xref: /freebsd/share/man/man4/route.4 (revision 4c8945a06b01a5c8122cdeb402af36bb46a06acc)
1.\" Copyright (c) 1990, 1991, 1993
2.\"	The Regents of the University of California.  All rights reserved.
3.\"
4.\" Redistribution and use in source and binary forms, with or without
5.\" modification, are permitted provided that the following conditions
6.\" are met:
7.\" 1. Redistributions of source code must retain the above copyright
8.\"    notice, this list of conditions and the following disclaimer.
9.\" 2. Redistributions in binary form must reproduce the above copyright
10.\"    notice, this list of conditions and the following disclaimer in the
11.\"    documentation and/or other materials provided with the distribution.
12.\" 3. All advertising materials mentioning features or use of this software
13.\"    must display the following acknowledgement:
14.\"	This product includes software developed by the University of
15.\"	California, Berkeley and its contributors.
16.\" 4. Neither the name of the University nor the names of its contributors
17.\"    may be used to endorse or promote products derived from this software
18.\"    without specific prior written permission.
19.\"
20.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
21.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
22.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
23.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
24.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
25.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
26.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
27.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
28.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
29.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
30.\" SUCH DAMAGE.
31.\"
32.\"     From: @(#)route.4	8.6 (Berkeley) 4/19/94
33.\" $FreeBSD$
34.\"
35.Dd November 4, 2004
36.Dt ROUTE 4
37.Os
38.Sh NAME
39.Nm route
40.Nd kernel packet forwarding database
41.Sh SYNOPSIS
42.In sys/types.h
43.In sys/time.h
44.In sys/socket.h
45.In net/if.h
46.In net/route.h
47.Ft int
48.Fn socket PF_ROUTE SOCK_RAW "int family"
49.Sh DESCRIPTION
50.Fx
51provides some packet routing facilities.
52The kernel maintains a routing information database, which
53is used in selecting the appropriate network interface when
54transmitting packets.
55.Pp
56A user process (or possibly multiple co-operating processes)
57maintains this database by sending messages over a special kind
58of socket.
59This supplants fixed size
60.Xr ioctl 2 Ns 's
61used in earlier releases.
62Routing table changes may only be carried out by the super user.
63.Pp
64The operating system may spontaneously emit routing messages in response
65to external events, such as receipt of a re-direct, or failure to
66locate a suitable route for a request.
67The message types are described in greater detail below.
68.Pp
69Routing database entries come in two flavors: for a specific
70host, or for all hosts on a generic subnetwork (as specified
71by a bit mask and value under the mask.
72The effect of wildcard or default route may be achieved by using
73a mask of all zeros, and there may be hierarchical routes.
74.Pp
75When the system is booted and addresses are assigned
76to the network interfaces, each protocol family
77installs a routing table entry for each interface when it is ready for traffic.
78Normally the protocol specifies the route
79through each interface as a
80.Dq direct
81connection to the destination host
82or network.
83If the route is direct, the transport layer of
84a protocol family usually requests the packet be sent to the
85same host specified in the packet.
86Otherwise, the interface
87is requested to address the packet to the gateway listed in the routing entry
88(i.e., the packet is forwarded).
89.Pp
90When routing a packet,
91the kernel will attempt to find
92the most specific route matching the destination.
93(If there are two different mask and value-under-the-mask pairs
94that match, the more specific is the one with more bits in the mask.
95A route to a host is regarded as being supplied with a mask of
96as many ones as there are bits in the destination).
97If no entry is found, the destination is declared to be unreachable,
98and a routing-miss message is generated if there are any
99listeners on the routing control socket described below.
100.Pp
101A wildcard routing entry is specified with a zero
102destination address value, and a mask of all zeroes.
103Wildcard routes will be used
104when the system fails to find other routes matching the
105destination.
106The combination of wildcard
107routes and routing redirects can provide an economical
108mechanism for routing traffic.
109.Pp
110One opens the channel for passing routing control messages
111by using the socket call shown in the synopsis above:
112.Pp
113The
114.Fa family
115parameter may be
116.Dv AF_UNSPEC
117which will provide
118routing information for all address families, or can be restricted
119to a specific address family by specifying which one is desired.
120There can be more than one routing socket open per system.
121.Pp
122Messages are formed by a header followed by a small
123number of sockaddrs (now variable length particularly
124in the
125.Tn ISO
126case), interpreted by position, and delimited
127by the new length entry in the sockaddr.
128An example of a message with four addresses might be an
129.Tn ISO
130redirect:
131Destination, Netmask, Gateway, and Author of the redirect.
132The interpretation of which address are present is given by a
133bit mask within the header, and the sequence is least significant
134to most significant bit within the vector.
135.Pp
136Any messages sent to the kernel are returned, and copies are sent
137to all interested listeners.
138The kernel will provide the process
139ID for the sender, and the sender may use an additional sequence
140field to distinguish between outstanding messages.
141However, message replies may be lost when kernel buffers are exhausted.
142.Pp
143The kernel may reject certain messages, and will indicate this
144by filling in the
145.Ar rtm_errno
146field.
147The routing code returns
148.Er EEXIST
149if
150requested to duplicate an existing entry,
151.Er ESRCH
152if
153requested to delete a non-existent entry,
154or
155.Er ENOBUFS
156if insufficient resources were available
157to install a new route.
158In the current implementation, all routing processes run locally,
159and the values for
160.Ar rtm_errno
161are available through the normal
162.Em errno
163mechanism, even if the routing reply message is lost.
164.Pp
165A process may avoid the expense of reading replies to
166its own messages by issuing a
167.Xr setsockopt 2
168call indicating that the
169.Dv SO_USELOOPBACK
170option
171at the
172.Dv SOL_SOCKET
173level is to be turned off.
174A process may ignore all messages from the routing socket
175by doing a
176.Xr shutdown 2
177system call for further input.
178.Pp
179If a route is in use when it is deleted,
180the routing entry will be marked down and removed from the routing table,
181but the resources associated with it will not
182be reclaimed until all references to it are released.
183User processes can obtain information about the routing
184entry to a specific destination by using a
185.Dv RTM_GET
186message, or by calling
187.Xr sysctl 3 .
188.Pp
189Messages include:
190.Bd -literal
191#define	RTM_ADD		0x1    /* Add Route */
192#define	RTM_DELETE	0x2    /* Delete Route */
193#define	RTM_CHANGE	0x3    /* Change Metrics, Flags, or Gateway */
194#define	RTM_GET		0x4    /* Report Information */
195#define	RTM_LOSING	0x5    /* Kernel Suspects Partitioning */
196#define	RTM_REDIRECT	0x6    /* Told to use different route */
197#define	RTM_MISS	0x7    /* Lookup failed on this address */
198#define	RTM_LOCK	0x8    /* fix specified metrics */
199#define	RTM_OLDADD	0x9    /* caused by SIOCADDRT */
200#define	RTM_OLDDEL	0xa    /* caused by SIOCDELRT */
201#define	RTM_RESOLVE	0xb    /* request to resolve dst to LL addr - unused */
202#define	RTM_NEWADDR	0xc    /* address being added to iface */
203#define	RTM_DELADDR	0xd    /* address being removed from iface */
204#define	RTM_IFINFO	0xe    /* iface going up/down etc. */
205#define	RTM_NEWMADDR	0xf    /* mcast group membership being added to if */
206#define	RTM_DELMADDR	0x10   /* mcast group membership being deleted */
207#define	RTM_IFANNOUNCE	0x11   /* iface arrival/departure */
208#define	RTM_IEEE80211	0x12   /* IEEE80211 wireless event */
209.Ed
210.Pp
211A message header consists of one of the following:
212.Bd -literal
213struct rt_msghdr {
214    u_short rtm_msglen;         /* to skip over non-understood messages */
215    u_char  rtm_version;        /* future binary compatibility */
216    u_char  rtm_type;           /* message type */
217    u_short rtm_index;          /* index for associated ifp */
218    int     rtm_flags;          /* flags, incl. kern & message, e.g. DONE */
219    int     rtm_addrs;          /* bitmask identifying sockaddrs in msg */
220    pid_t   rtm_pid;            /* identify sender */
221    int     rtm_seq;            /* for sender to identify action */
222    int     rtm_errno;          /* why failed */
223    int     rtm_fmask;          /* bitmask used in RTM_CHANGE message */
224    u_long  rtm_inits;          /* which metrics we are initializing */
225    struct  rt_metrics rtm_rmx;	/* metrics themselves */
226};
227
228struct if_msghdr {
229    u_short ifm_msglen;         /* to skip over non-understood messages */
230    u_char  ifm_version;        /* future binary compatibility */
231    u_char  ifm_type;           /* message type */
232    int     ifm_addrs;          /* like rtm_addrs */
233    int     ifm_flags;          /* value of if_flags */
234    u_short ifm_index;          /* index for associated ifp */
235    struct  if_data ifm_data;   /* statistics and other data about if */
236};
237
238struct ifa_msghdr {
239    u_short ifam_msglen;        /* to skip over non-understood messages */
240    u_char  ifam_version;       /* future binary compatibility */
241    u_char  ifam_type;          /* message type */
242    int     ifam_addrs;         /* like rtm_addrs */
243    int     ifam_flags;         /* value of ifa_flags */
244    u_short ifam_index;         /* index for associated ifp */
245    int     ifam_metric;        /* value of ifa_metric */
246};
247
248struct ifma_msghdr {
249    u_short ifmam_msglen;       /* to skip over non-understood messages */
250    u_char  ifmam_version;      /* future binary compatibility */
251    u_char  ifmam_type;         /* message type */
252    int     ifmam_addrs;        /* like rtm_addrs */
253    int     ifmam_flags;        /* value of ifa_flags */
254    u_short ifmam_index;        /* index for associated ifp */
255};
256
257struct if_announcemsghdr {
258	u_short	ifan_msglen;	/* to skip over non-understood messages */
259	u_char	ifan_version;	/* future binary compatibility */
260	u_char	ifan_type;	/* message type */
261	u_short	ifan_index;	/* index for associated ifp */
262	char	ifan_name[IFNAMSIZ]; /* if name, e.g. "en0" */
263	u_short	ifan_what;	/* what type of announcement */
264};
265.Ed
266.Pp
267The
268.Dv RTM_IFINFO
269message uses a
270.Ar if_msghdr
271header, the
272.Dv RTM_NEWADDR
273and
274.Dv RTM_DELADDR
275messages use a
276.Ar ifa_msghdr
277header, the
278.Dv RTM_NEWMADDR
279and
280.Dv RTM_DELMADDR
281messages use a
282.Vt ifma_msghdr
283header, the
284.Dv RTM_IFANNOUNCE
285message uses a
286.Vt if_announcemsghdr
287header,
288and all other messages use the
289.Ar rt_msghdr
290header.
291.Pp
292The
293.Dq Li "struct rt_metrics"
294and the flag bits are as defined in
295.Xr rtentry 9 .
296.Pp
297Specifiers for metric values in rmx_locks and rtm_inits are:
298.Bd -literal
299#define	RTV_MTU       0x1    /* init or lock _mtu */
300#define	RTV_HOPCOUNT  0x2    /* init or lock _hopcount */
301#define	RTV_EXPIRE    0x4    /* init or lock _expire */
302#define	RTV_RPIPE     0x8    /* init or lock _recvpipe */
303#define	RTV_SPIPE     0x10   /* init or lock _sendpipe */
304#define	RTV_SSTHRESH  0x20   /* init or lock _ssthresh */
305#define	RTV_RTT       0x40   /* init or lock _rtt */
306#define	RTV_RTTVAR    0x80   /* init or lock _rttvar */
307#define	RTV_WEIGHT    0x100  /* init or lock _weight */
308.Ed
309.Pp
310Specifiers for which addresses are present in the messages are:
311.Bd -literal
312#define RTA_DST       0x1    /* destination sockaddr present */
313#define RTA_GATEWAY   0x2    /* gateway sockaddr present */
314#define RTA_NETMASK   0x4    /* netmask sockaddr present */
315#define RTA_GENMASK   0x8    /* cloning mask sockaddr present - unused */
316#define RTA_IFP       0x10   /* interface name sockaddr present */
317#define RTA_IFA       0x20   /* interface addr sockaddr present */
318#define RTA_AUTHOR    0x40   /* sockaddr for author of redirect */
319#define RTA_BRD       0x80   /* for NEWADDR, broadcast or p-p dest addr */
320.Ed
321.Sh SEE ALSO
322.Xr sysctl 3 ,
323.Xr route 8 ,
324.Xr rtentry 9
325.Pp
326The constants for the
327.Va rtm_flags
328field are documented in the manual page for the
329.Xr route 8
330utility.
331.Sh HISTORY
332A
333.Dv PF_ROUTE
334protocol family first appeared in
335.Bx 4.3 reno .
336