1 .\" Copyright (c) 1990, 1991, 1993
2 .\" The Regents of the University of California. All rights reserved.
4 .\" Redistribution and use in source and binary forms, with or without
5 .\" modification, are permitted provided that the following conditions
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. Neither the name of the University nor the names of its contributors
13 .\" may be used to endorse or promote products derived from this software
14 .\" without specific prior written permission.
16 .\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
17 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
18 .\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
19 .\" ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
20 .\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
21 .\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
22 .\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
23 .\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
24 .\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
25 .\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
28 .\" From: @(#)route.4 8.6 (Berkeley) 4/19/94
29 .\" $FreeBSD: src/share/man/man4/route.4,v 1.9.2.6 2002/03/17 09:12:44 schweikh Exp $
30 .\" $DragonFly: src/share/man/man4/route.4,v 1.3 2004/03/11 12:28:55 hmp Exp $
37 .Nd kernel packet forwarding database
45 .Fn socket PF_ROUTE SOCK_RAW "int family"
48 provides some packet routing facilities.
49 The kernel maintains a routing information database, which
50 is used in selecting the appropriate network interface when
53 A user process (or possibly multiple co-operating processes)
54 maintains this database by sending messages over a special kind
56 This supplants fixed size
58 used in earlier releases.
59 Routing table changes may only be carried out by the super user.
61 The operating system may spontaneously emit routing messages in response
62 to external events, such as receipt of a re-direct, or failure to
63 locate a suitable route for a request.
64 The message types are described in greater detail below.
66 Routing database entries come in two flavors: for a specific
67 host, or for all hosts on a generic subnetwork (as specified
68 by a bit mask and value under the mask.
69 The effect of wildcard or default route may be achieved by using
70 a mask of all zeros, and there may be hierarchical routes.
72 When the system is booted and addresses are assigned
73 to the network interfaces, each protocol family
74 installs a routing table entry for each interface when it is ready for traffic.
75 Normally the protocol specifies the route
76 through each interface as a
78 connection to the destination host
79 or network. If the route is direct, the transport layer of
80 a protocol family usually requests the packet be sent to the
81 same host specified in the packet. Otherwise, the interface
82 is requested to address the packet to the gateway listed in the routing entry
83 (i.e. the packet is forwarded).
85 When routing a packet,
86 the kernel will attempt to find
87 the most specific route matching the destination.
88 (If there are two different mask and value-under-the-mask pairs
89 that match, the more specific is the one with more bits in the mask.
90 A route to a host is regarded as being supplied with a mask of
91 as many ones as there are bits in the destination).
92 If no entry is found, the destination is declared to be unreachable,
93 and a routing\-miss message is generated if there are any
94 listeners on the routing control socket described below.
96 A wildcard routing entry is specified with a zero
97 destination address value, and a mask of all zeroes.
98 Wildcard routes will be used
99 when the system fails to find other routes matching the
100 destination. The combination of wildcard
101 routes and routing redirects can provide an economical
102 mechanism for routing traffic.
104 One opens the channel for passing routing control messages
105 by using the socket call shown in the synopsis above:
112 routing information for all address families, or can be restricted
113 to a specific address family by specifying which one is desired.
114 There can be more than one routing socket open per system.
116 Messages are formed by a header followed by a small
117 number of sockaddrs (now variable length particularly
120 case), interpreted by position, and delimited
121 by the new length entry in the sockaddr.
122 An example of a message with four addresses might be an
125 Destination, Netmask, Gateway, and Author of the redirect.
126 The interpretation of which address are present is given by a
127 bit mask within the header, and the sequence is least significant
128 to most significant bit within the vector.
130 Any messages sent to the kernel are returned, and copies are sent
131 to all interested listeners. The kernel will provide the process
132 ID for the sender, and the sender may use an additional sequence
133 field to distinguish between outstanding messages. However,
134 message replies may be lost when kernel buffers are exhausted.
136 The kernel may reject certain messages, and will indicate this
140 The routing code returns
143 requested to duplicate an existing entry,
146 requested to delete a non-existent entry,
149 if insufficient resources were available
150 to install a new route.
151 In the current implementation, all routing processes run locally,
154 are available through the normal
156 mechanism, even if the routing reply message is lost.
158 A process may avoid the expense of reading replies to
159 its own messages by issuing a
161 call indicating that the
166 level is to be turned off.
167 A process may ignore all messages from the routing socket
170 system call for further input.
172 A process can specify which route message types it's interested in by passing
173 an array of route message types to the
180 For example, to only get specific messages:
181 .Bd -literal -offset indent
182 unsigned int rtfilter;
184 rtfilter = ROUTE_FILTER(RTM_IFINFO) | ROUTE_FILTER(RTM_IFANNOUNCE);
186 if (setsockopt(routefd, PF_ROUTE, ROUTE_MSGFILTER,
187 &rtfilter, sizeof(rtfilter)) == -1)
188 err(1, "setsockopt(ROUTE_MSGFILTER)");
191 A process can specify which RTM_MISS destination addresses it's interested in
192 by passing an array of struct sockaddr to the
199 For example, to only get RTM_MISS messages for specific destinations:
200 .Bd -literal -offset indent
201 char buf[1024] = { '\\0' }, *cp = buf;
202 struct sockaddr_in sin = {
203 .sin_family = AF_INET,
204 .sin_len = sizeof(sin),
207 inet_aton("192.168.0.1", &sin.sin_addr);
208 memcpy(cp, &sin, sin.sin_len);
209 cp += RT_ROUNDUP(sin.sin_len);
211 inet_aton("192.168.0.2", &sin.sin_addr);
212 memcpy(cp, &sin, sin.sin_len);
213 cp += RT_ROUNDUP(sin.sin_len);
215 if (setsockopt(routefd, PF_ROUTE, RO_MISSFILTER,
216 &sin, (socklen_t)(cp - buf)) == -1)
217 err(1, "setsockopt(RO_MISSFILTER)");
220 If a route is in use when it is deleted,
221 the routing entry will be marked down and removed from the routing table,
222 but the resources associated with it will not
223 be reclaimed until all references to it are released.
224 User processes can obtain information about the routing
225 entry to a specific destination by using a
227 message, or by calling
232 #define RTM_ADD 0x1 /* Add Route */
233 #define RTM_DELETE 0x2 /* Delete Route */
234 #define RTM_CHANGE 0x3 /* Change Metrics, Flags, or Gateway */
235 #define RTM_GET 0x4 /* Report Information */
236 #define RTM_LOSING 0x5 /* Kernel Suspects Partitioning */
237 #define RTM_REDIRECT 0x6 /* Told to use different route */
238 #define RTM_MISS 0x7 /* Lookup failed on this address */
239 #define RTM_LOCK 0x8 /* fix specified metrics */
240 #define RTM_RESOLVE 0xb /* request to resolve dst to LL addr */
241 #define RTM_NEWADDR 0xc /* address being added to iface */
242 #define RTM_DELADDR 0xd /* address being removed from iface */
243 #define RTM_IFINFO 0xe /* iface going up/down etc. */
244 #define RTM_NEWMADDR 0xf /* mcast group membership being added to if */
245 #define RTM_DELMADDR 0x10 /* mcast group membership being deleted */
246 #define RTM_IFANNOUNCE 0x11 /* iface arrival/departure */
249 A message header consists of one of the following:
252 u_short rtm_msglen; /* to skip over non-understood messages */
253 u_char rtm_version; /* future binary compatibility */
254 u_char rtm_type; /* message type */
255 u_short rtm_index; /* index for associated ifp */
256 int rtm_flags; /* flags, incl. kern & message, e.g. DONE */
257 int rtm_addrs; /* bitmask identifying sockaddrs in msg */
258 pid_t rtm_pid; /* identify sender */
259 int rtm_seq; /* for sender to identify action */
260 int rtm_errno; /* why failed */
261 int rtm_use; /* from rtentry */
262 u_long rtm_inits; /* which metrics we are initializing */
263 struct rt_metrics rtm_rmx; /* metrics themselves */
267 u_short ifm_msglen; /* to skip over non-understood messages */
268 u_char ifm_version; /* future binary compatibility */
269 u_char ifm_type; /* message type */
270 u_short ifm_index; /* index for associated ifp */
271 int ifm_flags; /* value of if_flags */
272 int ifm_addrs; /* like rtm_addrs */
273 struct if_data ifm_data; /* statistics and other data about if */
277 u_short ifam_msglen; /* to skip over non-understood messages */
278 u_char ifam_version; /* future binary compatibility */
279 u_char ifam_type; /* message type */
280 u_short ifam_index; /* index for associated ifp */
281 int ifam_flags; /* value of ifa_flags */
282 int ifam_addrs; /* like rtm_addrs */
283 int ifam_addrflags; /* family specific address flags */
284 int ifam_metric; /* value of ifa_metric */
288 u_short ifmam_msglen; /* to skip over non-understood messages */
289 u_char ifmam_version; /* future binary compatibility */
290 u_char ifmam_type; /* message type */
291 u_short ifmam_index; /* index for associated ifp */
292 int ifmam_flags; /* value of ifa_flags */
293 int ifmam_addrs; /* like rtm_addrs */
296 struct if_announcemsghdr {
297 u_short ifan_msglen; /* to skip over non-understood messages */
298 u_char ifan_version; /* future binary compatibility */
299 u_char ifan_type; /* message type */
300 u_short ifan_index; /* index for associated ifp */
301 char ifan_name[IFNAMSIZ]; /* if name, e.g. "en0" */
302 u_short ifan_what; /* what type of announcement */
325 .Vt if_announcemsghdr
327 and all other messages use the
332 .Dq Li "struct rt_metrics"
333 and the flag bits are as defined in
336 Specifiers for metric values in rmx_locks and rtm_inits are:
338 #define RTV_MTU 0x1 /* init or lock _mtu */
339 #define RTV_HOPCOUNT 0x2 /* init or lock _hopcount */
340 #define RTV_EXPIRE 0x4 /* init or lock _expire */
341 #define RTV_RPIPE 0x8 /* init or lock _recvpipe */
342 #define RTV_SPIPE 0x10 /* init or lock _sendpipe */
343 #define RTV_SSTHRESH 0x20 /* init or lock _ssthresh */
344 #define RTV_RTT 0x40 /* init or lock _rtt */
345 #define RTV_RTTVAR 0x80 /* init or lock _rttvar */
348 Specifiers for which addresses are present in the messages are:
350 #define RTA_DST 0x1 /* destination sockaddr present */
351 #define RTA_GATEWAY 0x2 /* gateway sockaddr present */
352 #define RTA_NETMASK 0x4 /* netmask sockaddr present */
353 #define RTA_GENMASK 0x8 /* cloning mask sockaddr present */
354 #define RTA_IFP 0x10 /* interface name sockaddr present */
355 #define RTA_IFA 0x20 /* interface addr sockaddr present */
356 #define RTA_AUTHOR 0x40 /* sockaddr for author of redirect */
357 #define RTA_BRD 0x80 /* for NEWADDR, broadcast or p-p dest addr */
366 protocol family first appeared in