xref: /freebsd/share/man/man4/inet.4 (revision 74bf4e164ba5851606a27d4feff27717452583e5)
1.\" Copyright (c) 1983, 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: @(#)inet.4	8.1 (Berkeley) 6/5/93
33.\" $FreeBSD$
34.\"
35.Dd August 15, 2004
36.Dt INET 4
37.Os
38.Sh NAME
39.Nm inet
40.Nd Internet protocol family
41.Sh SYNOPSIS
42.In sys/types.h
43.In netinet/in.h
44.Sh DESCRIPTION
45The Internet protocol family is a collection of protocols
46layered atop the
47.Em Internet Protocol
48.Pq Tn IP
49transport layer, and utilizing the Internet address format.
50The Internet family provides protocol support for the
51.Dv SOCK_STREAM , SOCK_DGRAM ,
52and
53.Dv SOCK_RAW
54socket types; the
55.Dv SOCK_RAW
56interface provides access to the
57.Tn IP
58protocol.
59.Sh ADDRESSING
60Internet addresses are four byte quantities, stored in
61network standard format (on little endian machines, such as the
62.Tn alpha ,
63.Tn amd64 ,
64.Tn i386
65and
66.Tn ia64
67these are word and byte reversed).
68The include file
69.In netinet/in.h
70defines this address
71as a discriminated union.
72.Pp
73Sockets bound to the Internet protocol family utilize
74the following addressing structure,
75.Bd -literal -offset indent
76struct sockaddr_in {
77	uint8_t		sin_len;
78	sa_family_t	sin_family;
79	in_port_t	sin_port;
80	struct in_addr	sin_addr;
81	char		sin_zero[8];
82};
83.Ed
84.Pp
85Sockets may be created with the local address
86.Dv INADDR_ANY
87to affect
88.Dq wildcard
89matching on incoming messages.
90The address in a
91.Xr connect 2
92or
93.Xr sendto 2
94call may be given as
95.Dv INADDR_ANY
96to mean
97.Dq this host .
98The distinguished address
99.Dv INADDR_BROADCAST
100is allowed as a shorthand for the broadcast address on the primary
101network if the first network configured supports broadcast.
102.Sh PROTOCOLS
103The Internet protocol family is comprised of
104the
105.Tn IP
106network protocol, Internet Control
107Message Protocol
108.Pq Tn ICMP ,
109Internet Group Management Protocol
110.Pq Tn IGMP ,
111Transmission Control
112Protocol
113.Pq Tn TCP ,
114and User Datagram Protocol
115.Pq Tn UDP .
116.Tn TCP
117is used to support the
118.Dv SOCK_STREAM
119abstraction while
120.Tn UDP
121is used to support the
122.Dv SOCK_DGRAM
123abstraction.
124A raw interface to
125.Tn IP
126is available
127by creating an Internet socket of type
128.Dv SOCK_RAW .
129The
130.Tn ICMP
131message protocol is accessible from a raw socket.
132.Pp
133The 32-bit Internet address contains both network and host parts.
134However, direct examination of addresses is discouraged.
135For those
136programs which absolutely need to break addresses into their component
137parts, the following
138.Xr ioctl 2
139commands are provided for a datagram socket in the Internet domain;
140they have the same form as the
141.Dv SIOCIFADDR
142command (see
143.Xr intro 4 ) .
144.Pp
145.Bl -tag -width SIOCSIFNETMASK
146.It Dv SIOCSIFNETMASK
147Set interface network mask.
148The network mask defines the network part of the address;
149if it contains more of the address than the address type would indicate,
150then subnets are in use.
151.It Dv SIOCGIFNETMASK
152Get interface network mask.
153.El
154.Sh ROUTING
155The current implementation of Internet protocols includes some routing-table
156adaptations to provide enhanced caching of certain end-to-end
157information necessary for Transaction TCP and Path MTU Discovery.
158The
159following changes are the most significant:
160.Bl -enum
161.It
162All IP routes, except those with the
163.Dv RTF_CLONING
164flag and those to multicast destinations, have the
165.Dv RTF_PRCLONING
166flag forcibly enabled (they are thus said to be
167.Dq "protocol cloning" ) .
168.It
169When the last reference to an IP route is dropped, the route is
170examined to determine if it was created by cloning such a route.
171If this is the case, the
172.Dv RTF_PROTO3
173flag is turned on, and the expiration timer is initialized to go off
174in net.inet.ip.rtexpire seconds.
175If such a route is re-referenced,
176the flag and expiration timer are reset.
177.It
178A kernel timeout runs once every ten minutes, or sooner if there are
179soon-to-expire routes in the kernel routing table, and deletes the
180expired routes.
181.El
182.Pp
183A dynamic process is in place to modify the value of
184net.inet.ip.rtexpire if the number of cached routes grows too large.
185If after an expiration run there are still more than
186net.inet.ip.rtmaxcache unreferenced routes remaining, the rtexpire
187value is multiplied by 3/4, and any routes which have longer
188expiration times have those times adjusted.
189This process is damped somewhat by specification of a minimum rtexpire value
190(net.inet.ip.rtminexpire), and by restricting the reduction to once in
191a ten-minute period.
192.Pp
193If some external process deletes the original route from which a
194protocol-cloned route was generated, the
195.Dq child route
196is deleted.
197(This is actually a generic mechanism in the routing code support for
198protocol-requested cloning.)
199.Pp
200No attempt is made to manage routes which were not created by protocol
201cloning; these are assumed to be static, under the management of an
202external routing process, or under the management of a link layer
203(e.g.,
204.Tn ARP
205for Ethernets).
206.Pp
207Only certain types of network activity will result in the cloning of a
208route using this mechanism.
209Specifically, those protocols (such as
210.Tn TCP
211and
212.Tn UDP )
213which themselves cache a long-lasting reference to route for a destination
214will trigger the mechanism; whereas raw
215.Tn IP
216packets, whether locally-generated or forwarded, will not.
217.Ss MIB Variables
218A number of variables are implemented in the net.inet branch of the
219.Xr sysctl 3
220MIB.
221In addition to the variables supported by the transport protocols
222(for which the respective manual pages may be consulted),
223the following general variables are defined:
224.Bl -tag -width IPCTL_FASTFORWARDING
225.It Dv IPCTL_FORWARDING
226.Pq ip.forwarding
227Boolean: enable/disable forwarding of IP packets.
228Defaults to off.
229.It Dv IPCTL_FASTFORWARDING
230.Pq ip.fastforwarding
231Boolean: enable/disable the use of fast IP forwarding code.
232Defaults to off.
233When fast forwarding is enabled, IP packets are forwarded directly to
234the appropriate network interface with a minimal validity checking, which
235greatly improves the throughput.
236On the other hand, they bypass the
237standard procedures, such as IP option processing and
238.Xr ipfirewall 4
239checking.
240It is not guaranteed that every packet will be fast-forwarded.
241.It Dv IPCTL_SENDREDIRECTS
242.Pq ip.redirect
243Boolean: enable/disable sending of ICMP redirects in response to
244unforwardable
245.Tn IP
246packets.
247Defaults to on.
248.It Dv IPCTL_DEFTTL
249.Pq ip.ttl
250Integer: default time-to-live
251.Pq Dq TTL
252to use for outgoing
253.Tn IP
254packets.
255.It Dv IPCTL_ACCEPTSOURCEROUTE
256.Pq ip.accept_sourceroute
257Boolean: enable/disable accepting of source-routed IP packets (default false).
258.It Dv IPCTL_SOURCEROUTE
259.Pq ip.sourceroute
260Boolean: enable/disable forwarding of source-routed IP packets (default false).
261.It Dv IPCTL_RTEXPIRE
262.Pq ip.rtexpire
263Integer: lifetime in seconds of protocol-cloned
264.Tn IP
265routes after the last reference drops (default one hour).
266This value varies dynamically as described above.
267.It Dv IPCTL_RTMINEXPIRE
268.Pq ip.rtminexpire
269Integer: minimum value of ip.rtexpire (default ten seconds).
270This value has no effect on user modifications, but restricts the dynamic
271adaptation described above.
272.It Dv IPCTL_RTMAXCACHE
273.Pq ip.rtmaxcache
274Integer: trigger level of cached, unreferenced, protocol-cloned routes
275which initiates dynamic adaptation (default 128).
276.It Va ip.process_options
277Integer: control IP options processing.
278By setting this variable to 0, all IP options in the incoming packets
279will be ignored, and the packets will be passed unmodified.
280By setting to 1, IP options in the incoming packets will be processed
281accordingly.
282By setting to 2, an
283.Tn ICMP
284.Dq "prohibited by filter"
285message will be sent back in respose to incoming packets with IP options.
286Default is 1.
287This
288.Xr sysctl 8
289variable affects packets destined for a local host as well as packets
290forwarded to some other host.
291.It Va ip.random_id
292Boolean: control IP IDs generation behaviour.
293Setting this
294.Xr sysctl 8
295to non-zero causes the ID field in IP packets to be randomized instead of
296incremented by 1 with each packet generated.
297This closes a minor information leak which allows remote observers to
298determine the rate of packet generation on the machine by watching the
299counter.
300In the same time, on high-speed links, it can decrease the ID reuse
301cycle greatly.
302Default is 0 (sequential IP IDs).
303IPv6 flow IDs and fragment IDs are always random.
304.El
305.Sh SEE ALSO
306.Xr ioctl 2 ,
307.Xr socket 2 ,
308.Xr sysctl 3 ,
309.Xr icmp 4 ,
310.Xr intro 4 ,
311.Xr ip 4 ,
312.Xr ipfirewall 4 ,
313.Xr tcp 4 ,
314.Xr ttcp 4 ,
315.Xr udp 4
316.Rs
317.%T "An Introductory 4.3 BSD Interprocess Communication Tutorial"
318.%B PS1
319.%N 7
320.Re
321.Rs
322.%T "An Advanced 4.3 BSD Interprocess Communication Tutorial"
323.%B PS1
324.%N 8
325.Re
326.Sh CAVEAT
327The Internet protocol support is subject to change as
328the Internet protocols develop.
329Users should not depend
330on details of the current implementation, but rather
331the services exported.
332.Sh HISTORY
333The
334.Nm
335protocol interface appeared in
336.Bx 4.2 .
337The
338.Dq protocol cloning
339code appeared in
340.Fx 2.1 .
341