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.Bl -tag -width indent 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.El 113.Pp 114If your network becomes glitchy, with long pauses in tcp sessions, then 115transparent bridging mode is likely the cause. This mode should only be 116used when you are bridging networks with devices that do MAC-based security 117or firewalling (for example, the supremely braindead at&t uverse router), 118or which impose severe limitations on MAC:IP assignments. 119.Pp 120If member interfaces constantly enter a 'blocked (link1)' state then the 121other end of those interfaces is not implementing the link1 keepalive. 122Both sides must implement the keepalive. 123.Pp 124If you get an enormous amount of packet loss and are using link2-based 125bonding, then the other side of those member interfaces are probably 126not implementing link2-based bonding. 127.Pp 128The 129.Nm 130driver implements the IEEE 802.1D Spanning Tree protocol (STP). 131Spanning Tree is used to detect and remove loops in a network topology. 132.Pp 133Packet filtering can be used with any firewall package that hooks in via the 134.Xr pfil 9 135framework. 136When filtering is enabled, bridged packets will pass through the filter 137inbound on the originating interface, on the bridge interface and outbound on 138the appropriate interfaces. 139Either stage can be disabled, this behaviour can be controlled using 140.Xr sysctl 8 : 141Set 142.Va net.link.bridge.pfil_member 143to 144.Li 1 145to enable filtering on the incoming and outgoing member interfaces 146and set 147.Va net.link.bridge.pfil_bridge 148to 149.Li 1 150to enable filtering on the bridge interface. 151.Pp 152ARP and REVARP packets are forwarded without being filtered and others 153that are not IP nor IPv6 packets are not forwarded when filtering is 154enabled. 155.Pp 156Note that packets to and from the bridging host will be seen by the 157filter on the interface with the appropriate address configured as well 158as on the interface on which the packet arrives or departs. 159.Pp 160The MTU of the first member interface to be added is used as the bridge MTU, 161all additional members are required to have exactly the same value. 162.Sh EXTRA FEATURES 163.Dx 164implements two additional features to make spanning tree operation more 165resilient. 166.Pp 167Specifying 168.Cm link0 169on the bridge interface places the bridge in transparent bridging mode. 170The bridge will make every attempt to retain the original source MAC in 171the ethernet link header. 172.Pp 173Specifying 174.Cm link1 175on the bridge interface forces the bridge to generate a 802.11d CFG 176message on every hello interval for all interfaces participating 177in the STP protocol. 178Normally CFG messages are only generated by the root bridge interface 179or during topology changes. 180In addition the bridge code expects to receive 802.11d frames from 181all interface participating in the STP protocol. 182.Pp 183An interface which fails to receive a 802.11d frame within 10 times 184the hello interval (usually 20 seconds) automatically goes into 185l1blocking mode, which can be observed in the ifconfig output for 186the bridge. This removes the interface from consideration and the 187bridge code automatically routes around it. 188.Pp 189Using 190.Cm link0 191and 192.Cm link1 193together between two 194.Dx 195boxes allows you to maintain multiple parallel vpns between those 196boxes via different networks (if you happen to be on more than one 197with internet access). 198Use separate openvpn instances and tap devices for each vpn link 199to accomplish this, placing them in the same bridge interface on 200the two endpoints. 201The tap devices do not need any IP configuration when bridged and 202can be assigned the same ether MAC (in fact they have to be 203if you want the failover to work nicely). 204.Sh SEE ALSO 205.Xr pf 4 , 206.Xr ifconfig 8 207.Sh HISTORY 208The 209.Nm 210driver first appeared in 211.Ox 2.5 212and found its way into 213.Dx 1.3 . 214Transparent bridging (link0) was added in 215.Dx 2.9 216in 2011. 217.Sh AUTHORS 218.An -nosplit 219The 220.Nm 221driver was originally written by 222.An Jason L. Wright Aq Mt 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 Aq Mt thorpej@wasabisystems.com . 230.Sh BUGS 231The 232.Nm 233driver currently supports only Ethernet and Ethernet-like (e.g. 802.11) 234network devices, with exactly the same interface MTU size as the bridge device. 235