| 1 |
27 |
unneback |
<html>
|
| 2 |
|
|
<body>
|
| 3 |
|
|
<pre>
|
| 4 |
|
|
NAME
|
| 5 |
|
|
bridge - Ethernet bridge interface
|
| 6 |
|
|
|
| 7 |
|
|
SYNOPSIS
|
| 8 |
|
|
pseudo-device bridge 2
|
| 9 |
|
|
|
| 10 |
|
|
DESCRIPTION
|
| 11 |
|
|
The bridge device creates a logical link between two or more Ethernet in
|
| 12 |
|
|
terfaces. This link between the interfaces selectively forwards frames
|
| 13 |
|
|
from each interface on the bridge to every other interface on the bridge.
|
| 14 |
|
|
A bridge can serve several services, including, isolation of traffic be
|
| 15 |
|
|
tween sets of machines so that traffic local to one set of machines is
|
| 16 |
|
|
not available on the wire of another set of machines, and it can act as a
|
| 17 |
|
|
transparent filter for ip(4) datagrams.
|
| 18 |
|
|
|
| 19 |
|
|
The bridges provided by this interface are learning bridges with IP fil
|
| 20 |
|
|
tering, see ipf(4). In general a bridge works like a hub, forwarding
|
| 21 |
|
|
traffic from one interface to another. It differs from a hub in that it
|
| 22 |
|
|
will "learn" which machines are on each of its attached segments by ac
|
| 23 |
|
|
tively listening to incoming traffic and examining the headers of each
|
| 24 |
|
|
frame. A table is built containing the MAC address and segment to which
|
| 25 |
|
|
the MAC address is attached. This allows a bridge to be more selective
|
| 26 |
|
|
about what it forwards, which can be used to reduce traffic on a set of
|
| 27 |
|
|
segments and also to provide an IP firewall without changing the topology
|
| 28 |
|
|
of the network.
|
| 29 |
|
|
|
| 30 |
|
|
The algorithm works as follows by default, but can be modified via
|
| 31 |
|
|
ioctl(2) or the utility brconfig(8). When a frame comes in, the origin
|
| 32 |
|
|
segment and the source address are recorded. If the bridge has no knowl
|
| 33 |
|
|
edge about where the destination is to be found, the bridge will forward
|
| 34 |
|
|
the frame to all attached segments. If the destination is known to be on
|
| 35 |
|
|
a different segment from its origin, the bridge will forward the packet
|
| 36 |
|
|
only to the destination segment. If the destination is on the same seg
|
| 37 |
|
|
ment as the origin segment, the bridge will drop the packet because the
|
| 38 |
|
|
receiver has already had a chance to see the frame. Before forwarding a
|
| 39 |
|
|
frame, the bridge will check to see if the packet contains an ip(4) data
|
| 40 |
|
|
gram; if so, the datagram is run through the ipf(4) interface so that it
|
| 41 |
|
|
can be filtered. Only the ipf(4) input rules for the source interface
|
| 42 |
|
|
are checked with the datagram; output rules have no effect.
|
| 43 |
|
|
|
| 44 |
|
|
IOCTLS
|
| 45 |
|
|
A bridge interface responds to all of the ioctl(2) calls specific to oth
|
| 46 |
|
|
er interfaces listed in netintro(4). The following ioctl(2) calls are
|
| 47 |
|
|
specific to bridge devices. They are defined in <sys/sockio.h>.
|
| 48 |
|
|
|
| 49 |
|
|
SIOCBRDGIFS (struct ifbifconf) Retrieve member interface list from a
|
| 50 |
|
|
bridge. This request takes an ifbifconf structure (see
|
| 51 |
|
|
below) as a value-result parameter. The ifbic_len field
|
| 52 |
|
|
should be initially set to the size of the buffer point
|
| 53 |
|
|
ed to by ifbic_buf. On return it will contain the
|
| 54 |
|
|
length, in bytes, of the configuration list. Alterna
|
| 55 |
|
|
tively, if the ifbic_len passed in is set to 0,
|
| 56 |
|
|
SIOCBRDGIFS will set ifbic_len to the size that
|
| 57 |
|
|
ifbic_buf needs to be to fit the entire configuration
|
| 58 |
|
|
list, and will not fill in the other parameters. This
|
| 59 |
|
|
is useful for determining the exact size that ifbic_buf
|
| 60 |
|
|
needs to be in advance.
|
| 61 |
|
|
|
| 62 |
|
|
The argument structure is defined as follows:
|
| 63 |
|
|
|
| 64 |
|
|
struct ifbreq {
|
| 65 |
|
|
char ifbr_name[IFNAMSIZ]; /* brdg nam */
|
| 66 |
|
|
char ifbr_ifsname[IFNAMSIZ]; /* if name */
|
| 67 |
|
|
u_int32_t ifbr_ifsflags; /* if flags */
|
| 68 |
|
|
};
|
| 69 |
|
|
|
| 70 |
|
|
#define IFBIF_LEARNING 0x1 /* learns addrs */
|
| 71 |
|
|
#define IFBIF_DISCOVER 0x2 /* gets fwd'd pkts */
|
| 72 |
|
|
|
| 73 |
|
|
struct ifbifconf {
|
| 74 |
|
|
char ifbic_name[IFNAMSIZ]; /* brdg name */
|
| 75 |
|
|
u_int32_t ifbic_len; /* buf size */
|
| 76 |
|
|
union {
|
| 77 |
|
|
caddr_t ifbicu_buf; /* buffer */
|
| 78 |
|
|
struct ifbreq *ifbicu_req;
|
| 79 |
|
|
} ifbic_ifbicu;
|
| 80 |
|
|
#define ifbic_buf ifbic_ifbicu.ifbicu_buf
|
| 81 |
|
|
#define ifbic_req ifbic_ifbicu.ifbicu_req
|
| 82 |
|
|
};
|
| 83 |
|
|
|
| 84 |
|
|
SIOCBRDGADD (struct ifbreq) Add the interface named in ifbr_ifsname
|
| 85 |
|
|
to the bridge named in ifbr_name.
|
| 86 |
|
|
|
| 87 |
|
|
SIOCBRDGDEL (struct ifbreq) Delete the interface named in
|
| 88 |
|
|
ifbr_ifsname from the bridge named in ifbr_name.
|
| 89 |
|
|
|
| 90 |
|
|
SIOCBRDGSIFFLGS (struct ifbreq) Set the bridge member interface flags
|
| 91 |
|
|
for the interface named in ifbr_ifsname attached to the
|
| 92 |
|
|
bridge ifbr_name. If the flag IFBIF_LEARNING is set on
|
| 93 |
|
|
an interface, source addresses from frames received on
|
| 94 |
|
|
the interface are recorded in the address cache. If the
|
| 95 |
|
|
flag IFBIF_DISCOVER is set, the interface will receive
|
| 96 |
|
|
packets destined for unknown destinations, otherwise a
|
| 97 |
|
|
frame that has a destination not found in the address
|
| 98 |
|
|
cache is not forwarded to this interface. The default
|
| 99 |
|
|
for newly added interfaces has both flags set. If the
|
| 100 |
|
|
flag IFBIF_BLOCKNONIP is set, packets that are one of
|
| 101 |
|
|
ip(4), ip6(4), arp(4), or Reverse ARP, will not be
|
| 102 |
|
|
bridged from and to the interface.
|
| 103 |
|
|
|
| 104 |
|
|
SIOCBRDGGIFFLGS Retrieve the bridge member interface flags for the in
|
| 105 |
|
|
terface named in ifbr_ifsname attached to the bridge
|
| 106 |
|
|
ifbr_name.
|
| 107 |
|
|
|
| 108 |
|
|
SIOCBRDGRTS (struct ifbaconf) Retrieve the address cache of the
|
| 109 |
|
|
bridge named in ifbac_name. This request takes an
|
| 110 |
|
|
ifbaconf structure (see below) as a value result parame
|
| 111 |
|
|
ter. The ifbac_len field should be initially set to the
|
| 112 |
|
|
size of the buffer pointed to by ifbac_buf. On return,
|
| 113 |
|
|
it will contain the length, in bytes, of the configura
|
| 114 |
|
|
tion list. Alternatively, if the ifbac_len passed in is
|
| 115 |
|
|
set to 0, SIOCBRDGRTS will set it to the size that
|
| 116 |
|
|
ifbac_buf needs to be to fit the entire configuration
|
| 117 |
|
|
list and not fill in the other parameters. As with
|
| 118 |
|
|
SIOCBRDGIFS, this is useful for determining the exact
|
| 119 |
|
|
size that ifbac_buf needs to be in advance.
|
| 120 |
|
|
|
| 121 |
|
|
The argument structure is defined as follows:
|
| 122 |
|
|
|
| 123 |
|
|
struct ifbareq {
|
| 124 |
|
|
char ifba_name[IFNAMSIZ]; /* brdg nam */
|
| 125 |
|
|
char ifba_ifsname[IFNAMSIZ];/* dest ifs */
|
| 126 |
|
|
u_int8_t ifba_age; /* addr age */
|
| 127 |
|
|
u_int8_t ifba_flags; /* addr flag */
|
| 128 |
|
|
struct ether_addr ifba_dst; /* dst addr */
|
| 129 |
|
|
};
|
| 130 |
|
|
|
| 131 |
|
|
#define IFBAF_TYPEMASK 0x03 /* addr type mask */
|
| 132 |
|
|
#define IFBAF_DYNAMIC 0x00 /* dynamic addr */
|
| 133 |
|
|
#define IFBAF_STATIC 0x01 /* static address */
|
| 134 |
|
|
|
| 135 |
|
|
struct ifbaconf {
|
| 136 |
|
|
char ifbac_name[IFNAMSIZ]; /* brdg name */
|
| 137 |
|
|
u_int32_t ifbac_len; /* buf size */
|
| 138 |
|
|
union {
|
| 139 |
|
|
caddr_t ifbacu_buf; /* buf */
|
| 140 |
|
|
struct ifbareq *ifbacu_req;
|
| 141 |
|
|
} ifbac_ifbacu;
|
| 142 |
|
|
#define ifbac_buf ifbac_ifbacu.ifbacu_buf
|
| 143 |
|
|
#define ifbac_req ifbac_ifbacu.ifbacu_req
|
| 144 |
|
|
};
|
| 145 |
|
|
Address cache entries with the type set to IFBAF_DYNAMIC
|
| 146 |
|
|
in ifba_flags are entries learned by the bridge. En
|
| 147 |
|
|
tries with the type set to IFBAF_STATIC are manually
|
| 148 |
|
|
added entries.
|
| 149 |
|
|
|
| 150 |
|
|
SIOCBRDGSADDR (struct ifbareq) Add an entry, manually, to the address
|
| 151 |
|
|
cache for the bridge named in ifba_name. The address and
|
| 152 |
|
|
its associated interface and flags are set in the
|
| 153 |
|
|
ifba_dst, ifba_ifsname, ifba_flags fields, respectively.
|
| 154 |
|
|
|
| 155 |
|
|
SIOCBRDGDADDR (struct ifbareq) Delete an entry from the address cache
|
| 156 |
|
|
of the bridge named in ifba_name. Entries are deleted
|
| 157 |
|
|
strictly based on the address field ifba_dst.
|
| 158 |
|
|
|
| 159 |
|
|
SIOCBRDGSCACHE (struct ifbcachereq) Set the maximum address cache size
|
| 160 |
|
|
for the bridge named in ifbc_name to ifbc_size entries.
|
| 161 |
|
|
|
| 162 |
|
|
The argument structure is as follows:
|
| 163 |
|
|
|
| 164 |
|
|
struct ifbcachereq {
|
| 165 |
|
|
char ifbc_name[IFNAMSIZ]; /* bridge */
|
| 166 |
|
|
u_int32_t ifbc_size; /* size */
|
| 167 |
|
|
};
|
| 168 |
|
|
|
| 169 |
|
|
SIOCBRDGGCACHE (struct ifbcachereq) Retrieve the maximum size of the
|
| 170 |
|
|
address cache for the bridge ifbc_name.
|
| 171 |
|
|
|
| 172 |
|
|
SIOCBRDGSTO (struct ifbcachetoreq) Set the time, in seconds, that
|
| 173 |
|
|
addresses which have not been seen on the network
|
| 174 |
|
|
(transmitted a packet) remain in the cache. If the time
|
| 175 |
|
|
is set to zero, no aging is performed on the address
|
| 176 |
|
|
cache. The argument structure is as follows:
|
| 177 |
|
|
|
| 178 |
|
|
struct ifbcachetoreq {
|
| 179 |
|
|
char ifbct_name[IFNAMSIZ]; /* bridge */
|
| 180 |
|
|
u_int32_t ifbct_time; /* time */
|
| 181 |
|
|
};
|
| 182 |
|
|
|
| 183 |
|
|
SIOCBRDGGTO (struct ifbcachetoreq) Retrieve the address cache expi
|
| 184 |
|
|
ration time (see above).
|
| 185 |
|
|
|
| 186 |
|
|
SIOCBRDGFLUSH (struct ifbreq) Flush addresses from the cache.
|
| 187 |
|
|
ifbr_name contains the name of the bridge device, and
|
| 188 |
|
|
ifbr_ifsflags should be set to IFBF_FLUSHALL to flush
|
| 189 |
|
|
all addresses from the cache or IFBF_FLUSHDYN to flush
|
| 190 |
|
|
only the dynamically learned addresses from the cache.
|
| 191 |
|
|
|
| 192 |
|
|
SIOCBRDGARL (struct ifbrlreq) Add a filtering rule to the bridge
|
| 193 |
|
|
named in ifbr_name on the interface named in
|
| 194 |
|
|
ifbr_ifsname. The argument structure is as follows:
|
| 195 |
|
|
|
| 196 |
|
|
struct ifbrlreq {
|
| 197 |
|
|
char ifbr_name[IFNAMSIZ]; /* bridge */
|
| 198 |
|
|
char ifbr_ifsname[IFNAMSIZ]; /* ifs */
|
| 199 |
|
|
u_int8_t ifbr_action; /* handling */
|
| 200 |
|
|
u_int8_t ifbr_flags; /* flags */
|
| 201 |
|
|
struct ether_addr ifbr_src; /* src mac */
|
| 202 |
|
|
struct ether_addr ifbr_dst; /* dst mac */
|
| 203 |
|
|
};
|
| 204 |
|
|
#define BRL_ACTION_BLOCK 0x01
|
| 205 |
|
|
#define BRL_ACTION_PASS 0x02
|
| 206 |
|
|
#define BRL_FLAG_IN 0x08
|
| 207 |
|
|
#define BRL_FLAG_OUT 0x04
|
| 208 |
|
|
#define BRL_FLAG_SRCVALID 0x02
|
| 209 |
|
|
#define BRL_FLAG_DSTVALID 0x01
|
| 210 |
|
|
|
| 211 |
|
|
Rules are applied in the order in which they were added
|
| 212 |
|
|
to the bridge, and the first matching rule's action pa
|
| 213 |
|
|
rameter determines the fate of the packet. The
|
| 214 |
|
|
ifbr_action parameter specifies whether a frame matching
|
| 215 |
|
|
the rule is to be blocked or passed.
|
| 216 |
|
|
|
| 217 |
|
|
If the BRL_FLAG_IN bit is set in ifbr_flags, then the
|
| 218 |
|
|
rule applies to frames received by the interface. If
|
| 219 |
|
|
the BRL_FLAG_OUT bit is set, then the rule applies to
|
| 220 |
|
|
frame transmitted by the interface. At least one of
|
| 221 |
|
|
BRL_FLAG_IN or BRL_FLAG_OUT must be set.
|
| 222 |
|
|
|
| 223 |
|
|
The source ethernet address in ifbr_src is checked if
|
| 224 |
|
|
the BRL_FLAG_SRCVALID bit is set in ifbr_flags. The des
|
| 225 |
|
|
tination address in ifbr_dst is check if the
|
| 226 |
|
|
BRL_FLAG_DSTVALID bit is set. If neither bit is set,
|
| 227 |
|
|
the rule is matches all frames.
|
| 228 |
|
|
|
| 229 |
|
|
SIOCBRDGFRL (struct ifbrlreq) Remove all filtering rules from a
|
| 230 |
|
|
bridge interface member. ifbr_name contains the name of
|
| 231 |
|
|
the bridge device, and ifbr_ifsname contains the name of
|
| 232 |
|
|
the bridge member interface.
|
| 233 |
|
|
|
| 234 |
|
|
SIOCBRDGGRL (struct ifbrlconf) Retrieve all of the rules from the
|
| 235 |
|
|
bridge, ifbrl_name, for the member interface,
|
| 236 |
|
|
ifbrl_ifsname.
|
| 237 |
|
|
|
| 238 |
|
|
This request takes an ifbrlconf structure (see below) as
|
| 239 |
|
|
a value result parameter. The ifbrl_len field should be
|
| 240 |
|
|
initially set to the size of the buffer pointed to by
|
| 241 |
|
|
ifbrl_buf. On return, it will contain the length, in
|
| 242 |
|
|
bytes, of the configuration list. Alternatively, if the
|
| 243 |
|
|
ifbrl_len passed in is set to 0, SIOCBRDGGRL will set it
|
| 244 |
|
|
to the size that ifbrl_buf needs to be to fit the entire
|
| 245 |
|
|
configuration list and not fill in the other parameters.
|
| 246 |
|
|
As with SIOCBRDGIFS, this is useful for determining the
|
| 247 |
|
|
exact size that ifbrl_buf needs to be in advance.
|
| 248 |
|
|
|
| 249 |
|
|
The argument structure is defined as follows:
|
| 250 |
|
|
|
| 251 |
|
|
struct ifbrlconf {
|
| 252 |
|
|
char ifbrl_name[IFNAMSIZ]; /* brdg nam */
|
| 253 |
|
|
char ifbrl_ifsname[IFNAMSIZ];/* ifs name */
|
| 254 |
|
|
u_int32_t ifbr_len; /* buf len */
|
| 255 |
|
|
union {
|
| 256 |
|
|
caddr_t ifbrlu_buf;
|
| 257 |
|
|
struct ifbrlreq *ifbrlu_req;
|
| 258 |
|
|
} ifrl_ifbrlu;
|
| 259 |
|
|
#define ifbrl_buf ifbrl_ifbrlu.ifbrlu_buf
|
| 260 |
|
|
#define ifbrl_req ifbrl_ifbrlu.ifbrlu_req
|
| 261 |
|
|
};
|
| 262 |
|
|
|
| 263 |
|
|
|
| 264 |
|
|
ERRORS
|
| 265 |
|
|
If the ioctl(2) call fails, errno(2) is set to one of the following val
|
| 266 |
|
|
ues:
|
| 267 |
|
|
|
| 268 |
|
|
[ENOENT] For an add request, this means that the named interface is
|
| 269 |
|
|
not configured into the system. For delete operation, it
|
| 270 |
|
|
means that the named interface is not a member of the
|
| 271 |
|
|
bridge. For a address cache deletion, the address was not
|
| 272 |
|
|
found in the table.
|
| 273 |
|
|
|
| 274 |
|
|
[ENOMEM] Memory could not be allocated for an interface or cache en
|
| 275 |
|
|
try to be added to the bridge.
|
| 276 |
|
|
|
| 277 |
|
|
[EEXIST] The named interface is already a member of the bridge.
|
| 278 |
|
|
|
| 279 |
|
|
[EBUSY] The named interface is already a member of another bridge.
|
| 280 |
|
|
|
| 281 |
|
|
[EINVAL] The named interface is not an Ethernet interface or an in
|
| 282 |
|
|
valid ioctl was performed on the bridge.
|
| 283 |
|
|
|
| 284 |
|
|
[ENETDOWN] Address cache operation (flush, add, delete) on a bridge
|
| 285 |
|
|
that is in the down state.
|
| 286 |
|
|
|
| 287 |
|
|
[EPERM] Super-user privilege is required to add and delete inter
|
| 288 |
|
|
faces to and from bridges and to set the bridge interface
|
| 289 |
|
|
flags.
|
| 290 |
|
|
|
| 291 |
|
|
[EFAULT] The buffer used in a SIOCBRDGIFS or SIOCBRDGRTS request
|
| 292 |
|
|
|
| 293 |
|
|
points outside of the process's allocated address space.
|
| 294 |
|
|
|
| 295 |
|
|
[ESRCH] No such member interface in the bridge.
|
| 296 |
|
|
|
| 297 |
|
|
SEE ALSO
|
| 298 |
|
|
errno(2), ioctl(2), ip(4), ipf(4), netintro(4), bridgename.if(5),
|
| 299 |
|
|
brconfig(8)
|
| 300 |
|
|
|
| 301 |
|
|
HISTORY
|
| 302 |
|
|
The brconfig(8) command and the bridge(4) kernel interface first appeared
|
| 303 |
|
|
in
|
| 304 |
|
|
|
| 305 |
|
|
AUTHOR
|
| 306 |
|
|
The brconfig(8) command and the bridge(4) kernel interface were written
|
| 307 |
|
|
by Jason L. Wright <jason@thought.net> as part of an undergraduate inde
|
| 308 |
|
|
pendent study at the University of North Carolina at Greensboro.
|
| 309 |
|
|
|
| 310 |
|
|
BUGS
|
| 311 |
|
|
There is currently no loop detection. Care must be taken to make sure
|
| 312 |
|
|
that loops are not created when a bridge is brought up.
|
| 313 |
|
|
|
| 314 |
|
|
Only ipf(4) input rules are checked with incoming packet; there is no
|
| 315 |
|
|
easy way to handle output rules.
|
| 316 |
|
|
|
| 317 |
|
|
</pre>
|
| 318 |
|
|
</body>
|
| 319 |
|
|
</html>
|
