1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
|
.\" "%W% %G%"
.TH brcfg\ 0.29 8
.SH NAME
brcfg \- Configure and monitor Linux Kernel Bridging functions
.SH SYNOPSIS
.B brcfg [ start ] [ stop ] [ status ] [ STP ] [ debug on | off ]
.TP
.B brcfg stats [ zero | enable | disable | show ]
.TP
.B brcfg [ policy accept|reject ] [ list ] [ exempt <prot1> <prot2> .. ]
.TP
.B brcfg [ port <nr> [ disable | enable | priority <val> | pathcost <val> ] ]
.TP
.B brcfg [ device <ifname> [ disable | enable | priority <val> | pathcost <val> ] ]
.SH DESCRIPTION
.I Brcfg
will display the current bridge status without parameters. Specifying
parameters may alter the bridge operation in the kernel.
.IBrcfg
will fail when the kernel has not been specifically compiled to
include bridging. This tool wll only work with later 2.2/2.3 kernels and
2.4 kernels.
.SH Theory of Operation
Bridging is possible if the machine you are running Linux on has multiple
Ethernet interfaces to make both ethernet seem to be one for all protocols
on both networks.
The Kernel uses an intelligent algorithm to perform bridging and to
syncronize/exchange information with other bridges on the Networks. The kernel will listen to MAC
addresses on all interfaces and build up tables describing which MAC address
is located on which interface. Broadcast packets or packets whose
destination interface is not known are forwarded to all attached interfaces
(they are "flooded").
The bridging happens at the lowest level of the Ethernet protocol and is
transparent for all higher protocols such as DecNet LAT or Netbeui or
whatever people have come up with to run on Ethernet.
.SH PROBLEMS
Trying to route regular protocols (TCP/IP, EtherTalk etc) will not work
correctly on interfaces that are bridged. Use the logical brg0 interface
to attach the machine to the bridged network.
Bridging uses promiscuous mode that means your machine will process
.B every
packet and check if its to be forwarded to other networks. That will cost
a lot in CPU resources. Be sure to use a good network card able to deliver
an incoming full network load into the TCP/Stack of Linux.
.SH OPTIONS
.TP
.B debug [ on | off ]
Switch the bridge / port into debugging mode. Huge amounts of information
will be logged via the kernel describing what the bridge is doing.
.TP
.B device <ifname> [ enable | disable | priority <value> | pathcost <value> ]
With additional arguments modifies the behavior of the port for interface,
otherwise it displays it. The priority is important when you have multiple
bridges between the same networks. Pathcost is the cost for forwarding over
this interface's segment to another bridge.
.TP
.B exempt <prot1> <prot2> ...
exempt specified protocols from the default policy (see policy option).
Usually this is used to specify which protocols should not be bridged (with
the default "accept" policy). You can get a list of supported protocols with
the "list" option. Please switch on protocol statistics (see stats option)
before exempting a protocol to know as what packet type your network
drivers recognize packets.
.br
The exempt option must be the final option on a line. It will assume all
following values on the commandline are protocol specifiers.
.br
If you know what you are doing then you can also specify a protocol number
from <net/if_ether.h> as a parameter to exempt.
.TP
.B list
Lists all protocols the brcfg program understands.
.TP
.B pathcost <value>
Assumes some priority to a path across the bridge. This is only effective
in a multi-bridge situation.
.TP
.B policy [ accept | reject ]
The policy of the bridge. "reject" means that no protocols will be bridged
unless specified with the "exempt" option. "accept" means that all protocols
will be bridged unless specified with the "exempt" option. The default is
"accept".
.br
Changing the policy will erase the list of protocols exempted.
.TP
.B port <nr> [ enable | disable | priority <value> | pathcost <value> ]
With additional arguments modifies the behavior of a port, otherwise it displays
it. The priority is important when you have multiple bridges between the same
networks. Pathcost is the cost for forwarding over this port's segment to another
bridge.
.TP
.B priority <value>
Assign a priority for this bridge. Applicable only to multi-bridge scenario.
.TP
.B start
Starts the bridge. On bootup the bridge is disabled so that options can be
set using the brcfg command before bringing the bridge up.
.TP
.B stop
Stops bridge operation.
.TP
.B stats [show]
Shows the current counters / protocols encountered on the networks.
.TP
.B stats zero
Zeros the list of protocols and counters in the kernel.
.TP
.B stats enable
Enables the recording of protocols and the counting of packets for each
protocol type in the kernel. This function is OFF by default.
.TP
.B stats disable
Disables the keeping of statistics.
.TP
With no options displays the current bridge status.
.TP
.SH AUTHOR
.PP
.I Unknown + Christoph Lameter + Matthew Grant
.RS
.nf
clameter@debian.org (Christoph Lameter)
.fi
.nf
grantma@anathoth.gen.nz (Matthew Grant)
.fi
.RE
.SH COPYRIGHT GPL
|
