Cm delete
165command, and globally (except those in set 31) with the
166.Cm flush
167command; displayed, optionally with the content of the
168counters, using the
169.Cm show
170and
171.Cm list
172commands.
173Finally, counters can be reset with the
174.Cm zero
175and
176.Cm resetlog
177commands.
178.Pp
179Also, each rule belongs to one of 32 different
180.Em sets
181, and there are
182.Nm
183commands to atomically manipulate sets, such as enable,
184disable, swap sets, move all rules in a set to another
185one, delete all rules in a set.
186These can be useful to
187install temporary configurations, or to test them.
188See Section
189.Sx SETS OF RULES
190for more information on
191.Em sets .
192.Pp
193The following options are available:
194.Bl -tag -width indent
195.It Fl a
196While listing, show counter values.
197The
198.Cm show
199command just implies this option.
200.It Fl b
201Only show the action and the comment, not the body of a rule.
202Implies
203.Fl c .
204.It Fl c
205When entering or showing rules, print them in compact form,
206i.e., without the optional "ip from any to any" string
207when this does not carry any additional information.
208.It Fl d
209While listing, show dynamic rules in addition to static ones.
210.It Fl e
211While listing, if the
212.Fl d
213option was specified, also show expired dynamic rules.
214.It Fl f
215Do not ask for confirmation for commands that can cause problems
216if misused,
217.No i.e. Cm flush .
218If there is no tty associated with the process, this is implied.
219.It Fl n
220Only check syntax of the command strings, without actually passing
221them to the kernel.
222.It Fl N
223Try to resolve addresses and service names in output.
224.It Fl q
225While
226.Cm add Ns ing ,
227.Cm zero Ns ing ,
228.Cm resetlog Ns ging
229or
230.Cm flush Ns ing ,
231be quiet about actions
232(implies
233.Fl f ) .
234This is useful for adjusting rules by executing multiple
235.Nm
236commands in a script
237(e.g.,
238.Ql sh\ /etc/rc.firewall ) ,
239or by processing a file of many
240.Nm
241rules across a remote login session.
242It also stops a table add or delete
243from failing if the entry already exists or is not present.
244If a
245.Cm flush
246is performed in normal (verbose) mode (with the default kernel
247configuration), it prints a message.
248Because all rules are flushed, the message might not be delivered
249to the login session, causing the remote login session to be closed
250and the remainder of the ruleset to not be processed.
251Access to the console would then be required to recover.
252.It Fl S
253While listing rules, show the
254.Em set
255each rule belongs to.
256If this flag is not specified, disabled rules will not be
257listed.
258.It Fl s Op Ar field
259While listing pipes, sort according to one of the four
260counters (total or current packets or bytes).
261.It Fl t
262While listing, show last match timestamp (converted with ctime()).
263.It Fl T
264While listing, show last match timestamp (as seconds from the epoch).
265This form can be more convenient for postprocessing by scripts.
266.El
267.Pp
268To ease configuration, rules can be put into a file which is
269processed using
270.Nm
271as shown in the last synopsis line.
272An absolute
273.Ar pathname
274must be used.
275The file will be read line by line and applied as arguments to the
276.Nm
277utility.
278.Pp
279Optionally, a preprocessor can be specified using
280.Fl p Ar preproc
281where
282.Ar pathname
283is to be piped through.
284Useful preprocessors include
285.Xr cpp 1
286and
287.Xr m4 1 .
288If
289.Ar preproc
290does not start with a slash
291.Pq Ql /
292as its first character, the usual
293.Ev PATH
294name search is performed.
295Care should be taken with this in environments where not all
296file systems are mounted (yet) by the time
297.Nm
298is being run (e.g.\& when they are mounted over NFS).
299Once
300.Fl p
301has been specified, any additional arguments as passed on to the preprocessor
302for interpretation.
303This allows for flexible configuration files (like conditionalizing
304them on the local hostname) and the use of macros to centralize
305frequently required arguments like IP addresses.
306.Pp
307The
308.Nm
309.Cm pipe
310and
311.Cm queue
312commands are used to configure the traffic shaper, as shown in the
313.Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
314Section below.
315.Pp
316If the world and the kernel get out of sync the
317.Nm
318ABI may break, preventing you from being able to add any rules.
319This can
320adversely effect the booting process.
321You can use
322.Nm
323.Cm disable
324.Cm firewall
325to temporarily disable the firewall to regain access to the network,
326allowing you to fix the problem.
327.Sh PACKET FLOW
328A packet is checked against the active ruleset in multiple places
329in the protocol stack, under control of several sysctl variables.
330These places and variables are shown below, and it is important to
331have this picture in mind in order to design a correct ruleset.
332.Bd -literal -offset indent
333       ^    to upper layers    V
334       |                       |
335       +----------->-----------+
336       ^                       V
337 [ip(6)_input]           [ip(6)_output]     net.inet(6).ip(6).fw.enable=1
338       |                       |
339       ^                       V
340 [ether_demux]        [ether_output_frame]  net.link.ether.ipfw=1
341       |                       |
342       +-->--[bdg_forward]-->--+            net.link.bridge.ipfw=1
343       ^                       V
344       |      to devices       |
345.Ed
346.Pp
347As can be noted from the above picture, the number of
348times the same packet goes through the firewall can
349vary between 0 and 4 depending on packet source and
350destination, and system configuration.
351.Pp
352Note that as packets flow through the stack, headers can be
353stripped or added to it, and so they may or may not be available
354for inspection.
355E.g., incoming packets will include the MAC header when
356.Nm
357is invoked from
358.Cm ether_demux() ,
359but the same packets will have the MAC header stripped off when
360.Nm
361is invoked from
362.Cm ip_input()
363or
364.Cm ip6_input() .
365.Pp
366Also note that each packet is always checked against the complete ruleset,
367irrespective of the place where the check occurs, or the source of the packet.
368If a rule contains some match patterns or actions which are not valid
369for the place of invocation (e.g.\& trying to match a MAC header within
370.Cm ip_input
371or
372.Cm ip6_input ),
373the match pattern will not match, but a
374.Cm not
375operator in front of such patterns
376.Em will
377cause the pattern to
378.Em always
379match on those packets.
380It is thus the responsibility of
381the programmer, if necessary, to write a suitable ruleset to
382differentiate among the possible places.
383.Cm skipto
384rules can be useful here, as an example:
385.Bd -literal -offset indent
386# packets from ether_demux or bdg_forward
387ipfw add 10 skipto 1000 all from any to any layer2 in
388# packets from ip_input
389ipfw add 10 skipto 2000 all from any to any not layer2 in
390# packets from ip_output
391ipfw add 10 skipto 3000 all from any to any not layer2 out
392# packets from ether_output_frame
393ipfw add 10 skipto 4000 all from any to any layer2 out
394.Ed
395.Pp
396(yes, at the moment there is no way to differentiate between
397ether_demux and bdg_forward).
398.Sh SYNTAX
399In general, each keyword or argument must be provided as
400a separate command line argument, with no leading or trailing
401spaces.
402Keywords are case-sensitive, whereas arguments may
403or may not be case-sensitive depending on their nature
404(e.g.\& uid's are, hostnames are not).
405.Pp
406In
407.Nm ipfw2
408you can introduce spaces after commas ',' to make
409the line more readable.
410You can also put the entire
411command (including flags) into a single argument.
412E.g., the following forms are equivalent:
413.Bd -literal -offset indent
414ipfw -q add deny src-ip 10.0.0.0/24,127.0.0.1/8
415ipfw -q add deny src-ip 10.0.0.0/24, 127.0.0.1/8
416ipfw "-q add deny src-ip 10.0.0.0/24, 127.0.0.1/8"
417.Ed
418.Sh RULE FORMAT
419The format of
420.Nm
421rules is the following:
422.Bd -ragged -offset indent
423.Bk -words
424.Op Ar rule_number
425.Op Cm set Ar set_number
426.Op Cm prob Ar match_probability
427.Ar action
428.Op Cm log Op Cm logamount Ar number
429.Op Cm altq Ar queue
430.Oo
431.Bro Cm tag | untag
432.Brc Ar number
433.Oc
434.Ar body
435.Ek
436.Ed
437.Pp
438where the body of the rule specifies which information is used
439for filtering packets, among the following:
440.Pp
441.Bl -tag -width "Source and dest. addresses and ports" -offset XXX -compact
442.It Layer-2 header fields
443When available
444.It IPv4 and IPv6 Protocol
445TCP, UDP, ICMP, etc.
446.It Source and dest. addresses and ports
447.It Direction
448See Section
449.Sx PACKET FLOW
450.It Transmit and receive interface
451By name or address
452.It Misc. IP header fields
453Version, type of service, datagram length, identification,
454fragment flag (non-zero IP offset),
455Time To Live
456.It IP options
457.It IPv6 Extension headers
458Fragmentation, Hop-by-Hop options,
459source routing, IPSec options.
460.It IPv6 Flow-ID
461.It Misc. TCP header fields
462TCP flags (SYN, FIN, ACK, RST, etc.),
463sequence number, acknowledgment number,
464window
465.It TCP options
466.It ICMP types
467for ICMP packets
468.It ICMP6 types
469for ICMP6 packets
470.It User/group ID
471When the packet can be associated with a local socket.
472.It Divert status
473Whether a packet came from a divert socket (e.g.,
474.Xr natd 8 ) .
475.El
476.Pp
477Note that some of the above information, e.g.\& source MAC or IP addresses and
478TCP/UDP ports, could easily be spoofed, so filtering on those fields
479alone might not guarantee the desired results.
480.Bl -tag -width indent
481.It Ar rule_number
482Each rule is associated with a
483.Ar rule_number
484in the range 1..65535, with the latter reserved for the
485.Em default
486rule.
487Rules are checked sequentially by rule number.
488Multiple rules can have the same number, in which case they are
489checked (and listed) according to the order in which they have
490been added.
491If a rule is entered without specifying a number, the kernel will
492assign one in such a way that the rule becomes the last one
493before the
494.Em default
495rule.
496Automatic rule numbers are assigned by incrementing the last
497non-default rule number by the value of the sysctl variable
498.Ar net.inet.ip.fw.autoinc_step
499which defaults to 100.
500If this is not possible (e.g.\& because we would go beyond the
501maximum allowed rule number), the number of the last
502non-default value is used instead.
503.It Cm set Ar set_number
504Each rule is associated with a
505.Ar set_number
506in the range 0..31.
507Sets can be individually disabled and enabled, so this parameter
508is of fundamental importance for atomic ruleset manipulation.
509It can be also used to simplify deletion of groups of rules.
510If a rule is entered without specifying a set number,
511set 0 will be used.
512.br
513Set 31 is special in that it cannot be disabled,
514and rules in set 31 are not deleted by the
515.Nm ipfw flush
516command (but you can delete them with the
517.Nm ipfw delete set 31
518command).
519Set 31 is also used for the
520.Em default
521rule.
522.It Cm prob Ar match_probability
523A match is only declared with the specified probability
524(floating point number between 0 and 1).
525This can be useful for a number of applications such as
526random packet drop or
527(in conjunction with
528.Xr dummynet 4 )
529to simulate the effect of multiple paths leading to out-of-order
530packet delivery.
531.Pp
532Note: this condition is checked before any other condition, including
533ones such as keep-state or check-state which might have side effects.
534.It Cm log Op Cm logamount Ar number
535When a packet matches a rule with the
536.Cm log
537keyword, a message will be
538logged to
539.Xr syslogd 8
540with a
541.Dv LOG_SECURITY
542facility.
543The logging only occurs if the sysctl variable
544.Em net.inet.ip.fw.verbose
545is set to 1
546(which is the default when the kernel is compiled with
547.Dv IPFIREWALL_VERBOSE )
548and the number of packets logged so far for that
549particular rule does not exceed the
550.Cm logamount
551parameter.
552If no
553.Cm logamount
554is specified, the limit is taken from the sysctl variable
555.Em net.inet.ip.fw.verbose_limit .
556In both cases, a value of 0 removes the logging limit.
557.Pp
558Once the limit is reached, logging can be re-enabled by
559clearing the logging counter or the packet counter for that entry, see the
560.Cm resetlog
561command.
562.Pp
563Note: logging is done after all other packet matching conditions
564have been successfully verified, and before performing the final
565action (accept, deny, etc.) on the packet.
566.It Cm tag Ar number
567When a packet matches a rule with the
568.Cm tag
569keyword, the numeric tag for the given
570.Ar number
571in the range 1..65534 will be attached to the packet.
572The tag acts as an internal marker (it is not sent out over
573the wire) that can be used to identify these packets later on.
574This can be used, for example, to provide trust between interfaces
575and to start doing policy-based filtering.
576A packet can have mutiple tags at the same time.
577Tags are "sticky", meaning once a tag is applied to a packet by a
578matching rule it exists until explicit removal.
579Tags are kept with the packet everywhere within the kernel, but are
580lost when packet leaves the kernel, for example, on transmitting
581packet out to the network or sending packet to a
582.Xr divert 4
583socket.
584.Pp
585To check for previously applied tags, use the
586.Cm tagged
587rule option.
588To delete previously applied tag, use the
589.Cm untag
590keyword.
591.Pp
592Note: since tags are kept with the packet everywhere in kernelspace,
593they can be set and unset anywhere in kernel network subsystem
594(using
595.Xr mbuf_tags 9
596facility), not only by means of
597.Xr ipfw 4
598.Cm tag
599and
600.Cm untag
601keywords.
602For example, there can be a specialized
603.Xr netgraph 4
604node doing traffic analyzing and tagging for later inspecting
605in firewall.
606.It Cm untag Ar number
607When a packet matches a rule with the
608.Cm untag
609keyword, the tag with the number
610.Ar number
611is searched among the tags attached to this packet and,
612if found, removed from it.
613Other tags bound to packet, if present, are left untouched.
614.It Cm altq Ar queue
615When a packet matches a rule with the
616.Cm altq
617keyword, the ALTQ identifier for the given
618.Ar queue
619(see
620.Xr altq 4 )
621will be attached.
622Note that this ALTQ tag is only meaningful for packets going "out" of IPFW,
623and not being rejected or going to divert sockets.
624Note that if there is insufficient memory at the time the packet is
625processed, it will not be tagged, so it is wise to make your ALTQ
626"default" queue policy account for this.
627If multiple
628.Cm altq
629rules match a single packet, only the first one adds the ALTQ classification
630tag.
631In doing so, traffic may be shaped by using
632.Cm count Cm altq Ar queue
633rules for classification early in the ruleset, then later applying
634the filtering decision.
635For example,
636.Cm check-state
637and
638.Cm keep-state
639rules may come later and provide the actual filtering decisions in
640addition to the fallback ALTQ tag.
641.Pp
642You must run
643.Xr pfctl 8
644to set up the queues before IPFW will be able to look them up by name,
645and if the ALTQ disciplines are rearranged, the rules in containing the
646queue identifiers in the kernel will likely have gone stale and need
647to be reloaded.
648Stale queue identifiers will probably result in misclassification.
649.Pp
650All system ALTQ processing can be turned on or off via
651.Nm
652.Cm enable Ar altq
653and
654.Nm
655.Cm disable Ar altq .
656The usage of
657.Em net.inet.ip.fw.one_pass
658is irrelevant to ALTQ traffic shaping, as the actual rule action is followed
659always after adding an ALTQ tag.
660.El
661.Ss RULE ACTIONS
662A rule can be associated with one of the following actions, which
663will be executed when the packet matches the body of the rule.
664.Bl -tag -width indent
665.It Cm allow | accept | pass | permit
666Allow packets that match rule.
667The search terminates.
668.It Cm check-state
669Checks the packet against the dynamic ruleset.
670If a match is found, execute the action associated with
671the rule which generated this dynamic rule, otherwise
672move to the next rule.
673.br
674.Cm Check-state
675rules do not have a body.
676If no
677.Cm check-state
678rule is found, the dynamic ruleset is checked at the first
679.Cm keep-state
680or
681.Cm limit
682rule.
683.It Cm count
684Update counters for all packets that match rule.
685The search continues with the next rule.
686.It Cm deny | drop
687Discard packets that match this rule.
688The search terminates.
689.It Cm divert Ar port
690Divert packets that match this rule to the
691.Xr divert 4
692socket bound to port
693.Ar port .
694The search terminates.
695.It Cm fwd | forward Ar ipaddr | tablearg Ns Op , Ns Ar port
696Change the next-hop on matching packets to
697.Ar ipaddr ,
698which can be an IP address or a host name.
699The next hop can also be supplied by the last table
700looked up for the packet by using the
701.Cm tablearg
702keyword instead of an explicit address.
703The search terminates if this rule matches.
704.Pp
705If
706.Ar ipaddr
707is a local address, then matching packets will be forwarded to
708.Ar port
709(or the port number in the packet if one is not specified in the rule)
710on the local machine.
711.br
712If
713.Ar ipaddr
714is not a local address, then the port number
715(if specified) is ignored, and the packet will be
716forwarded to the remote address, using the route as found in
717the local routing table for that IP.
718.br
719A
720.Ar fwd
721rule will not match layer-2 packets (those received
722on ether_input, ether_output, or bridged).
723.br
724The
725.Cm fwd
726action does not change the contents of the packet at all.
727In particular, the destination address remains unmodified, so
728packets forwarded to another system will usually be rejected by that system
729unless there is a matching rule on that system to capture them.
730For packets forwarded locally,
731the local address of the socket will be
732set to the original destination address of the packet.
733This makes the
734.Xr netstat 1
735entry look rather weird but is intended for
736use with transparent proxy servers.
737.Pp
738To enable
739.Cm fwd
740a custom kernel needs to be compiled with the option
741.Cd "options IPFIREWALL_FORWARD" .
742.It Cm nat Ar nat_nr
743Pass packet to a
744nat instance
745(for network address translation, address redirect, etc.):
746see the
747.Sx NETWORK ADDRESS TRANSLATION (NAT)
748Section for further information.
749.It Cm pipe Ar pipe_nr
750Pass packet to a
751.Xr dummynet 4
752.Dq pipe
753(for bandwidth limitation, delay, etc.).
754See the
755.Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
756Section for further information.
757The search terminates; however, on exit from the pipe and if
758the
759.Xr sysctl 8
760variable
761.Em net.inet.ip.fw.one_pass
762is not set, the packet is passed again to the firewall code
763starting from the next rule.
764.It Cm queue Ar queue_nr
765Pass packet to a
766.Xr dummynet 4
767.Dq queue
768(for bandwidth limitation using WF2Q+).
769.It Cm reject
770(Deprecated).
771Synonym for
772.Cm unreach host .
773.It Cm reset
774Discard packets that match this rule, and if the
775packet is a TCP packet, try to send a TCP reset (RST) notice.
776The search terminates.
777.It Cm reset6
778Discard packets that match this rule, and if the
779packet is a TCP packet, try to send a TCP reset (RST) notice.
780The search terminates.
781.It Cm skipto Ar number
782Skip all subsequent rules numbered less than
783.Ar number .
784The search continues with the first rule numbered
785.Ar number
786or higher.
787.It Cm tee Ar port
788Send a copy of packets matching this rule to the
789.Xr divert 4
790socket bound to port
791.Ar port .
792The search continues with the next rule.
793.It Cm unreach Ar code
794Discard packets that match this rule, and try to send an ICMP
795unreachable notice with code
796.Ar code ,
797where
798.Ar code
799is a number from 0 to 255, or one of these aliases:
800.Cm net , host , protocol , port ,
801.Cm needfrag , srcfail , net-unknown , host-unknown ,
802.Cm isolated , net-prohib , host-prohib , tosnet ,
803.Cm toshost , filter-prohib , host-precedence
804or
805.Cm precedence-cutoff .
806The search terminates.
807.It Cm unreach6 Ar code
808Discard packets that match this rule, and try to send an ICMPv6
809unreachable notice with code
810.Ar code ,
811where
812.Ar code
813is a number from 0, 1, 3 or 4, or one of these aliases:
814.Cm no-route, admin-prohib, address
815or
816.Cm port .
817The search terminates.
818.It Cm netgraph Ar cookie
819Divert packet into netgraph with given
820.Ar cookie .
821The search terminates.
822If packet is later returned from netgraph it is either
823accepted or continues with the next rule, depending on
824.Em net.inet.ip.fw.one_pass
825sysctl variable.
826.It Cm ngtee Ar cookie
827A copy of packet is diverted into netgraph, original
828packet is either accepted or continues with the next rule, depending on
829.Em net.inet.ip.fw.one_pass
830sysctl variable.
831See
832.Xr ng_ipfw 4
833for more information on
834.Cm netgraph
835and
836.Cm ngtee
837actions.
838.El
839.Ss RULE BODY
840The body of a rule contains zero or more patterns (such as
841specific source and destination addresses or ports,
842protocol options, incoming or outgoing interfaces, etc.)
843that the packet must match in order to be recognised.
844In general, the patterns are connected by (implicit)
845.Cm and
846operators -- i.e., all must match in order for the
847rule to match.
848Individual patterns can be prefixed by the
849.Cm not
850operator to reverse the result of the match, as in
851.Pp
852.Dl "ipfw add 100 allow ip from not 1.2.3.4 to any"
853.Pp
854Additionally, sets of alternative match patterns
855.Pq Em or-blocks
856can be constructed by putting the patterns in
857lists enclosed between parentheses ( ) or braces { }, and
858using the
859.Cm or
860operator as follows:
861.Pp
862.Dl "ipfw add 100 allow ip from { x or not y or z } to any"
863.Pp
864Only one level of parentheses is allowed.
865Beware that most shells have special meanings for parentheses
866or braces, so it is advisable to put a backslash \\ in front of them
867to prevent such interpretations.
868.Pp
869The body of a rule must in general include a source and destination
870address specifier.
871The keyword
872.Ar any
873can be used in various places to specify that the content of
874a required field is irrelevant.
875.Pp
876The rule body has the following format:
877.Bd -ragged -offset indent
878.Op Ar proto Cm from Ar src Cm to Ar dst
879.Op Ar options
880.Ed
881.Pp
882The first part (proto from src to dst) is for backward
883compatibility with earlier versions of
884.Fx .
885In modern
886.Fx
887any match pattern (including MAC headers, IP protocols,
888addresses and ports) can be specified in the
889.Ar options
890section.
891.Pp
892Rule fields have the following meaning:
893.Bl -tag -width indent
894.It Ar proto : protocol | Cm { Ar protocol Cm or ... }
895.It Ar protocol : Oo Cm not Oc Ar protocol-name | protocol-number
896An IP protocol specified by number or name
897(for a complete list see
898.Pa /etc/protocols ) ,
899or one of the following keywords:
900.Bl -tag -width indent
901.It Cm ip4 | ipv4
902Matches IPv4 packets.
903.It Cm ip6 | ipv6
904Matches IPv6 packets.
905.It Cm ip | all
906Matches any packet.
907.El
908.Pp
909The
910.Cm ipv6
911in
912.Cm proto
913option will be treated as inner protocol.
914And, the
915.Cm ipv4
916is not available in
917.Cm proto
918option.
919.Pp
920The
921.Cm { Ar protocol Cm or ... }
922format (an
923.Em or-block )
924is provided for convenience only but its use is deprecated.
925.It Ar src No and Ar dst : Bro Cm addr | Cm { Ar addr Cm or ... } Brc Op Oo Cm not Oc Ar ports
926An address (or a list, see below)
927optionally followed by
928.Ar ports
929specifiers.
930.Pp
931The second format
932.Em ( or-block
933with multiple addresses) is provided for convenience only and
934its use is discouraged.
935.It Ar addr : Oo Cm not Oc Bro
936.Cm any | me | me6 |
937.Cm table Ns Pq Ar number Ns Op , Ns Ar value
938.Ar | addr-list | addr-set
939.Brc
940.It Cm any
941matches any IP address.
942.It Cm me
943matches any IP address configured on an interface in the system.
944.It Cm me6
945matches any IPv6 address configured on an interface in the system.
946The address list is evaluated at the time the packet is
947analysed.
948.It Cm table Ns Pq Ar number Ns Op , Ns Ar value
949Matches any IPv4 address for which an entry exists in the lookup table
950.Ar number .
951If an optional 32-bit unsigned
952.Ar value
953is also specified, an entry will match only if it has this value.
954See the
955.Sx LOOKUP TABLES
956section below for more information on lookup tables.
957.It Ar addr-list : ip-addr Ns Op Ns , Ns Ar addr-list
958.It Ar ip-addr :
959A host or subnet address specified in one of the following ways:
960.Bl -tag -width indent
961.It Ar numeric-ip | hostname
962Matches a single IPv4 address, specified as dotted-quad or a hostname.
963Hostnames are resolved at the time the rule is added to the firewall list.
964.It Ar addr Ns / Ns Ar masklen
965Matches all addresses with base
966.Ar addr
967(specified as an IP address, a network number, or a hostname)
968and mask width of
969.Cm masklen
970bits.
971As an example, 1.2.3.4/25 or 1.2.3.0/25 will match
972all IP numbers from 1.2.3.0 to 1.2.3.127 .
973.It Ar addr Ns : Ns Ar mask
974Matches all addresses with base
975.Ar addr
976(specified as an IP address, a network number, or a hostname)
977and the mask of
978.Ar mask ,
979specified as a dotted quad.
980As an example, 1.2.3.4:255.0.255.0 or 1.0.3.0:255.0.255.0 will match
9811.*.3.*.
982This form is advised only for non-contiguous
983masks.
984It is better to resort to the
985.Ar addr Ns / Ns Ar masklen
986format for contiguous masks, which is more compact and less
987error-prone.
988.El
989.It Ar addr-set : addr Ns Oo Ns / Ns Ar masklen Oc Ns Cm { Ns Ar list Ns Cm }
990.It Ar list : Bro Ar num | num-num Brc Ns Op Ns , Ns Ar list
991Matches all addresses with base address
992.Ar addr
993(specified as an IP address, a network number, or a hostname)
994and whose last byte is in the list between braces { } .
995Note that there must be no spaces between braces and
996numbers (spaces after commas are allowed).
997Elements of the list can be specified as single entries
998or ranges.
999The
1000.Ar masklen
1001field is used to limit the size of the set of addresses,
1002and can have any value between 24 and 32.
1003If not specified,
1004it will be assumed as 24.
1005.br
1006This format is particularly useful to handle sparse address sets
1007within a single rule.
1008Because the matching occurs using a
1009bitmask, it takes constant time and dramatically reduces
1010the complexity of rulesets.
1011.br
1012As an example, an address specified as 1.2.3.4/24{128,35-55,89}
1013or 1.2.3.0/24{128,35-55,89}
1014will match the following IP addresses:
1015.br
10161.2.3.128, 1.2.3.35 to 1.2.3.55, 1.2.3.89 .
1017.It Ar addr6-list : ip6-addr Ns Op Ns , Ns Ar addr6-list
1018.It Ar ip6-addr :
1019A host or subnet specified one of the following ways:
1020.Pp
1021.Bl -tag -width indent
1022.It Ar numeric-ip | hostname
1023Matches a single IPv6 address as allowed by
1024.Xr inet_pton 3
1025or a hostname.
1026Hostnames are resolved at the time the rule is added to the firewall
1027list.
1028.It Ar addr Ns / Ns Ar masklen
1029Matches all IPv6 addresses with base
1030.Ar addr
1031(specified as allowed by
1032.Xr inet_pton
1033or a hostname)
1034and mask width of
1035.Cm masklen
1036bits.
1037.El
1038.Pp
1039No support for sets of IPv6 addresses is provided because IPv6 addresses
1040are typically random past the initial prefix.
1041.It Ar ports : Bro Ar port | port Ns \&- Ns Ar port Ns Brc Ns Op , Ns Ar ports
1042For protocols which support port numbers (such as TCP and UDP), optional
1043.Cm ports
1044may be specified as one or more ports or port ranges, separated
1045by commas but no spaces, and an optional
1046.Cm not
1047operator.
1048The
1049.Ql \&-
1050notation specifies a range of ports (including boundaries).
1051.Pp
1052Service names (from
1053.Pa /etc/services )
1054may be used instead of numeric port values.
1055The length of the port list is limited to 30 ports or ranges,
1056though one can specify larger ranges by using an
1057.Em or-block
1058in the
1059.Cm options
1060section of the rule.
1061.Pp
1062A backslash
1063.Pq Ql \e
1064can be used to escape the dash
1065.Pq Ql -
1066character in a service name (from a shell, the backslash must be
1067typed twice to avoid the shell itself interpreting it as an escape
1068character).
1069.Pp
1070.Dl "ipfw add count tcp from any ftp\e\e-data-ftp to any"
1071.Pp
1072Fragmented packets which have a non-zero offset (i.e., not the first
1073fragment) will never match a rule which has one or more port
1074specifications.
1075See the
1076.Cm frag
1077option for details on matching fragmented packets.
1078.El
1079.Ss RULE OPTIONS (MATCH PATTERNS)
1080Additional match patterns can be used within
1081rules.
1082Zero or more of these so-called
1083.Em options
1084can be present in a rule, optionally prefixed by the
1085.Cm not
1086operand, and possibly grouped into
1087.Em or-blocks .
1088.Pp
1089The following match patterns can be used (listed in alphabetical order):
1090.Bl -tag -width indent
1091.It Cm // this is a comment.
1092Inserts the specified text as a comment in the rule.
1093Everything following // is considered as a comment and stored in the rule.
1094You can have comment-only rules, which are listed as having a
1095.Cm count
1096action followed by the comment.
1097.It Cm bridged
1098Alias for
1099.Cm layer2 .
1100.It Cm diverted
1101Matches only packets generated by a divert socket.
1102.It Cm diverted-loopback
1103Matches only packets coming from a divert socket back into the IP stack
1104input for delivery.
1105.It Cm diverted-output
1106Matches only packets going from a divert socket back outward to the IP
1107stack output for delivery.
1108.It Cm dst-ip Ar ip-address
1109Matches IPv4 packets whose destination IP is one of the address(es)
1110specified as argument.
1111.It Bro Cm dst-ip6 | dst-ipv6 Brc Ar ip6-address
1112Matches IPv6 packets whose destination IP is one of the address(es)
1113specified as argument.
1114.It Cm dst-port Ar ports
1115Matches IP packets whose destination port is one of the port(s)
1116specified as argument.
1117.It Cm established
1118Matches TCP packets that have the RST or ACK bits set.
1119.It Cm ext6hdr Ar header
1120Matches IPv6 packets containing the extended header given by
1121.Ar header .
1122Supported headers are:
1123.Pp
1124Fragment,
1125.Pq Cm frag ,
1126Hop-to-hop options
1127.Pq Cm hopopt ,
1128Source routing
1129.Pq Cm route ,
1130Destination options
1131.Pq Cm dstopt ,
1132IPSec authentication headers
1133.Pq Cm ah ,
1134and IPSec encapsulated security payload headers
1135.Pq Cm esp .
1136.It Cm flow-id Ar labels
1137Matches IPv6 packets containing any of the flow labels given in
1138.Ar labels .
1139.Ar labels
1140is a comma seperate list of numeric flow labels.
1141.It Cm frag
1142Matches packets that are fragments and not the first
1143fragment of an IP datagram.
1144Note that these packets will not have
1145the next protocol header (e.g.\& TCP, UDP) so options that look into
1146these headers cannot match.
1147.It Cm gid Ar group
1148Matches all TCP or UDP packets sent by or received for a
1149.Ar group .
1150A
1151.Ar group
1152may be specified by name or number.
1153This option should be used only if debug.mpsafenet=0 to avoid possible
1154deadlocks due to layering violations in its implementation.
1155.It Cm jail Ar prisonID
1156Matches all TCP or UDP packets sent by or received for the
1157jail whos prison ID is
1158.Ar prisonID .
1159This option should be used only if debug.mpsafenet=0 to avoid possible
1160deadlocks due to layering violations in its implementation.
1161.It Cm icmptypes Ar types
1162Matches ICMP packets whose ICMP type is in the list
1163.Ar types .
1164The list may be specified as any combination of
1165individual types (numeric) separated by commas.
1166.Em Ranges are not allowed.
1167The supported ICMP types are:
1168.Pp
1169echo reply
1170.Pq Cm 0 ,
1171destination unreachable
1172.Pq Cm 3 ,
1173source quench
1174.Pq Cm 4 ,
1175redirect
1176.Pq Cm 5 ,
1177echo request
1178.Pq Cm 8 ,
1179router advertisement
1180.Pq Cm 9 ,
1181router solicitation
1182.Pq Cm 10 ,
1183time-to-live exceeded
1184.Pq Cm 11 ,
1185IP header bad
1186.Pq Cm 12 ,
1187timestamp request
1188.Pq Cm 13 ,
1189timestamp reply
1190.Pq Cm 14 ,
1191information request
1192.Pq Cm 15 ,
1193information reply
1194.Pq Cm 16 ,
1195address mask request
1196.Pq Cm 17
1197and address mask reply
1198.Pq Cm 18 .
1199.It Cm icmp6types Ar types
1200Matches ICMP6 packets whose ICMP6 type is in the list of
1201.Ar types .
1202The list may be specified as any combination of
1203individual types (numeric) separated by commas.
1204.Em Ranges are not allowed.
1205.It Cm in | out
1206Matches incoming or outgoing packets, respectively.
1207.Cm in
1208and
1209.Cm out
1210are mutually exclusive (in fact,
1211.Cm out
1212is implemented as
1213.Cm not in Ns No ).
1214.It Cm ipid Ar id-list
1215Matches IPv4 packets whose
1216.Cm ip_id
1217field has value included in
1218.Ar id-list ,
1219which is either a single value or a list of values or ranges
1220specified in the same way as
1221.Ar ports .
1222.It Cm iplen Ar len-list
1223Matches IP packets whose total length, including header and data, is
1224in the set
1225.Ar len-list ,
1226which is either a single value or a list of values or ranges
1227specified in the same way as
1228.Ar ports .
1229.It Cm ipoptions Ar spec
1230Matches packets whose IPv4 header contains the comma separated list of
1231options specified in
1232.Ar spec .
1233The supported IP options are:
1234.Pp
1235.Cm ssrr
1236(strict source route),
1237.Cm lsrr
1238(loose source route),
1239.Cm rr
1240(record packet route) and
1241.Cm ts
1242(timestamp).
1243The absence of a particular option may be denoted
1244with a
1245.Ql \&! .
1246.It Cm ipprecedence Ar precedence
1247Matches IPv4 packets whose precedence field is equal to
1248.Ar precedence .
1249.It Cm ipsec
1250Matches packets that have IPSEC history associated with them
1251(i.e., the packet comes encapsulated in IPSEC, the kernel
1252has IPSEC support and IPSEC_FILTERGIF option, and can correctly
1253decapsulate it).
1254.Pp
1255Note that specifying
1256.Cm ipsec
1257is different from specifying
1258.Cm proto Ar ipsec
1259as the latter will only look at the specific IP protocol field,
1260irrespective of IPSEC kernel support and the validity of the IPSEC data.
1261.Pp
1262Further note that this flag is silently ignored in kernels without
1263IPSEC support.
1264It does not affect rule processing when given and the
1265rules are handled as if with no
1266.Cm ipsec
1267flag.
1268.It Cm iptos Ar spec
1269Matches IPv4 packets whose
1270.Cm tos
1271field contains the comma separated list of
1272service types specified in
1273.Ar spec .
1274The supported IP types of service are:
1275.Pp
1276.Cm lowdelay
1277.Pq Dv IPTOS_LOWDELAY ,
1278.Cm throughput
1279.Pq Dv IPTOS_THROUGHPUT ,
1280.Cm reliability
1281.Pq Dv IPTOS_RELIABILITY ,
1282.Cm mincost
1283.Pq Dv IPTOS_MINCOST ,
1284.Cm congestion
1285.Pq Dv IPTOS_CE .
1286The absence of a particular type may be denoted
1287with a
1288.Ql \&! .
1289.It Cm ipttl Ar ttl-list
1290Matches IPv4 packets whose time to live is included in
1291.Ar ttl-list ,
1292which is either a single value or a list of values or ranges
1293specified in the same way as
1294.Ar ports .
1295.It Cm ipversion Ar ver
1296Matches IP packets whose IP version field is
1297.Ar ver .
1298.It Cm keep-state
1299Upon a match, the firewall will create a dynamic rule, whose
1300default behaviour is to match bidirectional traffic between
1301source and destination IP/port using the same protocol.
1302The rule has a limited lifetime (controlled by a set of
1303.Xr sysctl 8
1304variables), and the lifetime is refreshed every time a matching
1305packet is found.
1306.It Cm layer2
1307Matches only layer2 packets, i.e., those passed to
1308.Nm
1309from ether_demux() and ether_output_frame().
1310.It Cm limit Bro Cm src-addr | src-port | dst-addr | dst-port Brc Ar N
1311The firewall will only allow
1312.Ar N
1313connections with the same
1314set of parameters as specified in the rule.
1315One or more
1316of source and destination addresses and ports can be
1317specified.
1318Currently,
1319only IPv4 flows are supported.
1320.It Cm { MAC | mac } Ar dst-mac src-mac
1321Match packets with a given
1322.Ar dst-mac
1323and
1324.Ar src-mac
1325addresses, specified as the
1326.Cm any
1327keyword (matching any MAC address), or six groups of hex digits
1328separated by colons,
1329and optionally followed by a mask indicating the significant bits.
1330The mask may be specified using either of the following methods:
1331.Bl -enum -width indent
1332.It
1333A slash
1334.Pq /
1335followed by the number of significant bits.
1336For example, an address with 33 significant bits could be specified as:
1337.Pp
1338.Dl "MAC 10:20:30:40:50:60/33 any"
1339.Pp
1340.It
1341An ampersand
1342.Pq &
1343followed by a bitmask specified as six groups of hex digits separated
1344by colons.
1345For example, an address in which the last 16 bits are significant could
1346be specified as:
1347.Pp
1348.Dl "MAC 10:20:30:40:50:60&00:00:00:00:ff:ff any"
1349.Pp
1350Note that the ampersand character has a special meaning in many shells
1351and should generally be escaped.
1352.Pp
1353.El
1354Note that the order of MAC addresses (destination first,
1355source second) is
1356the same as on the wire, but the opposite of the one used for
1357IP addresses.
1358.It Cm mac-type Ar mac-type
1359Matches packets whose Ethernet Type field
1360corresponds to one of those specified as argument.
1361.Ar mac-type
1362is specified in the same way as
1363.Cm port numbers
1364(i.e., one or more comma-separated single values or ranges).
1365You can use symbolic names for known values such as
1366.Em vlan , ipv4, ipv6 .
1367Values can be entered as decimal or hexadecimal (if prefixed by 0x),
1368and they are always printed as hexadecimal (unless the
1369.Cm -N
1370option is used, in which case symbolic resolution will be attempted).
1371.It Cm proto Ar protocol
1372Matches packets with the corresponding IP protocol.
1373.It Cm recv | xmit | via Brq Ar ifX | Ar if Ns Cm * | Ar ipno | Ar any
1374Matches packets received, transmitted or going through,
1375respectively, the interface specified by exact name
1376.Ns No ( Ar ifX Ns No ),
1377by device name
1378.Ns No ( Ar if Ns Ar * Ns No ),
1379by IP address, or through some interface.
1380.Pp
1381The
1382.Cm via
1383keyword causes the interface to always be checked.
1384If
1385.Cm recv
1386or
1387.Cm xmit
1388is used instead of
1389.Cm via ,
1390then only the receive or transmit interface (respectively)
1391is checked.
1392By specifying both, it is possible to match packets based on
1393both receive and transmit interface, e.g.:
1394.Pp
1395.Dl "ipfw add deny ip from any to any out recv ed0 xmit ed1"
1396.Pp
1397The
1398.Cm recv
1399interface can be tested on either incoming or outgoing packets,
1400while the
1401.Cm xmit
1402interface can only be tested on outgoing packets.
1403So
1404.Cm out
1405is required (and
1406.Cm in
1407is invalid) whenever
1408.Cm xmit
1409is used.
1410.Pp
1411A packet may not have a receive or transmit interface: packets
1412originating from the local host have no receive interface,
1413while packets destined for the local host have no transmit
1414interface.
1415.It Cm setup
1416Matches TCP packets that have the SYN bit set but no ACK bit.
1417This is the short form of
1418.Dq Li tcpflags\ syn,!ack .
1419.It Cm src-ip Ar ip-address
1420Matches IPv4 packets whose source IP is one of the address(es)
1421specified as an argument.
1422.It Cm src-ip6 Ar ip6-address
1423Matches IPv6 packets whose source IP is one of the address(es)
1424specified as an argument.
1425.It Cm src-port Ar ports
1426Matches IP packets whose source port is one of the port(s)
1427specified as argument.
1428.It Cm tagged Ar tag-list
1429Matches packets whose tags are included in
1430.Ar tag-list ,
1431which is either a single value or a list of values or ranges
1432specified in the same way as
1433.Ar ports .
1434Tags can be applied to the packet using
1435.Cm tag
1436rule action parameter (see it's description for details on tags).
1437.It Cm tcpack Ar ack
1438TCP packets only.
1439Match if the TCP header acknowledgment number field is set to
1440.Ar ack .
1441.It Cm tcpdatalen Ar tcpdatalen-list
1442Matches TCP packets whose length of TCP data is
1443.Ar tcpdatalen-list ,
1444which is either a single value or a list of values or ranges
1445specified in the same way as
1446.Ar ports .
1447.It Cm tcpflags Ar spec
1448TCP packets only.
1449Match if the TCP header contains the comma separated list of
1450flags specified in
1451.Ar spec .
1452The supported TCP flags are:
1453.Pp
1454.Cm fin ,
1455.Cm syn ,
1456.Cm rst ,
1457.Cm psh ,
1458.Cm ack
1459and
1460.Cm urg .
1461The absence of a particular flag may be denoted
1462with a
1463.Ql \&! .
1464A rule which contains a
1465.Cm tcpflags
1466specification can never match a fragmented packet which has
1467a non-zero offset.
1468See the
1469.Cm frag
1470option for details on matching fragmented packets.
1471.It Cm tcpseq Ar seq
1472TCP packets only.
1473Match if the TCP header sequence number field is set to
1474.Ar seq .
1475.It Cm tcpwin Ar win
1476TCP packets only.
1477Match if the TCP header window field is set to
1478.Ar win .
1479.It Cm tcpoptions Ar spec
1480TCP packets only.
1481Match if the TCP header contains the comma separated list of
1482options specified in
1483.Ar spec .
1484The supported TCP options are:
1485.Pp
1486.Cm mss
1487(maximum segment size),
1488.Cm window
1489(tcp window advertisement),
1490.Cm sack
1491(selective ack),
1492.Cm ts
1493(rfc1323 timestamp) and
1494.Cm cc
1495(rfc1644 t/tcp connection count).
1496The absence of a particular option may be denoted
1497with a
1498.Ql \&! .
1499.It Cm uid Ar user
1500Match all TCP or UDP packets sent by or received for a
1501.Ar user .
1502A
1503.Ar user
1504may be matched by name or identification number.
1505This option should be used only if debug.mpsafenet=0 to avoid possible
1506deadlocks due to layering violations in its implementation.
1507.It Cm verrevpath
1508For incoming packets,
1509a routing table lookup is done on the packet's source address.
1510If the interface on which the packet entered the system matches the
1511outgoing interface for the route,
1512the packet matches.
1513If the interfaces do not match up,
1514the packet does not match.
1515All outgoing packets or packets with no incoming interface match.
1516.Pp
1517The name and functionality of the option is intentionally similar to
1518the Cisco IOS command:
1519.Pp
1520.Dl ip verify unicast reverse-path
1521.Pp
1522This option can be used to make anti-spoofing rules to reject all
1523packets with source addresses not from this interface.
1524See also the option
1525.Cm antispoof .
1526.It Cm versrcreach
1527For incoming packets,
1528a routing table lookup is done on the packet's source address.
1529If a route to the source address exists, but not the default route
1530or a blackhole/reject route, the packet matches.
1531Otherwise, the packet does not match.
1532All outgoing packets match.
1533.Pp
1534The name and functionality of the option is intentionally similar to
1535the Cisco IOS command:
1536.Pp
1537.Dl ip verify unicast source reachable-via any
1538.Pp
1539This option can be used to make anti-spoofing rules to reject all
1540packets whose source address is unreachable.
1541.It Cm antispoof
1542For incoming packets, the packet's source address is checked if it
1543belongs to a directly connected network.
1544If the network is directly connected, then the interface the packet
1545came on in is compared to the interface the network is connected to.
1546When incoming interface and directly connected interface are not the
1547same, the packet does not match.
1548Otherwise, the packet does match.
1549All outgoing packets match.
1550.Pp
1551This option can be used to make anti-spoofing rules to reject all
1552packets that pretend to be from a directly connected network but do
1553not come in through that interface.
1554This option is similar to but more restricted than
1555.Cm verrevpath
1556because it engages only on packets with source addresses of directly
1557connected networks instead of all source addresses.
1558.El
1559.Sh LOOKUP TABLES
1560Lookup tables are useful to handle large sparse address sets,
1561typically from a hundred to several thousands of entries.
1562There may be up to 128 different lookup tables, numbered 0 to 127.
1563.Pp
1564Each entry is represented by an
1565.Ar addr Ns Op / Ns Ar masklen
1566and will match all addresses with base
1567.Ar addr
1568(specified as an IP address or a hostname)
1569and mask width of
1570.Ar masklen
1571bits.
1572If
1573.Ar masklen
1574is not specified, it defaults to 32.
1575When looking up an IP address in a table, the most specific
1576entry will match.
1577Associated with each entry is a 32-bit unsigned
1578.Ar value ,
1579which can optionally be checked by a rule matching code.
1580When adding an entry, if
1581.Ar value
1582is not specified, it defaults to 0.
1583.Pp
1584An entry can be added to a table
1585.Pq Cm add ,
1586removed from a table
1587.Pq Cm delete ,
1588a table can be examined
1589.Pq Cm list
1590or flushed
1591.Pq Cm flush .
1592.Pp
1593Internally, each table is stored in a Radix tree, the same way as
1594the routing table (see
1595.Xr route 4 ) .
1596.Pp
1597Lookup tables currently support IPv4 addresses only.
1598.Pp
1599The
1600.Cm tablearg
1601feature provides the ability to use a value, looked up in the table, as
1602the argument for a rule action, action parameter or rule option.
1603This can significantly reduce number of rules in some configurations.
1604The
1605.Cm tablearg
1606argument can be used with the following actions:
1607.Cm pipe , queue, divert, tee, netgraph, ngtee, fwd
1608action parameters:
1609.Cm tag, untag,
1610rule options:
1611.Cm limit, tagged.
1612.Pp
1613When used with
1614.Cm fwd
1615it is possible to supply table entries with values
1616that are in the form of IP addresses or hostnames.
1617See the
1618.Sx EXAMPLES
1619Section for example usage of tables and the tablearg keyword.
1620.Sh SETS OF RULES
1621Each rule belongs to one of 32 different
1622.Em sets
1623, numbered 0 to 31.
1624Set 31 is reserved for the default rule.
1625.Pp
1626By default, rules are put in set 0, unless you use the
1627.Cm set N
1628attribute when entering a new rule.
1629Sets can be individually and atomically enabled or disabled,
1630so this mechanism permits an easy way to store multiple configurations
1631of the firewall and quickly (and atomically) switch between them.
1632The command to enable/disable sets is
1633.Bd -ragged -offset indent
1634.Nm
1635.Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
1636.Ed
1637.Pp
1638where multiple
1639.Cm enable
1640or
1641.Cm disable
1642sections can be specified.
1643Command execution is atomic on all the sets specified in the command.
1644By default, all sets are enabled.
1645.Pp
1646When you disable a set, its rules behave as if they do not exist
1647in the firewall configuration, with only one exception:
1648.Bd -ragged -offset indent
1649dynamic rules created from a rule before it had been disabled
1650will still be active until they expire.
1651In order to delete
1652dynamic rules you have to explicitly delete the parent rule
1653which generated them.
1654.Ed
1655.Pp
1656The set number of rules can be changed with the command
1657.Bd -ragged -offset indent
1658.Nm
1659.Cm set move
1660.Brq Cm rule Ar rule-number | old-set
1661.Cm to Ar new-set
1662.Ed
1663.Pp
1664Also, you can atomically swap two rulesets with the command
1665.Bd -ragged -offset indent
1666.Nm
1667.Cm set swap Ar first-set second-set
1668.Ed
1669.Pp
1670See the
1671.Sx EXAMPLES
1672Section on some possible uses of sets of rules.
1673.Sh STATEFUL FIREWALL
1674Stateful operation is a way for the firewall to dynamically
1675create rules for specific flows when packets that
1676match a given pattern are detected.
1677Support for stateful
1678operation comes through the
1679.Cm check-state , keep-state
1680and
1681.Cm limit
1682options of
1683.Nm rules .
1684.Pp
1685Dynamic rules are created when a packet matches a
1686.Cm keep-state
1687or
1688.Cm limit
1689rule, causing the creation of a
1690.Em dynamic
1691rule which will match all and only packets with
1692a given
1693.Em protocol
1694between a
1695.Em src-ip/src-port dst-ip/dst-port
1696pair of addresses
1697.Em ( src
1698and
1699.Em dst
1700are used here only to denote the initial match addresses, but they
1701are completely equivalent afterwards).
1702Dynamic rules will be checked at the first
1703.Cm check-state, keep-state
1704or
1705.Cm limit
1706occurrence, and the action performed upon a match will be the same
1707as in the parent rule.
1708.Pp
1709Note that no additional attributes other than protocol and IP addresses
1710and ports are checked on dynamic rules.
1711.Pp
1712The typical use of dynamic rules is to keep a closed firewall configuration,
1713but let the first TCP SYN packet from the inside network install a
1714dynamic rule for the flow so that packets belonging to that session
1715will be allowed through the firewall:
1716.Pp
1717.Dl "ipfw add check-state"
1718.Dl "ipfw add allow tcp from my-subnet to any setup keep-state"
1719.Dl "ipfw add deny tcp from any to any"
1720.Pp
1721A similar approach can be used for UDP, where an UDP packet coming
1722from the inside will install a dynamic rule to let the response through
1723the firewall:
1724.Pp
1725.Dl "ipfw add check-state"
1726.Dl "ipfw add allow udp from my-subnet to any keep-state"
1727.Dl "ipfw add deny udp from any to any"
1728.Pp
1729Dynamic rules expire after some time, which depends on the status
1730of the flow and the setting of some
1731.Cm sysctl
1732variables.
1733See Section
1734.Sx SYSCTL VARIABLES
1735for more details.
1736For TCP sessions, dynamic rules can be instructed to periodically
1737send keepalive packets to refresh the state of the rule when it is
1738about to expire.
1739.Pp
1740See Section
1741.Sx EXAMPLES
1742for more examples on how to use dynamic rules.
1743.Sh TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
1744.Nm
1745is also the user interface for the
1746.Xr dummynet 4
1747traffic shaper.
1748.Pp
1749.Nm dummynet
1750operates by first using the firewall to classify packets and divide them into
1751.Em flows ,
1752using any match pattern that can be used in
1753.Nm
1754rules.
1755Depending on local policies, a flow can contain packets for a single
1756TCP connection, or from/to a given host, or entire subnet, or a
1757protocol type, etc.
1758.Pp
1759Packets belonging to the same flow are then passed to either of two
1760different objects, which implement the traffic regulation:
1761.Bl -hang -offset XXXX
1762.It Em pipe
1763A pipe emulates a link with given bandwidth, propagation delay,
1764queue size and packet loss rate.
1765Packets are queued in front of the pipe as they come out from the classifier,
1766and then transferred to the pipe according to the pipe's parameters.
1767.Pp
1768.It Em queue
1769A queue
1770is an abstraction used to implement the WF2Q+
1771(Worst-case Fair Weighted Fair Queueing) policy, which is
1772an efficient variant of the WFQ policy.
1773.br
1774The queue associates a
1775.Em weight
1776and a reference pipe to each flow, and then all backlogged (i.e.,
1777with packets queued) flows linked to the same pipe share the pipe's
1778bandwidth proportionally to their weights.
1779Note that weights are not priorities; a flow with a lower weight
1780is still guaranteed to get its fraction of the bandwidth even if a
1781flow with a higher weight is permanently backlogged.
1782.Pp
1783.El
1784In practice,
1785.Em pipes
1786can be used to set hard limits to the bandwidth that a flow can use, whereas
1787.Em queues
1788can be used to determine how different flow share the available bandwidth.
1789.Pp
1790The
1791.Em pipe
1792and
1793.Em queue
1794configuration commands are the following:
1795.Bd -ragged -offset indent
1796.Cm pipe Ar number Cm config Ar pipe-configuration
1797.Pp
1798.Cm queue Ar number Cm config Ar queue-configuration
1799.Ed
1800.Pp
1801The following parameters can be configured for a pipe:
1802.Pp
1803.Bl -tag -width indent -compact
1804.It Cm bw Ar bandwidth | device
1805Bandwidth, measured in
1806.Sm off
1807.Op Cm K | M
1808.Brq Cm bit/s | Byte/s .
1809.Sm on
1810.Pp
1811A value of 0 (default) means unlimited bandwidth.
1812The unit must immediately follow the number, as in
1813.Pp
1814.Dl "ipfw pipe 1 config bw 300Kbit/s"
1815.Pp
1816If a device name is specified instead of a numeric value, as in
1817.Pp
1818.Dl "ipfw pipe 1 config bw tun0"
1819.Pp
1820then the transmit clock is supplied by the specified device.
1821At the moment only the
1822.Xr tun 4
1823device supports this
1824functionality, for use in conjunction with
1825.Xr ppp 8 .
1826.Pp
1827.It Cm delay Ar ms-delay
1828Propagation delay, measured in milliseconds.
1829The value is rounded to the next multiple of the clock tick
1830(typically 10ms, but it is a good practice to run kernels
1831with
1832.Dq "options HZ=1000"
1833to reduce
1834the granularity to 1ms or less).
1835Default value is 0, meaning no delay.
1836.El
1837.Pp
1838The following parameters can be configured for a queue:
1839.Pp
1840.Bl -tag -width indent -compact
1841.It Cm pipe Ar pipe_nr
1842Connects a queue to the specified pipe.
1843Multiple queues (with the same or different weights) can be connected to
1844the same pipe, which specifies the aggregate rate for the set of queues.
1845.Pp
1846.It Cm weight Ar weight
1847Specifies the weight to be used for flows matching this queue.
1848The weight must be in the range 1..100, and defaults to 1.
1849.El
1850.Pp
1851Finally, the following parameters can be configured for both
1852pipes and queues:
1853.Pp
1854.Bl -tag -width XXXX -compact
1855.Pp
1856.It Cm buckets Ar hash-table-size
1857Specifies the size of the hash table used for storing the
1858various queues.
1859Default value is 64 controlled by the
1860.Xr sysctl 8
1861variable
1862.Em net.inet.ip.dummynet.hash_size ,
1863allowed range is 16 to 65536.
1864.Pp
1865.It Cm mask Ar mask-specifier
1866Packets sent to a given pipe or queue by an
1867.Nm
1868rule can be further classified into multiple flows, each of which is then
1869sent to a different
1870.Em dynamic
1871pipe or queue.
1872A flow identifier is constructed by masking the IP addresses,
1873ports and protocol types as specified with the
1874.Cm mask
1875options in the configuration of the pipe or queue.
1876For each different flow identifier, a new pipe or queue is created
1877with the same parameters as the original object, and matching packets
1878are sent to it.
1879.Pp
1880Thus, when
1881.Em dynamic pipes
1882are used, each flow will get the same bandwidth as defined by the pipe,
1883whereas when
1884.Em dynamic queues
1885are used, each flow will share the parent's pipe bandwidth evenly
1886with other flows generated by the same queue (note that other queues
1887with different weights might be connected to the same pipe).
1888.br
1889Available mask specifiers are a combination of one or more of the following:
1890.Pp
1891.Cm dst-ip Ar mask ,
1892.Cm dst-ip6 Ar mask ,
1893.Cm src-ip Ar mask ,
1894.Cm src-ip6 Ar mask ,
1895.Cm dst-port Ar mask ,
1896.Cm src-port Ar mask ,
1897.Cm flow-id Ar mask ,
1898.Cm proto Ar mask
1899or
1900.Cm all ,
1901.Pp
1902where the latter means all bits in all fields are significant.
1903.Pp
1904.It Cm noerror
1905When a packet is dropped by a dummynet queue or pipe, the error
1906is normally reported to the caller routine in the kernel, in the
1907same way as it happens when a device queue fills up.
1908Setting this
1909option reports the packet as successfully delivered, which can be
1910needed for some experimental setups where you want to simulate
1911loss or congestion at a remote router.
1912.Pp
1913.It Cm plr Ar packet-loss-rate
1914Packet loss rate.
1915Argument
1916.Ar packet-loss-rate
1917is a floating-point number between 0 and 1, with 0 meaning no
1918loss, 1 meaning 100% loss.
1919The loss rate is internally represented on 31 bits.
1920.Pp
1921.It Cm queue Brq Ar slots | size Ns Cm Kbytes
1922Queue size, in
1923.Ar slots
1924or
1925.Cm KBytes .
1926Default value is 50 slots, which
1927is the typical queue size for Ethernet devices.
1928Note that for slow speed links you should keep the queue
1929size short or your traffic might be affected by a significant
1930queueing delay.
1931E.g., 50 max-sized ethernet packets (1500 bytes) mean 600Kbit
1932or 20s of queue on a 30Kbit/s pipe.
1933Even worse effects can result if you get packets from an
1934interface with a much larger MTU, e.g.\& the loopback interface
1935with its 16KB packets.
1936.Pp
1937.It Cm red | gred Ar w_q Ns / Ns Ar min_th Ns / Ns Ar max_th Ns / Ns Ar max_p
1938Make use of the RED (Random Early Detection) queue management algorithm.
1939.Ar w_q
1940and
1941.Ar max_p
1942are floating
1943point numbers between 0 and 1 (0 not included), while
1944.Ar min_th
1945and
1946.Ar max_th
1947are integer numbers specifying thresholds for queue management
1948(thresholds are computed in bytes if the queue has been defined
1949in bytes, in slots otherwise).
1950The
1951.Xr dummynet 4
1952also supports the gentle RED variant (gred).
1953Three
1954.Xr sysctl 8
1955variables can be used to control the RED behaviour:
1956.Bl -tag -width indent
1957.It Em net.inet.ip.dummynet.red_lookup_depth
1958specifies the accuracy in computing the average queue
1959when the link is idle (defaults to 256, must be greater than zero)
1960.It Em net.inet.ip.dummynet.red_avg_pkt_size
1961specifies the expected average packet size (defaults to 512, must be
1962greater than zero)
1963.It Em net.inet.ip.dummynet.red_max_pkt_size
1964specifies the expected maximum packet size, only used when queue
1965thresholds are in bytes (defaults to 1500, must be greater than zero).
1966.El
1967.El
1968.Pp
1969When used with IPv6 data, dummynet currently has several limitations.
1970First, debug.mpsafenet=0 must be set.
1971Second, the information necessicary to route link-local packets to an
1972interface is not avalable after processing by dummynet so those packets
1973are dropped in the output path.
1974Care should be taken to insure that link-local packets are not passed to
1975dummynet.
1976.Sh CHECKLIST
1977Here are some important points to consider when designing your
1978rules:
1979.Bl -bullet
1980.It
1981Remember that you filter both packets going
1982.Cm in
1983and
1984.Cm out .
1985Most connections need packets going in both directions.
1986.It
1987Remember to test very carefully.
1988It is a good idea to be near the console when doing this.
1989If you cannot be near the console,
1990use an auto-recovery script such as the one in
1991.Pa /usr/share/examples/ipfw/change_rules.sh .
1992.It
1993Do not forget the loopback interface.
1994.El
1995.Sh FINE POINTS
1996.Bl -bullet
1997.It
1998There are circumstances where fragmented datagrams are unconditionally
1999dropped.
2000TCP packets are dropped if they do not contain at least 20 bytes of
2001TCP header, UDP packets are dropped if they do not contain a full 8
2002byte UDP header, and ICMP packets are dropped if they do not contain
20034 bytes of ICMP header, enough to specify the ICMP type, code, and
2004checksum.
2005These packets are simply logged as
2006.Dq pullup failed
2007since there may not be enough good data in the packet to produce a
2008meaningful log entry.
2009.It
2010Another type of packet is unconditionally dropped, a TCP packet with a
2011fragment offset of one.
2012This is a valid packet, but it only has one use, to try
2013to circumvent firewalls.
2014When logging is enabled, these packets are
2015reported as being dropped by rule -1.
2016.It
2017If you are logged in over a network, loading the
2018.Xr kld 4
2019version of
2020.Nm
2021is probably not as straightforward as you would think.
2022I recommend the following command line:
2023.Bd -literal -offset indent
2024kldload ipfw && \e
2025ipfw add 32000 allow ip from any to any
2026.Ed
2027.Pp
2028Along the same lines, doing an
2029.Bd -literal -offset indent
2030ipfw flush
2031.Ed
2032.Pp
2033in similar surroundings is also a bad idea.
2034.It
2035The
2036.Nm
2037filter list may not be modified if the system security level
2038is set to 3 or higher
2039(see
2040.Xr init 8
2041for information on system security levels).
2042.El
2043.Sh PACKET DIVERSION
2044A
2045.Xr divert 4
2046socket bound to the specified port will receive all packets
2047diverted to that port.
2048If no socket is bound to the destination port, or if the divert module is
2049not loaded, or if the kernel was not compiled with divert socket support,
2050the packets are dropped.
2051.Sh NETWORK ADDRESS TRANSLATION (NAT)
2052The nat configuration command is the following:
2053.Bd -ragged -offset indent
2054.Bk -words
2055.Cm nat
2056.Ar nat_number
2057.Cm config
2058.Ar nat-configuration
2059.Ek
2060.Ed
2061.Pp
2062.
2063The following parameters can be configured:
2064.Bl -tag -width indent
2065.It Cm ip Ar ip_address
2066Define an ip address to use for aliasing.
2067.It Cm if Ar nic
2068Use ip addres of NIC for aliasing, dynamically changing
2069it if NIC's ip address change.
2070.It Cm log
2071Enable logging on this nat instance.
2072.It Cm deny_in
2073Deny any incoming connection from outside world.
2074.It Cm same_ports
2075Try to leave the alias port numbers unchanged from
2076the actual local port numbers.
2077.It Cm unreg_only
2078Traffic on the local network not originating from an
2079unregistered address spaces will be ignored.
2080.It Cm reset
2081Reset table of the packet aliasing engine on address change.
2082.It Cm reverse
2083Reverse the way libalias handles aliasing.
2084.It Cm proxy_only
2085Obey transparent proxy rules only, packet aliasing is not performed.
2086.El
2087.Pp
2088To let the packet continue after being (de)aliased, set the sysctl variable
2089.Em net.inet.ip.fw.one_pass
2090to 0.
2091For more information about aliasing modes, refer to
2092.Xr libalias 3
2093.
2094See Section
2095.Sx EXAMPLES
2096for some examples about nat usage.
2097.Sh REDIRECT AND LSNAT SUPPORT IN IPFW
2098Redirect and LSNAT support follow closely the syntax used in
2099.Xr natd 8
2100.
2101See Section
2102.Sx EXAMPLES
2103for some examples on how to do redirect and lsnat.
2104.Sh SYSCTL VARIABLES
2105A set of
2106.Xr sysctl 8
2107variables controls the behaviour of the firewall and
2108associated modules
2109.Pq Nm dummynet , bridge .
2110These are shown below together with their default value
2111(but always check with the
2112.Xr sysctl 8
2113command what value is actually in use) and meaning:
2114.Bl -tag -width indent
2115.It Em net.inet.ip.dummynet.expire : No 1
2116Lazily delete dynamic pipes/queue once they have no pending traffic.
2117You can disable this by setting the variable to 0, in which case
2118the pipes/queues will only be deleted when the threshold is reached.
2119.It Em net.inet.ip.dummynet.hash_size : No 64
2120Default size of the hash table used for dynamic pipes/queues.
2121This value is used when no
2122.Cm buckets
2123option is specified when configuring a pipe/queue.
2124.It Em net.inet.ip.dummynet.max_chain_len : No 16
2125Target value for the maximum number of pipes/queues in a hash bucket.
2126The product
2127.Cm max_chain_len*hash_size
2128is used to determine the threshold over which empty pipes/queues
2129will be expired even when
2130.Cm net.inet.ip.dummynet.expire=0 .
2131.It Em net.inet.ip.dummynet.red_lookup_depth : No 256
2132.It Em net.inet.ip.dummynet.red_avg_pkt_size : No 512
2133.It Em net.inet.ip.dummynet.red_max_pkt_size : No 1500
2134Parameters used in the computations of the drop probability
2135for the RED algorithm.
2136.It Em net.inet.ip.fw.autoinc_step : No 100
2137Delta between rule numbers when auto-generating them.
2138The value must be in the range 1..1000.
2139.It Em net.inet.ip.fw.curr_dyn_buckets : Em net.inet.ip.fw.dyn_buckets
2140The current number of buckets in the hash table for dynamic rules
2141(readonly).
2142.It Em net.inet.ip.fw.debug : No 1
2143Controls debugging messages produced by
2144.Nm .
2145.It Em net.inet.ip.fw.dyn_buckets : No 256
2146The number of buckets in the hash table for dynamic rules.
2147Must be a power of 2, up to 65536.
2148It only takes effect when all dynamic rules have expired, so you
2149are advised to use a
2150.Cm flush
2151command to make sure that the hash table is resized.
2152.It Em net.inet.ip.fw.dyn_count : No 3
2153Current number of dynamic rules
2154(read-only).
2155.It Em net.inet.ip.fw.dyn_keepalive : No 1
2156Enables generation of keepalive packets for
2157.Cm keep-state
2158rules on TCP sessions.
2159A keepalive is generated to both
2160sides of the connection every 5 seconds for the last 20
2161seconds of the lifetime of the rule.
2162.It Em net.inet.ip.fw.dyn_max : No 8192
2163Maximum number of dynamic rules.
2164When you hit this limit, no more dynamic rules can be
2165installed until old ones expire.
2166.It Em net.inet.ip.fw.dyn_ack_lifetime : No 300
2167.It Em net.inet.ip.fw.dyn_syn_lifetime : No 20
2168.It Em net.inet.ip.fw.dyn_fin_lifetime : No 1
2169.It Em net.inet.ip.fw.dyn_rst_lifetime : No 1
2170.It Em net.inet.ip.fw.dyn_udp_lifetime : No 5
2171.It Em net.inet.ip.fw.dyn_short_lifetime : No 30
2172These variables control the lifetime, in seconds, of dynamic
2173rules.
2174Upon the initial SYN exchange the lifetime is kept short,
2175then increased after both SYN have been seen, then decreased
2176again during the final FIN exchange or when a RST is received.
2177Both
2178.Em dyn_fin_lifetime
2179and
2180.Em dyn_rst_lifetime
2181must be strictly lower than 5 seconds, the period of
2182repetition of keepalives.
2183The firewall enforces that.
2184.It Em net.inet.ip.fw.enable : No 1
2185Enables the firewall.
2186Setting this variable to 0 lets you run your machine without
2187firewall even if compiled in.
2188.It Em net.inet6.ip6.fw.enable : No 1
2189provides the same functionality as above for the IPv6 case.
2190.It Em net.inet.ip.fw.one_pass : No 1
2191When set, the packet exiting from the
2192.Xr dummynet 4
2193pipe or from
2194.Xr ng_ipfw 4
2195node is not passed though the firewall again.
2196Otherwise, after an action, the packet is
2197reinjected into the firewall at the next rule.
2198.It Em net.inet.ip.fw.verbose : No 1
2199Enables verbose messages.
2200.It Em net.inet.ip.fw.verbose_limit : No 0
2201Limits the number of messages produced by a verbose firewall.
2202.It Em net.inet6.ip6.fw.deny_unknown_exthdrs : No 1
2203If enabled packets with unknown IPv6 Extension Headers will be denied.
2204.It Em net.link.ether.ipfw : No 0
2205Controls whether layer-2 packets are passed to
2206.Nm .
2207Default is no.
2208.It Em net.link.bridge.ipfw : No 0
2209Controls whether bridged packets are passed to
2210.Nm .
2211Default is no.
2212.El
2213.Pp
2214.Sh EXAMPLES
2215There are far too many possible uses of
2216.Nm
2217so this Section will only give a small set of examples.
2218.Pp
2219.Ss BASIC PACKET FILTERING
2220This command adds an entry which denies all tcp packets from
2221.Em cracker.evil.org
2222to the telnet port of
2223.Em wolf.tambov.su
2224from being forwarded by the host:
2225.Pp
2226.Dl "ipfw add deny tcp from cracker.evil.org to wolf.tambov.su telnet"
2227.Pp
2228This one disallows any connection from the entire cracker's
2229network to my host:
2230.Pp
2231.Dl "ipfw add deny ip from 123.45.67.0/24 to my.host.org"
2232.Pp
2233A first and efficient way to limit access (not using dynamic rules)
2234is the use of the following rules:
2235.Pp
2236.Dl "ipfw add allow tcp from any to any established"
2237.Dl "ipfw add allow tcp from net1 portlist1 to net2 portlist2 setup"
2238.Dl "ipfw add allow tcp from net3 portlist3 to net3 portlist3 setup"
2239.Dl "..."
2240.Dl "ipfw add deny tcp from any to any"
2241.Pp
2242The first rule will be a quick match for normal TCP packets,
2243but it will not match the initial SYN packet, which will be
2244matched by the
2245.Cm setup
2246rules only for selected source/destination pairs.
2247All other SYN packets will be rejected by the final
2248.Cm deny
2249rule.
2250.Pp
2251If you administer one or more subnets, you can take advantage
2252of the address sets and or-blocks and write extremely
2253compact rulesets which selectively enable services to blocks
2254of clients, as below:
2255.Pp
2256.Dl "goodguys=\*q{ 10.1.2.0/24{20,35,66,18} or 10.2.3.0/28{6,3,11} }\*q"
2257.Dl "badguys=\*q10.1.2.0/24{8,38,60}\*q"
2258.Dl ""
2259.Dl "ipfw add allow ip from ${goodguys} to any"
2260.Dl "ipfw add deny ip from ${badguys} to any"
2261.Dl "... normal policies ..."
2262.Pp
2263The
2264.Cm verrevpath
2265option could be used to do automated anti-spoofing by adding the
2266following to the top of a ruleset:
2267.Pp
2268.Dl "ipfw add deny ip from any to any not verrevpath in"
2269.Pp
2270This rule drops all incoming packets that appear to be coming to the
2271system on the wrong interface.
2272For example, a packet with a source
2273address belonging to a host on a protected internal network would be
2274dropped if it tried to enter the system from an external interface.
2275.Pp
2276The
2277.Cm antispoof
2278option could be used to do similar but more restricted anti-spoofing
2279by adding the following to the top of a ruleset:
2280.Pp
2281.Dl "ipfw add deny ip from any to any not antispoof in"
2282.Pp
2283This rule drops all incoming packets that appear to be coming from another
2284directly connected system but on the wrong interface.
2285For example, a packet with a source address of
2286.Li 192.168.0.0/24
2287, configured on
2288.Li fxp0
2289, but coming in on
2290.Li fxp1
2291would be dropped.
2292.Ss DYNAMIC RULES
2293In order to protect a site from flood attacks involving fake
2294TCP packets, it is safer to use dynamic rules:
2295.Pp
2296.Dl "ipfw add check-state"
2297.Dl "ipfw add deny tcp from any to any established"
2298.Dl "ipfw add allow tcp from my-net to any setup keep-state"
2299.Pp
2300This will let the firewall install dynamic rules only for
2301those connection which start with a regular SYN packet coming
2302from the inside of our network.
2303Dynamic rules are checked when encountering the first
2304.Cm check-state
2305or
2306.Cm keep-state
2307rule.
2308A
2309.Cm check-state
2310rule should usually be placed near the beginning of the
2311ruleset to minimize the amount of work scanning the ruleset.
2312Your mileage may vary.
2313.Pp
2314To limit the number of connections a user can open
2315you can use the following type of rules:
2316.Pp
2317.Dl "ipfw add allow tcp from my-net/24 to any setup limit src-addr 10"
2318.Dl "ipfw add allow tcp from any to me setup limit src-addr 4"
2319.Pp
2320The former (assuming it runs on a gateway) will allow each host
2321on a /24 network to open at most 10 TCP connections.
2322The latter can be placed on a server to make sure that a single
2323client does not use more than 4 simultaneous connections.
2324.Pp
2325.Em BEWARE :
2326stateful rules can be subject to denial-of-service attacks
2327by a SYN-flood which opens a huge number of dynamic rules.
2328The effects of such attacks can be partially limited by
2329acting on a set of
2330.Xr sysctl 8
2331variables which control the operation of the firewall.
2332.Pp
2333Here is a good usage of the
2334.Cm list
2335command to see accounting records and timestamp information:
2336.Pp
2337.Dl ipfw -at list
2338.Pp
2339or in short form without timestamps:
2340.Pp
2341.Dl ipfw -a list
2342.Pp
2343which is equivalent to:
2344.Pp
2345.Dl ipfw show
2346.Pp
2347Next rule diverts all incoming packets from 192.168.2.0/24
2348to divert port 5000:
2349.Pp
2350.Dl ipfw divert 5000 ip from 192.168.2.0/24 to any in
2351.Pp
2352.Ss TRAFFIC SHAPING
2353The following rules show some of the applications of
2354.Nm
2355and
2356.Xr dummynet 4
2357for simulations and the like.
2358.Pp
2359This rule drops random incoming packets with a probability
2360of 5%:
2361.Pp
2362.Dl "ipfw add prob 0.05 deny ip from any to any in"
2363.Pp
2364A similar effect can be achieved making use of dummynet pipes:
2365.Pp
2366.Dl "ipfw add pipe 10 ip from any to any"
2367.Dl "ipfw pipe 10 config plr 0.05"
2368.Pp
2369We can use pipes to artificially limit bandwidth, e.g.\& on a
2370machine acting as a router, if we want to limit traffic from
2371local clients on 192.168.2.0/24 we do:
2372.Pp
2373.Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
2374.Dl "ipfw pipe 1 config bw 300Kbit/s queue 50KBytes"
2375.Pp
2376note that we use the
2377.Cm out
2378modifier so that the rule is not used twice.
2379Remember in fact that
2380.Nm
2381rules are checked both on incoming and outgoing packets.
2382.Pp
2383Should we want to simulate a bidirectional link with bandwidth
2384limitations, the correct way is the following:
2385.Pp
2386.Dl "ipfw add pipe 1 ip from any to any out"
2387.Dl "ipfw add pipe 2 ip from any to any in"
2388.Dl "ipfw pipe 1 config bw 64Kbit/s queue 10Kbytes"
2389.Dl "ipfw pipe 2 config bw 64Kbit/s queue 10Kbytes"
2390.Pp
2391The above can be very useful, e.g.\& if you want to see how
2392your fancy Web page will look for a residential user who
2393is connected only through a slow link.
2394You should not use only one pipe for both directions, unless
2395you want to simulate a half-duplex medium (e.g.\& AppleTalk,
2396Ethernet, IRDA).
2397It is not necessary that both pipes have the same configuration,
2398so we can also simulate asymmetric links.
2399.Pp
2400Should we want to verify network performance with the RED queue
2401management algorithm:
2402.Pp
2403.Dl "ipfw add pipe 1 ip from any to any"
2404.Dl "ipfw pipe 1 config bw 500Kbit/s queue 100 red 0.002/30/80/0.1"
2405.Pp
2406Another typical application of the traffic shaper is to
2407introduce some delay in the communication.
2408This can significantly affect applications which do a lot of Remote
2409Procedure Calls, and where the round-trip-time of the
2410connection often becomes a limiting factor much more than
2411bandwidth:
2412.Pp
2413.Dl "ipfw add pipe 1 ip from any to any out"
2414.Dl "ipfw add pipe 2 ip from any to any in"
2415.Dl "ipfw pipe 1 config delay 250ms bw 1Mbit/s"
2416.Dl "ipfw pipe 2 config delay 250ms bw 1Mbit/s"
2417.Pp
2418Per-flow queueing can be useful for a variety of purposes.
2419A very simple one is counting traffic:
2420.Pp
2421.Dl "ipfw add pipe 1 tcp from any to any"
2422.Dl "ipfw add pipe 1 udp from any to any"
2423.Dl "ipfw add pipe 1 ip from any to any"
2424.Dl "ipfw pipe 1 config mask all"
2425.Pp
2426The above set of rules will create queues (and collect
2427statistics) for all traffic.
2428Because the pipes have no limitations, the only effect is
2429collecting statistics.
2430Note that we need 3 rules, not just the last one, because
2431when
2432.Nm
2433tries to match IP packets it will not consider ports, so we
2434would not see connections on separate ports as different
2435ones.
2436.Pp
2437A more sophisticated example is limiting the outbound traffic
2438on a net with per-host limits, rather than per-network limits:
2439.Pp
2440.Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
2441.Dl "ipfw add pipe 2 ip from any to 192.168.2.0/24 in"
2442.Dl "ipfw pipe 1 config mask src-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
2443.Dl "ipfw pipe 2 config mask dst-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
2444.Ss LOOKUP TABLES
2445In the following example, we need to create several traffic bandwidth
2446classes and we need different hosts/networks to fall into different classes.
2447We create one pipe for each class and configure them accordingly.
2448Then we create a single table and fill it with IP subnets and addresses.
2449For each subnet/host we set the argument equal to the number of the pipe
2450that it should use.
2451Then we classify traffic using a single rule:
2452.Pp
2453.Dl "ipfw pipe 1 config bw 1000Kbyte/s"
2454.Dl "ipfw pipe 4 config bw 4000Kbyte/s"
2455.Dl "..."
2456.Dl "ipfw table 1 add 192.168.2.0/24 1"
2457.Dl "ipfw table 1 add 192.168.0.0/27 4"
2458.Dl "ipfw table 1 add 192.168.0.2 1"
2459.Dl "..."
2460.Dl "ipfw pipe tablearg ip from table(1) to any"
2461.Pp
2462Using the
2463.Cm fwd
2464action, the table entries may include hostnames and IP addresses.
2465.Pp
2466.Dl "ipfw table 1 add 192.168.2.0/24 10.23.2.1"
2467.Dl "ipfw table 1 add 192.168.0.0/27 router1.dmz"
2468.Dl "..."
2469.Dl "ipfw add 100 fwd tablearg ip from any to table(1)"
2470.Ss SETS OF RULES
2471To add a set of rules atomically, e.g.\& set 18:
2472.Pp
2473.Dl "ipfw set disable 18"
2474.Dl "ipfw add NN set 18 ...         # repeat as needed"
2475.Dl "ipfw set enable 18"
2476.Pp
2477To delete a set of rules atomically the command is simply:
2478.Pp
2479.Dl "ipfw delete set 18"
2480.Pp
2481To test a ruleset and disable it and regain control if something goes wrong:
2482.Pp
2483.Dl "ipfw set disable 18"
2484.Dl "ipfw add NN set 18 ...         # repeat as needed"
2485.Dl "ipfw set enable 18; echo done; sleep 30 && ipfw set disable 18"
2486.Pp
2487Here if everything goes well, you press control-C before the "sleep"
2488terminates, and your ruleset will be left active.
2489Otherwise, e.g.\& if
2490you cannot access your box, the ruleset will be disabled after
2491the sleep terminates thus restoring the previous situation.
2492.Ss NAT, REDIRECT AND LSNAT
2493First redirect all the traffic to nat instance 123:
2494.Pp
2495.Dl "ipfw add nat 123 all from any to any"
2496.Pp
2497Then to configure nat instance 123 to alias all the outgoing traffic with ip
2498192.168.0.123, blocking all incoming connections, trying to keep
2499same ports on both sides, clearing aliasing table on address change
2500and keeping a log of traffic/link statistics:
2501.Pp
2502.Dl "ipfw nat 123 config ip 192.168.0.123 log deny_in reset same_ports"
2503.Pp
2504Or to change address of instance 123, aliasing table will be cleared (see
2505reset option):
2506.Pp
2507.Dl "ipfw nat 123 config ip 10.0.0.1"
2508.Pp
2509To see configuration of nat instance 123:
2510.Pp
2511.Dl "ipfw nat 123 show config"
2512.Pp
2513To show logs of all the instances in range 111-999:
2514.Pp
2515.Dl "ipfw nat 111-999 show"
2516.Pp
2517To see configurations of all instances:
2518.Pp
2519.Dl "ipfw nat show config"
2520.Pp
2521Or a redirect rule with mixed modes could looks like:
2522.Pp
2523.Dl "ipfw nat 123 config redirect_addr 10.0.0.1 10.0.0.66"
2524.Dl "			 redirect_port tcp 192.168.0.1:80 500"
2525.Dl "			 redirect_proto udp 192.168.1.43 192.168.1.1"
2526.Dl "			 redirect_addr 192.168.0.10,192.168.0.11"
2527.Dl "			 	    10.0.0.100	# LSNAT"
2528.Dl "			 redirect_port tcp 192.168.0.1:80,192.168.0.10:22"
2529.Dl "			 	    500		# LSNAT"
2530.Pp
2531or it could be splitted in:
2532.Pp
2533.Dl "ipfw nat 1 config redirect_addr 10.0.0.1 10.0.0.66"
2534.Dl "ipfw nat 2 config redirect_port tcp 192.168.0.1:80 500"
2535.Dl "ipfw nat 3 config redirect_proto udp 192.168.1.43 192.168.1.1"
2536.Dl "ipfw nat 4 config redirect_addr 192.168.0.10,192.168.0.11,192.168.0.12"
2537.Dl "				         10.0.0.100"
2538.Dl "ipfw nat 5 config redirect_port tcp"
2539.Dl "			192.168.0.1:80,192.168.0.10:22,192.168.0.20:25 500"
2540.Pp
2541.Sh SEE ALSO
2542.Xr cpp 1 ,
2543.Xr m4 1 ,
2544.Xr altq 4 ,
2545.Xr divert 4 ,
2546.Xr dummynet 4 ,
2547.Xr if_bridge 4 ,
2548.Xr ip 4 ,
2549.Xr ipfirewall 4 ,
2550.Xr ng_ipfw 4 ,
2551.Xr protocols 5 ,
2552.Xr services 5 ,
2553.Xr init 8 ,
2554.Xr kldload 8 ,
2555.Xr reboot 8 ,
2556.Xr sysctl 8 ,
2557.Xr syslogd 8
2558.Sh HISTORY
2559The
2560.Nm
2561utility first appeared in
2562.Fx 2.0 .
2563.Xr dummynet 4
2564was introduced in
2565.Fx 2.2.8 .
2566Stateful extensions were introduced in
2567.Fx 4.0 .
2568.Nm ipfw2
2569was introduced in Summer 2002.
2570.Sh AUTHORS
2571.An Ugen J. S. Antsilevich ,
2572.An Poul-Henning Kamp ,
2573.An Alex Nash ,
2574.An Archie Cobbs ,
2575.An Luigi Rizzo .
2576.Pp
2577.An -nosplit
2578API based upon code written by
2579.An Daniel Boulet
2580for BSDI.
2581.Pp
2582.An -nosplit
2583In-kernel NAT support written by
2584.An Paolo Pisati Aq piso@FreeBSD.org
2585as part of a Summer of Code 2005 project.
2586.Pp
2587Work on
2588.Xr dummynet 4
2589traffic shaper supported by Akamba Corp.
2590.Sh BUGS
2591Use of dummynet with IPv6 requires that debug.mpsafenet be set to 0.
2592.Pp
2593The syntax has grown over the years and sometimes it might be confusing.
2594Unfortunately, backward compatibility prevents cleaning up mistakes
2595made in the definition of the syntax.
2596.Pp
2597.Em !!! WARNING !!!
2598.Pp
2599Misconfiguring the firewall can put your computer in an unusable state,
2600possibly shutting down network services and requiring console access to
2601regain control of it.
2602.Pp
2603Incoming packet fragments diverted by
2604.Cm divert
2605are reassembled before delivery to the socket.
2606The action used on those packet is the one from the
2607rule which matches the first fragment of the packet.
2608.Pp
2609Packets diverted to userland, and then reinserted by a userland process
2610may lose various packet attributes.
2611The packet source interface name
2612will be preserved if it is shorter than 8 bytes and the userland process
2613saves and reuses the sockaddr_in
2614(as does
2615.Xr natd 8 ) ;
2616otherwise, it may be lost.
2617If a packet is reinserted in this manner, later rules may be incorrectly
2618applied, making the order of
2619.Cm divert
2620rules in the rule sequence very important.
2621.Pp
2622Dummynet drops all packets with IPv6 link-local addresses.
2623.Pp
2624Rules using
2625.Cm uid
2626or
2627.Cm gid
2628may not behave as expected.
2629In particular, incoming SYN packets may
2630have no uid or gid associated with them since they do not yet belong
2631to a TCP connection, and the uid/gid associated with a packet may not
2632be as expected if the associated process calls
2633.Xr setuid 2
2634or similar system calls.
2635.Pp
2636Rules which use uid, gid or jail based matching should be used only
2637if debug.mpsafenet=0 to avoid possible deadlocks due to layering
2638violations in its implementation.
2639.Pp
2640Rule syntax is subject to the command line environment and some patterns
2641may need to be escaped with the backslash character
2642or quoted appropriately.
2643.Pp
2644Due to the architecture of
2645.Xr libalias 3 ,
2646ipfw nat is not compatible with the tcp segmentation offloading
2647(TSO). Thus, to reliably nat your network traffic, please disable TSO
2648on your NICs using
2649.Xr ifconfig 8 .
2650