xref: /freebsd/sbin/pfctl/pfctl.8 (revision 190cef3d52236565eb22e18b33e9e865ec634aa3)
1.\" $OpenBSD: pfctl.8,v 1.138 2008/06/10 20:55:02 mcbride Exp $
2.\"
3.\" Copyright (c) 2001 Kjell Wooding.  All rights reserved.
4.\"
5.\" Redistribution and use in source and binary forms, with or without
6.\" modification, are permitted provided that the following conditions
7.\" are met:
8.\" 1. Redistributions of source code must retain the above copyright
9.\"    notice, this list of conditions and the following disclaimer.
10.\" 2. Redistributions in binary form must reproduce the above copyright
11.\"    notice, this list of conditions and the following disclaimer in the
12.\"    documentation and/or other materials provided with the distribution.
13.\" 3. The name of the author may not be used to endorse or promote products
14.\"    derived from this software without specific prior written permission.
15.\"
16.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
17.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
18.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
19.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
20.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
21.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
22.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
23.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
25.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
26.\"
27.\" $FreeBSD$
28.\"
29.Dd October 3, 2016
30.Dt PFCTL 8
31.Os
32.Sh NAME
33.Nm pfctl
34.Nd control the packet filter (PF) device
35.Sh SYNOPSIS
36.Nm pfctl
37.Bk -words
38.Op Fl AdeghmNnOPqRrvz
39.Op Fl a Ar anchor
40.Oo Fl D Ar macro Ns =
41.Ar value Oc
42.Op Fl F Ar modifier
43.Op Fl f Ar file
44.Op Fl i Ar interface
45.Op Fl K Ar host | network
46.Xo
47.Oo Fl k
48.Ar host | network | label | id
49.Oc Xc
50.Op Fl o Ar level
51.Op Fl p Ar device
52.Op Fl s Ar modifier
53.Xo
54.Oo Fl t Ar table
55.Fl T Ar command
56.Op Ar address ...
57.Oc Xc
58.Op Fl x Ar level
59.Ek
60.Sh DESCRIPTION
61The
62.Nm
63utility communicates with the packet filter device using the
64ioctl interface described in
65.Xr pf 4 .
66It allows ruleset and parameter configuration and retrieval of status
67information from the packet filter.
68.Pp
69Packet filtering restricts the types of packets that pass through
70network interfaces entering or leaving the host based on filter
71rules as described in
72.Xr pf.conf 5 .
73The packet filter can also replace addresses and ports of packets.
74Replacing source addresses and ports of outgoing packets is called
75NAT (Network Address Translation) and is used to connect an internal
76network (usually reserved address space) to an external one (the
77Internet) by making all connections to external hosts appear to
78come from the gateway.
79Replacing destination addresses and ports of incoming packets
80is used to redirect connections to different hosts and/or ports.
81A combination of both translations, bidirectional NAT, is also
82supported.
83Translation rules are described in
84.Xr pf.conf 5 .
85.Pp
86When the variable
87.Va pf
88is set to
89.Dv YES
90in
91.Xr rc.conf 5 ,
92the rule file specified with the variable
93.Va pf_rules
94is loaded automatically by the
95.Xr rc 8
96scripts and the packet filter is enabled.
97.Pp
98The packet filter does not itself forward packets between interfaces.
99Forwarding can be enabled by setting the
100.Xr sysctl 8
101variables
102.Em net.inet.ip.forwarding
103and/or
104.Em net.inet6.ip6.forwarding
105to 1.
106Set them permanently in
107.Xr sysctl.conf 5 .
108.Pp
109The
110.Nm
111utility provides several commands.
112The options are as follows:
113.Bl -tag -width Ds
114.It Fl A
115Load only the queue rules present in the rule file.
116Other rules and options are ignored.
117.It Fl a Ar anchor
118Apply flags
119.Fl f ,
120.Fl F ,
121and
122.Fl s
123only to the rules in the specified
124.Ar anchor .
125In addition to the main ruleset,
126.Nm
127can load and manipulate additional rulesets by name,
128called anchors.
129The main ruleset is the default anchor.
130.Pp
131Anchors are referenced by name and may be nested,
132with the various components of the anchor path separated by
133.Sq /
134characters, similar to how file system hierarchies are laid out.
135The last component of the anchor path is where ruleset operations are
136performed.
137.Pp
138Evaluation of
139.Ar anchor
140rules from the main ruleset is described in
141.Xr pf.conf 5 .
142.Pp
143For example, the following will show all filter rules (see the
144.Fl s
145flag below) inside the anchor
146.Dq authpf/smith(1234) ,
147which would have been created for user
148.Dq smith
149by
150.Xr authpf 8 ,
151PID 1234:
152.Bd -literal -offset indent
153# pfctl -a "authpf/smith(1234)" -s rules
154.Ed
155.Pp
156Private tables can also be put inside anchors, either by having table
157statements in the
158.Xr pf.conf 5
159file that is loaded in the anchor, or by using regular table commands, as in:
160.Bd -literal -offset indent
161# pfctl -a foo/bar -t mytable -T add 1.2.3.4 5.6.7.8
162.Ed
163.Pp
164When a rule referring to a table is loaded in an anchor, the rule will use the
165private table if one is defined, and then fall back to the table defined in the
166main ruleset, if there is one.
167This is similar to C rules for variable scope.
168It is possible to create distinct tables with the same name in the global
169ruleset and in an anchor, but this is often bad design and a warning will be
170issued in that case.
171.Pp
172By default, recursive inline printing of anchors applies only to unnamed
173anchors specified inline in the ruleset.
174If the anchor name is terminated with a
175.Sq *
176character, the
177.Fl s
178flag will recursively print all anchors in a brace delimited block.
179For example the following will print the
180.Dq authpf
181ruleset recursively:
182.Bd -literal -offset indent
183# pfctl -a 'authpf/*' -sr
184.Ed
185.Pp
186To print the main ruleset recursively, specify only
187.Sq *
188as the anchor name:
189.Bd -literal -offset indent
190# pfctl -a '*' -sr
191.Ed
192.It Fl D Ar macro Ns = Ns Ar value
193Define
194.Ar macro
195to be set to
196.Ar value
197on the command line.
198Overrides the definition of
199.Ar macro
200in the ruleset.
201.It Fl d
202Disable the packet filter.
203.It Fl e
204Enable the packet filter.
205.It Fl F Ar modifier
206Flush the filter parameters specified by
207.Ar modifier
208(may be abbreviated):
209.Pp
210.Bl -tag -width xxxxxxxxxxxx -compact
211.It Fl F Cm nat
212Flush the NAT rules.
213.It Fl F Cm queue
214Flush the queue rules.
215.It Fl F Cm rules
216Flush the filter rules.
217.It Fl F Cm states
218Flush the state table (NAT and filter).
219.It Fl F Cm Sources
220Flush the source tracking table.
221.It Fl F Cm info
222Flush the filter information (statistics that are not bound to rules).
223.It Fl F Cm Tables
224Flush the tables.
225.It Fl F Cm osfp
226Flush the passive operating system fingerprints.
227.It Fl F Cm all
228Flush all of the above.
229.El
230.It Fl f Ar file
231Load the rules contained in
232.Ar file .
233This
234.Ar file
235may contain macros, tables, options, and normalization, queueing,
236translation, and filtering rules.
237With the exception of macros and tables, the statements must appear in that
238order.
239.It Fl g
240Include output helpful for debugging.
241.It Fl h
242Help.
243.It Fl i Ar interface
244Restrict the operation to the given
245.Ar interface .
246.It Fl K Ar host | network
247Kill all of the source tracking entries originating from the specified
248.Ar host
249or
250.Ar network .
251A second
252.Fl K Ar host
253or
254.Fl K Ar network
255option may be specified, which will kill all the source tracking
256entries from the first host/network to the second.
257.It Xo
258.Fl k
259.Ar host | network | label | id
260.Xc
261Kill all of the state entries matching the specified
262.Ar host ,
263.Ar network ,
264.Ar label ,
265or
266.Ar id .
267.Pp
268For example, to kill all of the state entries originating from
269.Dq host :
270.Pp
271.Dl # pfctl -k host
272.Pp
273A second
274.Fl k Ar host
275or
276.Fl k Ar network
277option may be specified, which will kill all the state entries
278from the first host/network to the second.
279To kill all of the state entries from
280.Dq host1
281to
282.Dq host2 :
283.Pp
284.Dl # pfctl -k host1 -k host2
285.Pp
286To kill all states originating from 192.168.1.0/24 to 172.16.0.0/16:
287.Pp
288.Dl # pfctl -k 192.168.1.0/24 -k 172.16.0.0/16
289.Pp
290A network prefix length of 0 can be used as a wildcard.
291To kill all states with the target
292.Dq host2 :
293.Pp
294.Dl # pfctl -k 0.0.0.0/0 -k host2
295.Pp
296It is also possible to kill states by rule label or state ID.
297In this mode the first
298.Fl k
299argument is used to specify the type
300of the second argument.
301The following command would kill all states that have been created
302from rules carrying the label
303.Dq foobar :
304.Pp
305.Dl # pfctl -k label -k foobar
306.Pp
307To kill one specific state by its unique state ID
308(as shown by pfctl -s state -vv),
309use the
310.Ar id
311modifier and as a second argument the state ID and optional creator ID.
312To kill a state with ID 4823e84500000003 use:
313.Pp
314.Dl # pfctl -k id -k 4823e84500000003
315.Pp
316To kill a state with ID 4823e84500000018 created from a backup
317firewall with hostid 00000002 use:
318.Pp
319.Dl # pfctl -k id -k 4823e84500000018/2
320.It Fl m
321Merge in explicitly given options without resetting those
322which are omitted.
323Allows single options to be modified without disturbing the others:
324.Bd -literal -offset indent
325# echo "set loginterface fxp0" | pfctl -mf -
326.Ed
327.It Fl N
328Load only the NAT rules present in the rule file.
329Other rules and options are ignored.
330.It Fl n
331Do not actually load rules, just parse them.
332.It Fl O
333Load only the options present in the rule file.
334Other rules and options are ignored.
335.It Fl o Ar level
336Control the ruleset optimizer, overriding any rule file settings.
337.Pp
338.Bl -tag -width xxxxxxxxxxxx -compact
339.It Fl o Cm none
340Disable the ruleset optimizer.
341.It Fl o Cm basic
342Enable basic ruleset optimizations.
343This is the default behaviour.
344.It Fl o Cm profile
345Enable basic ruleset optimizations with profiling.
346.El
347For further information on the ruleset optimizer, see
348.Xr pf.conf 5 .
349.It Fl P
350Do not perform service name lookup for port specific rules,
351instead display the ports numerically.
352.It Fl p Ar device
353Use the device file
354.Ar device
355instead of the default
356.Pa /dev/pf .
357.It Fl q
358Only print errors and warnings.
359.It Fl R
360Load only the filter rules present in the rule file.
361Other rules and options are ignored.
362.It Fl r
363Perform reverse DNS lookups on states when displaying them.
364.It Fl s Ar modifier
365Show the filter parameters specified by
366.Ar modifier
367(may be abbreviated):
368.Pp
369.Bl -tag -width xxxxxxxxxxxxx -compact
370.It Fl s Cm nat
371Show the currently loaded NAT rules.
372.It Fl s Cm queue
373Show the currently loaded queue rules.
374When used together with
375.Fl v ,
376per-queue statistics are also shown.
377When used together with
378.Fl v v ,
379.Nm
380will loop and show updated queue statistics every five seconds, including
381measured bandwidth and packets per second.
382.It Fl s Cm rules
383Show the currently loaded filter rules.
384When used together with
385.Fl v ,
386the per-rule statistics (number of evaluations,
387packets and bytes) are also shown.
388Note that the
389.Dq skip step
390optimization done automatically by the kernel
391will skip evaluation of rules where possible.
392Packets passed statefully are counted in the rule that created the state
393(even though the rule is not evaluated more than once for the entire
394connection).
395.It Fl s Cm Anchors
396Show the currently loaded anchors directly attached to the main ruleset.
397If
398.Fl a Ar anchor
399is specified as well, the anchors loaded directly below the given
400.Ar anchor
401are shown instead.
402If
403.Fl v
404is specified, all anchors attached under the target anchor will be
405displayed recursively.
406.It Fl s Cm states
407Show the contents of the state table.
408.It Fl s Cm Sources
409Show the contents of the source tracking table.
410.It Fl s Cm info
411Show filter information (statistics and counters).
412When used together with
413.Fl v ,
414source tracking statistics are also shown.
415.It Fl s Cm Running
416Show the running status and provide a non-zero exit status when disabled.
417.It Fl s Cm labels
418Show per-rule statistics (label, evaluations, packets total, bytes total,
419packets in, bytes in, packets out, bytes out, state creations) of
420filter rules with labels, useful for accounting.
421.It Fl s Cm timeouts
422Show the current global timeouts.
423.It Fl s Cm memory
424Show the current pool memory hard limits.
425.It Fl s Cm Tables
426Show the list of tables.
427.It Fl s Cm osfp
428Show the list of operating system fingerprints.
429.It Fl s Cm Interfaces
430Show the list of interfaces and interface drivers available to PF.
431When used together with
432.Fl v ,
433it additionally lists which interfaces have skip rules activated.
434When used together with
435.Fl vv ,
436interface statistics are also shown.
437.Fl i
438can be used to select an interface or a group of interfaces.
439.It Fl s Cm all
440Show all of the above, except for the lists of interfaces and operating
441system fingerprints.
442.El
443.It Fl T Ar command Op Ar address ...
444Specify the
445.Ar command
446(may be abbreviated) to apply to the table.
447Commands include:
448.Pp
449.Bl -tag -width xxxxxxxxxxxx -compact
450.It Fl T Cm kill
451Kill a table.
452.It Fl T Cm flush
453Flush all addresses of a table.
454.It Fl T Cm add
455Add one or more addresses in a table.
456Automatically create a nonexisting table.
457.It Fl T Cm delete
458Delete one or more addresses from a table.
459.It Fl T Cm expire Ar number
460Delete addresses which had their statistics cleared more than
461.Ar number
462seconds ago.
463For entries which have never had their statistics cleared,
464.Ar number
465refers to the time they were added to the table.
466.It Fl T Cm replace
467Replace the addresses of the table.
468Automatically create a nonexisting table.
469.It Fl T Cm show
470Show the content (addresses) of a table.
471.It Fl T Cm test
472Test if the given addresses match a table.
473.It Fl T Cm zero
474Clear all the statistics of a table.
475.It Fl T Cm load
476Load only the table definitions from
477.Xr pf.conf 5 .
478This is used in conjunction with the
479.Fl f
480flag, as in:
481.Bd -literal -offset indent
482# pfctl -Tl -f pf.conf
483.Ed
484.El
485.Pp
486For the
487.Cm add ,
488.Cm delete ,
489.Cm replace ,
490and
491.Cm test
492commands, the list of addresses can be specified either directly on the command
493line and/or in an unformatted text file, using the
494.Fl f
495flag.
496Comments starting with a
497.Sq #
498are allowed in the text file.
499With these commands, the
500.Fl v
501flag can also be used once or twice, in which case
502.Nm
503will print the
504detailed result of the operation for each individual address, prefixed by
505one of the following letters:
506.Pp
507.Bl -tag -width XXX -compact
508.It A
509The address/network has been added.
510.It C
511The address/network has been changed (negated).
512.It D
513The address/network has been deleted.
514.It M
515The address matches
516.Po
517.Cm test
518operation only
519.Pc .
520.It X
521The address/network is duplicated and therefore ignored.
522.It Y
523The address/network cannot be added/deleted due to conflicting
524.Sq \&!
525attributes.
526.It Z
527The address/network has been cleared (statistics).
528.El
529.Pp
530Each table can maintain a set of counters that can be retrieved using the
531.Fl v
532flag of
533.Nm .
534For example, the following commands define a wide open firewall which will keep
535track of packets going to or coming from the
536.Ox
537FTP server.
538The following commands configure the firewall and send 10 pings to the FTP
539server:
540.Bd -literal -offset indent
541# printf "table <test> counters { ftp.openbsd.org }\en \e
542    pass out to <test>\en" | pfctl -f-
543# ping -qc10 ftp.openbsd.org
544.Ed
545.Pp
546We can now use the table
547.Cm show
548command to output, for each address and packet direction, the number of packets
549and bytes that are being passed or blocked by rules referencing the table.
550The time at which the current accounting started is also shown with the
551.Dq Cleared
552line.
553.Bd -literal -offset indent
554# pfctl -t test -vTshow
555   129.128.5.191
556    Cleared:     Thu Feb 13 18:55:18 2003
557    In/Block:    [ Packets: 0        Bytes: 0        ]
558    In/Pass:     [ Packets: 10       Bytes: 840      ]
559    Out/Block:   [ Packets: 0        Bytes: 0        ]
560    Out/Pass:    [ Packets: 10       Bytes: 840      ]
561.Ed
562.Pp
563Similarly, it is possible to view global information about the tables
564by using the
565.Fl v
566modifier twice and the
567.Fl s
568.Cm Tables
569command.
570This will display the number of addresses on each table,
571the number of rules which reference the table, and the global
572packet statistics for the whole table:
573.Bd -literal -offset indent
574# pfctl -vvsTables
575--a-r-C test
576    Addresses:   1
577    Cleared:     Thu Feb 13 18:55:18 2003
578    References:  [ Anchors: 0        Rules: 1        ]
579    Evaluations: [ NoMatch: 3496     Match: 1        ]
580    In/Block:    [ Packets: 0        Bytes: 0        ]
581    In/Pass:     [ Packets: 10       Bytes: 840      ]
582    In/XPass:    [ Packets: 0        Bytes: 0        ]
583    Out/Block:   [ Packets: 0        Bytes: 0        ]
584    Out/Pass:    [ Packets: 10       Bytes: 840      ]
585    Out/XPass:   [ Packets: 0        Bytes: 0        ]
586.Ed
587.Pp
588As we can see here, only one packet \- the initial ping request \- matched the
589table, but all packets passing as the result of the state are correctly
590accounted for.
591Reloading the table(s) or ruleset will not affect packet accounting in any way.
592The two
593.Dq XPass
594counters are incremented instead of the
595.Dq Pass
596counters when a
597.Dq stateful
598packet is passed but does not match the table anymore.
599This will happen in our example if someone flushes the table while the
600.Xr ping 8
601command is running.
602.Pp
603When used with a single
604.Fl v ,
605.Nm
606will only display the first line containing the table flags and name.
607The flags are defined as follows:
608.Pp
609.Bl -tag -width XXX -compact
610.It c
611For constant tables, which cannot be altered outside
612.Xr pf.conf 5 .
613.It p
614For persistent tables, which do not get automatically killed when no rules
615refer to them.
616.It a
617For tables which are part of the
618.Em active
619tableset.
620Tables without this flag do not really exist, cannot contain addresses, and are
621only listed if the
622.Fl g
623flag is given.
624.It i
625For tables which are part of the
626.Em inactive
627tableset.
628This flag can only be witnessed briefly during the loading of
629.Xr pf.conf 5 .
630.It r
631For tables which are referenced (used) by rules.
632.It h
633This flag is set when a table in the main ruleset is hidden by one or more
634tables of the same name from anchors attached below it.
635.It C
636This flag is set when per-address counters are enabled on the table.
637.El
638.It Fl t Ar table
639Specify the name of the table.
640.It Fl v
641Produce more verbose output.
642A second use of
643.Fl v
644will produce even more verbose output including ruleset warnings.
645See the previous section for its effect on table commands.
646.It Fl x Ar level
647Set the debug
648.Ar level
649(may be abbreviated) to one of the following:
650.Pp
651.Bl -tag -width xxxxxxxxxxxx -compact
652.It Fl x Cm none
653Do not generate debug messages.
654.It Fl x Cm urgent
655Generate debug messages only for serious errors.
656.It Fl x Cm misc
657Generate debug messages for various errors.
658.It Fl x Cm loud
659Generate debug messages for common conditions.
660.El
661.It Fl z
662Clear per-rule statistics.
663.El
664.Sh FILES
665.Bl -tag -width "/etc/pf.conf" -compact
666.It Pa /etc/pf.conf
667Packet filter rules file.
668.It Pa /etc/pf.os
669Passive operating system fingerprint database.
670.El
671.Sh SEE ALSO
672.Xr pf 4 ,
673.Xr pf.conf 5 ,
674.Xr pf.os 5 ,
675.Xr rc.conf 5 ,
676.Xr services 5 ,
677.Xr sysctl.conf 5 ,
678.Xr authpf 8 ,
679.Xr ftp-proxy 8 ,
680.Xr rc 8 ,
681.Xr sysctl 8
682.Sh HISTORY
683The
684.Nm
685program and the
686.Xr pf 4
687filter mechanism appeared in
688.Ox 3.0 .
689They first appeared in
690.Fx 5.3
691ported from the version in
692.Ox 3.5
693