xref: /dflybsd-src/share/man/man4/bridge.4 (revision 0eb2eccd5a86ef7dd7492d2651de55c3589f23d7)
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
45The
46.Nm
47driver creates a logical link between two or more IEEE 802 networks
48that use the same (or
49.Dq similar enough )
50framing format.
51For example, it is possible to bridge Ethernet and 802.11 networks together,
52but it is not possible to bridge Ethernet and Token Ring together.
53.Pp
54To use
55.Nm ,
56the administrator must first create the interface and configure
57the bridge parameters.
58The bridge is created using the
59.Xr ifconfig 8
60.Cm create
61subcommand.
62See the
63.Xr ifconfig 8
64manual page for further information on configuring bridges.
65.Pp
66A bridge can be used to provide several services, such as a simple
67802.11-to-Ethernet bridge for wireless hosts, and traffic isolation.
68.Pp
69A bridge works like a hub, forwarding traffic from one interface
70to another.
71Multicast and broadcast packets are always forwarded to all
72interfaces that are part of the bridge.
73For unicast traffic, the bridge learns which MAC addresses are associated
74with which interfaces and will forward the traffic selectively.
75.Pp
76The bridge operates in a safe mode by default, setting the MAC source in
77the link header on outgoing packets to the outgoing interface MAC.
78This reduces the chance that the layer-2 switching in your switches
79will become confused.
80.Pp
81The bridge supports various special features via
82.Cm link
83options.
84.Pp
85.It Cm link0
86The link0 option enables transparent bridging mode.
87The bridge will make every effort to retain the ethernet header
88when forwarding packets between interfaces, making the bridging
89function work more like a hardware bridge device.
90.It Cm link1
91The link1 option enables keepalive transmission and automatically
92places a member into a special blocked mode if no keepalive reception
93occurs.
94If either sides of the link uses this option then both sides must use
95this option.
96This option is impemented by sending CFG updates on the hello interval
97to the remote.
98The link is considered lost after 10 intervals (typically 20 seconds).
99.It Cm link2
100The link2 option enables channel bonding (see also ifbondweight).
101All member interfaces with the same mac address are considered to
102be in a bonding group.
103When something like
104.Xr tap 4
105is used, you can manually control or copy the mac to create bonding groups.
106When interface bonding is enabled normally blocked interfaces belonging
107to the same bonding group as an active forwarding interface will be
108changed to the bonding state.
109Both sides of link the member represents must operate in bonding mode
110for this to work, otherwise the remote end may decide to throw away
111half your packets.
112.Pp
113If your network becomes glitchy, with long pauses in tcp sessions, then
114transparent bridging mode is likely the cause.  This mode should only be
115used when you are bridging networks with devices that do MAC-based security
116or firewalling (for example, the supremely braindead at&t uverse router),
117or which impose severe limitations on MAC:IP assignments.
118.Pp
119If member interfaces constantly enter a 'blocked (link1)' state then the
120other end of those interfaces is not implementing the link1 keepalive.
121Both sides must implement the keepalive.
122.Pp
123If you get an enormous amount of packet loss and are using link2-based
124bonding, then the other side of those member interfaces are probably
125not implementing link2-based bonding.
126.Pp
127The
128.Nm
129driver implements the IEEE 802.1D Spanning Tree protocol (STP).
130Spanning Tree is used to detect and remove loops in a network topology.
131.Pp
132Packet filtering can be used with any firewall package that hooks in via the
133.Xr pfil 9
134framework.
135When filtering is enabled, bridged packets will pass through the filter
136inbound on the originating interface, on the bridge interface and outbound on
137the appropriate interfaces.
138Either stage can be disabled, this behaviour can be controlled using
139.Xr sysctl 8 :
140Set
141.Va net.link.bridge.pfil_member
142to
143.Li 1
144to enable filtering on the incoming and outgoing member interfaces
145and set
146.Va net.link.bridge.pfil_bridge
147to
148.Li 1
149to enable filtering on the bridge interface.
150.Pp
151ARP and REVARP packets are forwarded without being filtered and others
152that are not IP nor IPv6 packets are not forwarded when filtering is
153enabled.
154.Pp
155Note that packets to and from the bridging host will be seen by the
156filter on the interface with the appropriate address configured as well
157as on the interface on which the packet arrives or departs.
158.Pp
159The MTU of the first member interface to be added is used as the bridge MTU,
160all additional members are required to have exactly the same value.
161.Sh EXTRA FEATURES
162.Dx
163implements two additional features to make spanning tree operation more
164resilient.
165.Pp
166Specifying
167.Cm link0
168on the bridge interface places the bridge in transparent bridging mode.
169The bridge will make every attempt to retain the original source MAC in
170the ethernet link header.
171.Pp
172Specifying
173.Cm link1
174on the bridge interface forces the bridge to generate a 802.11d CFG
175message on every hello interval for all interfaces participating
176in the STP protocol.
177Normally CFG messages are only generated by the root bridge interface
178or during topology changes.
179In addition the bridge code expects to receive 802.11d frames from
180all interface participating in the STP protocol.
181.Pp
182An interface which fails to receive a 802.11d frame within 10 times
183the hello interval (usually 20 seconds) automatically goes into
184l1blocking mode, which can be observed in the ifconfig output for
185the bridge.  This removes the interface from consideration and the
186bridge code automatically routes around it.
187.Pp
188Using
189.Cm link0
190and
191.Cm link1
192together between two
193.Dx
194boxes allows you to maintain multiple parallel vpns between those
195boxes via different networks (if you happen to be on more than one
196with internet access).
197Use separate openvpn instances and tap devices for each vpn link
198to accomplish this, placing them in the same bridge interface on
199the two endpoints.
200The tap devices do not need any IP configuration when bridged and
201can be assigned the same ether MAC (in fact they have to be
202if you want the failover to work nicely).
203.Sh SEE ALSO
204.Xr pf 4 ,
205.Xr ifconfig 8
206.Sh HISTORY
207The
208.Nm
209driver first appeared in
210.Ox 2.5
211and found its way into
212.Dx 1.3 .
213Transparent bridging (link0) was added in
214.Dx 2.9
215in 2011.
216.Sh AUTHORS
217.An -nosplit
218The
219.Nm
220driver was originally written by
221.An Jason L. Wright
222.Aq jason@thought.net
223as part of an undergraduate independent study at the University of
224North Carolina at Greensboro.
225.Pp
226This version of the
227.Nm
228driver has been heavily modified from the original version by
229.An Jason R. Thorpe
230.Aq thorpej@wasabisystems.com .
231.Sh BUGS
232The
233.Nm
234driver currently supports only Ethernet and Ethernet-like (e.g. 802.11)
235network devices, with exactly the same interface MTU size as the bridge device.
236