.\" $OpenBSD: pf.4,v 1.51 2004/06/14 20:55:54 cedric Exp $ .\" .\" Copyright (C) 2001, Kjell Wooding. All rights reserved. .\" .\" Redistribution and use in source and binary forms, with or without .\" modification, are permitted provided that the following conditions .\" are met: .\" 1. Redistributions of source code must retain the above copyright .\" notice, this list of conditions and the following disclaimer. .\" 2. Redistributions in binary form must reproduce the above copyright .\" notice, this list of conditions and the following disclaimer in the .\" documentation and/or other materials provided with the distribution. .\" 3. Neither the name of the project nor the names of its contributors .\" may be used to endorse or promote products derived from this software .\" without specific prior written permission. .\" .\" THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE .\" ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF .\" SUCH DAMAGE. .\" .Dd June 24, 2001 .Dt PF 4 .Os .Sh NAME .Nm pf .Nd packet filter .Sh SYNOPSIS .Cd "pseudo-device pf" .Sh DESCRIPTION Packet filtering takes place in the kernel. A pseudo-device, .Pa /dev/pf , allows userland processes to control the behavior of the packet filter through an .Xr ioctl 2 interface. There are commands to enable and disable the filter, load rulesets, add and remove individual rules or state table entries, and retrieve statistics. The most commonly used functions are covered by .Xr pfctl 8 . .Pp Manipulations like loading a ruleset that involve more than a single ioctl call require a so-called ticket, which prevents the occurrence of multiple concurrent manipulations. .Pp Fields of ioctl parameter structures that refer to packet data (like addresses and ports) are generally expected in network byte-order. .Pp Anchor and ruleset manipulations can be supplied with empty strings for these parameters, in which case modifications will manipulate the .Dq main packet filter ruleset. .Sh IOCTL INTERFACE .Nm supports the following .Xr ioctl 2 commands, available through .Aq Pa net/pfvar.h : .Bl -tag -width xxxxxx .It Dv DIOCSTART Start the packet filter. .It Dv DIOCSTOP Stop the packet filter. .It Dv DIOCSTARTALTQ Start the ALTQ bandwidth control system (see .Xr altq 9 ) . .It Dv DIOCSTOPALTQ Stop the ALTQ bandwidth control system. .It Dv DIOCBEGINADDRS Fa "struct pfioc_pooladdr *pp" .Bd -literal struct pfioc_pooladdr { u_int32_t action; u_int32_t ticket; u_int32_t nr; u_int32_t r_num; u_int8_t r_action; u_int8_t r_last; u_int8_t af; char anchor[PF_ANCHOR_NAME_SIZE]; char ruleset[PF_RULESET_NAME_SIZE]; struct pf_pooladdr addr; }; .Ed .Pp Clear the buffer address pool and get a .Va ticket for subsequent .Dv DIOCADDADDR , .Dv DIOCADDRULE , and .Dv DIOCCHANGERULE calls. .It Dv DIOCADDADDR Fa "struct pfioc_pooladdr *pp" .Pp Add the pool address .Va addr to the buffer address pool to be used in the following .Dv DIOCADDRULE or .Dv DIOCCHANGERULE call. All other members of the structure are ignored. .It Dv DIOCADDRULE Fa "struct pfioc_rule *pr" .Bd -literal struct pfioc_rule { u_int32_t action; u_int32_t ticket; u_int32_t pool_ticket; u_int32_t nr; char anchor[PF_ANCHOR_NAME_SIZE]; char ruleset[PF_RULESET_NAME_SIZE]; struct pf_rule rule; }; .Ed .Pp Add .Va rule at the end of the inactive ruleset. This call requires a .Va ticket obtained through a preceding .Dv DIOCXBEGIN call and a .Va pool_ticket obtained through a .Dv DIOCBEGINADDRS call. .Dv DIOCADDADDR must also be called if any pool addresses are required. The optional .Va anchor and .Va ruleset names indicate the anchor and ruleset in which to append the rule. .Va nr and .Va action are ignored. .It Dv DIOCADDALTQ Fa "struct pfioc_altq *pa" Add an ALTQ discipline or queue. .Bd -literal struct pfioc_altq { u_int32_t action; u_int32_t ticket; u_int32_t nr; struct pf_altq altq; }; .Ed .It Dv DIOCGETRULES Fa "struct pfioc_rule *pr" Get a .Va ticket for subsequent .Dv DIOCGETRULE calls and the number .Va nr of rules in the active ruleset. .It Dv DIOCGETRULE Fa "struct pfioc_rule *pr" Get a .Va rule by its number .Va nr using the .Va ticket obtained through a preceding .Dv DIOCGETRULES call. .It Dv DIOCGETADDRS Fa "struct pfioc_pooladdr *pp" Get a .Va ticket for subsequent .Dv DIOCGETADDR calls and the number .Va nr of pool addresses in the rule specified with .Va r_action , .Va r_num , .Va anchor , and .Va ruleset . .It Dv DIOCGETADDR Fa "struct pfioc_pooladdr *pp" Get the pool address .Va addr by its number .Va nr from the rule specified with .Va r_action , .Va r_num , .Va anchor , and .Va ruleset using the .Va ticket obtained through a preceding .Dv DIOCGETADDRS call. .It Dv DIOCGETALTQS Fa "struct pfioc_altq *pa" Get a .Va ticket for subsequent .Dv DIOCGETALTQ calls and the number .Va nr of queues in the active list. .It Dv DIOCGETALTQ Fa "struct pfioc_altq *pa" Get the queueing discipline .Va altq by its number .Va nr using the .Va ticket obtained through a preceding .Dv DIOCGETALTQS call. .It Dv DIOCGETQSTATS Fa "struct pfioc_qstats *pq" Get the statistics on a queue. .Bd -literal struct pfioc_qstats { u_int32_t ticket; u_int32_t nr; void *buf; int nbytes; u_int8_t scheduler; }; .Ed .Pp This call fills in a pointer to the buffer of statistics .Va buf , of length .Va nbytes , for the queue specified by .Va nr . .It Dv DIOCGETANCHORS Fa "struct pfioc_anchor *pa" .Bd -literal struct pfioc_anchor { u_int32_t nr; char name[PF_ANCHOR_NAME_SIZE]; }; .Ed .Pp Get the number .Va nr of anchors in the active list for use in subsequent .Dv DIOCGETANCHOR calls. .It Dv DIOCGETANCHOR Fa "struct pfioc_anchor *pa" Get an anchor by its number .Va nr , the maximum number of which can be obtained from a preceding .Dv DIOCGETANCHORS call. This ioctl returns .Er EBUSY if another process is concurrently updating an anchor. .It Dv DIOCGETRULESETS Fa "struct pfioc_ruleset *pr" .Bd -literal struct pfioc_ruleset { u_int32_t nr; char anchor[PF_ANCHOR_NAME_SIZE]; char name[PF_RULESET_NAME_SIZE]; }; .Ed .Pp Get the number .Va nr of rulesets in the given .Va anchor for use in subsequent .Dv DIOCGETRULESET calls. This ioctl returns .Er EINVAL if the given anchor does not exist. .It Dv DIOCGETRULESET Fa "struct pfioc_ruleset *pr" Get a ruleset by its number .Va nr from the given .Va anchor , the maximum number of which can be obtained from a preceding .Dv DIOCGETRULESETS call. This ioctl returns .Er EINVAL if the given anchor does not exist or .Er EBUSY if another process is concurrently updating a ruleset. .It Dv DIOCADDSTATE Fa "struct pfioc_state *ps" Add a state entry. .Bd -literal struct pfioc_state { u_int32_t nr; struct pf_state state; }; .Ed .It Dv DIOCGETSTATE Fa "struct pfioc_state *ps" Extract the entry with the specified number .Va nr from the state table. .It Dv DIOCKILLSTATES Fa "struct pfioc_state_kill *psk" Remove matching entries from the state table. This ioctl returns the number of killed states in .Va psk_af . .Bd -literal struct pfioc_state_kill { sa_family_t psk_af; int psk_proto; struct pf_rule_addr psk_src; struct pf_rule_addr psk_dst; char psk_ifname[IFNAMSIZ]; }; .Ed .It Dv DIOCCLRSTATES Fa "struct pfioc_state_kill *psk" Clear all states. It works like .Dv DIOCKILLSTATES , but ignores the .Va psk_af , .Va psk_proto , .Va psk_src , and .Va psk_dst fields of the .Va pfioc_state_kill structure. .It Dv DIOCSETSTATUSIF Fa "struct pfioc_if *pi" Specify the interface for which statistics are accumulated. .Bd -literal struct pfioc_if { char ifname[IFNAMSIZ]; }; .Ed .It Dv DIOCGETSTATUS Fa "struct pf_status *s" Get the internal packet filter statistics. .Bd -literal struct pf_status { u_int64_t counters[PFRES_MAX]; u_int64_t fcounters[FCNT_MAX]; u_int64_t scounters[SCNT_MAX]; u_int64_t pcounters[2][2][3]; u_int64_t bcounters[2][2]; u_int64_t stateid; u_int32_t running; u_int32_t states; u_int32_t src_nodes; u_int32_t since; u_int32_t debug; u_int32_t hostid; char ifname[IFNAMSIZ]; }; .Ed .It Dv DIOCCLRSTATUS Clear the internal packet filter statistics. .It Dv DIOCNATLOOK Fa "struct pfioc_natlook *pnl" Look up a state table entry by source and destination addresses and ports. .Bd -literal struct pfioc_natlook { struct pf_addr saddr; struct pf_addr daddr; struct pf_addr rsaddr; struct pf_addr rdaddr; u_int16_t sport; u_int16_t dport; u_int16_t rsport; u_int16_t rdport; sa_family_t af; u_int8_t proto; u_int8_t direction; }; .Ed .It Dv DIOCSETDEBUG Fa "u_int32_t *level" Set the debug level. .Bd -literal enum { PF_DEBUG_NONE, PF_DEBUG_URGENT, PF_DEBUG_MISC, PF_DEBUG_NOISY }; .Ed .It Dv DIOCGETSTATES Fa "struct pfioc_states *ps" Get state table entries. .Bd -literal struct pfioc_states { int ps_len; union { caddr_t psu_buf; struct pf_state *psu_states; } ps_u; #define ps_buf ps_u.psu_buf #define ps_states ps_u.psu_states }; .Ed .Pp If .Va ps_len is zero, all states will be gathered into .Va pf_states and .Va ps_len will be set to the size they take in memory (i.e., .Li sizeof(struct pf_state) * nr ) . If .Va ps_len is non-zero, as many states that can fit into .Va ps_len as possible will be gathered, and .Va ps_len will be updated to the size those rules take in memory. .It Dv DIOCCHANGERULE Fa "struct pfioc_rule *pcr" Add or remove the .Va rule in the ruleset specified by .Va rule.action . .Pp The type of operation to be performed is indicated by .Va action , which can be any of the following: .Bd -literal enum { PF_CHANGE_NONE, PF_CHANGE_ADD_HEAD, PF_CHANGE_ADD_TAIL, PF_CHANGE_ADD_BEFORE, PF_CHANGE_ADD_AFTER, PF_CHANGE_REMOVE, PF_CHANGE_GET_TICKET }; .Ed .Pp .Va ticket must be set to the value obtained with .Dv PF_CHANGE_GET_TICKET for all actions except .Dv PF_CHANGE_GET_TICKET . .Va pool_ticket must be set to the value obtained with the .Dv DIOCBEGINADDRS call for all actions except .Dv PF_CHANGE_REMOVE and .Dv PF_CHANGE_GET_TICKET . .Pp .Va anchor and .Va ruleset indicate to which anchor and ruleset the operation applies. .Va nr indicates the rule number against which .Dv PF_CHANGE_ADD_BEFORE , .Dv PF_CHANGE_ADD_AFTER , or .Dv PF_CHANGE_REMOVE actions are applied. .\" It Dv DIOCCHANGEALTQ Fa "struct pfioc_altq *pcr" .It Dv DIOCCHANGEADDR Fa "struct pfioc_pooladdr *pca" Add or remove the pool address .Va addr from the rule specified by .Va r_action , .Va r_num , .Va anchor , and .Va ruleset . .It Dv DIOCSETTIMEOUT Fa "struct pfioc_tm *pt" .Bd -literal struct pfioc_tm { int timeout; int seconds; }; .Ed .Pp Set the state timeout of .Va timeout to .Va seconds . The old value will be placed into .Va seconds . For possible values of .Va timeout , consult the .Dv PFTM_* values in .Aq Pa net/pfvar.h . .It Dv DIOCGETTIMEOUT Fa "struct pfioc_tm *pt" Get the state timeout of .Va timeout . The value will be placed into the .Va seconds field. .It Dv DIOCCLRRULECTRS Clear per-rule statistics. .It Dv DIOCSETLIMIT Fa "struct pfioc_limit *pl" Set the hard limits on the memory pools used by the packet filter. .Bd -literal struct pfioc_limit { int index; unsigned limit; }; enum { PF_LIMIT_STATES, PF_LIMIT_SRC_NODES, PF_LIMIT_FRAGS }; .Ed .It Dv DIOCGETLIMIT Fa "struct pfioc_limit *pl" Get the hard .Va limit for the memory pool indicated by .Va index . .It Dv DIOCRCLRTABLES Fa "struct pfioc_table *io" Clear all tables. All the ioctls that manipulate radix tables use the same structure described below. For .Dv DIOCRCLRTABLES , .Va pfrio_ndel contains on exit the number of tables deleted. .Bd -literal struct pfioc_table { struct pfr_table pfrio_table; void *pfrio_buffer; int pfrio_esize; int pfrio_size; int pfrio_size2; int pfrio_nadd; int pfrio_ndel; int pfrio_nchange; int pfrio_flags; u_int32_t pfrio_ticket; }; #define pfrio_exists pfrio_nadd #define pfrio_nzero pfrio_nadd #define pfrio_nmatch pfrio_nadd #define pfrio_naddr pfrio_size2 #define pfrio_setflag pfrio_size2 #define pfrio_clrflag pfrio_nadd .Ed .It Dv DIOCRADDTABLES Fa "struct pfioc_table *io" Create one or more tables. On entry, .Va pfrio_buffer[pfrio_size] contains a table of .Va pfr_table structures. On exit, .Va pfrio_nadd contains the number of tables effectively created. .Bd -literal struct pfr_table { char pfrt_anchor[PF_ANCHOR_NAME_SIZE]; char pfrt_ruleset[PF_RULESET_NAME_SIZE]; char pfrt_name[PF_TABLE_NAME_SIZE]; u_int32_t pfrt_flags; u_int8_t pfrt_fback; }; .Ed .It Dv DIOCRDELTABLES Fa "struct pfioc_table *io" Delete one or more tables. On entry, .Va pfrio_buffer[pfrio_size] contains a table of .Va pfr_table structures. On exit, .Va pfrio_nadd contains the number of tables effectively deleted. .It Dv DIOCRGETTABLES Fa "struct pfioc_table *io" Get the list of all tables. On entry, .Va pfrio_buffer[pfrio_size] contains a valid writeable buffer for .Va pfr_table structures. On exit, .Va pfrio_size contains the number of tables written into the buffer. If the buffer is too small, the kernel does not store anything but just returns the required buffer size, without error. .It Dv DIOCRGETTSTATS Fa "struct pfioc_table *io" This call is like .Dv DIOCRGETTABLES but is used to get an array of .Va pfr_tstats structures. .Bd -literal struct pfr_tstats { struct pfr_table pfrts_t; u_int64_t pfrts_packets [PFR_DIR_MAX][PFR_OP_TABLE_MAX]; u_int64_t pfrts_bytes [PFR_DIR_MAX][PFR_OP_TABLE_MAX]; u_int64_t pfrts_match; u_int64_t pfrts_nomatch; long pfrts_tzero; int pfrts_cnt; int pfrts_refcnt[PFR_REFCNT_MAX]; }; #define pfrts_name pfrts_t.pfrt_name #define pfrts_flags pfrts_t.pfrt_flags .Ed .It Dv DIOCRCLRTSTATS Fa "struct pfioc_table *io" Clear the statistics of one or more tables. On entry, .Va pfrio_buffer[pfrio_size] contains a table of .Va pfr_table structures. On exit, .Va pfrio_nzero contains the number of tables effectively cleared. .It Dv DIOCRCLRADDRS Fa "struct pfioc_table *io" Clear all addresses in a table. On entry, .Va pfrio_table contains the table to clear. On exit, .Va pfrio_ndel contains the number of addresses removed. .It Dv DIOCRADDADDRS Fa "struct pfioc_table *io" Add one or more addresses to a table. On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains the list of .Va pfr_addr structures to add. On exit, .Va pfrio_nadd contains the number of addresses effectively added. .Bd -literal struct pfr_addr { union { struct in_addr _pfra_ip4addr; struct in6_addr _pfra_ip6addr; } pfra_u; u_int8_t pfra_af; u_int8_t pfra_net; u_int8_t pfra_not; u_int8_t pfra_fback; }; #define pfra_ip4addr pfra_u._pfra_ip4addr #define pfra_ip6addr pfra_u._pfra_ip6addr .Ed .It Dv DIOCRDELADDRS Fa "struct pfioc_table *io" Delete one or more addresses from a table. On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains the list of .Va pfr_addr structures to delete. On exit, .Va pfrio_ndel contains the number of addresses effectively deleted. .It Dv DIOCRSETADDRS Fa "struct pfioc_table *io" Replace the content of a table by a new address list. This is the most complicated command, which uses all the structure members. .Pp On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains the new list of .Va pfr_addr structures. Additionally, if .Va pfrio_size2 is non-zero, .Va pfrio_buffer[pfrio_size..pfrio_size2] must be a writeable buffer, into which the kernel can copy the addresses that have been deleted during the replace operation. On exit, .Va pfrio_ndel , .Va pfrio_nadd , and .Va pfrio_nchange contain the number of addresses deleted, added, and changed by the kernel. If .Va pfrio_size2 was set on entry, .Va pfrio_size2 will point to the size of the buffer used, exactly like .Dv DIOCRGETADDRS . .It Dv DIOCRGETADDRS Fa "struct pfioc_table *io" Get all the addresses of a table. On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains a valid writeable buffer for .Va pfr_addr structures. On exit, .Va pfrio_size contains the number of addresses written into the buffer. If the buffer was too small, the kernel does not store anything but just returns the required buffer size, without returning an error. .It Dv DIOCRGETASTATS Fa "struct pfioc_table *io" This call is like .Dv DIOCRGETADDRS but is used to get an array of .Va pfr_astats structures. .Bd -literal struct pfr_astats { struct pfr_addr pfras_a; u_int64_t pfras_packets [PFR_DIR_MAX][PFR_OP_ADDR_MAX]; u_int64_t pfras_bytes [PFR_DIR_MAX][PFR_OP_ADDR_MAX]; long pfras_tzero; }; .Ed .It Dv DIOCRCLRASTATS Fa "struct pfioc_table *io" Clear the statistics of one or more addresses. On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains a table of .Va pfr_addr structures to clear. On exit, .Va pfrio_nzero contains the number of addresses effectively cleared. .It Dv DIOCRTSTADDRS Fa "struct pfioc_table *io" Test if the given addresses match a table. On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains a table of .Va pfr_addr structures to test. On exit, the kernel updates the .Va pfr_addr table by setting the .Va pfra_fback member appropriately. .It Dv DIOCRSETTFLAGS Fa "struct pfioc_table *io" Change the .Dv PFR_TFLAG_CONST or .Dv PFR_TFLAG_PERSIST flags of a table. On entry, .Va pfrio_buffer[pfrio_size] contains a table of .Va pfr_table structures, and .Va pfrio_setflag contains the flags to add, while .Va pfrio_clrflag contains the flags to remove. On exit, .Va pfrio_nchange and .Va pfrio_ndel contain the number of tables altered or deleted by the kernel. Yes, tables can be deleted if one removes the .Dv PFR_TFLAG_PERSIST flag of an unreferenced table. .It Dv DIOCRINADEFINE Fa "struct pfioc_table *io" Defines a table in the inactive set. On entry, .Va pfrio_table contains the table ID and .Va pfrio_buffer[pfrio_size] contains the list of .Va pfr_addr structures to put in the table. A valid ticket must also be supplied to .Va pfrio_ticket . On exit, .Va pfrio_nadd contains 0 if the table was already defined in the inactive list or 1 if a new table has been created. .Va pfrio_naddr contains the number of addresses effectively put in the table. .It Dv DIOCXBEGIN Fa "struct pfioc_trans *io" .Bd -literal struct pfioc_trans { int size; /* number of elements */ int esize; /* size of each element in bytes */ struct pfioc_trans_e { int rs_num; char anchor[PF_ANCHOR_NAME_SIZE]; char ruleset[PF_RULESET_NAME_SIZE]; u_int32_t ticket; } *array; }; .Ed .Pp Clear all the inactive rulesets specified in the .Va pfioc_trans_e array. For each ruleset, a ticket is returned for subsequent "add rule" ioctls, as well as for the .Dv DIOCXCOMMIT and .Dv DIOCXROLLBACK calls. .Pp Ruleset types, identified by .Va rs_num , include the following: .Pp .Bl -tag -width PF_RULESET_FILTER -offset ind -compact .It Dv PF_RULESET_SCRUB Scrub (packet normalization) rules. .It Dv PF_RULESET_FILTER Filter rules. .It Dv PF_RULESET_NAT NAT (Network Address Translation) rules. .It Dv PF_RULESET_BINAT Bidirectional NAT rules. .It Dv PF_RULESET_RDR Redirect rules. .It Dv PF_RULESET_ALTQ ALTQ disciplines. .It Dv PF_RULESET_TABLE Address tables. .El .It Dv DIOCXCOMMIT Fa "struct pfioc_trans *io" Atomically switch a vector of inactive rulesets to the active rulesets. This call is implemented as a standard two-phase commit, which will either fail for all rulesets or completely succeed. All tickets need to be valid. This ioctl returns .Er EBUSY if another process is concurrently updating some of the same rulesets. .It Dv DIOCXROLLBACK Fa "struct pfioc_trans *io" Clean up the kernel by undoing all changes that have taken place on the inactive rulesets since the last .Dv DIOCXBEGIN . .Dv DIOCXROLLBACK will silently ignore rulesets for which the ticket is invalid. .It Dv DIOCSETHOSTID Fa "u_int32_t *hostid" Set the host ID, which is used by .Xr pfsync 4 to identify which host created state table entries. .It Dv DIOCOSFPFLUSH Flush the passive OS fingerprint table. .It Dv DIOCOSFPADD Fa "struct pf_osfp_ioctl *io" .Bd -literal struct pf_osfp_ioctl { struct pf_osfp_entry { SLIST_ENTRY(pf_osfp_entry) fp_entry; pf_osfp_t fp_os; char fp_class_nm[PF_OSFP_LEN]; char fp_version_nm[PF_OSFP_LEN]; char fp_subtype_nm[PF_OSFP_LEN]; } fp_os; pf_tcpopts_t fp_tcpopts; u_int16_t fp_wsize; u_int16_t fp_psize; u_int16_t fp_mss; u_int16_t fp_flags; u_int8_t fp_optcnt; u_int8_t fp_wscale; u_int8_t fp_ttl; int fp_getnum; }; .Ed .Pp Add a passive OS fingerprint to the table. Set .Va fp_os.fp_os to the packed fingerprint, .Va fp_os.fp_class_nm to the name of the class (Linux, Windows, etc), .Va fp_os.fp_version_nm to the name of the version (NT, 95, 98), and .Va fp_os.fp_subtype_nm to the name of the subtype or patchlevel. The members .Va fp_mss , .Va fp_wsize , .Va fp_psize , .Va fp_ttl , .Va fp_optcnt , and .Va fp_wscale are set to the TCP MSS, the TCP window size, the IP length, the IP TTL, the number of TCP options, and the TCP window scaling constant of the TCP SYN packet, respectively. .Pp The .Va fp_flags member is filled according to the .Aq Pa net/pfvar.h include file .Dv PF_OSFP_* defines. The .Va fp_tcpopts member contains packed TCP options. Each option uses .Dv PF_OSFP_TCPOPT_BITS bits in the packed value. Options include any of .Dv PF_OSFP_TCPOPT_NOP , .Dv PF_OSFP_TCPOPT_SACK , .Dv PF_OSFP_TCPOPT_WSCALE , .Dv PF_OSFP_TCPOPT_MSS , or .Dv PF_OSFP_TCPOPT_TS . .Pp The .Va fp_getnum member is not used with this ioctl. .Pp The structure's slack space must be zeroed for correct operation; .Xr memset 3 the whole structure to zero before filling and sending to the kernel. .It Dv DIOCOSFPGET Fa "struct pf_osfp_ioctl *io" Get the passive OS fingerprint number .Va fp_getnum from the kernel's fingerprint list. The rest of the structure members will come back filled. Get the whole list by repeatedly incrementing the .Va fp_getnum number until the ioctl returns .Er EBUSY . .It Dv DIOCGETSRCNODES Fa "struct pfioc_src_nodes *psn" .Bd -literal struct pfioc_src_nodes { int psn_len; union { caddr_t psu_buf; struct pf_src_node *psu_src_nodes; } psn_u; #define psn_buf psn_u.psu_buf #define psn_src_nodes psn_u.psu_src_nodes }; .Ed .Pp Get the list of source nodes kept by sticky addresses and source tracking. The ioctl must be called once with .Va psn_len set to 0. If the ioctl returns without error, .Va psn_len will be set to the size of the buffer required to hold all the .Va pf_src_node structures held in the table. A buffer of this size should then be allocated, and a pointer to this buffer placed in .Va psn_buf . The ioctl must then be called again to fill this buffer with the actual source node data. After that call, .Va psn_len will be set to the length of the buffer actually used. .It Dv DIOCCLRSRCNODES Clear the tree of source tracking nodes. .It Dv DIOCIGETIFACES Fa "struct pfioc_iface *io" Get the list of interfaces and interface drivers known to .Nm . All the ioctls that manipulate interfaces use the same structure described below: .Bd -literal struct pfioc_iface { char pfiio_name[IFNAMSIZ]; void *pfiio_buffer; int pfiio_esize; int pfiio_size; int pfiio_nzero; int pfiio_flags; }; #define PFI_FLAG_GROUP 0x0001 /* gets groups of interfaces */ #define PFI_FLAG_INSTANCE 0x0002 /* gets single interfaces */ #define PFI_FLAG_ALLMASK 0x0003 .Ed .Pp If not empty, .Va pfiio_name can be used to restrict the search to a specific interface or driver. .Va pfiio_buffer[pfiio_size] is the user-supplied buffer for returning the data. On entry, .Va pfiio_size represents the number of .Va pfi_if entries that can fit into the buffer. The kernel will replace this value by the real number of entries it wants to return. .Va pfiio_esize should be set to .Li sizeof(struct pfi_if) . .Va pfiio_flags should be set to .Dv PFI_FLAG_GROUP , .Dv PFI_FLAG_INSTANCE , or both, to tell the kernel to return a group of interfaces (drivers, like "fxp"), real interface instances (like "fxp1") or both. The data is returned in the .Va pfi_if structure described below: .Bd -literal struct pfi_if { char pfif_name[IFNAMSIZ]; u_int64_t pfif_packets[2][2][2]; u_int64_t pfif_bytes[2][2][2]; u_int64_t pfif_addcnt; u_int64_t pfif_delcnt; long pfif_tzero; int pfif_states; int pfif_rules; int pfif_flags; }; #define PFI_IFLAG_GROUP 0x0001 /* group of interfaces */ #define PFI_IFLAG_INSTANCE 0x0002 /* single instance */ #define PFI_IFLAG_CLONABLE 0x0010 /* clonable group */ #define PFI_IFLAG_DYNAMIC 0x0020 /* dynamic group */ #define PFI_IFLAG_ATTACHED 0x0040 /* interface attached */ .Ed .It Dv DIOCICLRISTATS Fa "struct pfioc_iface *io" Clear the statistics counters of one or more interfaces. .Va pfiio_name and .Va pfiio_flags can be used to select which interfaces need to be cleared. The filtering process is the same as for .Dv DIOCIGETIFACES . .Va pfiio_nzero will be set by the kernel to the number of interfaces and drivers that have been cleared. .El .Sh FILES .Bl -tag -width /dev/pf -compact .It Pa /dev/pf packet filtering device. .El .Sh EXAMPLES The following example demonstrates how to use the .Dv DIOCNATLOOK command to find the internal host/port of a NATed connection: .Bd -literal #include #include #include #include #include #include #include #include #include #include u_int32_t read_address(const char *s) { int a, b, c, d; sscanf(s, "%i.%i.%i.%i", &a, &b, &c, &d); return htonl(a << 24 | b << 16 | c << 8 | d); } void print_address(u_int32_t a) { a = ntohl(a); printf("%d.%d.%d.%d", a >> 24 & 255, a >> 16 & 255, a >> 8 & 255, a & 255); } int main(int argc, char *argv[]) { struct pfioc_natlook nl; int dev; if (argc != 5) { printf("%s \\n", argv[0]); return 1; } dev = open("/dev/pf", O_RDWR); if (dev == -1) err(1, "open(\\"/dev/pf\\") failed"); memset(&nl, 0, sizeof(struct pfioc_natlook)); nl.saddr.v4.s_addr = read_address(argv[1]); nl.sport = htons(atoi(argv[2])); nl.daddr.v4.s_addr = read_address(argv[3]); nl.dport = htons(atoi(argv[4])); nl.af = AF_INET; nl.proto = IPPROTO_TCP; nl.direction = PF_IN; if (ioctl(dev, DIOCNATLOOK, &nl)) err(1, "DIOCNATLOOK"); printf("internal host "); print_address(nl.rsaddr.v4.s_addr); printf(":%u\\n", ntohs(nl.rsport)); return 0; } .Ed .Sh SEE ALSO .Xr ioctl 2 , .Xr bridge 4 , .Xr pflog 4 , .Xr pfsync 4 , .Xr pfctl 8 , .Xr altq 9 .Sh HISTORY The .Nm packet filtering mechanism first appeared in .Ox 3.0 .