1 .\" $OpenBSD: pfctl.8,v 1.139 2008/06/11 07:23:36 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.
32 .Nd control the packet filter (PF) device
38 .Oo Fl D Ar macro Ns =
43 .Op Fl K Ar host | network
46 .Ar host | network | label | id
61 utility communicates with the packet filter device using the
62 ioctl interface described in
64 It allows ruleset and parameter configuration and retrieval of status
65 information from the packet filter.
67 Packet filtering restricts the types of packets that pass through
68 network interfaces entering or leaving the host based on filter
71 The packet filter can also replace addresses and ports of packets.
72 Replacing source addresses and ports of outgoing packets is called
73 NAT (Network Address Translation) and is used to connect an internal
74 network (usually reserved address space) to an external one (the
75 Internet) by making all connections to external hosts appear to
76 come from the gateway.
77 Replacing destination addresses and ports of incoming packets
78 is used to redirect connections to different hosts and/or ports.
79 A combination of both translations, bidirectional NAT, is also
81 Translation rules are described in
90 the rule file specified with the variable
92 is loaded automatically by the
94 scripts and the packet filter is enabled.
96 The packet filter does not itself forward packets between interfaces.
97 Forwarding can be enabled by setting the
100 .Em net.inet.ip.forwarding
102 .Em net.inet6.ip6.forwarding
104 Set them permanently in
109 utility provides several commands.
110 The options are as follows:
113 Load only the queue rules present in the rule file.
114 Other rules and options are ignored.
121 only to the rules in the specified
123 In addition to the main ruleset,
125 can load and manipulate additional rulesets by name,
127 The main ruleset is the default anchor.
129 Anchors are referenced by name and may be nested,
130 with the various components of the anchor path separated by
132 characters, similar to how file system hierarchies are laid out.
133 The last component of the anchor path is where ruleset operations are
138 rules from the main ruleset is described in
141 For example, the following will show all filter rules (see the
143 flag below) inside the anchor
144 .Dq authpf/smith(1234) ,
145 which would have been created for user
150 .Bd -literal -offset indent
151 # pfctl -a "authpf/smith(1234)" -s rules
154 Private tables can also be put inside anchors, either by having table
157 file that is loaded in the anchor, or by using regular table commands, as in:
158 .Bd -literal -offset indent
159 # pfctl -a foo/bar -t mytable -T add 1.2.3.4 5.6.7.8
162 When a rule referring to a table is loaded in an anchor, the rule will use the
163 private table if one is defined, and then fall back to the table defined in the
164 main ruleset, if there is one.
165 This is similar to C rules for variable scope.
166 It is possible to create distinct tables with the same name in the global
167 ruleset and in an anchor, but this is often bad design and a warning will be
170 By default, recursive inline printing of anchors applies only to unnamed
171 anchors specified inline in the ruleset.
172 If the anchor name is terminated with a
176 flag will recursively print all anchors in a brace delimited block.
177 For example the following will print the
180 .Bd -literal -offset indent
181 # pfctl -a 'authpf/*' -sr
184 To print the main ruleset recursively, specify only
187 .Bd -literal -offset indent
190 .It Fl D Ar macro Ns = Ns Ar value
196 Overrides the definition of
200 Disable the packet filter.
202 Enable the packet filter.
204 Flush the filter parameters specified by
206 (may be abbreviated):
208 .Bl -tag -width xxxxxxxxxxxx -compact
212 Flush the queue rules.
214 Flush the filter rules.
216 Flush the state table (NAT and filter).
218 Flush the source tracking table.
220 Flush the filter information (statistics that are not bound to rules).
224 Flush the passive operating system fingerprints.
226 Flush all of the above.
229 Load the rules contained in
233 may contain macros, tables, options, and normalization, queueing,
234 translation, and filtering rules.
235 With the exception of macros and tables, the statements must appear in that
238 Include output helpful for debugging.
241 .It Fl i Ar interface
242 Restrict the operation to the given
244 .It Fl K Ar host | network
245 Kill all of the source tracking entries originating from the specified
253 option may be specified, which will kill all the source tracking
254 entries from the first host/network to the second.
257 .Ar host | network | label | id
259 Kill all of the state entries matching the specified
266 For example, to kill all of the state entries originating from
275 option may be specified, which will kill all the state entries
276 from the first host/network to the second.
277 To kill all of the state entries from
282 .Dl # pfctl -k host1 -k host2
284 To kill all states originating from 192.168.1.0/24 to 172.16.0.0/16:
286 .Dl # pfctl -k 192.168.1.0/24 -k 172.16.0.0/16
288 A network prefix length of 0 can be used as a wildcard.
289 To kill all states with the target
292 .Dl # pfctl -k 0.0.0.0/0 -k host2
294 It is also possible to kill states by rule label or state ID.
295 In this mode the first
297 argument is used to specify the type
298 of the second argument.
299 The following command would kill all states that have been created
300 from rules carrying the label
303 .Dl # pfctl -k label -k foobar
305 To kill one specific state by its unique state ID
306 (as shown by pfctl -s state -vv),
309 modifier and as a second argument the state ID and optional creator ID.
310 To kill a state with ID 4823e84500000003 use:
312 .Dl # pfctl -k id -k 4823e84500000003
314 To kill a state with ID 4823e84500000018 created from a backup
315 firewall with hostid 00000002 use:
317 .Dl # pfctl -k id -k 4823e84500000018/2
320 Merge in explicitly given options without resetting those
322 Allows single options to be modified without disturbing the others:
323 .Bd -literal -offset indent
324 # echo "set loginterface fxp0" | pfctl -mf -
327 Load only the NAT rules present in the rule file.
328 Other rules and options are ignored.
330 Do not actually load rules, just parse them.
332 Load only the options present in the rule file.
333 Other rules and options are ignored.
335 Control the ruleset optimizer, overriding any rule file settings.
337 .Bl -tag -width xxxxxxxxxxxx -compact
339 Disable the ruleset optimizer.
341 Enable basic ruleset optimizations.
342 This is the default behaviour.
344 Enable basic ruleset optimizations with profiling.
346 For further information on the ruleset optimizer, see
351 instead of the default
354 Only print errors and warnings.
356 Load only the filter rules present in the rule file.
357 Other rules and options are ignored.
359 Perform reverse DNS lookups on states when displaying them.
361 Show the filter parameters specified by
363 (may be abbreviated):
365 .Bl -tag -width xxxxxxxxxxxxx -compact
367 Show the currently loaded NAT rules.
369 Show the currently loaded queue rules.
370 When used together with
372 per-queue statistics are also shown.
373 When used together with
376 will loop and show updated queue statistics every five seconds, including
377 measured bandwidth and packets per second.
379 Show the currently loaded filter rules.
380 When used together with
382 the per-rule statistics (number of evaluations,
383 packets and bytes) are also shown.
386 optimization done automatically by the kernel
387 will skip evaluation of rules where possible.
388 Packets passed statefully are counted in the rule that created the state
389 (even though the rule isn't evaluated more than once for the entire
392 Show the currently loaded anchors directly attached to the main ruleset.
395 is specified as well, the anchors loaded directly below the given
400 is specified, all anchors attached under the target anchor will be
401 displayed recursively.
403 Show the contents of the state table.
405 Show the contents of the source tracking table.
407 Show filter information (statistics and counters).
408 When used together with
410 source tracking statistics are also shown.
412 Show per-rule statistics (label, evaluations, packets total, bytes total,
413 packets in, bytes in, packets out, bytes out, state creations) of
414 filter rules with labels, useful for accounting.
416 Show the current global timeouts.
418 Show the current pool memory hard limits.
420 Show the list of tables.
422 Show the list of operating system fingerprints.
423 .It Fl s Cm Interfaces
424 Show the list of interfaces and interface drivers available to PF.
425 When used together with
427 it additionally lists which interfaces have skip rules activated.
428 When used together with
430 interface statistics are also shown.
432 can be used to select an interface or a group of interfaces.
434 Show all of the above, except for the lists of interfaces and operating
437 .It Fl T Ar command Op Ar address ...
440 (may be abbreviated) to apply to the table.
443 .Bl -tag -width xxxxxxxxxxxx -compact
447 Flush all addresses of a table.
449 Add one or more addresses in a table.
450 Automatically create a nonexisting table.
452 Delete one or more addresses from a table.
453 .It Fl T Cm expire Ar number
454 Delete addresses which had their statistics cleared more than
457 For entries which have never had their statistics cleared,
459 refers to the time they were added to the table.
461 Replace the addresses of the table.
462 Automatically create a nonexisting table.
464 Show the content (addresses) of a table.
466 Test if the given addresses match a table.
468 Clear all the statistics of a table.
470 Load only the table definitions from
472 This is used in conjunction with the
475 .Bd -literal -offset indent
476 # pfctl -Tl -f pf.conf
486 commands, the list of addresses can be specified either directly on the command
487 line and/or in an unformatted text file, using the
490 Comments starting with a
492 are allowed in the text file.
493 With these commands, the
495 flag can also be used once or twice, in which case
498 detailed result of the operation for each individual address, prefixed by
499 one of the following letters:
501 .Bl -tag -width XXX -compact
503 The address/network has been added.
505 The address/network has been changed (negated).
507 The address/network has been deleted.
515 The address/network is duplicated and therefore ignored.
517 The address/network cannot be added/deleted due to conflicting
521 The address/network has been cleared (statistics).
524 Each table can maintain a set of counters that can be retrieved using the
528 For example, the following commands define a wide open firewall which will keep
529 track of packets going to or coming from the
532 The following commands configure the firewall and send 10 pings to the FTP
534 .Bd -literal -offset indent
535 # printf "table <test> counters { ftp.openbsd.org }\en \e
536 pass out to <test>\en" | pfctl -f-
537 # ping -qc10 ftp.openbsd.org
540 We can now use the table
542 command to output, for each address and packet direction, the number of packets
543 and bytes that are being passed or blocked by rules referencing the table.
544 The time at which the current accounting started is also shown with the
547 .Bd -literal -offset indent
548 # pfctl -t test -vTshow
550 Cleared: Thu Feb 13 18:55:18 2003
551 In/Block: [ Packets: 0 Bytes: 0 ]
552 In/Pass: [ Packets: 10 Bytes: 840 ]
553 Out/Block: [ Packets: 0 Bytes: 0 ]
554 Out/Pass: [ Packets: 10 Bytes: 840 ]
557 Similarly, it is possible to view global information about the tables
560 modifier twice and the
564 This will display the number of addresses on each table,
565 the number of rules which reference the table, and the global
566 packet statistics for the whole table:
567 .Bd -literal -offset indent
571 Cleared: Thu Feb 13 18:55:18 2003
572 References: [ Anchors: 0 Rules: 1 ]
573 Evaluations: [ NoMatch: 3496 Match: 1 ]
574 In/Block: [ Packets: 0 Bytes: 0 ]
575 In/Pass: [ Packets: 10 Bytes: 840 ]
576 In/XPass: [ Packets: 0 Bytes: 0 ]
577 Out/Block: [ Packets: 0 Bytes: 0 ]
578 Out/Pass: [ Packets: 10 Bytes: 840 ]
579 Out/XPass: [ Packets: 0 Bytes: 0 ]
582 As we can see here, only one packet \- the initial ping request \- matched the
583 table, but all packets passing as the result of the state are correctly
585 Reloading the table(s) or ruleset will not affect packet accounting in any way.
588 counters are incremented instead of the
592 packet is passed but doesn't match the table anymore.
593 This will happen in our example if someone flushes the table while the
597 When used with a single
600 will only display the first line containing the table flags and name.
601 The flags are defined as follows:
603 .Bl -tag -width XXX -compact
605 For constant tables, which cannot be altered outside
608 For persistent tables, which don't get automatically killed when no rules
611 For tables which are part of the
614 Tables without this flag do not really exist, cannot contain addresses, and are
619 For tables which are part of the
622 This flag can only be witnessed briefly during the loading of
625 For tables which are referenced (used) by rules.
627 This flag is set when a table in the main ruleset is hidden by one or more
628 tables of the same name from anchors attached below it.
630 This flag is set when per-address counters are enabled on the table.
633 Specify the name of the table.
635 Produce more verbose output.
638 will produce even more verbose output including ruleset warnings.
639 See the previous section for its effect on table commands.
643 (may be abbreviated) to one of the following:
645 .Bl -tag -width xxxxxxxxxxxx -compact
647 Don't generate debug messages.
649 Generate debug messages only for serious errors.
651 Generate debug messages for various errors.
653 Generate debug messages for common conditions.
656 Clear per-rule statistics.
659 .Bl -tag -width "/etc/pf.conf" -compact
661 Packet filter rules file.
663 Passive operating system fingerprint database.
680 filter mechanism first appeared in