[apple/xnu.git] / bsd / man / man4 / gif.4

.\"	$FreeBSD: src/share/man/man4/gif.4,v 1.3.2.9 2002/03/25 10:44:05 brooks Exp $
.\"	$KAME: gif.4,v 1.28 2001/05/18 13:15:56 itojun Exp $
.\"
.\" Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
.\" 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 April 10, 1999
.Dt GIF 4
.Os
.Sh NAME
.Nm gif
.Nd generic tunnel interface
.Sh SYNOPSIS
.Cd "pseudo-device gif"
.Sh DESCRIPTION
The
.Nm
interface is a generic tunnelling pseudo device for IPv4 and IPv6.
It can tunnel IPv[46] traffic over IPv[46].
Therefore, there can be four possible configurations.
The behavior of
.Nm
is mainly based on RFC2893 IPv6-over-IPv4 configured tunnel.
On
.Nx ,
.Nm
can also tunnel ISO traffic over IPv[46] using EON encapsulation.
.Pp
Each
.Nm
interface is created at runtime using interface cloning.
This is
most easily done with the
.Xr ifconfig 8
.Cm create
command or using the
.Va gifconfig_ Ns Aq Ar interface
variable in
.Xr rc.conf 5 .
.Pp
To use
.Nm ,
administrator needs to configure protocol and addresses used for the outer
header.
This can be done by using
.Xr gifconfig 8 ,
or
.Dv SIOCSIFPHYADDR
ioctl.
Also, administrator needs to configure protocol and addresses used for the
inner header, by using
.Xr ifconfig 8 .
Note that IPv6 link-local address
(those start with
.Li fe80:: )
will be automatically configured whenever possible.
You may need to remove IPv6 link-local address manually using
.Xr ifconfig 8 ,
when you would like to disable the use of IPv6 as inner header
(like when you need pure IPv4-over-IPv6 tunnel).
Finally, use routing table to route the packets toward
.Nm
interface.
.Pp
.Nm
can be configured to be ECN friendly.
This can be configured by
.Dv IFF_LINK1 .
.Ss ECN friendly behavior
.Nm
can be configured to be ECN friendly, as described in
.Dv draft-ietf-ipsec-ecn-02.txt .
This is turned off by default, and can be turned on by
.Dv IFF_LINK1
interface flag.
.Pp
Without
.Dv IFF_LINK1 ,
.Nm
will show a normal behavior, like described in RFC2893.
This can be summarized as follows:
.Bl -tag -width "Ingress" -offset indent
.It Ingress
Set outer TOS bit to
.Dv 0 .
.It Egress
Drop outer TOS bit.
.El
.Pp
With
.Dv IFF_LINK1 ,
.Nm
will copy ECN bits
.Dv ( 0x02
and
.Dv 0x01
on IPv4 TOS byte or IPv6 traffic class byte)
on egress and ingress, as follows:
.Bl -tag -width "Ingress" -offset indent
.It Ingress
Copy TOS bits except for ECN CE
(masked with
.Dv 0xfe )
from
inner to outer.
Set ECN CE bit to
.Dv 0 .
.It Egress
Use inner TOS bits with some change.
If outer ECN CE bit is
.Dv 1 ,
enable ECN CE bit on the inner.
.El
.Pp
Note that the ECN friendly behavior violates RFC2893.
This should be used in mutual agreement with the peer.
.Ss Security
Malicious party may try to circumvent security filters by using
tunnelled packets.
For better protection,
.Nm
performs martian filter and ingress filter against outer source address,
on egress.
Note that martian/ingress filters are no way complete.
You may want to secure your node by using packet filters.
Ingress filter can be turned off by
.Dv IFF_LINK2
bit.
.\"
.Ss Miscellaneous
By default,
.Nm
tunnels may not be nested.
This behavior may be modified at runtime by setting the
.Xr sysctl 8
variable
.Va net.link.gif.max_nesting
to the desired level of nesting.
Additionally,
.Nm
tunnels are restricted to one per pair of end points.
Parallel tunnels may be enabled by setting the
.Xr sysctl 8
variable
.Va net.link.gif.parallel_tunnels
to 1.
.Sh SEE ALSO
.Xr inet 4 ,
.Xr inet6 4 ,
.Xr gifconfig 8
.Rs
.%A	R. Gilligan
.%A	E. Nordmark
.%B	RFC2893
.%T	Transition Mechanisms for IPv6 Hosts and Routers
.%D	August 2000
.%O	ftp://ftp.isi.edu/in-notes/rfc2893.txt
.Re
.Rs
.%A	Sally Floyd
.%A	David L. Black
.%A	K. K. Ramakrishnan
.%T	"IPsec Interactions with ECN"
.%D	December 1999
.%O	draft-ietf-ipsec-ecn-02.txt
.Re
.\"
.Sh HISTORY
The
.Nm
device first appeared in WIDE hydrangea IPv6 kit.
.\"
.Sh BUGS
There are many tunnelling protocol specifications,
defined differently from each other.
.Nm
may not interoperate with peers which are based on different specifications,
and are picky about outer header fields.
For example, you cannot usually use
.Nm
to talk with IPsec devices that use IPsec tunnel mode.
.Pp
The current code does not check if the ingress address
(outer source address)
configured to
.Nm
makes sense.
Make sure to configure an address which belongs to your node.
Otherwise, your node will not be able to receive packets from the peer,
and your node will generate packets with a spoofed source address.
.Pp
If the outer protocol is IPv4,
.Nm
does not try to perform path MTU discovery for the encapsulated packet
(DF bit is set to 0).
.Pp
If the outer protocol is IPv6, path MTU discovery for encapsulated packet
may affect communication over the interface.
The first bigger-than-pmtu packet may be lost.
To avoid the problem, you may want to set the interface MTU for
.Nm
to 1240 or smaller, when outer header is IPv6 and inner header is IPv4.
.Pp
.Nm
does not translate ICMP messages for outer header into inner header.
.Pp
In the past,
.Nm
had a multi-destination behavior, configurable via
.Dv IFF_LINK0
flag.
The behavior was obsoleted and is no longer supported.
.Pp
It is thought that this is not actually a bug in gif, but rather lies
somewhere around a manipulation of an IPv6 routing table.
Commit	Line	Data
9bccf70c A	1	.\" $FreeBSD: src/share/man/man4/gif.4,v 1.3.2.9 2002/03/25 10:44:05 brooks Exp $
	2	.\" $KAME: gif.4,v 1.28 2001/05/18 13:15:56 itojun Exp $
	3	.\"
	4	.\" Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
	5	.\" All rights reserved.
	6	.\"
	7	.\" Redistribution and use in source and binary forms, with or without
	8	.\" modification, are permitted provided that the following conditions
	9	.\" are met:
	10	.\" 1. Redistributions of source code must retain the above copyright
	11	.\" notice, this list of conditions and the following disclaimer.
	12	.\" 2. Redistributions in binary form must reproduce the above copyright
	13	.\" notice, this list of conditions and the following disclaimer in the
	14	.\" documentation and/or other materials provided with the distribution.
	15	.\" 3. Neither the name of the project nor the names of its contributors
	16	.\" may be used to endorse or promote products derived from this software
	17	.\" without specific prior written permission.
	18	.\"
	19	.\" THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
	20	.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
	21	.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
	22	.\" ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
	23	.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
	24	.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
	25	.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
	26	.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
	27	.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
	28	.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
	29	.\" SUCH DAMAGE.
	30	.\"
	31	.Dd April 10, 1999
	32	.Dt GIF 4
	33	.Os
	34	.Sh NAME
	35	.Nm gif
	36	.Nd generic tunnel interface
	37	.Sh SYNOPSIS
	38	.Cd "pseudo-device gif"
	39	.Sh DESCRIPTION
	40	The
	41	.Nm
	42	interface is a generic tunnelling pseudo device for IPv4 and IPv6.
	43	It can tunnel IPv[46] traffic over IPv[46].
	44	Therefore, there can be four possible configurations.
	45	The behavior of
	46	.Nm
	47	is mainly based on RFC2893 IPv6-over-IPv4 configured tunnel.
	48	On
	49	.Nx ,
	50	.Nm
	51	can also tunnel ISO traffic over IPv[46] using EON encapsulation.
	52	.Pp
	53	Each
	54	.Nm
	55	interface is created at runtime using interface cloning.
	56	This is
	57	most easily done with the
	58	.Xr ifconfig 8
	59	.Cm create
	60	command or using the
	61	.Va gifconfig_ Ns Aq Ar interface
	62	variable in
	63	.Xr rc.conf 5 .
	64	.Pp
65	To use
66	.Nm ,
67	administrator needs to configure protocol and addresses used for the outer
68	header.
69	This can be done by using
70	.Xr gifconfig 8 ,
71	or
72	.Dv SIOCSIFPHYADDR
73	ioctl.
74	Also, administrator needs to configure protocol and addresses used for the
75	inner header, by using
76	.Xr ifconfig 8 .
77	Note that IPv6 link-local address
78	(those start with
79	.Li fe80:: )
80	will be automatically configured whenever possible.
81	You may need to remove IPv6 link-local address manually using
82	.Xr ifconfig 8 ,
83	when you would like to disable the use of IPv6 as inner header
84	(like when you need pure IPv4-over-IPv6 tunnel).
85	Finally, use routing table to route the packets toward
86	.Nm
87	interface.
88	.Pp
89	.Nm
90	can be configured to be ECN friendly.
91	This can be configured by
92	.Dv IFF_LINK1 .
93	.Ss ECN friendly behavior
94	.Nm
95	can be configured to be ECN friendly, as described in
96	.Dv draft-ietf-ipsec-ecn-02.txt .
97	This is turned off by default, and can be turned on by
98	.Dv IFF_LINK1
99	interface flag.
100	.Pp
101	Without
102	.Dv IFF_LINK1 ,
103	.Nm
104	will show a normal behavior, like described in RFC2893.
105	This can be summarized as follows:
106	.Bl -tag -width "Ingress" -offset indent
107	.It Ingress
108	Set outer TOS bit to
109	.Dv 0 .
110	.It Egress
111	Drop outer TOS bit.
112	.El
113	.Pp
114	With
115	.Dv IFF_LINK1 ,
116	.Nm
117	will copy ECN bits
118	.Dv ( 0x02
119	and
120	.Dv 0x01
121	on IPv4 TOS byte or IPv6 traffic class byte)
122	on egress and ingress, as follows:
123	.Bl -tag -width "Ingress" -offset indent
124	.It Ingress
125	Copy TOS bits except for ECN CE
126	(masked with
127	.Dv 0xfe )
128	from
129	inner to outer.
130	Set ECN CE bit to
131	.Dv 0 .
132	.It Egress
133	Use inner TOS bits with some change.
134	If outer ECN CE bit is
135	.Dv 1 ,
136	enable ECN CE bit on the inner.
137	.El
138	.Pp
139	Note that the ECN friendly behavior violates RFC2893.
140	This should be used in mutual agreement with the peer.
141	.Ss Security
142	Malicious party may try to circumvent security filters by using
143	tunnelled packets.
144	For better protection,
145	.Nm
146	performs martian filter and ingress filter against outer source address,
147	on egress.
148	Note that martian/ingress filters are no way complete.
149	You may want to secure your node by using packet filters.
150	Ingress filter can be turned off by
151	.Dv IFF_LINK2
152	bit.
153	.\"
154	.Ss Miscellaneous
155	By default,
156	.Nm
157	tunnels may not be nested.
158	This behavior may be modified at runtime by setting the
159	.Xr sysctl 8
160	variable
161	.Va net.link.gif.max_nesting
162	to the desired level of nesting.
163	Additionally,
164	.Nm
165	tunnels are restricted to one per pair of end points.
166	Parallel tunnels may be enabled by setting the
167	.Xr sysctl 8
168	variable
169	.Va net.link.gif.parallel_tunnels
170	to 1.
171	.Sh SEE ALSO
172	.Xr inet 4 ,
173	.Xr inet6 4 ,
174	.Xr gifconfig 8
175	.Rs
176	.%A R. Gilligan
177	.%A E. Nordmark
178	.%B RFC2893
179	.%T Transition Mechanisms for IPv6 Hosts and Routers
180	.%D August 2000
181	.%O ftp://ftp.isi.edu/in-notes/rfc2893.txt
182	.Re
183	.Rs
184	.%A Sally Floyd
185	.%A David L. Black
186	.%A K. K. Ramakrishnan
187	.%T "IPsec Interactions with ECN"
188	.%D December 1999
189	.%O draft-ietf-ipsec-ecn-02.txt
190	.Re
191	.\"
192	.Sh HISTORY
193	The
194	.Nm
195	device first appeared in WIDE hydrangea IPv6 kit.
196	.\"
197	.Sh BUGS
198	There are many tunnelling protocol specifications,
199	defined differently from each other.
200	.Nm
201	may not interoperate with peers which are based on different specifications,
202	and are picky about outer header fields.
203	For example, you cannot usually use
204	.Nm
205	to talk with IPsec devices that use IPsec tunnel mode.
206	.Pp
207	The current code does not check if the ingress address
208	(outer source address)
209	configured to
210	.Nm
211	makes sense.
212	Make sure to configure an address which belongs to your node.
213	Otherwise, your node will not be able to receive packets from the peer,
214	and your node will generate packets with a spoofed source address.
215	.Pp
216	If the outer protocol is IPv4,
217	.Nm
218	does not try to perform path MTU discovery for the encapsulated packet
219	(DF bit is set to 0).
220	.Pp
221	If the outer protocol is IPv6, path MTU discovery for encapsulated packet
222	may affect communication over the interface.
223	The first bigger-than-pmtu packet may be lost.
224	To avoid the problem, you may want to set the interface MTU for
225	.Nm
226	to 1240 or smaller, when outer header is IPv6 and inner header is IPv4.
227	.Pp
228	.Nm
229	does not translate ICMP messages for outer header into inner header.
230	.Pp
231	In the past,
232	.Nm
233	had a multi-destination behavior, configurable via
234	.Dv IFF_LINK0
235	flag.
236	The behavior was obsoleted and is no longer supported.
237	.Pp
238	It is thought that this is not actually a bug in gif, but rather lies
239	somewhere around a manipulation of an IPv6 routing table.