WO2011127736A1

WO2011127736A1 - Method and apparatus for maintaining saved information of node in ethernet ring network

Info

Publication number: WO2011127736A1
Application number: PCT/CN2010/079422
Authority: WO
Inventors: 吴少勇
Original assignee: 中兴通讯股份有限公司
Priority date: 2010-04-12
Filing date: 2010-12-03
Publication date: 2011-10-20
Also published as: CN102215119A; CN102215119B

Abstract

The present invention discloses a method and an apparatus for maintaining the saved information of the node in an Ethernet ring network, wherein when a protection switching request on a non-virtual channel sub-ring needs to be eliminated, the node on the non-virtual channel sub-ring deletes the node saved information in all ports on the non-virtual channel sub-ring of the current node, and does not save the received node saved information in the no-request protocol frame. The method and apparatus of the present invention solve the problem that the address table can not be updated properly when the non-virtual channel sub-ring switches back.

Description

Maintenance method and device for saving information in Ethernet ring network node

The present invention relates to Ethernet technology, and in particular, to a method and apparatus for maintaining information in an Ethernet ring node. Background technique

With the development of Ethernet towards multi-service bearer, especially for some services, the reliability and real-time requirements of the network are getting higher and higher. Ethernet is widely used in ring networking, which improves the reliability of the network. In this ring network protection scheme, fast protection switching is usually required to reach 50ms or less. Currently, this fast protection switching technology can be implemented based on the G.8032 protocol of the International Telecommunication Union (ITU-T).

Based on the above G.8032 protocol, an automatic protection switching protocol and mechanism is defined for the Ethernet layer of the ring topology Ethernet. The network protection method is applicable to the ring topology Ethernet. The implementation process is as follows: In the ring topology Ethernet, select a link as a ring protection link. When the link of the Ethernet ring network is faultless, At least one of the two adjacent nodes of the ring protection link blocks the port connected to the ring protection link, preventing the protected data from passing through the ring protection link. The node responsible for blocking the ring protection link port is called The ring protects the node to which the link belongs. In this way, there is only a unique communication path between any two nodes on the Ethernet ring network. Therefore, the closed loop of the communication path is not generated in the Ethernet ring network, preventing the closed loop and the network storm; If the faulty link is not a ring protection link, block the faulty port adjacent to the faulty link and open the blocked port on the ring protection link so that the protected data can be protected from the ring protection link. Through, a new communication path is generated between any two nodes on the ring protection link, which ensures the reconnection of the communication path and improves the network. Reliability.

During the switching process of the Ethernet ring network, the communication path of data traffic needs to be changed due to The address table information recorded by each node is no longer applicable after the communication path of the data traffic changes. Therefore, each node needs to refresh the address table. The specific action is to delete the address entry of the port on the ring. In the G.8032 protocol, when a node sends a protocol frame, the protocol frame includes its own node number and blocked port index number. After receiving the protocol frame, other nodes extract the node number and blocked port in the protocol frame. The index number is compared with the node number and the blocked port index number stored in the receiving port. If the two are not the same, the node number of the receiving port and the blocked port index number are updated to the extracted node number and the blocked port index number. If the node number and the blocked port index number in the protocol frame are different from the node number and blocked port index number saved by the peer port of the receiving port, the node may need to refresh the address table. When the protocol frame received by the node is an NR (No Request) message, the node deletes the node number and the blocked port index number saved by the receiving port, and does not save the node number and the blocked port index number in the receiving protocol frame, that is, The node number saved by the node receiving port and the blocked port index number are empty.

The above-mentioned Ethernet ring address table refresh scheme ensures that the single ring network of the Ethernet ring network can quickly delete the wrong address entries when switching, but also brings the virtual channel of the multi-ring network in the Ethernet ring network. The ring does not correctly delete the wrong address entry when switching.

For example, in an Ethernet ring network multi-ring network, when no virtual channel sub-ring is in a normal state, the node to which the ring protection link belongs sends an NRRB (No Request, RPL Blocked) protocol frame to the non-virtual channel sub-ring. The protocol frame includes the node number of the node to which the ring protection link belongs and the blocked port index number. After receiving the NRRB protocol frame, the other node extracts the node number and the blocked port index number in the protocol frame, and saves it on the receiving port. The node number is compared with the blocked port index number. If the two are not the same, the node number of the receiving port and the blocked port index number are updated to the extracted node number and the blocked port index number. Therefore, when no virtual channel subring is stable, Each node records the node number of the node to which the ring protection link belongs and the blocked port index number. When the link on the sub-ring without a virtual channel fails, the adjacent node of the faulty link detects the fault, blocks the faulty port, and sends an SF (Signal Fail) protocol frame to the sub-ring without the virtual channel. The frame includes the node number of the sending node and the blocked port index number. After receiving the SF protocol frame, the other nodes extract the node number and the blocked port index number in the protocol frame, and the node number and blocked port index stored in the receiving port. For comparison, if the two are not the same, update the node number of the receiving port and the blocked port index number to the extracted node number and blocked port index number, and also refresh the address table. Therefore, when there is a link failure in the virtual channel subring, there is no node in the virtual channel subring that is distributed between the ring protection link and the faulty link. On both of the ring ports, one side records the ring protection chain. The node number of the node to which the road belongs and the index number of the blocked port. The other side records the node number of the adjacent node of the faulty link and the index number of the blocked port. When a faulty link with no virtual channel sub-ring is restored, the node to which the ring protection link belongs re-blocks the ring protection link port, and sends an NRRB protocol frame to the non-virtual channel sub-ring. The protocol frame includes the node to which the ring protection link belongs. Node number and blocked port index number, nodes in the virtual channel subring that are distributed between the ring protection link and the faulty link. After receiving the NRRB protocol frame, the node number and the blocked port index number in the protocol frame are extracted. , compared with the node number saved on the receiving port and the blocked port index number, the two are the same, so the address table will not be refreshed. In fact, when the virtual channel subring is not restored, the address table needs to be refreshed due to the change of the communication path of the data traffic.

Therefore, in the switching scenario without the virtual channel sub-ring, the existing refreshing scheme of the Ethernet ring address table may not properly refresh the address table; and, there is currently no solution to this problem. Summary of the invention

In view of this, the main purpose of the present invention is to provide a method and apparatus for maintaining information in an Ethernet ring node to solve the problem that the address table cannot be refreshed normally when the virtual channel sub-ring is switched back.

In order to solve the above problems, the technical solution of the present invention is implemented as follows:

A method for maintaining information stored in an Ethernet ring network node, the method comprising:

When the protection switching request on the no virtual channel subring needs to be cleared, there is no virtual channel sub The node on the ring deletes the node save information of all the ports on the non-virtual channel subring, and does not save the node save information in the received requestless protocol frame.

The method further includes the process of confirming that the protection switching request on the no virtual channel subring needs to be cleared:

When the node on the no virtual channel subring receives the no-request protocol frame, it is confirmed that the protection switching request needs to be cleared.

The method further includes: the node on the no virtual channel subring determining whether it is a virtual channel subring node, and performing the deleting operation when the determination result is yes.

The method further includes: when the determination result is negative, performing a subsequent operation using the prior art.

The method further includes:

Updating the node save information saved by the node receiving port according to the node save information received by the node, and refreshing the address table based on the node.

The node save information includes a node number and a blocked port index number included in the transmitted protocol frame.

A maintenance device for storing information in an Ethernet ring network node, where the device includes a trigger unit and a storage information maintenance unit;

The triggering unit is configured to confirm that the protection switching request on the virtual channelless subring needs to be cleared, and notify the saving information maintenance unit of the situation;

The save information maintenance unit is configured to delete a node on the managed virtual channelless subring to save information on all ports of the virtual channel subring without saving the node save information in the received unrequested protocol frame. .

The device further includes a node determining unit, configured to determine whether the managed node is a non-virtual channel sub-ring node, and notify the save information maintenance unit to perform the deleting operation when the determination result is yes. The save information maintenance unit is further configured to:

And updating, according to the node save information received by the node, the node save information saved by the node receiving port, and refreshing the address table based on the node.

The method and device of the present invention solve the problem that the address table cannot be refreshed normally when there is no virtual channel sub-ring back-cut. DRAWINGS

FIG. 3 is a schematic diagram of operation when the Ethernet ring network has no virtual channel sub-ring link recovery;

FIG. 4 is a schematic diagram of operation when the Ethernet ring network has no virtual channel sub-ring back-cut;

FIG. 5 is a flowchart of maintenance of information stored in an Ethernet ring network node according to an embodiment of the present invention; FIG. 6 is a schematic diagram of operation when an Ethernet ring network has no virtual channel sub-ring link recovery according to an embodiment of the present invention;

FIG. 8 is a diagram of a maintenance apparatus for storing information in an Ethernet ring network node according to an embodiment of the present invention. detailed description

In general, a node on a sub-ring without a virtual channel in the Ethernet ring network can delete the node number and blocked port index number of all ports on the sub-ring without the virtual channel when the node receives the unrequested protocol frame, and does not save. The node number and the blocked port index number in the received request-free protocol frame enable the address table to be refreshed normally when the node number and the blocked port index number are received.

The implementation of the technical solution will be described in detail below with reference to the accompanying drawings. Figure 1 is a schematic diagram of the protection structure of the Ethernet ring network based on G.8032. As shown in Figure 1, nodes S1, S2, S3, S4, S5, and S6 form a virtual channel sub-ring of the Ethernet ring network, and node S2 is ring protection. The link to which the link belongs, the link between the nodes S1 and S2 is the ring protection link, and the index of the ring protection link port of the node S2 is 0. When the virtual channel sub-ring is in the normal state, the node S2 sends the NRRB protocol frame on the non-virtual channel sub-ring. The NRRB protocol frame includes the node number S2 and the blocked port index number 0 of the node to which the ring protection link belongs. After receiving the NRRB protocol frame, the node S4 extracts the node number and the blocked port index number in the NRRB protocol frame, and compares them with the node number and the blocked port index number stored in the receiving port respectively, and the comparison result is not the same. At the same time, the node number of the receiving port and the blocked port index number are updated to the extracted node number and the blocked port index number.

Therefore, when the virtual channel subring is not stable, the node number recorded on the port that the node S4 receives the NRRB protocol frame is S2, and the blocked port index number is 0.

2 is a schematic diagram of a protection structure when a link failure occurs in a virtual channel sub-ring in FIG. 1, as shown in FIG. 2, assuming that a link between nodes S5 and S6 is faulty, and node S5 detects that the link is faulty and blocks. The faulty port sends an SF protocol frame on the sub-ring without a virtual channel. The SF protocol frame includes the node number S5 of the sending node and the blocked port index number 1. After receiving the SF protocol frame, the node S4 extracts the node number and the blocked port index number in the SF protocol frame, and compares them with the node number and the blocked port index number stored in the receiving port respectively. When the comparison result is different. The node number of the receiving port and the blocked port index number are updated to the extracted node number and the blocked port index number, and the address table is also refreshed.

Therefore, after the link failure occurs, the node S4 refreshes the address table when receiving the SF protocol frame, and the node number of the port record of the SF protocol frame is S5, and the blocked port index number is 1. In fact, node S2 has already opened the port at this time, but since node S2 does not issue a protocol frame when opening the port, node S4 still records the information of node S2 blocking the port, which is actually the wrong information. FIG. 3 is a schematic diagram of the protection structure of the virtual link sub-ring failure link recovery in FIG. 2, as shown in FIG. 3, the link failure between nodes S5 and S6 is recovered, and the node S5 detects that the link failure is restored after the failure. The NR protocol frame is sent on the virtual channel sub-ring. After receiving the NR protocol frame, the node S4 deletes the node number and the blocked port index number saved by the receiving port, that is, deletes the saved node number S5 and the blocked port index number 1 information, and does not save the information. The node number and blocked port index number in the received NR protocol frame.

4 is a schematic diagram of a protection structure when the virtual channel sub-ring is switched back in FIG. 3, as shown in FIG. 4, the node S2 re-blocks the ring protection link port, and sends an NRRB protocol frame on the no-virtual channel sub-ring, the NRRB protocol. The frame includes the node number S2 and the blocked port index number 0 of the node to which the ring protection link belongs. After receiving the NRRB protocol frame, the node S4 extracts the node number S2 and the blocked port index number 0, and compares them with the node number S2 and the blocked port index number 0 stored in the receiving port respectively, because the comparison result is the same, Node S4 does not refresh the address table. In fact, when the virtual channel subring is not switched back, the address table needs to be refreshed due to the change of the communication path of the data traffic. S4 does not refresh the address table, which will cause communication failure in the subsequent communication process.

Therefore, the present invention has emerged. With the port blocking information deletion scheme of the present invention, when a frame with no request protocol is received, the node on the sub-ring without the virtual channel deletes the node on the sub-ring without the virtual channel. The node number and blocked port index number saved by the port, and the node number and blocked port index number in the received unsolicited protocol frame are not saved.

The scheme for updating the blocked port information used in the present invention will be specifically described below.

Referring to FIG. 5, FIG. 5 is a maintenance flow diagram of information stored in an Ethernet ring network node according to an embodiment of the present invention, where the process includes the following steps:

Step 510: The Ethernet ring network node receives the requestless protocol frame.

Step 520: The Ethernet ring network node that receives the requestless protocol frame determines whether it is a node on the virtual channel subring. If yes, go to step 530; otherwise, go to step 540.

Step 530: The node deletes the node number of all ports on the node without the virtual channel subring. And block the port index number, and confirm that the node number and blocked port index number in the received unrequested protocol frame are not saved, and the process ends.

Step 540: Apply the current technology to process the subsequent process.

The technical solutions of the present invention are further described in detail below with reference to specific examples.

See Figure 6, Figure 6 shows the situation when the virtual link sub-ring failure link is restored in Figure 2. In Figure 6, the link between node S5 and node S6 is recovered, and node S5 detects that the link is restored and sends an NR protocol frame on the virtual channelless subring. After receiving the NR protocol frame, the node S4 first confirms that it is a non-virtual channel sub-ring node, and then deletes the node number and the blocked port index number of all ports on the non-virtual channel sub-ring of the local node, and does not save the received The node number and blocked port index number in the NR protocol frame. At this point, the node number and blocked port index number recorded on the ports on both rings in node S4 are empty.

See Figure 7, Figure 7 for the protection structure of Figure 6 without the virtual channel sub-ring back-cut. In Figure 7, node S2 re-blocks the ring protection link port because there is no virtual channel sub-ring back-cut, and sends an NRRB protocol frame on the non-virtual channel sub-ring. The NRRB protocol frame includes the node number of the node to which the ring protection link belongs. S2 and blocked port index number 0. After receiving the NRRB protocol frame, the node S4 extracts the node number S2 and the blocked port index number 0 in the NRRB protocol frame, and compares them with the node number and the blocked port index number stored in the receiving port respectively. The node number and the blocked port index number recorded in the NRRB protocol frame are the same as the node number recorded on the two ring ports of the node S4. The blocked port index numbers are different, so that the node S4 needs to update the node number saved by the receiving port and the blocked port index number, and refresh the address table based on this.

It can be seen from the above that when the Ethernet ring network has no virtual channel sub-ring back-cut, the node on the no-virtual channel sub-ring receives the unrequested protocol frame. After receiving a no-request protocol frame, the node on the sub-ring without the virtual channel can delete the node number of all the ports on the ring without the virtual channel sub-ring. And blocking the port index number, and The node number and the blocked port index number in the received unrequested protocol frame are not saved, so the incorrectly recorded node number and blocked port index number that can cause the address table to fail to be refreshed are deleted. When there is no virtual channel sub-ring back-cut, the address table can be refreshed correctly, which ensures the correct switching of data traffic.

In practical applications, in order to ensure the smooth implementation of the above operations, the settings shown in Figure 8 can be performed. Referring to FIG. 8, FIG. 8 is a diagram of a maintenance device for storing information in an Ethernet ring node according to an embodiment of the present invention. The device includes a trigger unit, a node judgment unit, and a save information maintenance unit that can be connected in sequence or connected in pairs. These three units can be set in the node or in other functional entities outside the node.

In the specific application, the trigger unit is used to confirm that the sub-loop back-cut phase is currently in the virtual channel, and the situation is notified to the node judging unit. It should be noted that there are various methods for confirming that there is no virtual channel sub-ring back-cutting stage, such as: judging whether a non-requested protocol frame is received, and determining that there is no virtual channel sub-loop back when receiving the no-request protocol frame. The phase is determined; or, it is determined whether other indication information indicating that the sub-loop back-cut phase is currently in the virtual channel is received, and if it is received, it is determined that the sub-ring back-cut phase is currently in the virtual channel.

The node judging unit can receive the notification from the trigger unit, and judge whether the managed node is a non-virtual channel sub-ring node, and notify the save information maintenance unit when the judgment result is YES. It should be noted that there are various methods for determining whether a node is a sub-ring node without a virtual channel, such as: obtaining a node identifier, and searching for a member identifier of the obtained node identifier indicating that the node belongs to a sub-ring without a virtual channel, if Find, determine that the node is a virtual channel subring node.

When the save information maintenance unit receives the notification from the node judging unit, deletes the node number and the blocked port index number of all the ports of the managed node on the non-virtual channel subring, and does not save the received requestless protocol frame. Node number and blocked port index number.

After that, the save information maintenance unit can receive the node number and resistance according to the managed node. The port index number is updated, the node number saved by the receiving port of the node and the blocked port index number are updated, and the address table is refreshed based on the refresh address table, or the related address table management unit is notified to refresh the address table.

In addition, when the judgment result of the node judgment unit is that the managed node is not a virtual channel sub-ring node, the situation may be notified to the save information maintenance unit, and the save information maintenance unit applies the current technical process when receiving the above notification. Follow-up process.

In summary, it can be seen that no node of the virtual channel sub-ring can receive the unsolicited protocol frame when the Ethernet ring has no virtual channel sub-ring back-cut in both the method and the device. In addition, after receiving the no-request protocol frame, the node on the sub-ring without the virtual channel can delete all the ports on the ring of the virtual channel sub-ring without determining the virtual sub-ring node. The node number and the blocked port index number, and the node number and blocked port index number in the received unsolicited protocol frame are not saved, so the incorrectly recorded node number that can cause the address table to fail to be refreshed normally is deleted. The port index number is blocked. When the node is switched back without a virtual channel sub-ring, the address table can be correctly refreshed, ensuring correct switching of data traffic.

The above node number and blocked port index number may be collectively referred to as node save information. In addition, when receiving the no-request protocol frame, the node determines that the protection switching request on the non-virtual channel sub-ring needs to be cleared, and then deletes the node save information of all the ports on the non-virtual channel sub-ring of the local node, and confirms that the information is not saved. The nodes in the unsolicited protocol frame are saved. In fact, the node determines that there are multiple ways in which the protection switching request on the virtual channelless subring needs to be cleared, such as: receiving a specific update protocol frame during the failback or failback phase.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included. Within the scope of protection of the present invention.

Claims

Claim

When the protection switching request on the virtual channelless subring does not need to be cleared, the node on the non-virtual channel subring deletes the node save information of all ports on the virtual channelless subring, and does not save the received requestless protocol. The nodes in the frame hold the information.

2. The method of claim 1, wherein the method further comprises the step of confirming that the protection switching request on the no virtual channel subring needs to be cleared:

The method according to claim 2, wherein the method further comprises: determining, by the node on the no-virtual channel sub-ring, whether the node is a virtual channel sub-ring node, and executing the Describe the delete operation.

The method according to claim 3, wherein the method further comprises: when the determination result is no, applying a prior art to perform a subsequent operation.

The method according to any one of claims 1 to 4, characterized in that the method further comprises:

The method according to any one of claims 1 to 4, wherein the node save information includes a node number and a blocked port index number included in the transmitted protocol frame.

A maintenance device for storing information in an Ethernet ring network node, wherein the device includes a trigger unit and a save information maintenance unit;

The triggering unit is configured to confirm that the protection switching request on the virtual channel subring does not need to be cleared, and notify the saving information maintenance unit of the situation; The save information maintenance unit is configured to delete a node on the managed virtual channelless subring to save information on all ports of the virtual channel subring without saving the node save information in the received unrequested protocol frame. .

The device according to claim 7, wherein the device further comprises a node determining unit, configured to determine whether the managed node is a virtual channel sub-ring node, and notify the station when the determination result is yes The save information maintenance unit performs the delete operation.

The device according to claim 7 or 8, wherein the save information maintenance unit is further configured to:

10. The apparatus according to claim 7 or 8, wherein the node saving information comprises a node number and a blocked port index number included in the transmitted protocol frame.