1 .\" $OpenBSD: pfctl.8,v 1.128 2007/01/30 21:01:56 jmc Exp $
3 .\" Copyright (c) 2001 Kjell Wooding. All rights reserved.
5 .\" Redistribution and use in source and binary forms, with or without
6 .\" modification, are permitted provided that the following conditions
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.
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.
34 .Nd "control the packet filter (PF) and network address translation (NAT) device"
40 .Oo Fl D Ar macro Ns =
45 .Op Fl K Ar host | network
46 .Op Fl k Ar host | network
60 utility communicates with the packet filter device using the
61 ioctl interface described in
63 It allows ruleset and parameter configuration and retrieval of status
64 information from the packet filter.
66 Packet filtering restricts the types of packets that pass through
67 network interfaces entering or leaving the host based on filter
70 The packet filter can also replace addresses and ports of packets.
71 Replacing source addresses and ports of outgoing packets is called
72 NAT (Network Address Translation) and is used to connect an internal
73 network (usually reserved address space) to an external one (the
74 Internet) by making all connections to external hosts appear to
75 come from the gateway.
76 Replacing destination addresses and ports of incoming packets
77 is used to redirect connections to different hosts and/or ports.
78 A combination of both translations, bidirectional NAT, is also
80 Translation rules are described in
89 the rule file specified with the variable
91 is loaded automatically by the
93 scripts and the packet filter is enabled.
95 The packet filter does not itself forward packets between interfaces.
96 Forwarding can be enabled by setting the
99 .Em net.inet.ip.forwarding
101 .Em net.inet6.ip6.forwarding
103 Set them permanently in
108 utility provides several commands.
109 The options are as follows:
112 Load only the queue rules present in the rule file.
113 Other rules and options are ignored.
120 only to the rules in the specified
122 In addition to the main ruleset,
124 can load and manipulate additional rulesets by name,
126 The main ruleset is the default anchor.
128 Anchors are referenced by name and may be nested,
129 with the various components of the anchor path separated by
131 characters, similar to how file system hierarchies are laid out.
132 The last component of the anchor path is where ruleset operations are
137 rules from the main ruleset is described in
140 For example, the following will show all filter rules (see the
142 flag below) inside the anchor
143 .Dq authpf/smith(1234) ,
144 which would have been created for user
149 .Bd -literal -offset indent
150 # pfctl -a "authpf/smith(1234)" -s rules
153 Private tables can also be put inside anchors, either by having table
156 file that is loaded in the anchor, or by using regular table commands, as in:
157 .Bd -literal -offset indent
158 # pfctl -a foo/bar -t mytable -T add 1.2.3.4 5.6.7.8
161 When a rule referring to a table is loaded in an anchor, the rule will use the
162 private table if one is defined, and then fall back to the table defined in the
163 main ruleset, if there is one.
164 This is similar to C rules for variable scope.
165 It is possible to create distinct tables with the same name in the global
166 ruleset and in an anchor, but this is often bad design and a warning will be
169 By default, recursive inline printing of anchors applies only to unnamed
170 anchors specified inline in the ruleset.
171 If the anchor name is terminated with a
175 flag will recursively print all anchors in a brace delimited block.
176 For example the following will print the
179 .Bd -literal -offset indent
180 # pfctl -a 'authpf/*' -sr
183 To print the main ruleset recursively, specify only
186 .Bd -literal -offset indent
189 .It Fl D Ar macro Ns = Ns Ar value
195 Overrides the definition of
199 Disable the packet filter.
201 Enable the packet filter.
203 Flush the filter parameters specified by
205 (may be abbreviated):
207 .Bl -tag -width xxxxxxxxxxxx -compact
211 Flush the queue rules.
213 Flush the filter rules.
215 Flush the state table (NAT and filter).
217 Flush the source tracking table.
219 Flush the filter information (statistics that are not bound to rules).
223 Flush the passive operating system fingerprints.
225 Flush all of the above.
228 Load the rules contained in
232 may contain macros, tables, options, and normalization, queueing,
233 translation, and filtering rules.
234 With the exception of macros and tables, the statements must appear in that
237 Include output helpful for debugging.
240 .It Fl i Ar interface
241 Restrict the operation to the given
243 .It Fl K Ar host | network
244 Kill all of the source tracking entries originating from the specified
252 option may be specified, which will kill all the source tracking
253 entries from the first host/network to the second.
254 .It Fl k Ar host | network
255 Kill all of the state entries originating from the specified
263 option may be specified, which will kill all the state entries
264 from the first host/network to the second.
265 For example, to kill all of the state entries originating from
270 To kill all of the state entries from
275 .Dl # pfctl -k host1 -k host2
277 To kill all states originating from 192.168.1.0/24 to 172.16.0.0/16:
279 .Dl # pfctl -k 192.168.1.0/24 -k 172.16.0.0/16
281 A network prefix length of 0 can be used as a wildcard.
282 To kill all states with the target
285 .Dl # pfctl -k 0.0.0.0/0 -k host2
287 Merge in explicitly given options without resetting those
289 Allows single options to be modified without disturbing the others:
290 .Bd -literal -offset indent
291 # echo "set loginterface fxp0" | pfctl -mf -
294 Load only the NAT rules present in the rule file.
295 Other rules and options are ignored.
297 Do not actually load rules, just parse them.
299 Load only the options present in the rule file.
300 Other rules and options are ignored.
302 Control the ruleset optimizer.
303 The ruleset optimizer attempts to improve rulesets by removing rule
304 duplication and making better use of rule ordering.
306 .Bl -tag -width xxxxxxxxxxxx -compact
308 Disable the ruleset optimizer.
310 Enable basic ruleset optimizations.
312 Enable basic ruleset optimizations with profiling.
316 optimization does does four things:
320 remove duplicate rules
322 remove rules that are a subset of another rule
324 combine multiple rules into a table when advantageous
326 re-order the rules to improve evaluation performance
331 is specified, the currently loaded ruleset will be examined as a feedback
332 profile to tailor the optimization of the
334 rules to the actual network behavior.
336 It is important to note that the ruleset optimizer will modify the ruleset
337 to improve performance.
338 A side effect of the ruleset modification is that per-rule accounting
339 statistics will have different meanings than before.
340 If per-rule accounting is important for billing purposes or whatnot, either
341 the ruleset optimizer should not be used or a
343 field should be added to all of the accounting rules to act as optimization
346 To retain compatibility with previous behaviour, a single
348 without any options will enable
350 optimizations, and a second
352 will enable profiling.
356 instead of the default
359 Only print errors and warnings.
361 Load only the filter rules present in the rule file.
362 Other rules and options are ignored.
364 Perform reverse DNS lookups on states when displaying them.
366 Show the filter parameters specified by
368 (may be abbreviated):
370 .Bl -tag -width xxxxxxxxxxxxx -compact
372 Show the currently loaded NAT rules.
374 Show the currently loaded queue rules.
375 When used together with
377 per-queue statistics are also shown.
378 When used together with
381 will loop and show updated queue statistics every five seconds, including
382 measured bandwidth and packets per second.
384 Show the currently loaded filter rules.
385 When used together with
387 the per-rule statistics (number of evaluations,
388 packets and bytes) are also shown.
391 optimization done automatically by the kernel
392 will skip evaluation of rules where possible.
393 Packets passed statefully are counted in the rule that created the state
394 (even though the rule isn't evaluated more than once for the entire
397 Show the currently loaded anchors directly attached to the main ruleset.
400 is specified as well, the anchors loaded directly below the given
405 is specified, all anchors attached under the target anchor will be
406 displayed recursively.
408 Show the contents of the state table.
410 Show the contents of the source tracking table.
412 Show filter information (statistics and counters).
413 When used together with
415 source tracking statistics are also shown.
417 Show per-rule statistics (label, evaluations, packets total, bytes total,
418 packets in, bytes in, packets out, bytes out) of
419 filter rules with labels, useful for accounting.
421 Show the current global timeouts.
423 Show the current pool memory hard limits.
425 Show the list of tables.
427 Show the list of operating system fingerprints.
428 .It Fl s Cm Interfaces
429 Show the list of interfaces and interface drivers available to PF.
430 When used together with
432 it additionally lists which interfaces have skip rules activated.
433 When used together with
435 interface statistics are also shown.
437 can be used to select an interface or a group of interfaces.
439 Show all of the above, except for the lists of interfaces and operating
442 .It Fl T Ar command Op Ar address ...
445 (may be abbreviated) to apply to the table.
448 .Bl -tag -width xxxxxxxxxxxx -compact
452 Flush all addresses of a table.
454 Add one or more addresses in a table.
455 Automatically create a nonexisting table.
457 Delete one or more addresses from a table.
458 .It Fl T Cm expire Ar number
459 Delete addresses which had their statistics cleared more than
462 For entries which have never had their statistics cleared,
464 refers to the time they were added to the table.
466 Replace the addresses of the table.
467 Automatically create a nonexisting table.
469 Show the content (addresses) of a table.
471 Test if the given addresses match a table.
473 Clear all the statistics of a table.
475 Load only the table definitions from
477 This is used in conjunction with the
480 .Bd -literal -offset indent
481 # pfctl -Tl -f pf.conf
491 commands, the list of addresses can be specified either directly on the command
492 line and/or in an unformatted text file, using the
495 Comments starting with a
497 are allowed in the text file.
498 With these commands, the
500 flag can also be used once or twice, in which case
503 detailed result of the operation for each individual address, prefixed by
504 one of the following letters:
506 .Bl -tag -width XXX -compact
508 The address/network has been added.
510 The address/network has been changed (negated).
512 The address/network has been deleted.
520 The address/network is duplicated and therefore ignored.
522 The address/network cannot be added/deleted due to conflicting
526 The address/network has been cleared (statistics).
529 Each table maintains a set of counters that can be retrieved using the
533 For example, the following commands define a wide open firewall which will keep
534 track of packets going to or coming from the
537 The following commands configure the firewall and send 10 pings to the FTP
539 .Bd -literal -offset indent
540 # printf "table <test> { ftp.openbsd.org }\en \e
541 pass out to <test>\en" | pfctl -f-
542 # ping -qc10 ftp.openbsd.org
545 We can now use the table
547 command to output, for each address and packet direction, the number of packets
548 and bytes that are being passed or blocked by rules referencing the table.
549 The time at which the current accounting started is also shown with the
552 .Bd -literal -offset indent
553 # pfctl -t test -vTshow
555 Cleared: Thu Feb 13 18:55:18 2003
556 In/Block: [ Packets: 0 Bytes: 0 ]
557 In/Pass: [ Packets: 10 Bytes: 840 ]
558 Out/Block: [ Packets: 0 Bytes: 0 ]
559 Out/Pass: [ Packets: 10 Bytes: 840 ]
562 Similarly, it is possible to view global information about the tables
565 modifier twice and the
569 This will display the number of addresses on each table,
570 the number of rules which reference the table, and the global
571 packet statistics for the whole table:
572 .Bd -literal -offset indent
576 Cleared: Thu Feb 13 18:55:18 2003
577 References: [ Anchors: 0 Rules: 1 ]
578 Evaluations: [ NoMatch: 3496 Match: 1 ]
579 In/Block: [ Packets: 0 Bytes: 0 ]
580 In/Pass: [ Packets: 10 Bytes: 840 ]
581 In/XPass: [ Packets: 0 Bytes: 0 ]
582 Out/Block: [ Packets: 0 Bytes: 0 ]
583 Out/Pass: [ Packets: 10 Bytes: 840 ]
584 Out/XPass: [ Packets: 0 Bytes: 0 ]
587 As we can see here, only one packet \- the initial ping request \- matched the
588 table, but all packets passing as the result of the state are correctly
590 Reloading the table(s) or ruleset will not affect packet accounting in any way.
593 counters are incremented instead of the
597 packet is passed but doesn't match the table anymore.
598 This will happen in our example if someone flushes the table while the
602 When used with a single
605 will only display the first line containing the table flags and name.
606 The flags are defined as follows:
608 .Bl -tag -width XXX -compact
610 For constant tables, which cannot be altered outside
613 For persistent tables, which don't get automatically killed when no rules
616 For tables which are part of the
619 Tables without this flag do not really exist, cannot contain addresses, and are
624 For tables which are part of the
627 This flag can only be witnessed briefly during the loading of
630 For tables which are referenced (used) by rules.
632 This flag is set when a table in the main ruleset is hidden by one or more
633 tables of the same name from anchors attached below it.
636 Specify the name of the table.
638 Produce more verbose output.
641 will produce even more verbose output including ruleset warnings.
642 See the previous section for its effect on table commands.
646 (may be abbreviated) to one of the following:
648 .Bl -tag -width xxxxxxxxxxxx -compact
650 Don't generate debug messages.
652 Generate debug messages only for serious errors.
654 Generate debug messages for various errors.
656 Generate debug messages for common conditions.
659 Clear per-rule statistics.
662 .Bl -tag -width "/etc/pf.conf" -compact
664 Packet filter rules file.
666 Passive operating system fingerprint database.
683 filter mechanism first appeared in