i386 removal, part 1/x: Remove 'real' APM and associated stuff.
[dragonfly.git] / share / man / man4 / bridge.4
1 .\" Copyright 2001 Wasabi Systems, Inc.
2 .\" All rights reserved.
3 .\"
4 .\" Written by Jason R. Thorpe for Wasabi Systems, Inc.
5 .\" Spanning tree modifications by Matthew Dillon
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. All advertising materials mentioning features or use of this software
16 .\"    must display the following acknowledgement:
17 .\"     This product includes software developed for the NetBSD Project by
18 .\"     Wasabi Systems, Inc.
19 .\" 4. The name of Wasabi Systems, Inc. may not be used to endorse
20 .\"    or promote products derived from this software without specific prior
21 .\"    written permission.
22 .\"
23 .\" THIS SOFTWARE IS PROVIDED BY WASABI SYSTEMS, INC. ``AS IS'' AND
24 .\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
25 .\" TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
26 .\" PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL WASABI SYSTEMS, INC
27 .\" BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
28 .\" CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
29 .\" SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
30 .\" INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
31 .\" CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
32 .\" ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
33 .\" POSSIBILITY OF SUCH DAMAGE.
34 .\"
35 .\"
36 .Dd February 22, 2011
37 .Dt BRIDGE 4
38 .Os
39 .Sh NAME
40 .Nm bridge
41 .Nd network bridge device
42 .Sh SYNOPSIS
43 .Cd "pseudo-device bridge"
44 .Sh DESCRIPTION
45 The
46 .Nm
47 driver creates a logical link between two or more IEEE 802 networks
48 that use the same (or
49 .Dq similar enough )
50 framing format.
51 For example, it is possible to bridge Ethernet and 802.11 networks together,
52 but it is not possible to bridge Ethernet and Token Ring together.
53 .Pp
54 To use
55 .Nm ,
56 the administrator must first create the interface and configure
57 the bridge parameters.
58 The bridge is created using the
59 .Xr ifconfig 8
60 .Cm create
61 subcommand.
62 See the
63 .Xr ifconfig 8
64 manual page for further information on configuring bridges.
65 .Pp
66 A bridge can be used to provide several services, such as a simple
67 802.11-to-Ethernet bridge for wireless hosts, and traffic isolation.
68 .Pp
69 A bridge works like a hub, forwarding traffic from one interface
70 to another.
71 Multicast and broadcast packets are always forwarded to all
72 interfaces that are part of the bridge.
73 For unicast traffic, the bridge learns which MAC addresses are associated
74 with which interfaces and will forward the traffic selectively.
75 By default the bridge can keep track of 4096 MAC addresses.
76 .Pp
77 The bridge operates in a safe mode by default, setting the MAC source in
78 the link header on outgoing packets to the outgoing interface MAC.
79 This reduces the chance that the layer-2 switching in your switches
80 will become confused.
81 .Pp
82 The bridge supports various special features via
83 .Cm link
84 options.
85 .Bl -tag -width indent
86 .It Cm link0
87 The link0 option enables transparent bridging mode.
88 The bridge will make every effort to retain the ethernet header
89 when forwarding packets between interfaces, making the bridging
90 function work more like a hardware bridge device.
91 .It Cm link1
92 The link1 option enables keepalive transmission and automatically
93 places a member into a special blocked mode if no keepalive reception
94 occurs.
95 If either sides of the link uses this option then both sides must use
96 this option.
97 This option is impemented by sending CFG updates on the hello interval
98 to the remote.
99 The link is considered lost after 10 intervals (typically 20 seconds).
100 .It Cm link2
101 The link2 option enables channel bonding (see also ifbondweight).
102 All member interfaces with the same mac address are considered to
103 be in a bonding group.
104 When something like
105 .Xr tap 4
106 is used, you can manually control or copy the mac to create bonding groups.
107 When interface bonding is enabled normally blocked interfaces belonging
108 to the same bonding group as an active forwarding interface will be
109 changed to the bonding state.
110 Both sides of link the member represents must operate in bonding mode
111 for this to work, otherwise the remote end may decide to throw away
112 half your packets.
113 .El
114 .Pp
115 If your network becomes glitchy, with long pauses in tcp sessions, then
116 transparent bridging mode is likely the cause.  This mode should only be
117 used when you are bridging networks with devices that do MAC-based security
118 or firewalling (for example, the supremely braindead at&t uverse router),
119 or which impose severe limitations on MAC:IP assignments.
120 .Pp
121 If member interfaces constantly enter a 'blocked (link1)' state then the
122 other end of those interfaces is not implementing the link1 keepalive.
123 Both sides must implement the keepalive.
124 .Pp
125 If you get an enormous amount of packet loss and are using link2-based
126 bonding, then the other side of those member interfaces are probably
127 not implementing link2-based bonding.
128 .Pp
129 The
130 .Nm
131 driver implements the IEEE 802.1D Spanning Tree protocol (STP).
132 Spanning Tree is used to detect and remove loops in a network topology.
133 .Pp
134 Packet filtering can be used with any firewall package that hooks in via the
135 .Xr pfil 9
136 framework.
137 When filtering is enabled, bridged packets will pass through the filter
138 inbound on the originating interface, on the bridge interface and outbound on
139 the appropriate interfaces.
140 Either stage can be disabled, this behaviour can be controlled using
141 .Xr sysctl 8 :
142 Set
143 .Va net.link.bridge.pfil_member
144 to
145 .Li 1
146 to enable filtering on the incoming and outgoing member interfaces
147 and set
148 .Va net.link.bridge.pfil_bridge
149 to
150 .Li 1
151 to enable filtering on the bridge interface.
152 .Pp
153 ARP and REVARP packets are forwarded without being filtered and others
154 that are not IP nor IPv6 packets are not forwarded when filtering is
155 enabled.
156 .Pp
157 Note that packets to and from the bridging host will be seen by the
158 filter on the interface with the appropriate address configured as well
159 as on the interface on which the packet arrives or departs.
160 .Pp
161 The MTU of the first member interface to be added is used as the bridge MTU,
162 all additional members are required to have exactly the same value.
163 .Sh EXTRA FEATURES
164 .Dx
165 implements two additional features to make spanning tree operation more
166 resilient.
167 .Pp
168 Specifying
169 .Cm link0
170 on the bridge interface places the bridge in transparent bridging mode.
171 The bridge will make every attempt to retain the original source MAC in
172 the ethernet link header.
173 .Pp
174 Specifying
175 .Cm link1
176 on the bridge interface forces the bridge to generate a 802.11d CFG
177 message on every hello interval for all interfaces participating
178 in the STP protocol.
179 Normally CFG messages are only generated by the root bridge interface
180 or during topology changes.
181 In addition the bridge code expects to receive 802.11d frames from
182 all interface participating in the STP protocol.
183 .Pp
184 An interface which fails to receive a 802.11d frame within 10 times
185 the hello interval (usually 20 seconds) automatically goes into
186 l1blocking mode, which can be observed in the ifconfig output for
187 the bridge.  This removes the interface from consideration and the
188 bridge code automatically routes around it.
189 .Pp
190 Using
191 .Cm link0
192 and
193 .Cm link1
194 together between two
195 .Dx
196 boxes allows you to maintain multiple parallel vpns between those
197 boxes via different networks (if you happen to be on more than one
198 with internet access).
199 Use separate openvpn instances and tap devices for each vpn link
200 to accomplish this, placing them in the same bridge interface on
201 the two endpoints.
202 The tap devices do not need any IP configuration when bridged and
203 can be assigned the same ether MAC (in fact they have to be
204 if you want the failover to work nicely).
205 .Sh SEE ALSO
206 .Xr pf 4 ,
207 .Xr ifconfig 8
208 .Sh HISTORY
209 The
210 .Nm
211 driver first appeared in
212 .Ox 2.5
213 and found its way into
214 .Dx 1.3 .
215 Transparent bridging (link0) was added in
216 .Dx 2.9
217 in 2011.
218 .Sh AUTHORS
219 .An -nosplit
220 The
221 .Nm
222 driver was originally written by
223 .An Jason L. Wright Aq Mt jason@thought.net
224 as part of an undergraduate independent study at the University of
225 North Carolina at Greensboro.
226 .Pp
227 This version of the
228 .Nm
229 driver has been heavily modified from the original version by
230 .An Jason R. Thorpe Aq Mt thorpej@wasabisystems.com .
231 .Sh BUGS
232 The
233 .Nm
234 driver currently supports only Ethernet and Ethernet-like (e.g. 802.11)
235 network devices, with exactly the same interface MTU size as the bridge device.