WO2012106925A1 - Method, apparatus and system for locating faults in ip network - Google Patents

Abstract

Disclosed in the present invention are a method, an apparatus and a system for locating faults in an IP network. Wherein, the method for locating faults in the IP network includes: establishing a network model of the IP network, and obtaining each path for transmitting service data flows in the IP network, wherein the network model comprises: a plurality of router nodes and a plurality of connection links among the router nodes; monitoring real-time transmission protocol (RTP) flows on each path passing the preset router node, determining the quality of service (QoS) parameter values of each RTP flow, and determining the path with lowest QoS value as the fault path; estimating the QoS parameter values of each hop connection link included in the fault path, and determining the connection link with the lowest QoS value as the fault link. The embodiments of the present invention enable accurate locating for the fault link in an IP network.

Description

 IP network fault location method, device and system

 The present invention relates to the field of communications technologies, and in particular, to an IP network fault location method, apparatus, and system.

Background technique

 During the operation of the network, faults such as communication interruption or network service quality degradation often occur, which affects the normal operation of the network. Therefore, network fault location is an important issue in network management and operation.

 Often, the causes of network failures are more complex. For example, for an IP network, the service data flow and the routing information are two important characteristic information of the network. The change of the route causes the distribution of the service data flow to change, thereby causing the service QoS (Quality of Service). On the other hand, in the case that the route does not change, the service data flow itself may cause a change in the service QoS for various reasons, thereby causing a network failure.

 The traditional network management system based on SNMP (Simple Network Management Protocol) can discover and display the status and topology of a single network element, but it cannot identify the service data packets containing various applications and transmit them in the network. The actual routing route. Many network problems cannot be diagnosed by means of querying SNMP objects. For example, a network failure may be caused by a software vulnerability that causes a software component on the network element to be reset repeatedly. In this case, although the surface of the network element device may appear to be very healthy, it may hide a huge network failure.

 More common network failures, such as: route jitter, routing configuration errors, due to the ability of the IP network to self-heal, in order to avoid impact on other unrelated services, the traffic may be transmitted along the malformed routing route. This inadvertently leads to network congestion problems.

 Due to the above characteristics of the IP network and the SNMP-based network management system, in the prior art, the fault query for the IP network can only be diagnosed manually after the fault occurs, making the diagnosis time lengthy, and, due to the service data flow and The route itself is dynamic, resulting in inaccuracy in network fault measurement and unpredictability of faults. In addition, the existing fault diagnosis method can only judge whether the network link is on or off, and cannot locate the fault.

Summary of the invention

The present invention solves the above technical problems existing in the background art, and proposes an IP network failure. The positioning method, device and system can accurately locate the faulty link in the IP network. The technical solution of the present invention is:

 An embodiment of the present invention provides a method for locating an IP network fault, including:

 Establishing a network model of the IP network, and acquiring each path of the service data flow in the IP network, where the network model includes: multiple connection links between multiple router nodes and router nodes;

 Monitoring a real-time transport protocol RTP stream of each path of the preset router node, determining a quality of service QoS parameter value of each RTP stream, determining a path having the lowest QoS value as a fault path; and performing each hop included in the fault path The QoS parameter value of the link is estimated to determine that the link with the lowest QoS value is the faulty link.

 An IP network fault locating device includes:

 a network model establishing module, configured to establish a network model of the IP network, and obtain each path that the service data stream transmits in the IP network, where the network model includes: multiple pieces between multiple router nodes and router nodes Connection link

 a fault path determining module, configured to monitor a real-time transport protocol RTP flow of each path of the preset router node, determine a quality of service QoS parameter value of each RTP stream, and determine a path with the lowest QoS value as a fault path;

 The fault link determining module is configured to estimate a QoS parameter value of each hop connection link included in the fault path, and determine that the link link with the lowest QoS value is a fault link.

 An IP network transmission system includes: a plurality of connection links between a plurality of router nodes and router nodes; wherein the IP network fault location device is set on a preset router node.

 In the embodiment of the present invention, since there are multiple paths in the network, there are crossovers and shared links between different paths. By deploying measurement points in the network, the RTP flows of each path passing through the measurement points are monitored, and each RTP is detected. The service quality QoS parameter value of the stream is calculated, and the fault path is initially located; then, the QoS parameter value calculation result of each hop connection link in the fault path is analyzed, and a part of the healthy shared chain can be eliminated in the path of the fault. The road realizes the gradual reduction of the fault range and finally realizes the positioning of the fault link.

DRAWINGS

FIG. 1 is a schematic flowchart of a method for locating an IP network fault according to an embodiment of the present invention; FIG. 2 is a schematic diagram of an implementation process for determining a faulty link according to a fault path according to an embodiment of the present invention;

 FIG. 3 is a schematic diagram of a specific network scenario according to an embodiment of the present disclosure;

 FIG. 4 is a schematic structural diagram of an IP network fault locating device according to an embodiment of the present invention; FIG. 5 is a schematic structural diagram of a fault link determining module in FIG.

detailed description

 The technical solutions in the embodiments of the present invention will be described in detail with reference to the accompanying drawings. All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present invention without the creative work are all within the scope of the present invention.

 A computer network can be seen as an undirected graph of a number of backbone nodes (router nodes), a number of leaf nodes (host nodes), and a number of edges (network links, Link), which can be collectively referred to as Entity. The router and the host are independent devices, and the network link is used to establish a connection between the router and the router, and between the router and the host to form a computer network. The path between any two nodes in the network is called a network path. A network path consists of a number of router nodes and links along the path. Each packet transmitted in the network passes through a router node and is called a hop. When the network transmission output is faulty, it often appears that the quality of the network path is degraded, such as an increase in packet loss rate, an increase in delay, etc., or an interruption in the path.

 Network fault location, that is, finding the faulty link in the faulty path, but relying only on the fault information of one network path, it is difficult to define which hop link the problem is. Therefore, if a part of the healthy link can be removed from the path in question, the fault range can be further narrowed; by eliminating the multiple healthy links, the faulty link can be finally located.

 To this end, an embodiment of the present invention provides a method for locating an IP network fault, as shown in FIG. 1, including the following steps:

 Step 101: Establish a network model of the IP network, and obtain each path that the service data stream transmits in the IP network, where the network model includes: multiple connection links between multiple router nodes and router nodes;

Step 102: Monitor a real-time transport protocol RTP stream of each path of the preset router node, determine a quality of service QoS parameter value of each RTP stream, and determine a path with the lowest QoS value as Fault path

 In the IP network, with the popularity of VoIP (voice over Internet Protocol) applications, users' perception of the quality of service of VoIP sessions also greatly affects their overall evaluation of network service quality. Most of the VoIP sessions are in the form of RTP (Real-Time Transport Protocol). During a VoIP session, the source sends a large number of RTP packets to the destination. These successive RTP packets are called RTP. flow.

 In addition, in order to objectively evaluate the QoS of the VoIP session, an evaluation criterion, R value, is defined, which ranges from 0 to 100. When the R value is higher, the service quality of the VoIP is higher.

 Because there are multiple paths in the network, there are crossovers and shared links between different paths. Therefore, in this step, measurement points are deployed in the network, and RTP flows of each path passing through the measurement points are monitored, and each RTP stream is detected. The quality of service QoS parameter values are calculated, so that the path with the lowest QoS value among these paths can be determined as the fault path, and the location of the fault path is initially implemented.

 Step 103: Estimate a QoS parameter value of each hop connection link included in the fault path, and determine that the connection link with the lowest QoS value is a fault link.

 In this step, the calculation result of the QoS parameter value of each hop connection link in the fault path is continuously analyzed, and the connection link having the lowest QoS value in each connection link is determined as the fault link.

 In the embodiment of the present invention, the RTP flow of each path passing through the measurement point is monitored by deploying measurement points in the network, because there are multiple paths in the network, and there is a crossover and a shared link between the different paths. The quality of service QoS parameter values of each RTP stream are calculated, and the fault path is initially located; then, the QoS parameter value calculation result of each hop connection link in the fault path is continuously analyzed, and a part of the healthy path can be eliminated in the path of the fault. The link is shared, and the fault range is gradually reduced, and finally the fault link is located.

 In order to facilitate a full understanding of the technical solutions of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described.

In the embodiment of the present invention, a planar network model is first established for an IP network, and the network model includes a plurality of entity models (Entity Models) respectively representing respective entities in the real IP network. There are two types of entity models, namely Router Entity Model (REM) and Link Entity Model (LEM). Each REM represents a physical router in the IP network, and each LEM represents a link in the real IP network, and the LEM records it with REM. The connection relationship between the links, and record the latest state of the link (including: R value, fault status, status update time, etc.).

 The above two entity models can simulate the properties and actions of real entities, and can read the latest attribute values from the entities they represent to update their corresponding attributes. For example: REM can read the latest routing table from the router it represents through SNMP (Simple Network Management Protocol) and simulate its routing table lookup and route selection.

 By establishing the above two physical models, the planar network model can achieve the following two functions: 1. Simulate the attributes and actions of each network entity; 2. Simulate the router action hop by hop and derive the network path between any pair of IP addresses. .

 In the embodiment of the present invention, the specific method for deriving the network path between the IP addresses may be: searching the routing table of all the REMs, and finding the REM closest to the source IP address. (From the routing table of the REM, the IP sub-source is located. The network is directly connected to the REM, and the REM is recorded as REM1). Then, starting from REM1, the next hop REM of the target IP (denoted as REM2) is searched in its routing table in turn, until the next hop REM is found, and the search is performed. The process forms a REM sequence (REMl, REM2, REM3... REMn). Then find all LEM models, according to their connection with REM, find the link between REM1 and REM2 (denoted as LEM1), the link between REM2 and REM3 (denoted as LEM2), until REMn-1 and REMn The link between (marked as LEMn-1). These links form a new sequence (LEM1, LEM2, LEM3...LEMn-1), which represents the network path from the source IP to the destination IP.

 After the network model is established, the RTP stream measurement point can be deployed at the preset router node, and the quality of service (R value) of all VoIP sessions passing through the measurement point is calculated by the RTP stream traffic analysis means; Finding the routing path of these VoIP sessions, and calculating the QoS parameters of the network path to which each session belongs according to the QoS parameters of the VoIP session, and thereby discovering the network path with degraded service quality. It should be noted that the quality of service (QoS) here is the R value of the RTP stream, and the R value is 0-100. The higher the R value, the higher the path quality. For example, the R value can be specified to be 70 or more. Qualified, 70 points or less indicates failures such as congestion. The calculation method of the R value refers to the ITU-T standard, and the specific calculation formula is:

R=100-(0.024*d+0.11*(d-177.3)*H(d-177.3)+16.68*ln(l+0.0426*p)) where d is the unidirectional transmission delay and p is the packet loss rate. Both of these parameters can be measured by known measurements The method is obtained.

 Where H(x) is a function, defined as:

HW=0 ( x<0)

 H (x) = l ( x ≥ 0) The deployment of the above measurement points can be set on the shared link of multiple paths or the link with the largest traffic of traffic data, through which the pair is connected Or RTP service data packets in multiple paths of the link with the largest traffic flow of the service data are captured, and QoS monitoring is performed on the RTP service data packet; for a certain target network, the measurement point may be set to be highly suspected and faulty. A link, or a link that requires a higher quality of service, or a link between a local area network and an external network. Furthermore, when there are multiple router nodes on the link where the above measurement points are deployed, the measurement points can be set to any of the router nodes.

 After determining the fault path, the judgment of the fault link can be continued. As shown in FIG. 2, the implementation manner of determining a faulty link according to a fault path according to an embodiment of the present invention includes:

 Step 201: Acquire a connection link that overlaps with other paths in the fault path, and determine the superposed link as a non-faulty connection link.

 Step 202: Perform QoS parameter value estimation on each hop connection link remaining after removing the non-faulty connection link in the fault path, and connect the remaining hop connection link with the lowest QoS value. The road is determined to be a faulty link.

 In the foregoing measurement process, the QoS parameters of different network paths have been calculated separately. Because there are intersections and overlaps between network paths, and multiple network paths may have multiple shared links, the quality of multiple network paths is performed. In addition, the connection link that overlaps with other paths in the fault path is obtained, and the superimposed link is determined as a non-faulty link, and all the links along the high quality path (which can be considered as a normal path) can be excluded. The possibility of failure, thereby culling these links in the fault path. Finally, the QoS parameter value estimation is performed on the unremoved links in the fault path, and the connection link having the lowest QoS value is determined as the fault link. At this point, the positioning of the network failure is completed.

The process of superimposing the path quality is the process of analyzing the R value of the path, and follows the principle of “being high”, that is, if two paths share the same link, one of the paths has a higher quality (higher R value), and the other If the quality of one path is poor (R value is low), the quality of the shared link is still judged to be high. For example: Two paths (A and B) share the same link L. The R value of path A is 100 points, and the R value of path B is 50 points. The R value of L takes a higher value of 100 points.

 The technical solution of the above embodiment is described in the following by a specific network scenario. For example, in the IP network shown in Figure 3, there are four routing nodes eight, B, C, and D. There are two VoIP sessions in the network. The end-to-end paths of the two VoIP sessions are A-B-C and D-B-C. At this point, the A-B-C session quality is good, but the D-B-C session quality is slightly worse, so the fault needs to be located.

 According to the deployment method of the above measurement points, the measurement points are deployed at point C on the shared link BC, and the VoIP sessions on the ABC and DBC paths are monitored. According to the RTP traffic measurement, it is assumed that the R value of the VoIP session on the ABC path is calculated. 100, and the R value of the VoIP session on the DBC path is 50. Therefore, the faulty link is initially determined to be faulty on the DBC path, that is, both the DB link and the BC link, and after deduction, two VoIP sessions are found. The shared link is BC. Due to the high quality of the ABC session, the possibility of a BC link failure can be ruled out, and the faulty link can be located to the DB link.

 It can be seen that, in the embodiment of the present invention, by using the R value of the RTP stream of the VoIP session as the basis of the link evaluation, the fault path and the fault link in the IP network can be quickly and automatically found, which is larger and affects the specific service. Early warning is given before the problem occurs, thereby improving the stability of the network and the satisfaction of the user. In the process of fault location, no manual intervention or active testing is required, and the established planar network model of the IP network can automatically track the network entity. State-of-the-art status and attributes for unattended accurate fault location. Corresponding to the foregoing method embodiments, the present invention further provides an IP network fault locating device, as shown in FIG. 4, which may include:

 The network model establishing module 401 is configured to establish a network model of the IP network, and obtain each path that the service data stream transmits in the IP network, where the network model includes: multiple between multiple router nodes and router nodes Strip connection link;

The fault path determining module 402 is configured to monitor a real-time transport protocol RTP stream of each path of the preset router node, determine a quality of service QoS parameter value of each RTP stream, and determine a path with the lowest QoS value as a fault path; The fault link determining module 403 is configured to estimate a QoS parameter value of each hop connection link included in the fault path, and determine that the link link having the lowest QoS value is a fault link.

 In the embodiment of the present invention, the fault path determination module monitors each path passing through the measurement point by deploying measurement points in the network, because there are multiple paths in the network, and there is a crossover and a shared link between the different paths. RTP flow, and calculating the quality of service QoS parameter values of each RTP stream, initially locating the fault path; then, using the fault link determination module to continue analyzing the QoS parameter value calculation result of each hop connection link in the fault path, Then, a part of the healthy shared link can be eliminated in the path of the fault, and the fault range is gradually reduced, and finally the fault link is located.

 For the network model building module, the routing data carried by each router node in the IP network can be obtained, and the service data stream transmission path between any pair of IP addresses can be obtained according to the routing table information. The specific method may be: searching the routing table of all REMs, and finding the REM closest to the source IP address. (From the routing table of the REM, it can be seen that the IP subnet where the source IP is located is directly connected to the REM, and the REM is recorded as REM1). Then, starting from REM1, it searches for the next hop REM of the target IP (denoted as REM2) in its routing table, until the next hop REM is not found, and the search process forms a REM sequence (REM1, REM2, REM3... REMn). Then find all LEM models, according to their connection with REM, find the link between REM1 and REM2 (denoted as LEM1), the link between REM2 and REM3 (denoted as LEM2), until REMn-1 and REMn The link between them (denoted as LEMn-l). These links form a new sequence (LEM1, LEM2, LEM3... LEMn-1), which represents the network path from the source IP to the destination IP.

 After the network model is established, the RTP flow measurement point can be deployed at the preset router node, and the quality of service (R value) of all VoIP sessions passing the measurement point is calculated by the RTP flow traffic analysis means; Derivation, find the routing path of these VoIP sessions, and calculate the QoS parameters of the network path to which each session belongs according to the QoS parameters of the VoIP session, and discover the network path with degraded service quality.

In the embodiment of the present invention, the deployment of the measurement points may be set on a shared link of multiple paths or a link with the largest traffic of service data flows, and the traffic passing through the shared link or service data flows through the measurement point. The RTP service data packets in multiple paths of the largest link are captured, and QoS monitoring is performed on the RTP service data packets; for a certain target network, the measurement points may be set on a link suspected to be faulty, or A link with a higher quality of service, or a LAN and an external network On the interconnected link. In addition, when there are multiple router nodes on the link where the above measurement points are deployed, the measurement points can be set at any one of the router nodes.

 In the IP network fault locating device provided by the embodiment of the present invention, as shown in FIG. 5, the fault link determining module 403 may include:

 The superimposed link acquiring unit 501 is configured to acquire a connection link that is superimposed with other paths in the fault path, and determine the superposed link as a non-faulty connection link;

 The QoS estimation unit 502 is configured to perform QoS parameter value estimation on each hop connection link remaining after removing the non-faulty connection link in the fault path.

 The result unit 503 is configured to use the connection link having the lowest QoS value among the remaining hop connection links as a fault link.

 Through the fault path determination module 402, the QoS parameters of the different network paths can be separately calculated; because there are intersections and overlaps between the network paths, and different network paths may have multiple shared links, the obtained The unit 501 is configured to superimpose the quality of the plurality of network paths, obtain an overlapping connection link with the other normal paths in the fault path, and determine the superimposed link as a non-faulty connection link, and the high quality path may be excluded. It is considered to be the possibility of failure of all links along the route in the normal path, so that these links are eliminated in the fault path. Finally, the QoS estimation unit 502 performs QoS parameter value estimation on the unremoved links in the fault path, and determines the connection link having the lowest QoS value as the fault link. At this point, the positioning of the network failure is completed. For the IP network transmission system, the IP network fault location device described in the foregoing embodiment can be applied to locate the IP network fault. Generally, an IP network transmission system usually includes multiple connection links between multiple router nodes and router nodes. In specific implementation, the above IP network fault location device may be set on a preset router node, so that the router node has The function of fault location, so that the final network fault location is achieved through the router node.

It should be noted that the preset router node is a shared link of multiple paths or any router node deployed on a link with the largest traffic of traffic data. For detailed network fault location principles, the embodiments of the present invention are not described. For the device and system embodiment, since it basically corresponds to the method embodiment, it is described Compare the cartridges, see the section for a description of the method examples. The device embodiments described above are merely illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, ie may be located One place, or it can be distributed to multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the embodiment. Those of ordinary skill in the art can understand and implement without any creative effort.

 The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments are obvious to those skilled in the art, and the general principles defined herein may be made without departing from the spirit or scope of the embodiments of the invention, in other embodiments. achieve. Therefore, the present embodiments of the invention are not to be limited to the embodiments shown herein, but are to be accorded to the broadest scope of the principles and novel features disclosed herein.

Claims

Rights request  A method for locating an IP network fault, characterized in that it comprises:  Establishing a network model of the IP network, and acquiring each path of the service data flow in the IP network, where the network model includes: multiple connection links between multiple router nodes and router nodes;  Monitoring a real-time transport protocol RTP stream of each path of the preset router node, determining a quality of service QoS parameter value of each RTP stream, determining a path having the lowest QoS value as a fault path; and performing each hop included in the fault path The QoS parameter value of the link is estimated to determine that the link with the lowest QoS value is the faulty link.  The method for locating an IP network fault according to claim 1, wherein the real-time transport protocol RTP stream that monitors each path of the preset router node includes:  RTP service packets in multiple paths through the shared link or the link with the largest traffic flow of the service data flow through the shared link of multiple paths or the router node deployed on the link with the largest traffic flow of the service data flow Capture, and perform QoS monitoring on the RTP service data packet.  The IP network fault location method according to claim 2, wherein the QoS parameter value of each hop connection link included in the fault path is estimated, and the connection link having the lowest QoS value is determined as The fault link includes:  Obtaining a connection link that overlaps with other paths in the fault path, and determining the superposed link as a non-faulty connection link;  Determining QoS parameter values for each hop connection link remaining after removing the non-faulty connection link in the fault path, and determining a connection link having the lowest QoS value among the remaining hop connection links as Faulty link.  The method for locating an IP network fault according to claim 2, wherein the shared node or the router node deployed on the link with the largest traffic of the service data flows through the shared link The RTP service data packet in the multiple paths of the link with the largest service traffic flow is captured, and the QoS monitoring is performed on the RTP service data packet, including: RTP in multiple paths of the link with the largest traffic through the shared link or service data flow through the shared link of the multiple paths or any router node deployed on the link with the largest traffic flow The service data packet is captured, and QoS monitoring is performed on the RTP service data packet. The IP network fault locating method according to claim 1, wherein the acquiring the path of the service data stream transmitted in the IP network specifically includes:  Obtain routing table information carried by each router node in the IP network, and obtain a service data transmission path between any pair of IP addresses according to each routing table information.  6. An IP network fault locating device, comprising:  a network model establishing module, configured to establish a network model of the IP network, and obtain each path that the service data stream transmits in the IP network, where the network model includes: multiple pieces between multiple router nodes and router nodes Connection link  a fault path determining module, configured to monitor a real-time transport protocol RTP flow of each path of the preset router node, determine a quality of service QoS parameter value of each RTP stream, and determine a path with the lowest QoS value as a fault path;  The fault link determining module is configured to estimate a QoS parameter value of each hop connection link included in the fault path, and determine that the link link with the lowest QoS value is a fault link.  The IP network fault locating device according to claim 6, wherein the fault path determining module is configured by a router node deployed on a shared link of multiple paths or a link with the largest traffic of traffic data. Capturing RTP service data packets in multiple paths of the link through which the shared link or service data flow is the largest, and performing QoS monitoring on the RTP service data packets. The IP network fault locating device according to claim 7, wherein the fault link determining module comprises:  And a superimposed link acquiring unit, configured to acquire a connection link that overlaps with other paths in the fault path, and determine the superposed link as a non-faulty connection link;  a QoS estimation unit, configured to perform QoS parameter value estimation on each hop connection link remaining after removing the non-faulty connection link in the fault path;  And a result unit, configured to use the connection link having the lowest QoS value among the remaining hop connection links as a fault link. The IP network fault locating device according to claim 7, wherein the fault path determining module is configured by using the shared link of the multiple paths or the link with the largest traffic of the service data flow. a router node that captures RTP service data packets in multiple paths of a link that has the largest traffic through the shared link or service data flow, and performs QoS on the RTP service data packet monitor.  The IP network fault locating device according to claim 6, wherein the network model establishing module obtains any pair of routing table information carried by each router node in the IP network according to routing information. The service data stream transmission path between IP addresses.  An IP network transmission system, comprising: a plurality of connection links between a plurality of router nodes and a router node; wherein, the preset router node is set as claimed The IP network fault locating device according to any one of the items 6-8, 10.  The IP network transmission system according to claim 11, wherein the preset router node is a shared link of multiple paths or any router node deployed on a link with the largest traffic of traffic data.