3Com Router User Manual

							18
CONFIGURING HDLC
This chapter contains information on the following topics:
■Configure HDLC
■Display and Debug HDLC
Configure HDLCHDLC (High Data Link Control) is a bit-oriented link layer protocol. Its most 
prominent feature is that it can transparently transmit any kind of bit flow without 
the restriction that the data must be character set. Protocols of standard HDLC 
protocol group operate upon the synchronous serial lines, e.g., DDN. The address 
field of HDLC is 8 bits, its control field is 8 bits, and the protocol field is 16 bits, 
which are used to represent all kinds of control information of HDLC protocol and 
to mark whether they are data. The 3Com Router supports the HDLC protocol and 
can connect with HDLC protocol of other popular devices.HDLC configuration 
includes:
■Configure the link layer protocol of the interface to HDLC
1Configure the Link Layer Protocol of the Interface to HDLC
In synchronous interface view, perform the following task.
Ta b l e 318   Configure the link layer protocol of the interface to HDLC
By default, the link layer protocol of the interface is PPP.
Only when the interface operates in the synchronous mode, can the link layer 
protocol be configured to HDLC.
When the interface link layer protocol is SLIP, its physical attribute cannot be 
changed to synchronous mode. At this time, you should first change the link layer 
protocol of the interface to PPP before you change the interface attribute to 
synchronous mode.
Display and Debug 
HDLCTa b l e 319   Display and debug HDLC
OperationCommand
Configure the link layer protocol of the interface to 
HDLClink-protocol hdlc
OperationCommand
Enable all the debugging of HDLC protocoldebugging hdlc all [ interface 
type number ]
Enable HDLC event debuggingdebugging hdlc event [ 
interface type number ] 
						

							288CHAPTER 18: CONFIGURING HDLC
Enable HDLC packet debuggingdebugging hdlc packet [ 
interface type number ] 
						

							19
CONFIGURING BRIDGE
This chapter contains information on the following topics:
■Bridge Overview
■Configure Bridge’s Routing Function
■Display and Debug Bridge
■Typical Bridge Configuration
Bridge OverviewBridge is a type of network device on the data link layer, which interconnects Local 
Area Networks (LANs) and transfers data between them.
 In some small-sized 
networks
, especially in the networks widely dispersed, using bridges can reduce 
the network maintenance cost, and the network terminal users do not need to 
make special settings for the devices, since the bridges interconnect networks just 
like hubs.
In practice, there are four types of bridging:
■Transparent Bridging: Such bridging is used to interconnect networks of the 
same medium. It is mainly applied in the Ethernet environment. Usually, 
transparent bridging keeps a bridging table that records the correlation 
between destination MAC addresses and interfaces.
■Source-route Bridging: Such bridging forwards frames based on the routing 
indicators contained in the frames. The table of correlation between 
destination MAC addresses and routing indicators will be determined and 
maintained by the end stations (the starting and the ending point). This 
bridging is found primarily in the Token Ring environments.
■Translational Bridging: Such bridging is used to interconnect LANs of different 
physical media. It is typically applied to interconnect different types of 
networks, such as Ethernet, Fiber Distributed Data Interface (FDDI) and Token 
Ring.
■Source-route Translational Bridging: As the name implies, such bridging is the 
hybrid of “Source-route Bridging” and “Translational Bridging”. They allow the 
communication in mixed Toke Ring and Ethernet environments.
The transparent bridging supported by the 3Com Router series has the following 
features:
■Conforms to the IEEE 802.1d standards and supports the STP and bridging 
functions specified in IEEE 802.1d.
■Supports bridging on the links of PPP and HDLC.
■Supports bridging on X.25 links. 
						

							290CHAPTER 19: CONFIGURING BRIDGE
■Supports bridging on the Frame Relay links.
■Supports bridging on the sub-interfaces of VLAN.
■Supports bridging on BDR and dialing standby.
■Supports binding of multiple ports and load sharing.
■Support both routing and bridging function for specified protocol.
■Support filtering Ethernet frames according to the MAC address or Ethernet 
frame format.
■Provides command configuration and management functions.
■Provides functions of logging, alarming and debugging.
Main Functions of 
BridgingObtain address table
Bridging implements forwarding in accordance with the bridging table comprised 
of MAC addresses and interfaces. A bridge should obtain the correlation between 
MAC addresses and ports. When the bridge connects with a LAN segment, it will 
detect all the Ethernet frames on this segment. Once the Ethernet frame sent from 
a node is detected, the source MAC address of this frame will be picked up and 
the correlation between this MAC address and the interface receiving this frame 
will be added to the bridging address table.
As shown in the following figure, four workstations A, B, C and D are distributed 
in two LANs: Ethernet segment 1 connected with Bridge port 1 and Ethernet 
segment 2 connected with Bridge port 2. At a certain moment, when Workstation 
A transmits an Ethernet frame to Workstation B, both the bridge and Workstation 
B will receive this frame.
Figure 101   Workstation A transmits information to workstation B on the Ethernet 
segment 1
Upon receiving the Ethernet frame, the bridge learns that Workstation A is 
connected with Bridge port 1 since the frame received is from Port 1. As a result, 
the correlation between the MAC address of Workstation A and Bridge port 1 will 
be added to the bridging table, as shown in the following figure:
Bridge
00e0.fcaa.aaaa00e0.fcbb.bbbb Source addressDestination address
Ethernet segment 1
Bridge port 1
Bridge port 2 00e0.fccc.cccc
00e0.fcdd.dddd
Ethernet segment 2 00e0.fcaa.aaaa00e0.fcbb.bbbb
Workstation AWorkstation B
Workstation CWorkstation D 
						

							Bridge Overview291
Figure 102   Bridge learns that Workstation A is connected with Port 1
Once Workstation B responds to Workstation A, the bridge can detect the 
responding Ethernet frame from Workstation B and learn that Workstation B is 
also connected to Bridge port 1 because the frame is detected on port 1 too. As a 
result, the correlation between the MAC address of Workstation B and Bridge port 
1 is added to the bridging table too, as shown in the following figure:
Figure 103   Bridge learns that Workstation B is connected with the port 1 too.
At last, given that all the workstations are in use, the bridge will obtain all 
correlation between the MAC addresses and the bridge ports as shown in the 
following figure:
Bridge
Ethernet segment 1
Bridge port 1
Bridge port 2
Ethernet segment 2 00e0.fcbb.bbbb
00e0.fcaa.aaaa00e0.fcbb.bbbb Source addressDestination address
Bridging table
MAC
address00e0.fcaa.aaaaPort1
00e0.fcaa.aaaa
00e0.fccc.cccc
00e0.fcdd.dddd Workstation AWorkstation B
Workstation C
Workstation D
Bridge
Ethernet segment 1
Bridge port 1
Bridge port 2 00e0.fccc.cccc
00e0.fcdd.dddd
Ethernet segment 2 Bridging table
MAC
address
00e0.fcbb.bbbb
00e0.fcaa.aaaa
Port
11
00e0.fcbb.bbbb00e0.fcaa.aaaa Source addressDestination address
00e0.fcbb.bbbb 00e0.fcaa.aaaa
Workstation A
Workstation CWorkstation B
Workstation D 
						

							292CHAPTER 19: CONFIGURING BRIDGE
Figure 104   Final bridging address table
Forward and Filter
The bridge will make the decision to forward frames or not (that is, to filter 
frames) depending on the following three conditions:
■If Workstation A sends an Ethernet frame whose destination is Workstation C, 
the bridge will detect this frame and learn that Workstation C corresponds to 
Bridge port 2 by looking up its bridging table. So, it will forward the frame to 
Bridge port 2, as shown in the following figure.
Figure 105   Forward
Note that the bridge will forward the broadcast or multicast frames received on 
one port to the other ports.
Given that Workstation A sends an Ethernet frame to Workstation B, the 
bridge will filter this frame rather than forwarding it, since Workstation B and 
Workstation A are located on the same physical network segment.
Bridge
Ethernet segment 1
Bridge port 1
Bridge port 2
Ethernet segment 2 00e0.fcaa.aaaa
Bridging table
MAC
address
00e0.fcbb.bbbb00e0.fccc.cccc00e0.fcdd.dddd 00e0.fcaa.aaaa
Port
1122
00e0.fcbb.bbbb
00e0.fccc.cccc00e0.fcdd.dddd Workstation A
Workstation CWorkstation B
Workstation D
00e0.fccc.cccc00e0.fcaa.aaaa Destination addressSource address
Bridge Source addressDestination address
Ethernet segment 1
Bridge port 1
Bridge port 2 00e0.fccc.cccc
00e0.fcdd.dddd
Ethernet segment 2 00e0.fcaa.aaaa
00e0.fcbb.bbbbBridging table
MAC
address
00e0.fcbb.bbbb
00e0.fccc.cccc
00e0.fcdd.dddd 00e0.fcaa.aaaa
Port
1
1
2
2
00e0.fcaa.aaaa00e0.fccc.cccc
Forwarding Workstation A Workstation B
Workstation CWorkstation D 
						

							Bridge Overview293
Figure 106   Filter (not forward)
■Suppose that Workstation A sends an Ethernet frame to Workstation C, and 
the bridge does not find the correlation between the MAC address of 
Workstation C and the port in the bridging address table, what will the bridge 
do? The bridge will forward this frame destined to an unknown MAC address 
to all ports except the one on which it is received. In this case, the bridge 
actually plays the role of a hub to make sure the continuous information 
transmission, as shown in the following figure:
Figure 107   No matched MAC address is found in the bridging table
Eliminating loop
As shown in the following figure, both bridges X and Y are connected with 
Ethernet segment 1. Once detecting a broadcasting frame, both bridges will send 
it to all ports except the source port on which the frame is detected. That is, both 
bridges X and Y will forward this broadcast frame.
No forwarding
Bridge
Ethernet segment 1
Bridge port 1
Bridge port 2
Ethernet segment 2
00e0.fcaa.aaaa00e0.fcbb.bbbb Source addressDestination address
Bridging table
MAC
address
00e0.fcbb. bbbb00e0. fccc . cccc00e0.fcdd . dddd00e0.fcaa. aaaa
Port
1122
Workstation A Workstation B
Workstation C Workstation D 00e0.fcaa.aaaa00e0.fcbb.bbbb
00e0.fcdd.dddd
Bridge
Ethernet segment 1
Bridge port 1
Bridge port 2 00e0.fccc.cccc
00e0.fcdd.dddd
Ethernet segment 2 00e0.fcaa.aaaa00e0.fcbb.bbbb
00e0.fcaa.aaaa00e0.fccc.cccc Source addressDestination address
Bridging table
MAC
address
00e0. fcaa. aaaaPort
1100e0. fcbb .bbbb 
						

							294CHAPTER 19: CONFIGURING BRIDGE
Figure 108   Preliminary examination state of bridging loops
As shown in the following figure, the broadcast frame is forwarded over Ethernet 
segment 2 and Ethernet segment 3 that are connected with Bridge Z. Upon 
detecting two copies of this frame on two different ports, Bridge Z forwards them 
to Ethernet segment 3 and Ethernet segment 2 again. Thus, Ethernet segment 2 
and Ethernet segment 3 receive a copy of this frame for the second time. Like this, 
the frame is repeatedly forwarded over the network, which is called bridging loop.
Figure 109   Bridging loop
In practice, if there are hundreds of physical segments, bridging loops will cause a 
sharp decline to the network performance. After the location where loops occur is 
detected, the only solution is to cut off all connections. It is obvious that 
eliminating loops is an essential requirement for ensuring the bridge working 
normally. Therefore, the third function of bridge is to locate loops and block 
redundant ports.
Spanning Tree ProtocolSpanning Tree Protocol (STP) is used to prevent redundant paths through certain 
algorithms. A loop network is thus pruned to be a loop-free tree network so as to 
avoid the infinite cycling of data frames in the loop network.
STP transmits a type of special data frame called Bridge Protocol Data Unit (BPDU) 
between bridges. The overall network will compute a minimum spanning tree 
describing the distribution of bridges in the network. This minimum spanning tree 
Broadcast address
Bridge XBridge Y
Bridge Z
Ethernet segment 1
Ethernet segment 2
Ethernet segment 3
FFFFFFFFFFFF
Bridge X
Bridge Y
Bridge Z
Broadcast frame
Ethernet segment 1
Ethernet segment 2
Ethernet segment 3 Forwarding broadcast frame
Forwarding broadcast frame
Forwarding broadcast frame againForwarding broadcast frame again
FFFFFFFFFFFF
FFFFFFFFFFFF
FFFFFFFFFFFF
FFFFFFFFFFFF
FFFFFFFFFFFF 
						

							Bridge Overview295
will also specify which bridge to be the “root bridge” and which bridges to be the 
“leaf nodes”.
A BPDU contains the following information:
■Root Identifier: Consists of the Bridge Priority and the MAC address of the root 
bridge.
■Root Path Cost: Path cost from the individual leaf nodes to the root bridge.
■Bridge Identifier: Consists of the Bridge priority and the MAC address of the 
current bridge.
■Port Identifier: Consists of the Port Priority and the Port Number.
■Message Age of BPDU
■Max Age of BPDU
■Hello Time of BPDU
■Forward Delay of port state transition
Spanning Tree Topology
■Specify the root bridge. The bridge with the smallest Bridge Identifier will be the 
root bridge of the local network.
■Specify the designated bridge. Designated bridge is the one directly connected 
with the current (subordinate) bridge and responsible for forwarding data to 
the current (subordinate) bridge. The path cost via a designated bridge is the 
lowest between the leaf nodes and root bridge.
■Specify the designated port. Designated ports are those on the designated 
bridge and responsible for forwarding data to the subordinate bridges. The 
path cost of BPDUs sent on a designated port will be the lowest.
■Specify the root port. Root port refers to the one on the current bridge and 
responsible for receiving the data forwarded by the designated bridge.
■Specify blocked ports. Except the designated ports and the root ports, all other 
ports will be blocked and are called blocked ports.
Upon the computation of the minimum spanning tree, the newly generated root 
port and designated ports begin to forward packets after a period of forward 
delay. After all the bridges on the network accomplish the spanning tree 
computation, the network topology will be stabilized and will remain the same 
until the network takes changes.
The following figure illustrates the topology of the minimum spanning tree on a 
network: 
						

							296CHAPTER 19: CONFIGURING BRIDGE
Figure 110   Spanning tree topology
BPDU Forwarding Mechanism
Upon the initiation of the network, all the bridges assume themselves as the root 
bridge. The designated interface of the bridge regularly sends its BPDU once a 
Hello Time. If it is the root port receives the BPDU, it will increase the Message Age 
carried in the BPDU and enable the timer to time this BPDU. If a path fails, the root 
port on this path will not receive new BPDUs any more and old BPDUs will be 
discarded due to timeout, which will result in the spanning tree recompilation. A 
new path will thus be generated to replace the failed one.
However, the recomputed new BPDU will not be propagated throughout the 
network right away, so the old root port and designated ports that have not 
detected the topology changes will still forward the data through the old path. If 
the newly elected root port and designated ports begin to forward data 
immediately, a temporary loop may be introduced. In STP, a transitional state 
mechanism is thus adopted. Specifically, the root port and the designated ports 
will undergo a transitional state for an interval of forward delay to enter the 
forwarding state to resume the data forwarding. Such a delay ensures that the 
new BPDU has already been propagated throughout the network before the data 
frames are forwarded according to the latest topology.
Multi-Protocol RouterGenerally, a router is called multi-protocol router when it can implement the 
routed protocols like IP and IPX, as well as the bridging protocol. For a 
multi-protocol router, the bridging protocol can be either enabled or disabled. 
However, if both the routed protocols and the bridging protocols are enabled on a 
router, the router will be taken as a multi-protocol router. In this case, whether a 
packet should be routed through IP or IPX or forwarded via the bridge will depend 
on the protocol type of the packet. For example, bridging protocol and IP are 
concurrently enabled on a router. If the packet to be processed is an IP packet, it 
Bridge 1
Bridge 2Bridge 4 Bridge 3
Bridge 5
Hub HubDP
DP DP
RP
DPRP
RP
RP
DP
DPDP
DP
DP
DP
DP
DP
DP
RP = Root Port
DP= Designated Port  Designated
Bridge  Designated
Bridge  Designated
Bridge
 Designated
Bridge  Root Bridge/
Designed Bridge
DP