2 .\" $FreeBSD: src/sbin/ipfw/ipfw.8,v 1.63.2.33 2003/02/04 01:36:02 brueffer Exp $
3 .\" $DragonFly: src/sbin/ipfw/ipfw.8,v 1.14 2007/11/02 12:50:20 sephe Exp $
10 .Nd IP firewall and traffic shaper control program
25 .Brq Cm delete | zero | resetlog
30 .Brq Cm firewall | one_pass | debug | verbose | dyn_keepalive
33 .Brq Cm firewall | one_pass | debug | verbose | dyn_keepalive
36 .Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
40 .Ar number Cm to Ar number
42 .Cm set swap Ar number number
54 .Brq Cm delete | list | show
62 .Ar macro Ns Op = Ns Ar value
70 utility is the user interface for controlling the
77 .Bd -ragged -offset XXXX
79 this manual page documents the newer version of
83 CURRENT in July 2002, also known as
86 is a superset of the old firewall,
88 The differences between the two are listed in Section
89 .Sx IPFW2 ENHANCEMENTS ,
90 which you are encouraged to read to revise older rulesets and possibly
91 write them more efficiently.
100 numbered from 1 to 65535.
101 Packets are passed to
103 from a number of different places in the protocol stack
104 (depending on the source and destination of the packet,
107 is invoked multiple times on the same packet).
108 The packet passed to the firewall is compared
109 against each of the rules in the firewall
111 When a match is found, the action corresponding to the
112 matching rule is performed.
114 Depending on the action and certain system settings, packets
115 can be reinjected into the firewall at some rule after the
116 matching one for further processing.
120 ruleset always includes a
122 rule (numbered 65535) which cannot be modified,
123 and matches all packets.
124 The action associated with the
130 depending on how the kernel is configured.
132 If the ruleset includes one or more rules with the
140 behaviour, i.e. upon a match it will create dynamic rules matching
141 the exact parameters (addresses and ports) of the matching packet.
143 These dynamic rules, which have a limited lifetime, are checked
144 at the first occurrence of a
149 rule, and are typically used to open the firewall on-demand to
150 legitimate traffic only.
152 .Sx STATEFUL FIREWALL
155 Sections below for more information on the stateful behaviour of
158 All rules (including dynamic ones) have a few associated counters:
159 a packet count, a byte count, a log count and a timestamp
160 indicating the time of the last match.
161 Counters can be displayed or reset with
165 Rules can be added with the
167 command; deleted individually or in groups with the
169 command, and globally with the
171 command; displayed, optionally with the content of the
177 Finally, counters can be reset with the
183 Also, each rule belongs to one of 32 different
187 commands to atomically manipulate sets, such as enable,
188 disable, swap sets, move all rules in a set to another
189 one, delete all rules in a set. These can be useful to
190 install temporary configurations, or to test them.
193 for more information on
196 The following options are available:
197 .Bl -tag -width indent
199 While listing, show counter values.
202 command just implies this option.
204 When entering or showing rules, print them in compact form,
205 i.e. without the optional "ip from any to any" string
206 when this does not carry any additional information.
208 While listing, show dynamic rules in addition to static ones.
210 While listing, if the
212 option was specified, also show expired dynamic rules.
214 Don't ask for confirmation for commands that can cause problems
217 If there is no tty associated with the process, this is implied.
219 Try to resolve addresses and service names in output.
227 be quiet about actions
230 This is useful for adjusting rules by executing multiple
234 .Ql sh\ /etc/rc.firewall ) ,
235 or by processing a file of many
237 rules across a remote login session.
240 is performed in normal (verbose) mode (with the default kernel
241 configuration), it prints a message.
242 Because all rules are flushed, the message might not be delivered
243 to the login session, causing the remote login session to be closed
244 and the remainder of the ruleset to not be processed.
245 Access to the console would then be required to recover.
247 While listing rules, show the
249 each rule belongs to.
250 If this flag is not specified, disabled rules will not be
253 While listing pipes, sort according to one of the four
254 counters (total or current packets or bytes).
256 While listing, show last match timestamp.
259 To ease configuration, rules can be put into a file which is
262 as shown in the last synopsis line.
266 The file will be read line by line and applied as arguments to the
270 Optionally, a preprocessor can be specified using
274 is to be piped through.
275 Useful preprocessors include
281 doesn't start with a slash
283 as its first character, the usual
285 name search is performed.
286 Care should be taken with this in environments where not all
287 file systems are mounted (yet) by the time
289 is being run (e.g. when they are mounted over NFS).
292 has been specified, optional
296 specifications can follow and will be passed on to the preprocessor.
297 This allows for flexible configuration files (like conditionalizing
298 them on the local hostname) and the use of macros to centralize
299 frequently required arguments like IP addresses.
306 commands are used to configure the traffic shaper, as shown in the
307 .Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
310 If the world and the kernel get out of sync the
312 ABI may break, preventing you from being able to add any rules. This can
313 adversely affect the booting process. You can use
317 to temporarily disable the firewall to regain access to the network,
318 allowing you to fix the problem.
320 A packet is checked against the active ruleset in multiple places
321 in the protocol stack, under control of several sysctl variables.
322 These places and variables are shown below, and it is important to
323 have this picture in mind in order to design a correct ruleset.
324 .Bd -literal -offset indent
327 +----------->-----------+
329 [ip_input] [ip_output] net.inet.ip.fw.enable=1
332 [ether_demux] [ether_output_frame] net.link.ether.ipfw=1
337 As can be noted from the above picture, the number of
338 times the same packet goes through the firewall can
339 vary between 0 and 4 depending on packet source and
340 destination, and system configuration.
342 Note that as packets flow through the stack, headers can be
343 stripped or added to it, and so they may or may not be available
345 E.g., incoming packets will include the MAC header when
349 but the same packets will have the MAC header stripped off when
354 Also note that each packet is always checked against the complete ruleset,
355 irrespective of the place where the check occurs, or the source of the packet.
356 If a rule contains some match patterns or actions which are not valid
357 for the place of invocation (e.g. trying to match a MAC header within
359 the match pattern will not match, but a
361 operator in front of such patterns
365 match on those packets.
366 It is thus the responsibility of
367 the programmer, if necessary, to write a suitable ruleset to
368 differentiate among the possible places.
370 rules can be useful here, as an example:
371 .Bd -literal -offset indent
372 # packets from ether_demux
373 ipfw add 10 skipto 1000 all from any to any layer2 in
374 # packets from ip_input
375 ipfw add 10 skipto 2000 all from any to any not layer2 in
376 # packets from ip_output
377 ipfw add 10 skipto 3000 all from any to any not layer2 out
378 # packets from ether_output_frame
379 ipfw add 10 skipto 4000 all from any to any layer2 out
384 rules is the following:
385 .Bd -ragged -offset indent
387 .Op Cm set Ar set_number
388 .Op Cm prob Ar match_probability
391 .Op Cm log Op Cm logamount Ar number
395 where the body of the rule specifies which information is used
396 for filtering packets, among the following:
398 .Bl -tag -width "Source and dest. addresses and ports" -offset XXX -compact
399 .It Layer-2 header fields
403 .It Source and dest. addresses and ports
407 .It Transmit and receive interface
409 .It Misc. IP header fields
410 Version, type of service, datagram length, identification,
411 fragment flag (non-zero IP offset),
414 .It Misc. TCP header fields
415 TCP flags (SYN, FIN, ACK, RST, etc.),
416 sequence number, acknowledgment number,
422 When the packet can be associated with a local socket.
425 Note that some of the above information, e.g. source MAC or IP addresses and
426 TCP/UDP ports, could easily be spoofed, so filtering on those fields
427 alone might not guarantee the desired results.
428 .Bl -tag -width indent
430 Each rule is associated with a
432 in the range 1..65535, with the latter reserved for the
435 Rules are checked sequentially by rule number.
436 Multiple rules can have the same number, in which case they are
437 checked (and listed) according to the order in which they have
439 If a rule is entered without specifying a number, the kernel will
440 assign one in such a way that the rule becomes the last one
444 Automatic rule numbers are assigned by incrementing the last
445 non-default rule number by the value of the sysctl variable
446 .Ar net.inet.ip.fw.autoinc_step
447 which defaults to 100.
448 If this is not possible (e.g. because we would go beyond the
449 maximum allowed rule number), the number of the last
450 non-default value is used instead.
451 .It Cm set Ar set_number
452 Each rule is associated with a
454 in the range 0..31, with the latter reserved for the
457 Sets can be individually disabled and enabled, so this parameter
458 is of fundamental importance for atomic ruleset manipulation.
459 It can be also used to simplify deletion of groups of rules.
460 If a rule is entered without specifying a set number,
462 .It Cm prob Ar match_probability
463 A match is only declared with the specified probability
464 (floating point number between 0 and 1).
465 This can be useful for a number of applications such as
466 random packet drop or
469 to simulate the effect of multiple paths leading to out-of-order
471 .It Cm log Op Cm logamount Ar number
472 When a packet matches a rule with the
474 keyword, a message will be
480 The logging only occurs if the sysctl variable
481 .Em net.inet.ip.fw.verbose
483 (which is the default when the kernel is compiled with
484 .Dv IPFIREWALL_VERBOSE
485 ) and the number of packets logged so far for that
486 particular rule does not exceed the
491 is specified, the limit is taken from the sysctl variable
492 .Em net.inet.ip.fw.verbose_limit .
493 In both cases, a value of 0 removes the logging limit.
495 Once the limit is reached, logging can be re-enabled by
496 clearing the logging counter or the packet counter for that entry, see the
501 A rule can be associated with one of the following actions, which
502 will be executed when the packet matches the body of the rule.
503 .Bl -tag -width indent
504 .It Cm allow | accept | pass | permit
505 Allow packets that match rule.
506 The search terminates.
508 Checks the packet against the dynamic ruleset.
509 If a match is found, execute the action associated with
510 the rule which generated this dynamic rule, otherwise
511 move to the next rule.
514 rules do not have a body.
517 rule is found, the dynamic ruleset is checked at the first
523 Update counters for all packets that match rule.
524 The search continues with the next rule.
526 Discard packets that match this rule.
527 The search terminates.
528 .It Cm divert Ar port
529 Divert packets that match this rule to the
533 The search terminates.
534 .It Cm fwd | forward Ar ipaddr Ns Op , Ns Ar port
535 Change the next-hop on matching packets to
537 which can be an IP address in dotted quad format or a host name.
538 The search terminates if this rule matches.
542 is a local address, then matching packets will be forwarded to
544 (or the port number in the packet if one is not specified in the rule)
545 on the local machine.
549 is not a local address, then the port number
550 (if specified) is ignored, and the packet will be
551 forwarded to the remote address, using the route as found in
552 the local routing table for that IP.
556 rule will not match layer-2 packets (those received
557 on ether_input or ether_output).
561 action does not change the contents of the packet at all.
562 In particular, the destination address remains unmodified, so
563 packets forwarded to another system will usually be rejected by that system
564 unless there is a matching rule on that system to capture them.
565 For packets forwarded locally,
566 the local address of the socket will be
567 set to the original destination address of the packet.
570 entry look rather weird but is intended for
571 use with transparent proxy servers.
572 .It Cm pipe Ar pipe_nr
576 (for bandwidth limitation, delay, etc.).
578 .Sx TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
579 Section for further information.
580 The search terminates; however, on exit from the pipe and if
584 .Em net.inet.ip.fw.one_pass
585 is not set, the packet is passed again to the firewall code
586 starting from the next rule.
587 .It Cm queue Ar queue_nr
591 (for bandwidth limitation using WF2Q+).
597 Discard packets that match this rule, and if the
598 packet is a TCP packet, try to send a TCP reset (RST) notice.
599 The search terminates.
600 .It Cm skipto Ar number
601 Skip all subsequent rules numbered less than
603 The search continues with the first rule numbered
607 Send a copy of packets matching this rule to the
611 The search terminates and the original packet is accepted
615 .It Cm unreach Ar code
616 Discard packets that match this rule, and try to send an ICMP
617 unreachable notice with code
621 is a number from 0 to 255, or one of these aliases:
622 .Cm net , host , protocol , port ,
623 .Cm needfrag , srcfail , net-unknown , host-unknown ,
624 .Cm isolated , net-prohib , host-prohib , tosnet ,
625 .Cm toshost , filter-prohib , host-precedence
627 .Cm precedence-cutoff .
628 The search terminates.
631 The body of a rule contains zero or more patterns (such as
632 specific source and destination addresses or ports,
633 protocol options, incoming or outgoing interfaces, etc.)
634 that the packet must match in order to be recognised.
635 In general, the patterns are connected by (implicit)
637 operators -- i.e. all must match in order for the
639 Individual patterns can be prefixed by the
641 operator to reverse the result of the match, as in
643 .Dl "ipfw add 100 allow ip from not 1.2.3.4 to any"
645 Additionally, sets of alternative match patterns (
647 ) can be constructed by putting the patterns in
648 lists enclosed between parentheses ( ) or braces { }, and
653 .Dl "ipfw add 100 allow ip from { x or not y or z } to any"
655 Only one level of parentheses is allowed.
656 Beware that most shells have special meanings for parentheses
657 or braces, so it is advisable to put a backslash \\ in front of them
658 to prevent such interpretations.
660 The body of a rule must in general include a source and destination
664 can be used in various places to specify that the content of
665 a required field is irrelevant.
667 The rule body has the following format:
668 .Bd -ragged -offset indent
669 .Op Ar proto Cm from Ar src Cm to Ar dst
673 The first part (protocol from src to dst) is for backward
678 any match pattern (including MAC headers, IPv4 protocols,
679 addresses and ports) can be specified in the
683 Rule fields have the following meaning:
684 .Bl -tag -width indent
685 .It Ar proto : protocol | Cm { Ar protocol Cm or ... }
686 An IPv4 protocol (or an
688 with multiple protocols) specified by number or name
689 (for a complete list see
690 .Pa /etc/protocols ) .
695 keywords mean any protocol will match.
696 .It Ar src No and Ar dst : ip-address | Cm { Ar ip-address Cm or ... } Op Ar ports
701 containing one or more of them,
702 optionally followed by
706 An address (or set of addresses) specified in one of the following
707 ways, optionally preceded by a
710 .Bl -tag -width indent
712 matches any IP address.
714 matches any IP address configured on an interface in the system.
715 The address list is evaluated at the time the packet is
717 .It Ar numeric-ip | hostname
718 Matches a single IPv4 address, specified as dotted-quad or a hostname.
719 Hostnames are resolved at the time the rule is added to the firewall list.
720 .It Ar addr Ns / Ns Ar masklen
721 Matches all addresses with base
723 (specified as a dotted quad or a hostname)
727 As an example, 1.2.3.4/25 will match
728 all IP numbers from 1.2.3.0 to 1.2.3.127 .
729 .It Ar addr Ns / Ns Ar masklen Ns Cm { Ns Ar num,num,... Ns Cm }
730 Matches all addresses with base address
732 (specified as a dotted quad or a hostname)
733 and whose last byte is in the list between braces { } .
734 Note that there must be no spaces between braces, commas and
738 field is used to limit the size of the set of addresses,
739 and can have any value between 24 and 32.
741 As an example, an address specified as 1.2.3.4/24{128,35,55,89}
742 will match the following IP addresses:
744 1.2.3.128 1.2.3.35 1.2.3.55 1.2.3.89 .
746 This format is particularly useful to handle sparse address sets
747 within a single rule. Because the matching occurs using a
748 bitmask, it takes constant time and dramatically reduces
749 the complexity of rulesets.
750 .It Ar addr Ns : Ns Ar mask
751 Matches all addresses with base
753 (specified as a dotted quad or a hostname)
756 specified as a dotted quad.
757 As an example, 1.2.3.4/255.0.255.0 will match
759 We suggest to use this form only for non-contiguous
760 masks, and resort to the
761 .Ar addr Ns / Ns Ar masklen
762 format for contiguous masks, which is more compact and less
765 .It Ar ports : Oo Cm not Oc Bro Ar port | port Ns \&- Ns Ar port Ns Brc Op , Ns Ar ...
766 For protocols which support port numbers (such as TCP and UDP), optional
768 may be specified as one or more ports or port ranges, separated
769 by commas but no spaces, and an optional
774 notation specifies a range of ports (including boundaries).
778 may be used instead of numeric port values.
779 The length of the port list is limited to 30 ports or ranges,
780 though one can specify larger ranges by using an
788 can be used to escape the dash
790 character in a service name (from a shell, the backslash must be
791 typed twice to avoid the shell itself interpreting it as an escape
794 .Dl "ipfw add count tcp from any ftp\e\e-data-ftp to any"
796 Fragmented packets which have a non-zero offset (i.e. not the first
797 fragment) will never match a rule which has one or more port
801 option for details on matching fragmented packets.
803 .Ss RULE OPTIONS (MATCH PATTERNS)
804 Additional match patterns can be used within
805 rules. Zero or more of these so-called
807 can be present in a rule, optionally prefixed by the
809 operand, and possibly grouped into
812 The following match patterns can be used (listed in alphabetical order):
813 .Bl -tag -width indent
814 .It Cm dst-ip Ar ip address
815 Matches IP packets whose destination IP is one of the address(es)
816 specified as argument.
817 .It Cm dst-port Ar source ports
818 Matches IP packets whose destination port is one of the port(s)
819 specified as argument.
821 Matches TCP packets that have the RST or ACK bits set.
823 Matches packets that are fragments and not the first
824 fragment of an IP datagram. Note that these packets will not have
825 the next protocol header (e.g. TCP, UDP) so options that look into
826 these headers cannot match.
828 Matches all TCP or UDP packets sent by or received for a
832 may be specified by name or number.
833 .It Cm icmptypes Ar types
834 Matches ICMP packets whose ICMP type is in the list
836 The list may be specified as any combination of ranges or
837 individual types separated by commas.
838 The supported ICMP types are:
842 destination unreachable
854 time-to-live exceeded
868 and address mask reply
871 Matches incoming or outgoing packets, respectively.
875 are mutually exclusive (in fact,
880 Matches IP packets whose
885 Matches IP packets whose total length, including header and data, is
888 .It Cm ipoptions Ar spec
889 Matches packets whose IP header contains the comma separated list of
892 The supported IP options are:
895 (strict source route),
897 (loose source route),
899 (record packet route) and
902 The absence of a particular option may be denoted
905 .It Cm ipprecedence Ar precedence
906 Matches IP packets whose precedence field is equal to
909 Matches IP packets whose
911 field contains the comma separated list of
912 service types specified in
914 The supported IP types of service are:
917 .Pq Dv IPTOS_LOWDELAY ,
919 .Pq Dv IPTOS_THROUGHPUT ,
921 .Pq Dv IPTOS_RELIABILITY ,
923 .Pq Dv IPTOS_MINCOST ,
926 The absence of a particular type may be denoted
930 Matches IP packets whose time to live is
932 .It Cm ipversion Ar ver
933 Matches IP packets whose IP version field is
936 Upon a match, the firewall will create a dynamic rule, whose
937 default behaviour is to match bidirectional traffic between
938 source and destination IP/port using the same protocol.
939 The rule has a limited lifetime (controlled by a set of
941 variables), and the lifetime is refreshed every time a matching
944 Matches only layer2 packets, i.e. those passed to
946 from ether_demux() and ether_output_frame().
947 .It Cm limit Bro Cm src-addr | src-port | dst-addr | dst-port Brc Ar N
948 The firewall will only allow
950 connections with the same
951 set of parameters as specified in the rule.
953 of source and destination addresses and ports can be
955 .It Cm { MAC | mac } Ar dst-mac src-mac
956 Match packets with a given
960 addresses, specified as the
962 keyword (matching any MAC address), or six groups of hex digits
964 and optionally followed by a mask indicating how many bits are
967 .Dl "MAC 10:20:30:40:50:60/33 any"
969 Note that the order of MAC addresses (destination first,
971 the same as on the wire, but the opposite of the one used for
973 .It Cm mac-type Ar mac-type
974 Matches packets whose Ethernet Type field
975 corresponds to one of those specified as argument.
977 is specified in the same way as
979 (i.e. one or more comma-separated single values or ranges).
980 You can use symbolic names for known values such as
981 .Em vlan , ipv4, ipv6 .
982 Values can be entered as decimal or hexadecimal (if prefixed by 0x),
983 and they are always printed as hexadecimal (unless the
985 option is used, in which case symbolic resolution will be attempted).
986 .It Cm proto Ar protocol
987 Matches packets with the corresponding IPv4 protocol.
988 .It Cm recv | xmit | via Brq Ar ifX | Ar if Ns Cm * | Ar ipno | Ar any
989 Matches packets received, transmitted or going through,
990 respectively, the interface specified by exact name
991 .Ns No ( Ar ifX Ns No ),
993 .Ns No ( Ar if Ns Ar * Ns No ),
994 by IP address, or through some interface.
998 keyword causes the interface to always be checked.
1005 then only the receive or transmit interface (respectively)
1007 By specifying both, it is possible to match packets based on
1008 both receive and transmit interface, e.g.:
1010 .Dl "ipfw add deny ip from any to any out recv ed0 xmit ed1"
1014 interface can be tested on either incoming or outgoing packets,
1017 interface can only be tested on outgoing packets.
1022 is invalid) whenever
1026 A packet may not have a receive or transmit interface: packets
1027 originating from the local host have no receive interface,
1028 while packets destined for the local host have no transmit
1031 Matches TCP packets that have the SYN bit set but no ACK bit.
1032 This is the short form of
1033 .Dq Li tcpflags\ syn,!ack .
1034 .It Cm src-ip Ar ip-address
1035 Matches IP packets whose source IP is one of the address(es)
1036 specified as argument.
1037 .It Cm src-port Ar ports
1038 Matches IP packets whose source port is one of the port(s)
1039 specified as argument.
1040 .It Cm tcpack Ar ack
1042 Match if the TCP header acknowledgment number field is set to
1044 .It Cm tcpflags Ar spec
1046 Match if the TCP header contains the comma separated list of
1049 The supported TCP flags are:
1058 The absence of a particular flag may be denoted
1061 A rule which contains a
1063 specification can never match a fragmented packet which has
1067 option for details on matching fragmented packets.
1068 .It Cm tcpseq Ar seq
1070 Match if the TCP header sequence number field is set to
1072 .It Cm tcpwin Ar win
1074 Match if the TCP header window field is set to
1076 .It Cm tcpoptions Ar spec
1078 Match if the TCP header contains the comma separated list of
1079 options specified in
1081 The supported TCP options are:
1084 (maximum segment size),
1086 (tcp window advertisement),
1090 (rfc1323 timestamp) and
1092 (rfc1644 t/tcp connection count).
1093 The absence of a particular option may be denoted
1097 Match all TCP or UDP packets sent by or received for a
1101 may be matched by name or identification number.
1104 Each rule belongs to one of 32 different
1107 Set 31 is reserved for the default rule.
1109 By default, rules are put in set 0, unless you use the
1111 attribute when entering a new rule.
1112 Sets can be individually and atomically enabled or disabled,
1113 so this mechanism permits an easy way to store multiple configurations
1114 of the firewall and quickly (and atomically) switch between them.
1115 The command to enable/disable sets is
1116 .Bd -ragged -offset indent
1118 .Cm set Oo Cm disable Ar number ... Oc Op Cm enable Ar number ...
1125 sections can be specified.
1126 Command execution is atomic on all the sets specified in the command.
1127 By default, all sets are enabled.
1129 When you disable a set, its rules behave as if they do not exist
1130 in the firewall configuration, with only one exception:
1131 .Bd -ragged -offset indent
1132 dynamic rules created from a rule before it had been disabled
1133 will still be active until they expire. In order to delete
1134 dynamic rules you have to explicitly delete the parent rule
1135 which generated them.
1138 The set number of rules can be changed with the command
1139 .Bd -ragged -offset indent
1142 .Brq Cm rule Ar rule-number | old-set
1146 Also, you can atomically swap two rulesets with the command
1147 .Bd -ragged -offset indent
1149 .Cm set swap Ar first-set second-set
1154 Section on some possible uses of sets of rules.
1155 .Sh STATEFUL FIREWALL
1156 Stateful operation is a way for the firewall to dynamically
1157 create rules for specific flows when packets that
1158 match a given pattern are detected. Support for stateful
1159 operation comes through the
1160 .Cm check-state , keep-state
1167 Dynamic rules are created when a packet matches a
1171 rule, causing the creation of a
1173 rule which will match all and only packets with
1177 .Em src-ip/src-port dst-ip/dst-port
1182 are used here only to denote the initial match addresses, but they
1183 are completely equivalent afterwards).
1184 Dynamic rules will be checked at the first
1185 .Cm check-state, keep-state
1188 occurrence, and the action performed upon a match will be the same
1189 as in the parent rule.
1191 Note that no additional attributes other than protocol and IP addresses
1192 and ports are checked on dynamic rules.
1194 The typical use of dynamic rules is to keep a closed firewall configuration,
1195 but let the first TCP SYN packet from the inside network install a
1196 dynamic rule for the flow so that packets belonging to that session
1197 will be allowed through the firewall:
1199 .Dl "ipfw add check-state"
1200 .Dl "ipfw add allow tcp from my-subnet to any setup keep-state"
1201 .Dl "ipfw add deny tcp from any to any"
1203 A similar approach can be used for UDP, where an UDP packet coming
1204 from the inside will install a dynamic rule to let the response through
1207 .Dl "ipfw add check-state"
1208 .Dl "ipfw add allow udp from my-subnet to any keep-state"
1209 .Dl "ipfw add deny udp from any to any"
1211 Dynamic rules expire after some time, which depends on the status
1212 of the flow and the setting of some
1216 .Sx SYSCTL VARIABLES
1218 For TCP sessions, dynamic rules can be instructed to periodically
1219 send keepalive packets to refresh the state of the rule when it is
1224 for more examples on how to use dynamic rules.
1225 .Sh TRAFFIC SHAPER (DUMMYNET) CONFIGURATION
1227 is also the user interface for the
1232 operates by first using the firewall to classify packets and divide them into
1234 using any match pattern that can be used in
1237 Depending on local policies, a flow can contain packets for a single
1238 TCP connection, or from/to a given host, or entire subnet, or a
1241 Packets belonging to the same flow are then passed to either of two
1242 different objects, which implement the traffic regulation:
1243 .Bl -hang -offset XXXX
1245 A pipe emulates a link with given bandwidth, propagation delay,
1246 queue size and packet loss rate.
1247 Packets are queued in front of the pipe as they come out from the classifier,
1248 and then transferred to the pipe according to the pipe's parameters.
1252 is an abstraction used to implement the WF2Q+
1253 (Worst-case Fair Weighted Fair Queueing) policy, which is
1254 an efficient variant of the WFQ policy.
1256 The queue associates a
1258 and a reference pipe to each flow, and then all backlogged (i.e.,
1259 with packets queued) flows linked to the same pipe share the pipe's
1260 bandwidth proportionally to their weights.
1261 Note that weights are not priorities; a flow with a lower weight
1262 is still guaranteed to get its fraction of the bandwidth even if a
1263 flow with a higher weight is permanently backlogged.
1267 can be used to set hard limits to the bandwidth that a flow can use, whereas
1269 can be used to determine how different flow share the available bandwidth.
1275 configuration commands are the following:
1276 .Bd -ragged -offset indent
1277 .Cm pipe Ar number Cm config Ar pipe-configuration
1279 .Cm queue Ar number Cm config Ar queue-configuration
1282 The following parameters can be configured for a pipe:
1284 .Bl -tag -width indent -compact
1285 .It Cm bw Ar bandwidth
1286 Bandwidth, measured in
1289 .Brq Cm bit/s | Byte/s .
1292 A value of 0 (default) means unlimited bandwidth.
1293 The unit must immediately follow the number, as in
1295 .Dl "ipfw pipe 1 config bw 300Kbit/s"
1297 .It Cm delay Ar ms-delay
1298 Propagation delay, measured in milliseconds.
1299 The value is rounded to the next multiple of the clock tick
1300 (typically 10ms, but it is a good practice to run kernels
1302 .Cd "options HZ=1000"
1304 the granularity to 1ms or less).
1305 Default value is 0, meaning no delay.
1308 The following parameters can be configured for a queue:
1310 .Bl -tag -width indent -compact
1311 .It Cm pipe Ar pipe_nr
1312 Connects a queue to the specified pipe.
1313 Multiple queues (with the same or different weights) can be connected to
1314 the same pipe, which specifies the aggregate rate for the set of queues.
1316 .It Cm weight Ar weight
1317 Specifies the weight to be used for flows matching this queue.
1318 The weight must be in the range 1..100, and defaults to 1.
1321 Finally, the following parameters can be configured for both
1324 .Bl -tag -width XXXX -compact
1326 .It Cm buckets Ar hash-table-size
1327 Specifies the size of the hash table used for storing the
1329 Default value is 64 controlled by the
1332 .Em net.inet.ip.dummynet.hash_size ,
1333 allowed range is 16 to 65536.
1335 .It Cm mask Ar mask-specifier
1336 Packets sent to a given pipe or queue by an
1338 rule can be further classified into multiple flows, each of which is then
1342 A flow identifier is constructed by masking the IP addresses,
1343 ports and protocol types as specified with the
1345 options in the configuration of the pipe or queue.
1346 For each different flow identifier, a new pipe or queue is created
1347 with the same parameters as the original object, and matching packets
1352 are used, each flow will get the same bandwidth as defined by the pipe,
1355 are used, each flow will share the parent's pipe bandwidth evenly
1356 with other flows generated by the same queue (note that other queues
1357 with different weights might be connected to the same pipe).
1359 Available mask specifiers are a combination of one or more of the following:
1361 .Cm dst-ip Ar mask ,
1362 .Cm src-ip Ar mask ,
1363 .Cm dst-port Ar mask ,
1364 .Cm src-port Ar mask ,
1369 where the latter means all bits in all fields are significant.
1372 When a packet is dropped by a dummynet queue or pipe, the error
1373 is normally reported to the caller routine in the kernel, in the
1374 same way as it happens when a device queue fills up. Setting this
1375 option reports the packet as successfully delivered, which can be
1376 needed for some experimental setups where you want to simulate
1377 loss or congestion at a remote router.
1379 .It Cm plr Ar packet-loss-rate
1382 .Ar packet-loss-rate
1383 is a floating-point number between 0 and 1, with 0 meaning no
1384 loss, 1 meaning 100% loss.
1385 The loss rate is internally represented on 31 bits.
1387 .It Cm queue Brq Ar slots | size Ns Cm Kbytes
1392 Default value is 50 slots, which
1393 is the typical queue size for Ethernet devices.
1394 Note that for slow speed links you should keep the queue
1395 size short or your traffic might be affected by a significant
1397 E.g., 50 max-sized ethernet packets (1500 bytes) mean 600Kbit
1398 or 20s of queue on a 30Kbit/s pipe.
1399 Even worse effect can result if you get packets from an
1400 interface with a much larger MTU, e.g. the loopback interface
1401 with its 16KB packets.
1403 .It Cm red | gred Ar w_q Ns / Ns Ar min_th Ns / Ns Ar max_th Ns / Ns Ar max_p
1404 Make use of the RED (Random Early Detection) queue management algorithm.
1409 point numbers between 0 and 1 (0 not included), while
1413 are integer numbers specifying thresholds for queue management
1414 (thresholds are computed in bytes if the queue has been defined
1415 in bytes, in slots otherwise).
1418 also supports the gentle RED variant (gred).
1421 variables can be used to control the RED behaviour:
1422 .Bl -tag -width indent
1423 .It Em net.inet.ip.dummynet.red_lookup_depth
1424 specifies the accuracy in computing the average queue
1425 when the link is idle (defaults to 256, must be greater than zero)
1426 .It Em net.inet.ip.dummynet.red_avg_pkt_size
1427 specifies the expected average packet size (defaults to 512, must be
1429 .It Em net.inet.ip.dummynet.red_max_pkt_size
1430 specifies the expected maximum packet size, only used when queue
1431 thresholds are in bytes (defaults to 1500, must be greater than zero).
1435 Here are some important points to consider when designing your
1439 Remember that you filter both packets going
1443 Most connections need packets going in both directions.
1445 Remember to test very carefully.
1446 It is a good idea to be near the console when doing this.
1447 If you cannot be near the console,
1448 use an auto-recovery script such as the one in
1449 .Pa /usr/share/examples/ipfw/change_rules.sh .
1451 Don't forget the loopback interface.
1456 There are circumstances where fragmented datagrams are unconditionally
1458 TCP packets are dropped if they do not contain at least 20 bytes of
1459 TCP header, UDP packets are dropped if they do not contain a full 8
1460 byte UDP header, and ICMP packets are dropped if they do not contain
1461 4 bytes of ICMP header, enough to specify the ICMP type, code, and
1463 These packets are simply logged as
1465 since there may not be enough good data in the packet to produce a
1466 meaningful log entry.
1468 Another type of packet is unconditionally dropped, a TCP packet with a
1469 fragment offset of one.
1470 This is a valid packet, but it only has one use, to try
1471 to circumvent firewalls.
1472 When logging is enabled, these packets are
1473 reported as being dropped by rule -1.
1475 If you are logged in over a network, loading the
1479 is probably not as straightforward as you would think.
1480 I recommend the following command line:
1481 .Bd -literal -offset indent
1482 kldload /modules/ipfw.ko && \e
1483 ipfw add 32000 allow ip from any to any
1486 Along the same lines, doing an
1487 .Bd -literal -offset indent
1491 in similar surroundings is also a bad idea.
1495 filter list may not be modified if the system security level
1496 is set to 3 or higher
1499 for information on system security levels).
1501 .Sh PACKET DIVERSION
1504 socket bound to the specified port will receive all packets
1505 diverted to that port.
1506 If no socket is bound to the destination port, or if the kernel
1507 wasn't compiled with divert socket support, the packets are
1509 .Sh SYSCTL VARIABLES
1512 variables controls the behaviour of the firewall and
1513 associated modules (
1516 These are shown below together with their default value
1517 (but always check with the
1519 command what value is actually in use) and meaning:
1520 .Bl -tag -width indent
1521 .It Em net.inet.ip.dummynet.expire : No 1
1522 Lazily delete dynamic pipes/queue once they have no pending traffic.
1523 You can disable this by setting the variable to 0, in which case
1524 the pipes/queues will only be deleted when the threshold is reached.
1525 .It Em net.inet.ip.dummynet.hash_size : No 64
1526 Default size of the hash table used for dynamic pipes/queues.
1527 This value is used when no
1529 option is specified when configuring a pipe/queue.
1530 .It Em net.inet.ip.dummynet.max_chain_len : No 16
1531 Target value for the maximum number of pipes/queues in a hash bucket.
1533 .Cm max_chain_len*hash_size
1534 is used to determine the threshold over which empty pipes/queues
1535 will be expired even when
1536 .Cm net.inet.ip.dummynet.expire=0 .
1537 .It Em net.inet.ip.dummynet.red_lookup_depth : No 256
1538 .It Em net.inet.ip.dummynet.red_avg_pkt_size : No 512
1539 .It Em net.inet.ip.dummynet.red_max_pkt_size : No 1500
1540 Parameters used in the computations of the drop probability
1541 for the RED algorithm.
1542 .It Em net.inet.ip.fw.autoinc_step : No 100
1543 Delta between rule numbers when auto-generating them.
1544 The value must be in the range 1..1000.
1545 .It Em net.inet.ip.fw.curr_dyn_buckets : Em net.inet.ip.fw.dyn_buckets
1546 The current number of buckets in the hash table for dynamic rules
1548 .It Em net.inet.ip.fw.debug : No 1
1549 Controls debugging messages produced by
1551 .It Em net.inet.ip.fw.dyn_buckets : No 256
1552 The number of buckets in the hash table for dynamic rules.
1553 Must be a power of 2, up to 65536.
1554 It only takes effect when all dynamic rules have expired, so you
1555 are advised to use a
1557 command to make sure that the hash table is resized.
1558 .It Em net.inet.ip.fw.dyn_count : No 3
1559 Current number of dynamic rules
1561 .It Em net.inet.ip.fw.dyn_keepalive : No 1
1562 Enables generation of keepalive packets for
1564 rules on TCP sessions. A keepalive is generated to both
1565 sides of the connection every 5 seconds for the last 20
1566 seconds of the lifetime of the rule.
1567 .It Em net.inet.ip.fw.dyn_max : No 8192
1568 Maximum number of dynamic rules.
1569 When you hit this limit, no more dynamic rules can be
1570 installed until old ones expire.
1571 .It Em net.inet.ip.fw.dyn_ack_lifetime : No 300
1572 .It Em net.inet.ip.fw.dyn_syn_lifetime : No 20
1573 .It Em net.inet.ip.fw.dyn_fin_lifetime : No 1
1574 .It Em net.inet.ip.fw.dyn_rst_lifetime : No 1
1575 .It Em net.inet.ip.fw.dyn_udp_lifetime : No 5
1576 .It Em net.inet.ip.fw.dyn_short_lifetime : No 30
1577 These variables control the lifetime, in seconds, of dynamic
1579 Upon the initial SYN exchange the lifetime is kept short,
1580 then increased after both SYN have been seen, then decreased
1581 again during the final FIN exchange or when a RST is received.
1583 .Em dyn_fin_lifetime
1585 .Em dyn_rst_lifetime
1586 must be strictly lower than 5 seconds, the period of
1587 repetition of keepalives. The firewall enforces that.
1588 .It Em net.inet.ip.fw.enable : No 1
1589 Enables the firewall.
1590 Setting this variable to 0 lets you run your machine without
1591 firewall even if compiled in.
1592 .It Em net.inet.ip.fw.one_pass : No 1
1593 When set, the packet exiting from the
1595 pipe is not passed though the firewall again.
1596 Otherwise, after a pipe action, the packet is
1597 reinjected into the firewall at the next rule.
1599 Note: layer 2 packets coming out of a pipe
1600 are never reinjected in the firewall irrespective of the
1601 value of this variable.
1602 .It Em net.inet.ip.fw.verbose : No 1
1603 Enables verbose messages.
1604 .It Em net.inet.ip.fw.verbose_limit : No 0
1605 Limits the number of messages produced by a verbose firewall.
1606 .It Em net.link.ether.ipfw : No 0
1607 Controls whether layer-2 packets are passed to
1611 .Sh IPFW2 ENHANCEMENTS
1612 This Section lists the features that have been introduced in
1614 which were not present in
1616 We list them in order of the potential impact that they can
1617 have in writing your rulesets.
1618 You might want to consider using these features in order to
1619 write your rulesets in a more efficient way.
1620 .Bl -tag -width indent
1621 .It Handling of non-IPv4 packets
1623 will silently accept all non-IPv4 packets.
1625 will filter all packets (including non-IPv4 ones) according to the ruleset.
1626 To achieve the same behaviour as
1628 you can use the following as the very first rule in your ruleset:
1630 .Dl "ipfw add 1 allow layer2 not mac-type ip"
1634 option might seem redundant, but it is necessary -- packets
1635 passed to the firewall from layer3 will not have a MAC header,
1638 pattern will always fail on them, and the
1640 operator will make this rule into a pass-all.
1643 does not supports address sets (those in the form
1644 .Ar addr/masklen{num,num,...}
1647 .It Port specifications
1649 only allows one port range when specifying TCP and UDP ports, and
1650 is limited to 10 entries instead of the 15 allowed by
1654 you can only specify ports when the rule is requesting
1660 you can put port specifications in rules matching all packets,
1661 and the match will be attempted only on those packets carrying
1662 protocols which include port identifiers.
1666 allowed the first port entry to be specified as
1670 can be an arbitrary 16-bit mask.
1671 This syntax is of questionable usefulness and it is not
1672 supported anymore in
1676 does not support Or-blocks.
1679 does not generate keepalives for stateful sessions.
1680 As a consequence, it might cause idle sessions to drop because
1681 the lifetime of the dynamic rules expires.
1684 does not implement sets of rules.
1685 .It MAC header filtering and Layer-2 firewalling.
1687 does not implement filtering on MAC header fields, nor is it
1688 invoked on packets from
1691 .Fn ether_output_frame .
1693 .Em net.link.ether.ipfw
1694 has no effect there.
1696 The following options are not supported in
1699 .Cm dst-ip, dst-port, layer2, mac, mac-type, src-ip, src-port.
1701 Additionally, the following options are not supported in
1706 .Cm ipid, iplen, ipprecedence, iptos, ipttl,
1707 .Cm ipversion, tcpack, tcpseq, tcpwin .
1708 .It Dummynet options
1709 The following option for
1711 pipes/queues is not supported:
1715 There are far too many possible uses of
1717 so this Section will only give a small set of examples.
1718 .Ss BASIC PACKET FILTERING
1719 This command adds an entry which denies all tcp packets from
1720 .Em cracker.evil.org
1721 to the telnet port of
1723 from being forwarded by the host:
1725 .Dl "ipfw add deny tcp from cracker.evil.org to wolf.tambov.su telnet"
1727 This one disallows any connection from the entire cracker's
1730 .Dl "ipfw add deny ip from 123.45.67.0/24 to my.host.org"
1732 A first and efficient way to limit access (not using dynamic rules)
1733 is the use of the following rules:
1735 .Dl "ipfw add allow tcp from any to any established"
1736 .Dl "ipfw add allow tcp from net1 portlist1 to net2 portlist2 setup"
1737 .Dl "ipfw add allow tcp from net3 portlist3 to net3 portlist3 setup"
1739 .Dl "ipfw add deny tcp from any to any"
1741 The first rule will be a quick match for normal TCP packets,
1742 but it will not match the initial SYN packet, which will be
1745 rules only for selected source/destination pairs.
1746 All other SYN packets will be rejected by the final
1750 If you administer one or more subnets, you can take advantage of the
1752 syntax to specify address sets and or-blocks and write extremely
1753 compact rulesets which selectively enable services to blocks
1754 of clients, as below:
1756 .Dl "goodguys=\*q{ 10.1.2.0/24{20,35,66,18} or 10.2.3.0/28{6,3,11} }\*q"
1757 .Dl "badguys=\*q10.1.2.0/24{8,38,60}\*q"
1759 .Dl "ipfw add allow ip from ${goodguys} to any"
1760 .Dl "ipfw add deny ip from ${badguys} to any"
1761 .Dl "... normal policies ..."
1765 syntax would require a separate rule for each IP in the above
1768 In order to protect a site from flood attacks involving fake
1769 TCP packets, it is safer to use dynamic rules:
1771 .Dl "ipfw add check-state"
1772 .Dl "ipfw add deny tcp from any to any established"
1773 .Dl "ipfw add allow tcp from my-net to any setup keep-state"
1775 This will let the firewall install dynamic rules only for
1776 those connection which start with a regular SYN packet coming
1777 from the inside of our network.
1778 Dynamic rules are checked when encountering the first
1785 rule should usually be placed near the beginning of the
1786 ruleset to minimize the amount of work scanning the ruleset.
1787 Your mileage may vary.
1789 To limit the number of connections a user can open
1790 you can use the following type of rules:
1792 .Dl "ipfw add allow tcp from my-net/24 to any setup limit src-addr 10"
1793 .Dl "ipfw add allow tcp from any to me setup limit src-addr 4"
1795 The former (assuming it runs on a gateway) will allow each host
1796 on a /24 network to open at most 10 TCP connections.
1797 The latter can be placed on a server to make sure that a single
1798 client does not use more than 4 simultaneous connections.
1801 stateful rules can be subject to denial-of-service attacks
1802 by a SYN-flood which opens a huge number of dynamic rules.
1803 The effects of such attacks can be partially limited by
1806 variables which control the operation of the firewall.
1808 Here is a good usage of the
1810 command to see accounting records and timestamp information:
1814 or in short form without timestamps:
1818 which is equivalent to:
1822 Next rule diverts all incoming packets from 192.168.2.0/24
1823 to divert port 5000:
1825 .Dl ipfw divert 5000 ip from 192.168.2.0/24 to any in
1827 The following rules show some of the applications of
1831 for simulations and the like.
1833 This rule drops random incoming packets with a probability
1836 .Dl "ipfw add prob 0.05 deny ip from any to any in"
1838 A similar effect can be achieved making use of dummynet pipes:
1840 .Dl "ipfw add pipe 10 ip from any to any"
1841 .Dl "ipfw pipe 10 config plr 0.05"
1843 We can use pipes to artificially limit bandwidth, e.g. on a
1844 machine acting as a router, if we want to limit traffic from
1845 local clients on 192.168.2.0/24 we do:
1847 .Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
1848 .Dl "ipfw pipe 1 config bw 300Kbit/s queue 50KBytes"
1850 note that we use the
1852 modifier so that the rule is not used twice.
1853 Remember in fact that
1855 rules are checked both on incoming and outgoing packets.
1857 Should we want to simulate a bidirectional link with bandwidth
1858 limitations, the correct way is the following:
1860 .Dl "ipfw add pipe 1 ip from any to any out"
1861 .Dl "ipfw add pipe 2 ip from any to any in"
1862 .Dl "ipfw pipe 1 config bw 64Kbit/s queue 10Kbytes"
1863 .Dl "ipfw pipe 2 config bw 64Kbit/s queue 10Kbytes"
1865 The above can be very useful, e.g. if you want to see how
1866 your fancy Web page will look for a residential user who
1867 is connected only through a slow link.
1868 You should not use only one pipe for both directions, unless
1869 you want to simulate a half-duplex medium (e.g. AppleTalk,
1871 It is not necessary that both pipes have the same configuration,
1872 so we can also simulate asymmetric links.
1874 Should we want to verify network performance with the RED queue
1875 management algorithm:
1877 .Dl "ipfw add pipe 1 ip from any to any"
1878 .Dl "ipfw pipe 1 config bw 500Kbit/s queue 100 red 0.002/30/80/0.1"
1880 Another typical application of the traffic shaper is to
1881 introduce some delay in the communication.
1882 This can significantly affect applications which do a lot of Remote
1883 Procedure Calls, and where the round-trip-time of the
1884 connection often becomes a limiting factor much more than
1887 .Dl "ipfw add pipe 1 ip from any to any out"
1888 .Dl "ipfw add pipe 2 ip from any to any in"
1889 .Dl "ipfw pipe 1 config delay 250ms bw 1Mbit/s"
1890 .Dl "ipfw pipe 2 config delay 250ms bw 1Mbit/s"
1892 Per-flow queueing can be useful for a variety of purposes.
1893 A very simple one is counting traffic:
1895 .Dl "ipfw add pipe 1 tcp from any to any"
1896 .Dl "ipfw add pipe 1 udp from any to any"
1897 .Dl "ipfw add pipe 1 ip from any to any"
1898 .Dl "ipfw pipe 1 config mask all"
1900 The above set of rules will create queues (and collect
1901 statistics) for all traffic.
1902 Because the pipes have no limitations, the only effect is
1903 collecting statistics.
1904 Note that we need 3 rules, not just the last one, because
1907 tries to match IP packets it will not consider ports, so we
1908 would not see connections on separate ports as different
1911 A more sophisticated example is limiting the outbound traffic
1912 on a net with per-host limits, rather than per-network limits:
1914 .Dl "ipfw add pipe 1 ip from 192.168.2.0/24 to any out"
1915 .Dl "ipfw add pipe 2 ip from any to 192.168.2.0/24 in"
1916 .Dl "ipfw pipe 1 config mask src-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
1917 .Dl "ipfw pipe 2 config mask dst-ip 0x000000ff bw 200Kbit/s queue 20Kbytes"
1919 To add a set of rules atomically, e.g. set 18:
1921 .Dl "ipfw disable set 18"
1922 .Dl "ipfw add NN set 18 ... # repeat as needed"
1923 .Dl "ipfw enable set 18"
1925 To delete a set of rules atomically the command is simply:
1927 .Dl "ipfw delete set 18"
1929 To test a ruleset and disable it and regain control if something goes wrong:
1931 .Dl "ipfw disable set 18"
1932 .Dl "ipfw add NN set 18 ... # repeat as needed"
1933 .Dl "ipfw enable set 18 ; echo done; sleep 30 && ipfw disable set 18"
1935 Here if everything goes well, you press control-C before the "sleep"
1936 terminates, and your ruleset will be left active. Otherwise, e.g. if
1937 you cannot access your box, the ruleset will be disabled after
1938 the sleep terminates thus restoring the previous situation.
1956 utility first appeared in
1961 Stateful extensions were introduced in
1964 was introduced in Summer 2002.
1966 .An Ugen J. S. Antsilevich ,
1967 .An Poul-Henning Kamp ,
1973 API based upon code written by
1979 traffic shaper supported by Akamba Corp.
1981 The syntax has grown over the years and sometimes it might be confusing.
1982 Unfortunately, backward compatibility prevents cleaning up mistakes
1983 made in the definition of the syntax.
1987 Misconfiguring the firewall can put your computer in an unusable state,
1988 possibly shutting down network services and requiring console access to
1989 regain control of it.
1991 Incoming packet fragments diverted by
1995 are reassembled before delivery to the socket.
1996 The action used on those packet is the one from the
1997 rule which matches the first fragment of the packet.
1999 Packets that match a
2001 rule should not be immediately accepted, but should continue
2002 going through the rule list.
2003 This may be fixed in a later version.
2005 Packets diverted to userland, and then reinserted by a userland process
2008 will lose various packet attributes, including their source interface.
2009 If a packet is reinserted in this manner, later rules may be incorrectly
2010 applied, making the order of
2012 rules in the rule sequence very important.