WO2012068866A1

WO2012068866A1 - Method and system for refreshing media access control in ethernet ring network

Info

Publication number: WO2012068866A1
Application number: PCT/CN2011/075522
Authority: WO
Inventors: 曾红李
Original assignee: 中兴通讯股份有限公司
Priority date: 2010-11-24
Filing date: 2011-06-09
Publication date: 2012-05-31
Also published as: CN102025561B; CN102025561A

Abstract

The present invention provides a method and system for refreshing Media Access Control (MAC) in an Ethernet ring network. The Ethernet ring network includes first type nodes comprising intermediate nodes and boundary nodes which perform the same private ring network protocol and public protocol, and at least one second type node which can perform the public protocol. The method includes the following steps: all of nodes in the Ethernet ring network detect the states of Ethernet ring network links (S201); when a first type node detects that a fault occurs in an Ethernet ring network link, the first type node transmits a topology change message using the public protocol to the second type nodes through the boundary nodes of the first type node, and a MAC forwarding address table of the second type node is refreshed (S202); when a second type node detects that the fault occurs in the Ethernet ring network link, by transmitting a topology change message using the public protocol to the boundary nodes of the first type node, the MAC forwarding address table of the first type node is refreshed (S203). The present invention solves the problem that handover time cannot meet the carrier-class requirement due to inability to successfully perform butt joint and inability to rapidly refresh MAC, when the devices of different manufacturers form a ring network.

Description

Method and system for refreshing MAC in Ethernet ring network

The present invention relates to the field of communications, and in particular, to a method and system for refreshing a MAC (Media Access Control) in an Ethernet ring network. Background technique

With the widespread use of Ethernet in metropolitan area networks, the size of Layer 2 networks is also growing. The performance of Ethernet networks, such as reliability, has become the focus of telecom operators. In a switched network, the primary means of improving network reliability is to deploy redundant links so that the backup link can be used when the primary link fails. In the case of redundant link backup, the protection and recovery capabilities of the network become the key, especially the development of real-time services such as voice and data puts higher requirements on the network's fault recovery capability. Traditional Ethernet resilience is usually implemented through the STP (Spanning Tree Protocol) protocol, but their self-healing speed is too slow and needs to be in minutes and seconds to meet the requirements of metro services. In order to enable Ethernet to be applied to metropolitan area networks, the current focus on Ethernet ring protection technology.

In the practical application of the Ethernet ring network, there are usually multiple communication paths, which implements redundant backup between the primary path and the alternate path. When both the primary path and the alternate path are intact, the protected data forwarding function of the standby path is blocked, and the protection data between the networks is transmitted on the primary path. When the primary path fails, the protection data forwarding function of the alternate path is enabled, and the protection data of the network is switched to the alternate path for transmission, thereby realizing the network to switch from the normal state of the path to the fault state, thereby improving the network's anti-fault capability.

Figure la is a schematic diagram of the protection technology of the Ethernet ring network. As shown in Figure la, nodes A1, A2, A3, and B are nodes supporting the Ethernet ring switching function. Network N and Node B are connected, and network M and node A2 are connected. . There are two physical paths that can be selected when communicating between Network N and Network M. Bar, one N<->B<->A1<->A2<->M, the other one is N<->B<->A3<->A2<-> M. In the Ethernet ring protection technology, the master node, the transit node, the master port of the master node, the slave port of the master node, and the ring protection link are usually involved. If the Ethernet ring network is fault-free, the data packet is blocked on the ring to prevent the loop from being formed. The link that is formed by the loop protection link can be used to perform the primary path and backup of the ring network. The switching of the path. The master node is directly connected to the ring protection link, and blocks the data packets forwarded by the port directly connected to the ring protection link if the ring network is not faulty. The primary node is also known as the control node. On the Ethernet ring network, you can select a node with Ethernet switching function as the master node. The port that the master node directly connects to the ring protection link is the slave port when the ring network is faulty. The other port that sends the probe loop packet is the master port. As shown in Figure la, the nodes protected by the ring network are Al, A2, A3, and B, and the links included are <A1, A2>, <A2, A3>, <A3, B>, and <8, into 1 >. The node A2 is the master node, and the link <input 2, incoming 3> directly connected to the port 22 of the node A2 is a ring protection link. Port 21 of A2 is the master port, and port 22 is the slave port. The other nodes on the ring except A2 are transit nodes.

When the link on the ring is intact, the master node blocks the data forwarding function of the slave port connected to the ring protection link. No loop occurs in the network, which prevents the "broadcast storm" caused by the network loop. As shown in Figure lb, the communication paths of networks N and M at this time are: N<->B->A1<->A2<->M.

When the link on the ring fails, the master node releases the data packet forwarding function of the slave port, and informs other transit nodes to refresh their MAC forwarding address table.

The master node can learn the fault status of the link on the ring through the alarm packet. When a link of the ring network fails, the two transmission nodes directly connected to the faulty link detect the fault, and send fault alarms (LINK-DOWN) on the ring network through the faultless port. After receiving the LINK-DOWN packet, the node does not process the LINK-DOWN packet and forwards the LINK-DOWN packet to the master node. After receiving the LINK-DOWN packet, the master node learns that the link on the ring is faulty. As shown in Figure lc, when link <1, in 2> fails When the nodes A1 and A2 detect a link fault, A1 periodically sends a LINK-DOWN packet on the ring through port 11 to notify the master node that there is a link failure on the ring of A2. A2 is the master node. Therefore, LINK-DOWN packets are no longer sent to other transit nodes.

When the master node learns that the link on the ring is faulty, it notifies other transit nodes on the ring to refresh its MAC forwarding address table. After receiving the LINK-DOWN packet, the master node releases the data packet forwarding function of the slave port, refreshes the MAC address forwarding table, and periodically sends ring network refresh packets on the ring network through the master port and the slave port. RING_DOWN_FLUSH_FDB ), when the other transit nodes on the ring receive the RING_DOWN_FLUSH_FDB packet, they refresh their MAC forwarding address table to implement Ethernet ring network protection. As shown in Figure lc, the <A1-A2> link on the ring is faulty. The master node A2 releases the data packet forwarding function from port 22 and periodically sends the ring network through the slave port 22 on the ring network. After the packet is refreshed, the other transit nodes A3, B, and A1 on the ring network refresh their MAC address forwarding table after receiving the ring refresh packet. The communication paths of networks N and M are updated as follows: N<->B<->A3<->A2<->M.

In the prior art, when a link fault occurs in the ring network and the ring network is switched, if the nodes on the ring are composed of devices of different manufacturers, the switching time usually fails to meet the carrier-class requirements. It can be seen from the Ethernet ring switching process that refreshing the MAC forwarding address table is the key to determining the speed of the Ethernet ring switching. However, when the devices of different manufacturers form a ring network, the ring network protocols of different manufacturers have different implementation mechanisms, and the devices of different manufacturers cannot be successfully connected. Therefore, the MAC cannot be quickly refreshed, thus making the switching time. It is difficult to be less than 50ms required by the carrier class.

From the above analysis, it can be seen that in the ring network composed of devices of different manufacturers, due to different ring network protocols of different manufacturers, the switching time of the Ethernet ring network is less than 50 ms required by the carrier class. Therefore, it is very meaningful to propose a new method and a new system for refreshing the MAC in the Ethernet ring network. Summary of the invention

An object of the present invention is to provide a method and system for refreshing a MAC in an Ethernet ring network, which can better solve the problem that when devices of different manufacturers form an Ethernet ring network, the MAC cannot be quickly refreshed. The problem of not meeting the requirements of the carrier level is not achieved.

According to an aspect of the present invention, there is provided a method for refreshing a MAC in an Ethernet ring network, the Ethernet ring network comprising a first type of node having intermediate nodes and boundary nodes performing the same private ring network protocol and a public protocol, and executable At least one second type of node of the public protocol; the method comprises: when the first type of node detects that the Ethernet ring link is faulty, sending a topology change of the public protocol to the second type of node through the boundary node of the first type of node Packet, refreshing the second type of node MAC forwarding address table;

When the second type of node detects that the Ethernet ring link is faulty, the first type of node MAC forwarding address table is refreshed by sending a topology change packet of the public protocol to the border node of the first type node.

Preferably, the method further includes: before the first type of node detects that the Ethernet ring link is faulty, before the border node of the first type node sends the topology change packet of the public protocol to the second type node, the method further includes: The boundary node of the first type of node sends a ring refresh message to all nodes except the boundary node in the first type of node on the Ethernet ring network, and refreshes the MAC forwarding address table of all the first type nodes.

Preferably, when the first type of node that detects the failure of the tonnage network link is a border node, the topology change message sent by the border node of the first type node to the second type node is: The border node generates and sends a ring refresh message, and generates and sends a topology change message of the public protocol.

Preferably, when the first type of node that detects the failure of the Ethernet ring link is an intermediate node, the topology change message sent by the border node of the first type node to the second type node is: The intermediate node generates and sends a ring network refresh message, and the border node of the intermediate node generates and sends a topology change message of the public protocol after receiving the ring network refresh message. Preferably, after the topology change message of the public protocol is sent to the border node of the first type of node, and the MAC address forwarding address table of the first type node is refreshed, the method further includes: receiving, by the boundary node of the first type of node When the topology of the public protocol sent by the second type of node changes the message, it refreshes itself.

The MAC forwards the address table, and generates and sends a ring refresh message for refreshing the MAC forwarding address table of the other nodes except the border node in the first type of node.

Preferably, the public protocol is an MSTP (Multi-Spaning Tree Protocol) protocol.

According to another aspect of the present invention, a system for refreshing a MAC in an Ethernet ring network is provided, the system comprising:

a first type of node having intermediate nodes and boundary nodes that perform the same private ring network protocol and public protocol; and at least one second type node that can execute the public protocol;

The first type of node is configured to detect the link state of the Ethernet ring network, and when detecting that the Ethernet ring link is faulty, send a topology change packet of the public protocol to the second type node by using the border node of the first type node. Refreshing the second class node MAC forwarding address table;

The second type of node is configured to detect the link state of the Ethernet ring network, and when detecting the fault of the Ethernet ring link, send the topology change packet of the public protocol to the border node of the first type node, and refresh the first Class node MAC forwarding address table.

Preferably, the intermediate nodes of the first type of nodes include:

a detecting module, configured to detect an Ethernet ring link state;

The ring refresh message module is configured to generate and send a ring network refresh when the detecting module detects that the Ethernet ring network is faulty;

The address refreshing module is configured to refresh the MAC forwarding address table when the detecting module detects that the link is faulty or receives a ring refresh message sent by a node other than itself in the first type of node.

Preferably, the boundary node of the first type of node includes: a detecting module, configured to detect an Ethernet ring link state;

The ring network refresh packet module is configured to generate and send a ring network refresh when the detecting module detects that the Ethernet ring network is faulty.

The topology change message module is configured to generate and send a topology change when the detection module detects that the Ethernet ring network is faulty or receives a ring refresh message sent by a node other than itself in the first type of node. Obituary; and

An address refreshing module, configured to detect, when the detecting module detects that the link is faulty, or receives a ring refresh message sent by a node other than itself in the first type of node, or receives a topology change sent by the second type of node. When the message is received, the MAC forwarding address table is refreshed.

Preferably, the second type of node comprises:

a detecting module, configured to detect an Ethernet ring link state;

The topology change message module is configured to generate and send a topology change message when the detection module detects that the Ethernet ring network is faulty;

The address refreshing module is configured to refresh the MAC forwarding address table when the detecting module detects that the link is faulty or receives a topology change packet sent by the border node of the first type of node.

Compared with the prior art, the invention has the beneficial effects that: when devices of different manufacturers form an Ethernet ring network, the MAC can be quickly refreshed, so that the switching time reaches the carrier level requirement. DRAWINGS

Figure la is a topology diagram of an Ethernet ring network provided by the prior art; Figure lc is a topology diagram of a communication path when a link failure occurs in an Ethernet ring network provided by the prior art; Figure 2 is an Ethernet ring network provided by the present invention. A flowchart of a method for refreshing a MAC; FIG. 3a is a flowchart of a method for refreshing a MAC in an Ethernet ring network according to an embodiment of the present invention; Figure

Figure 3d is the information of the master node in the MAC address forwarding table of the second type of node in the Ethernet ring network according to the embodiment of the present invention;

3e is a flowchart of triggering an MSTP topology change message when the Ethernet ring network is switched according to an embodiment of the present invention;

4 is a schematic structural diagram of a system for refreshing a MAC in an Ethernet ring network according to an embodiment of the present invention;

5 is a schematic structural diagram of an intermediate node of a first type of node in an Ethernet ring network according to an embodiment of the present invention;

6 is a schematic structural diagram of a boundary node of a first type of node in an Ethernet ring network according to an embodiment of the present invention;

FIG. 7 is a schematic structural diagram of a second type of node in an Ethernet ring network according to an embodiment of the present invention. detailed description

The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

2 is a flowchart of a method for refreshing a MAC in an Ethernet ring network provided by the present invention. As shown in FIG. 2, the method includes the following steps:

Step S201: All nodes in the Ethernet ring network detect the link state of the Ethernet ring network. Step S202: When the first type node detects that the Ethernet ring link is faulty, the node sends a public protocol to the second type node through the border node. The topology changes the message, and refreshes the second class node MAC forwarding address table;

Step S203: When the second type of node detects that the Ethernet ring network is faulty, the second type of node MAC forwarding address table is refreshed by sending a topology change packet of the public protocol to the border node of the first type of node. The Ethernet ring network includes a first type of node having intermediate nodes and border nodes that perform the same private ring network protocol and a public protocol, and at least one second type node that can execute a public protocol.

FIG. 3 is a flowchart of a method for refreshing a MAC in an Ethernet ring network according to an embodiment of the present invention. As shown in FIG. 3a, an embodiment of the method includes the following steps:

Step S301: Configure a ring network protocol in the Ethernet ring network.

For example, in Figure 3b, the ring network includes the first type of nodes Al, A2, A3 and the second type of node B, and the links included are <A1, A2>, <A2, A3>, <A3, B; <B, A1> link. The first type of nodes are devices of the same manufacturer that run the same private ring network protocol and public protocol simultaneously on the Ethernet ring network. The second type of node is a device of another manufacturer that does not support the private ring network protocol run by the first type of node but can run the public protocol. Al, A2, and A3 run a private ring network protocol with each other and run the MSTP protocol. Unlike the manufacturers of Al, A2, and A3, B does not support the private ring network protocol running between Al, A2, and A3, but can run the MSTP protocol.

Step S302: Configure the master node, the master port of the master node, the slave port of the master node, the ring protection link, the border node, and the border port in the Ethernet ring network.

For example, in Figure 3b, node A2 is the master node, port 21 is the master port, port 22 is the slave port, and link <2, 3> directly connected to port 22 is the ring protection link. Nodes A1 and A3 are boundary nodes, and A2 is an intermediate node. A boundary node is a node of a first type of node that is directly connected to a second type of node. The intermediate node is the first type of node between two boundary nodes. Port 11 of node A1 is the boundary port, and port 32 of node A3 is the boundary port. The boundary port is the port of the border node that directly connects to other vendors' devices (the second type of node) on the ring network.

Step S303, the node in the Ethernet ring network learns the MAC.

For example, in FIG. 3b, after the private ring network protocol and the MSTP protocol are started, in the Ethernet ring, the port 21 of the master node A2 is the master port, and is in the forwarding state, and the port 22 of the A2 port is the slave port and is in the block state. At this time, Node B learns the MAC of the A2 node from port 21 of A2. At this time, the traffic from the Node B to the A2 node is forwarded from the port 42 of the Node B. For example, if the MAC address of the A2 node is 00:00:00:00:00:02, the forwarding information on the Node B is as shown in Figure 3d.

Step S304: The node in the Ethernet ring network detects a link fault, and the border node sends an MSTP topology change packet to the second type node. When the link in the ring network fails, the ring network needs to be switched. The nodes at both ends of the fault link first determine whether it is a boundary node. If yes, send the MSTP topology change message to the second type node. If not, then The MSTP topology is not triggered to change the message.

Specifically, for example, in Fig. 3c, assuming that the link between A1 and A2 is broken, A2 checks that the link <A1, A2> directly connected to it fails, and the master node A2 releases the slave port. The data packet forwarding function of 22, and the ring network refresh packet is sent to A3 by using the private ring network protocol, and the A3 ring network is notified to switch. A3 is a border node. After receiving the ring-net refresh message of A2, it uses the private ring network protocol to send the topology change of MSTP to node B. The message is notified to the B-node ring network. At the same time, the border node A1 also detects that the link <8, 2> directly connected to it is faulty, and sends a topology change of the MSTP to the second type of node B to notify the B-node that the ring network has been switched.

Step S305, receiving a node of the Ethernet ring network of the topology change message, and refreshing the MAC. For example, in FIG. 3c, after receiving the MSTP topology change message sent by A3 and A1, the Node B refreshes its MAC address forwarding table. Since the Node B runs MSTP, it can recognize the topology change message of the MSTP. After receiving the topology change message of the MSTP sent by A3 and A1, the node B refreshes the MAC forwarding address table and deletes the data in the MAC forwarding address table. The communication path is updated to: B<->A3<->A2. At this time, the traffic from the Node B to the A2 node will be sent out from the port 41 through the broadcast form instead of being forwarded from the original port 42, so that the traffic can be quickly switched from the failed link to the backup link. FIG. 4 is a schematic structural diagram of a system for refreshing a MAC in an Ethernet ring network according to an embodiment of the present invention. As shown in FIG. 4, the system includes: a first type of nodes A1, A2, A3 and a second type of node B. The first type of nodes A1, A2, and A3 are devices of the same manufacturer that perform the same private ring network protocol and public protocol on the ring network. The second type of Node B is a device of another manufacturer that does not support the private ring network protocol operated by the first type of node on the ring network. The nodes of the first type of nodes directly connected to the second type of node B are boundary nodes A1 and A3, and the first type of nodes between the two boundary nodes A1 and A3 are intermediate nodes A2.

When the first type of node detects that the Ethernet ring link is faulty, it sends a topology change packet of the public protocol to the second type node through its border node, and refreshes the MAC forwarding address table of the second type node. Assume that the link between A1 and A2 is faulty. After detecting the link fault, the border node A1 sends an MSTP topology change packet to B to notify B to refresh its MAC forwarding address table. At the same time, A2 also detects the link fault. A2 sends a ring refresh packet to A1 through the private ring network protocol. After receiving the ring refresh packet of A2, the border node A1 refreshes its MAC address forwarding table and sends MSTP to B. Topology changes the message.

When the second type of node detects that the Ethernet ring link is faulty, the first type of node MAC forwarding address table is refreshed by sending a topology change packet of the public protocol to the border node of the first type node. Assume that the link between B and A1 fails. After B detects the fault, it sends a topology change message to A3 to notify A3 to refresh the MAC forwarding address table. At the same time, A1 also detects the link fault between A1 and B. A1 sends a ring refresh message to the ring away from the fault direction through the private ring network protocol, and notifies other nodes A2 and A3 of the first type node that remove A1. Refresh the MAC forwarding address table.

FIG. 5 is a schematic structural diagram of an intermediate node of a first type of node in an Ethernet ring network according to an embodiment of the present invention. As shown in FIG. 5, an intermediate node of a first type of node includes a detection module 1, an address refresh module 2, and a ring network refresh. Message module 3. The detecting module 1 is configured to detect an Ethernet ring link state; the ring refresh message module 3 is configured to generate a sum when the detecting module 1 detects that the Ethernet ring network is faulty. The ring refreshing packet is sent; the address refreshing module 2 is configured to: when the detecting module 1 detects that the link is faulty or receives a ring refresh message sent by a node other than itself in the first type of node, refreshing the MAC Forward the address table.

6 is a schematic structural diagram of a boundary node of a first type of node in an Ethernet ring network according to an embodiment of the present invention. As shown in FIG. 6, a boundary node of a first type of node includes: a detection module 4, an address refreshing module 5, and a ring network. The message module 6 and the topology change message module 7 are refreshed. The detecting module 4 is configured to detect an Ethernet ring network link state, and the ring network refresh message module 6 is configured to generate and send a ring network refresh message when the detecting module 4 detects that the Ethernet ring network is faulty; The text module 7 is configured to generate and send a topology change text when the detection module 4 detects that the Ethernet ring network is faulty or receives a ring network refresh message sent by a node other than itself in the first type node; The module 5 is configured to detect, in the detection module 4, that the link is faulty or receives a ring refresh message sent by a node other than itself in the first type of node, or receives a topology change sent by the second type of node. When the message is received, the MAC forwarding address table is refreshed.

FIG. 7 is a schematic structural diagram of a second type of node in an Ethernet ring network according to an embodiment of the present invention. As shown in FIG. 7, the second type of node includes: a detection module 8, an address refreshing module 9, and a topology change message 10. The detecting module 8 is configured to detect an Ethernet ring network link state; the topology change message module 10 is configured to generate and send a topology change message when the detecting module 8 detects that the Ethernet ring network is faulty; the address refreshing module 9 And, when the detecting module 8 detects that the link is faulty or receives a topology change message sent by the border node of the first type node, refreshing the MAC forwarding address table.

In summary, the present invention solves the device composition of different manufacturers by configuring the boundary node in the Ethernet ring network and transmitting the MSTP topology change message to the device of the other manufacturer (the second type node) through the border node. When the ring network fails to connect successfully, the MAC address cannot be refreshed quickly, resulting in the problem that the switching time cannot meet the carrier-class requirements. The invention has the following technical effects: When the devices of different manufacturers form a ring network, the MAC can be quickly refreshed, thereby achieving the requirements of the carrier class.

Although the present invention has been described in detail above, the present invention is not limited thereto, and the present technology is Various modifications may be made by those skilled in the art in light of the principles of the invention. Therefore, modifications made in accordance with the principles of the invention should be construed as falling within the scope of the invention.

Claims

Claim

A method for refreshing a medium access control MAC in an Ethernet ring network, the Ethernet ring network comprising: a first type node having an intermediate node and a boundary node executing the same private ring network protocol and a public protocol, and an executable public protocol At least one second type of node; wherein the method comprises:

Ether i ψ , ^π ^

When the first type of node detects that the Ethernet ring link is faulty, the border node of the first type node sends a topology change packet of the public protocol to the second type node, and refreshes the MAC forwarding address table of the second type node;

When the second type of node detects that the Ethernet ring link is faulty, the first type of node MAC forwarding address table is refreshed by sending a topology change text of the public protocol to the border node of the first type of node.

The method according to claim 1, wherein when the first type of node detects that the Ethernet ring link is faulty, the boundary node of the first type of node sends the public protocol to the second type of node. Before the topology changes the message, the method further includes:

The boundary node of the first type of node refreshes the MAC forwarding address table of all the first type nodes by sending ring refresh messages to other nodes except the boundary node in the first type of nodes on the Ethernet ring network.

The method according to claim 2, wherein when it is detected that the first type of node in which the Ethernet ring link fails is a boundary node, the boundary node passing through the first type of node is to the second type of node The topology change message sent by the public protocol is: the border node generates and sends the ring network refresh>3⁄4 text, and generates and sends a topology change text of the public protocol.

The method according to claim 2, wherein when it is detected that the first type of node in which the Ethernet ring link fails is an intermediate node, the boundary node passing through the first type of node is to the second type of node The topology change message for sending the public protocol is: the intermediate node generation and transmission station The ring network refreshes the message, and the border node of the intermediate node generates and sends the topology change message of the public protocol after receiving the ring network refresh message.

The method according to claim 1, wherein the sending a topology change of the public protocol to the border node of the first type of node, after refreshing the first class node MAC forwarding address table, The method further includes: when receiving the topology change message of the public protocol sent by the second type node, the boundary node of the first type node refreshes its own MAC forwarding address table, and generates and sends the first to refresh the first The ring network of the MAC forwarding address table of the node other than the border node in the class node is refreshed.

The method according to any one of claims 1 to 5, wherein the public protocol is a multiple spanning tree MSTP protocol.

7. A system for refreshing a MAC in an Ethernet ring network, comprising: a first type of node having intermediate nodes and boundary nodes that perform the same private ring network protocol and a public protocol; and at least one second type node that can execute a public protocol; It is characterized in that

The first type of node is configured to detect an Ethernet ring link state, and when the Ethernet ring link fails, send a topology change of the public protocol to the second type node by using the boundary node of the first type node. Packet, refreshing the second class node MAC forwarding address table;

The second type of node is configured to detect the link state of the Ethernet ring network, and when detecting that the Ethernet ring link is faulty, send a topology change packet of the public protocol to the border node of the first type node. , refresh the first class node MAC forwarding address table.

8. The system according to claim 7, wherein the intermediate node of the first type of node comprises:

a detecting module, configured to detect an Ethernet ring link state;

The ring refresh message module is configured to generate and send the ring refresh message when the detecting module detects that the Ethernet ring network is faulty;

An address refreshing module, configured to detect, in the detecting module, that a link is faulty or is received The MAC forwarding address table is refreshed when the ring-network refresh message sent by the node other than itself is sent by the first-type node.

9. The system according to claim 7, wherein the boundary node of the first type of node comprises:

a detecting module, configured to detect an Ethernet ring link state;

The ring refresh message module is configured to generate and send the ring network to refresh when the detecting module detects that the Ethernet ring network is faulty;

The topology change message module is configured to generate and send the topology when the detecting module detects that the Ethernet ring network is faulty or receives a ring refresh message sent by a node other than itself in the first type of node. Change 4 essays; and

An address refreshing module, configured to: when the detecting module detects a link failure or receives a ring refresh message sent by a node other than itself in the first type node, or receives a extension sent by the second type node When Park changes the message, the MAC forwarding address table is refreshed.

The system of claim 7, wherein the second type of node comprises: a detecting module, configured to detect an Ethernet ring link state;

And a topology change message module, configured to generate and send the topology change when the detection module detects that the Ethernet ring network is faulty;