WO2008106883A1

WO2008106883A1 - Method and device for disposing inner connecting faults of resilient packet ring nodes

Info

Publication number: WO2008106883A1
Application number: PCT/CN2008/070372
Authority: WO
Inventors: Yue Wang
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2007-03-06
Filing date: 2008-02-28
Publication date: 2008-09-12
Also published as: CN100461737C; CN101022389A

Abstract

A method for disposing inner connecting faults of resilient packet ring (RPR) nodes, includes: when the first node of the RPR ring detects an inner connecting fault, the link analysis result is obtained from the link state information of the current RPR ring; the first SPAN is selected from the two SPANs of the first node based on the link analysis result to simulate the generation of the signal failure (SF), and the SF signals are sent to the other nodes of the ring. Also a device for disposing inner connecting faults of resilient packet ring nodes is provided.

Description

Method and device for processing internal connection fault of elastic packet ring node

The present invention relates to network communication technologies, and in particular, to a method and a device for processing internal connection faults of a Resilient Packet Ring (RPR) node. Background of the invention

RPR is a reverse double-ring structure. It can transmit data simultaneously on two rings. It is a new metropolitan area network standard because of its interface bandwidth, carrier-class fault switching capability, and efficient bandwidth utilization. RPR has gradually been favored by more and more operators due to its many advantages such as physical layer independence.

Referring to the RPR ring shown in Figure 1, the ring includes 6 RPR nodes, the outer ring clockwise single ring is called 0 ring, and the inner counterclockwise single ring is called 1 ring; between two adjacent RPR nodes. A link is called a segment (SPAN), and a SPAN consists of a pair of unidirectional links in opposite directions. For each node, there are two SPANs, called westbound SPAN and eastbound SPAN, respectively, where westbound SPAN is used. Receive 0 ring data and send 1 ring data. East SPAN is used to receive 1 ring data and send 0 ring data.

RPR's fault self-healing capability is very strong, enabling fault protection switching within 50ms. Referring to the RPR ring disconnected due to the fault shown in Figure 2, after rapid fault processing, all data streams can be avoided by disconnecting the link due to the fault, and the normal transmission of the data stream is realized. In Figure 2, a SPAN that cannot be used to transfer data is called an edge. There are many cases of link failure, including signal failure (SF, Signal Fail) and signal degradation (SD, Signal degrade). Among them, the SF fault is the most serious, and the edge will be generated, such as fiber breakage; SD indicates that the line quality is poor, but the data can be transmitted, and the edge can be generated under certain conditions.

In order to implement fast protection switching, each node on the RPR ring detects its own SPAN in real time. State, and obtain the status of other SPANs on the ring through information published by other nodes. When a change is detected, the analysis is processed based on all the data collected, and a quick response is made.

In an RPR node, the transmission and reception functions of the east-facing SPAN and the west-direction SPAN may be distributed on different hardware, and the internals are connected by fiber or cable. This internal connection is also part of the RPR ring, which can also cause breakage of the RPR ring if the internal connection fails.

However, the current agreement is most concerned with failures on SPAN, and the handling of internal connection failures for RPR nodes is too numerous. For example, a single simulation simulates the occurrence of SF for the westward SPAN, or the SF for the eastward SPAN, or the SF for both SPANs. According to such a processing method, if an internal connection failure of the RPR node occurs in an actual application, a serious accident may occur under certain circumstances. The most simple example is: Assume that the westward SPAN of node A already has an SF fault and generate an edge. After an internal connection fault occurs, if only the eastward SPAN simulation generates SF, the result of this processing is that two of node A The SPANs all generate edges and cannot send data. Node A is completely disconnected from the ring, which damages the connectivity of the ring network and causes the ring network to be completely unavailable. Summary of the invention

The main purpose of the embodiments of the present invention is to provide an RPR node internal connection fault processing method and device, which can ensure the connectivity and availability of the ring network to the greatest extent in the event of an internal connection failure of the RPR node.

To achieve the above objective, the RPR node internal connection fault processing method provided by the present invention is as follows: When the first node on the RPR ring detects an internal connection fault, the link analysis result is obtained from the link state information of the current RPR ring;

Selecting the first SPAN analog generation signal failure SF from the two segments SPAN of the first node according to the link analysis result, and transmitting the SF signal to other nodes on the ring, and selecting the number of nodes for the SF signal transmission when the first SPAN simulation generates the SF It is greater than the number of nodes of the SF signal transmission when the second SPAN analog generation SF is selected. The link analysis result is: the first node has a SPAN generation edge;

The selecting the first SPAN simulation to generate the SF from the two SPANs of the first node according to the link analysis result includes: selecting a SPAN simulated generated SF of the generated edge from the two SPANs of the first node.

The link analysis result is: the node except the first node has a SPAN generation edge; the selecting the first SPAN simulation generation SF from the two SPANs of the first node according to the link analysis result includes: The SPAN of the node's two SPANs that selects the nearest edge on the ring simulates the generation of the SF.

The link analysis result is: the first node has a SPAN fault;

The selecting the first SPAN simulation to generate the SF from the two SPANs of the first node according to the link analysis result includes: selecting the faulty SPAN simulation generation SF from the two SPANs of the first node.

The link analysis result is: the node except the first node has a SPAN fault on the ring; the selecting the first SPAN simulation to generate the SF from the two SPANs of the first node according to the link analysis result includes: The SPAN of the node's two SPANs that selects the closest link to the faulty link on the ring generates the SF.

The method further includes: when detecting that the selected SPAN does not have an SF failure, and another SPAN of the first node that is not selected has an SF failure, simulating the selected SPAN on the SF disappears, and selecting the Another SPAN of one node simulates the generation of SF.

The RPR node internal connection fault processing apparatus provided by the present invention includes: a storage unit and a processing unit, wherein

a storage unit, configured to store link state information of the RPR ring;

The processing unit is configured to: when the internal connection fault occurs in the node, obtain the RPR ring link state information from the storage unit, obtain the link analysis result from the link state information, and obtain the two segments of the internal connection fault node according to the link analysis result. SPAN selects the first SPAN analog to generate signal failure SF, And sending the SF signal to other nodes on the ring, and selecting the first SPAN analog to generate the SF, the number of nodes for transmitting the SF signal is greater than the number of nodes for transmitting the SF signal when the second SPAN is simulated to generate the SF.

The processing unit includes: an obtaining unit, a selecting unit, and a sending unit, where

An obtaining unit, configured to obtain RPR ring link state information from the storage unit when the internal connection failure occurs, obtain link analysis result from the link state information, and send the link analysis result to the selecting unit;

a selecting unit, configured to select a first SPAN simulation to generate an SF from two SPANs of the internal connection faulty node according to the received link analysis result, and send the generated SF information to the sending unit;

And a sending unit, configured to send, according to the received SF information, an SF signal to other nodes on the ring. The processing unit further includes: an updating unit, configured to simulate that the SF on the first SPAN disappears when detecting that the selected first SPAN does not have an SF failure, and the second SPAN that is not selected has an SF failure, and Select the second SPAN simulation to generate the SF.

It can be seen that, in the embodiment of the present invention, when an internal connection failure of the RPR node is detected, instead of randomly selecting a SPAN simulation to generate the SF, it is required to follow the link state of the RPR ring from the node where the internal connection failure occurs. Select one of the two SPANs to simulate the generation of SF. For example, from the two SPANs where the internal connected faulty node occurs, select the SPAN simulation to generate the SF that has generated the edge or the fault may generate the edge, or select the SPAN that is closest to the link that has generated the edge on the ring or that has the fault that may generate the edge. The simulation generates SF to ensure that nodes with internal connection failures can be connected with as many nodes as possible. In this way, the connectivity and availability of the ring network can be maximized in the event of a fault in the internal connection of the node, so that the ring network can provide more reliable services. BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic diagram of an RPR ring in the prior art.

2 is a schematic diagram of a disconnected RPR ring in the prior art. FIG. 3 is a flowchart of a method for processing an internal connection failure of an RPR node according to an embodiment of the present invention. 4 is a flow chart of replacing a SPAN simulation generated SF in an embodiment of the present invention.

FIG. 5 is a schematic structural diagram of an RPR node internal connection fault processing apparatus according to an embodiment of the present invention. Mode for carrying out the invention

The present invention will be further described in detail below with reference to the drawings and embodiments.

The method for processing an internal connection fault of an RPR node provided by the embodiment of the present invention mainly includes: when the first node on the RPR ring detects an internal connection fault, the link analysis result is obtained by the link state information of the current RPR ring; and according to the link analysis As a result, the first SPAN analog generation SF is selected from the two SPANs of the first node, and then the SF signal is sent to other nodes on the ring, and the number of nodes of the SF signal transmission when the first SPAN simulation is generated to generate the SF is greater than the selection of the second SPAN simulation. The number of nodes for SF signal transmission when generating SF.

That is to say, when an internal connection fault occurs on a node on the RPR ring, the SPAN analog generation SF needs to be combined with the current link state information of the RPR ring to ensure that the node with internal connection failure can be connected with as many nodes as possible. In this way, the connectivity and availability of the ring network can be maximized in the event of an internal connection failure in the node, enabling the ring network to provide more reliable services.

The link state of the RPR ring includes the status of two SPANs of the node that has an internal connection failure, and the status of other links on the ring, that is, other SPANs. Each node on the RPR ring detects the status of its own SPAN in real time and obtains the status of other SPANs on the ring through information published by other nodes. The link status includes information such as whether the link has generated an edge and whether the link has a fault.

The link analysis result includes: whether there are edges or faults in the two SPANs of the first node, and whether other nodes on the ring except the first node have edges or faults.

The RPR node internal connection fault processing method provided by the present invention is described in detail below with reference to FIG. Description, as shown in Figure 3, the method mainly includes the following steps:

Step 301: Each node on the RPR ring detects its internal connection status in real time. When the first node detects that an internal connection fault occurs, the node checks the link state information of the RPR ring and determines the two SPANs of the node. If there is an edge, if yes, go to step 302; otherwise, go to step 303.

Step 302: Select a SPAN simulation of the generated edge from the two SPANs of the first node to generate the SF, and then end the processing of the process.

Since the SPAN is already an edge, selecting the SPAN analog to generate the SF does not cause a topological change of the RPR ring, and does not affect the transmission of the data stream on the ring.

If the two SPANs of the first node are edges, you can choose any of the SPAN simulations to generate the SF.

Step 303: The first node learns the status of other links on the ring according to the information advertised by other nodes on the ring, and determines whether there are edges in other links on the ring, and if yes, performs step 304; otherwise, performs Step 305.

Step 304: Select the SPAN simulation closest to the edge on the ring from the two SPANs of the first node to generate the SF, and then end the processing of the process.

The SPAN closest to the edge on the distance ring refers to the SPAN in the direction that can reach the edge with the shortest hop count. For example, in Figure 1, if node A has an internal connection failure, and it is known that there are no edges in the two SPANs of the node, and the westward SPAN of the node B generates an edge, then node A will select its own westward SPAN to simulate the generation of SF. Because the westward SPAN of node A is closest to the westward SPAN of node B. Such a selection method can connect the node in which the internal connection failure occurs with as many nodes as possible, thereby improving the connectivity of the ring network.

Step 305: The first node determines whether there is a fault in the two SPANs of the node. If yes, step 306 is performed; otherwise, step 307 is performed.

Step 306: The first node selects a faulty SPAN mode from two SPANs of the node. It is planned to generate SF, and then the processing of this process is ended.

If both SPANs of the first node are faulty, you can choose one of the SPAN simulations to generate the SF, because the faulty SPAN is likely to generate edges.

Step 307: The first node determines whether there is a fault in other links on the ring. If yes, go to step 308; otherwise, go to step 309.

Step 308: The first node selects, from the two SPANs of the node, the link with the fault on the ring, that is, the SPAN simulation to generate the SF after the shortest hop reaches the faulty link, and then ends the process. deal with.

Step 309: If there is no edge and no fault in all links on the ring, the first node may arbitrarily select a SPAN simulation to generate the SF from the two SPANs of the node, and then end the processing of the process.

After the SF is generated by the simulation, the first node can broadcast the information to other nodes on the ring, so that the data stream is no longer transmitted through the SPAN of the analog generation SF to ensure the availability of the ring network.

It should be noted that, in step 301, after the first node determines that there are no edges in the two SPANs, step 305 may be performed first; when step 305 returns a negative judgment result, step 303 is performed; After the judgment result, step 307 is performed. That is to say, the first node first determines whether there are edges in the two SPANs of the node, or whether there is a fault. When the two SPANs of the node are neither edged nor faulty, whether the other links on the ring are judged whether There is an edge, or there is a fault.

Moreover, after the negative determination result is obtained in step 301, 303 or 305, the first node may also directly select one SPAN simulation generation SF from the two SPANs of the node, and end the processing of the process.

It can be seen that, in this embodiment, when an internal connection failure of the RPR node is detected, instead of randomly selecting a SPAN analog generation SF, the link state of the RPR ring needs to be from two SPANs of the node where the internal connection failure occurs. Select a SPAN simulation to generate the SF. For example, from hair In the two SPANs of the internal connection fault node, select the SPAN simulation to generate the SF that has generated the edge or the fault may generate the edge, or select the SPAN simulation to generate the SF from the link that has generated the edge on the ring or the link that may have the fault. To ensure that nodes with internal connection failures can be connected with as many nodes as possible. In this way, the connectivity and availability of the ring network can be maximized in the event of a fault in the internal connection of the node, so that the ring network provides reliable service.

In addition, in order to ensure the accuracy of the SF generated by the simulation, the ring network can provide a more reliable service. After simulating the generation of the SF by a certain SPAN, the SF can be generated for the simulation according to the change of the topology information and the link state of the ring network in real time. The SPAN is replaced.

The following takes the case where the first node detects that the SPAN state of the node changes, as an example, and describes the process of replacing the SPAN simulation to generate the SF. Referring to Figure 4, the process mainly includes the following steps:

Step 401: The first node detects the status of the two SPANs of the node in real time, and determines whether the change is detected. If yes, step 402 is performed; otherwise, the process of the process ends.

Step 402: The first node determines whether there is an internal connection fault. If yes, step 403 is performed; otherwise, the processing of the flow ends.

Step 403: The first node determines whether the SPAN of the analog generated SF selected in the previous processing of the internal connection fault currently has an SF fault, and if yes, ends the processing of the flow; if not, the step 404 is performed.

Step 404: The first node determines whether another SPAN of the node (relative to the SPAN of the previously selected simulated SF) has an SF fault. If yes, step 405 is performed; otherwise, the process ends. .

Step 405: The SF on the SPAN selected before the first node simulation disappears, and another SPAN simulation is selected to generate the SF.

In addition, corresponding to the RPR node internal connection fault processing method provided by the present invention, the present invention further provides an RPR node internal connection fault processing apparatus. Referring to FIG. 5, the device includes: Storage unit and processing unit. The storage unit is configured to store link state information of the RPR ring, and the processing unit is configured to obtain RPR ring link state information from the storage unit when the RPR node has an internal connection fault, and obtain link analysis by the link state information. As a result, the first SPAN analog generation SF is selected from the two SPANs of the internal connection fault node according to the link analysis result, and the SF signal is sent to other nodes on the ring, and the node that transmits the SF signal when the first SPAN simulation generates the SF is selected. The number is greater than the number of nodes for which the SF signal is transmitted when the second SPAN is simulated to generate the SF.

The processing unit may specifically include: an obtaining unit, a selecting unit, and a sending unit. The acquiring unit is configured to acquire RPR ring link state information from the storage unit when the node has an internal connection fault, obtain link analysis result from the link state information, and send the link analysis result to the selecting unit; And configured to select, according to the received link analysis result, the first SPAN simulation to generate the SF from the two SPANs of the internal connection faulty node, and send the generated SF information to the sending unit, and send the unit, according to the received The SF information sends SF signals to other nodes on the ring.

In addition, the processing unit may further include: an updating unit, configured to simulate the SF on the first SPAN when detecting that the selected first SPAN does not have an SF failure, and the second SPAN that is not selected has an SF failure Disappears and selects the second SPAN simulation to generate the SF.

The above-mentioned objects, technical solutions, and advantageous effects of the present invention are further described in detail. It is to be understood that the above description is not intended to limit the present invention, and is within the spirit and principles of the present invention. Any modifications, equivalent substitutions, improvements, etc., are intended to be included within the scope of the present invention.

Claims

Claim

An elastic packet ring RPR node internal connection fault processing method, characterized in that the method comprises:

When the first node on the RPR ring detects an internal connection fault, the link analysis result is obtained by the link state information of the current RPR ring.

According to the link analysis result, the first SPAN analog generation signal failure SF is selected from the two segments SPAN of the first node, and the SF signal is sent to other nodes on the ring.

The method according to claim 1, wherein when the first SPAN analog generation SF is selected, the number of nodes transmitted by the SF signal is greater than the number of nodes transmitted by the SF signal when the second SPAN analog generation SF is selected.

The method according to claim 1, wherein the link analysis result is: the first node has a SPAN generation edge;

The method according to claim 1, wherein the result of the link analysis is: a node other than the first node on the ring has a SPAN generation edge;

The selecting the first SPAN simulation from the two SPANs of the first node according to the link analysis result to generate the SF includes: selecting a SPAN analog generation SF that is closest to the edge on the distance ring from the two SPANs of the first node.

The method according to claim 1, wherein the result of the link analysis is: the first node has a SPAN fault;

The selecting the first SPAN simulation to generate the SF from the two SPANs of the first node according to the link analysis result includes: selecting the SPAN simulated generation SF with the fault from the two SPANs of the first node.

The method according to claim 1, wherein the result of the link analysis is: a node other than the first node on the ring has a SPAN fault;

The selecting the first SPAN simulation to generate the SF from the two SPANs of the first node according to the link analysis result includes: selecting, from the two SPANs of the first node, a SPAN analog generation SF that is closest to the link with the fault on the ring.

The method according to any one of claims 1 to 6, wherein the method further comprises:

When it is detected that there is no SF fault in the selected SPAN, and another SPAN of the first node that is not selected has an SF fault, the SF on the selected SPAN is simulated to disappear, and another one of the first node is selected. SPAN simulates the generation of SF.

8. An elastic packet ring RPR node internal connection fault processing device, which is characterized in that: a storage unit and a processing unit, wherein

The processing unit is configured to obtain the RPR ring link state information when the internal connection fault occurs, and obtain the link analysis result by the link state information, and select, according to the link analysis result, the two segments SPAN of the internal connection fault node The first SPAN analog generates a signal failure SF and sends an SF signal to other nodes on the ring.

The device according to claim 8, wherein the storage unit is configured to store RPR ring link state information, and provide RPR ring link state information to the processing unit.

10. The apparatus according to claim 8, wherein when the first SPAN analog generation SF is selected, the number of nodes transmitted by the SF signal is greater than the number of nodes transmitted by the SF signal when the second SPAN analog generation SF is selected.

The device according to claim 8, wherein the processing unit comprises: an obtaining unit, a selecting unit, and a sending unit, where

The obtaining unit is configured to obtain RPR ring link state information from the storage unit when the internal connection failure occurs, obtain link analysis result from the link state information, and send the link analysis result to Selection unit

And a sending unit, configured to send, according to the received SF information, an SF signal to other nodes on the ring.

The device according to claim 11, wherein the processing unit further comprises: an updating unit, configured to detect, in the selected first SPAN, that there is no SF fault, and the second one is not selected When there is an SF fault in the SPAN, the SF on the first SPAN disappears, and the second SPAN is selected to generate the SF.