Method, apparatus and system for locating faults in ip network

FIELD ip network fault location method, apparatus and system

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

Background technique

Network running will always be a communication breakdown, or network service degradation and other fault conditions, affecting the normal operation of the network. Therefore, the network fault location is a network management and operations is an important issue.

Typically, the network failure that caused the reason is more complex. For example, the IP network traffic data stream and routing information is to describe two important features information about the network, where the route change will lead to changes in the business data flow distribution, thus resulting in service QoS (Quality of Service, Quality of Service) change; on the contrary, the route does not change the situation, the traffic data stream itself will for various reasons resulting in service QoS changes, causing network failures.

Based on the traditional SNMP (Simple Network Management Protocol, tube Network Management Protocol) network management system can discover and display status and topology of individual network elements, but it does not recognize contain a variety of applications business data packet transmission in the network the actual routing path. Many network problems can not be connected by querying SNMP object mode single-cylinder to diagnose. For example, a network failure may be due to software bugs caused by a software component NE repeatedly reset. In this case, although the NE device surface may seem very healthy, but may hide a huge network hidden faults.

More common network fault conditions, such as: Routing jitter routing configuration errors due to IP network self-healing ability, in order to avoid other unrelated businesses an impact, traffic data stream may be along malformed routing path for delivery, thus inadvertently cause a malfunction problem of network congestion.

Since the IP network and based on the characteristics of SNMP network management system, the existing art, for troubleshooting IP networks only after the failure occurred through artificial way to diagnose such diagnosis time lengthy, and, since the traffic data stream and route itself is dynamic, resulting in network failure measurement inaccuracies and faults unpredictability; in addition, the existing fault diagnosis method can only determine the network link-off or not, can not be achieved for fault localization.

SUMMARY

The present invention is to solve the above technical problems in the background art exist, while presenting an IP network fault Positioning method, apparatus and system, it is possible to occur IP network failed link accurate positioning. Technical solutions of the present invention are:

Embodiments of the present invention provides a IP network fault location method, comprising:

Establish a network model of IP networks, access to traffic data stream at the IP network transmission of each path, wherein the network model comprises: a plurality of connecting link between the plurality of router node and router nodes;

Monitoring through real-time transport protocol (RTP) streams of each path preset router nodes determine the QoS parameters for each RTP stream, the path having the lowest QoS value is determined as the fault path; on the failed path included hops connection QoS parameters link estimate, determined to have links connecting the lowest QoS value is the failed link.

Of an IP network fault location device, comprising:

Network modeling module for establishing IP network network model to obtain traffic data stream at the IP network transmission of each path, wherein the network model includes: a plurality of router node and router nodes between multiple connecting link;

Fault path determining module for monitoring through real-time transport protocol of each path preset router nodes RTP stream, determine the quality of service (QoS) parameters for each RTP stream, path having the lowest QoS value is determined as the fault path;

Failed link determination module for QoS parameter value of the fault path comprises a respective hop connection link be estimated in a connecting link having a minimum QoS value is the failed link.

Of an IP network transmission system, comprising: a plurality of router node and router nodes between the plurality of connection links; wherein the pre-router nodes on which the above-described IP network fault location device.

Embodiment of the invention, since there are multiple paths in the network, between different paths cross and shared links through the RTP deployed measuring points in the network, monitoring through the various paths of the measuring point of the RTP stream, and Quality of Service QoS parameter values ​​flow is calculated, the initial identification of a failed path; then continue for QoS parameter value calculation failures along the path of each hop connection link analysis, it can be in the path fails in removing part of a healthy sharing chain Road, fail-ranges gradually narrowing, and ultimately failed link positioning.

BRIEF DESCRIPTION

Figure 1 is a schematic diagram of an IP network fault location process according to an embodiment; 2 is a schematic embodiment of the invention provides according to the fault path determining the failed link to achieve a schematic flow diagram;

Figure 3 of the present invention a specific network scenarios embodiment provides a schematic embodiment;

Figure 4 is a schematic embodiment of the invention is provided a IP network fault location device structural diagram; Fig. 5 is 4 failed link determination module is a structural schematic view.

detailed description

The present invention will now be combined with the implementation of the accompanying drawings, were clear examples of technical solutions of the present invention, a complete description of embodiments, it is clear that the described embodiments are merely part of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments by those of ordinary skill in the creative work did not make the premise obtained, are within the scope of the present invention is protected.

Computer networks can be seen as a number of backbone node (router nodes), a number of leaf node (master node) and a plurality of sides (network links, Link) composed of an undirected graph above nodes and links can be referred to as an entity (Entity). Routers and hosts are independent devices, and network links are used in routers and router, the router and the hosts establish a connection between, constitute a computer network. Network any two nodes between the path known as the network path (Path), a network path consists of several along the router nodes and links composed of network packets transmitted each through a router node, called the hop. When the network transmission failure, is often manifested in the network path degradation, such as packet loss rate increases, the delay increases, etc., or the path is interrupted.

Network fault location, that is, in the event of failure of the path to find the faulty link, but it is relying on a network path failure information 4 Gen difficult to define in the end the problem lies in what hop link. Therefore, if a problem occurs in the path removed a healthy part of the link, you can fault range further refine; through multiple health link is removed, you can finally locate the fault link.

To this end, the present invention provides a method IP network fault location method, shown in Figure 1, comprising the steps of the process:

Step 101, to establish a network model IP network, access traffic data stream at the IP network transmission of each path, wherein the network model comprises: a plurality of connecting link between the plurality of router node and router nodes;

Step 102, real-time transport protocol (RTP) streams of each path monitoring through preset router nodes determine the QoS parameters for each RTP stream, the path having the lowest QoS value is determined as Failed path;

In IP networks, along with VoIP (Voice over Internet Protocol, IP telephone network) the popularity of applications, users of VoIP session quality of service perceived also greatly affect the network quality of service overall rating. VoIP conversation they use RTP (Real-time Transport Protocol, Real-time Transport Protocol) form, in a VoIP session, the source has sent a large number of RTP data packets to the destination port, these are successively RTP packet is called RTP flow.

In addition, in order to QoS VoIP session objective evaluation, define an evaluation criterion - R value, which ranges from 0 to 100, when the higher the R-value, the higher the quality of service of VoIP.

Because of the network there are multiple paths, different paths between the presence of the cross and the shared link, so this step, in the network deployment of measurement points, monitoring through the measurement point of each path RTP stream, and each RTP stream the quality of service QoS parameter values ​​are calculated, thereby, possible to have a path of minimum QoS values ​​of these paths is determined in the fault path, the initial realization locate the fault path.

Step 103, for the fault path includes each hop connection link QoS parameter values ​​are estimated, determined to have the lowest value of QoS connection links to the failed link.

In this step, continue to fault the path of each hop connection link QoS parameter value calculation results were analyzed to determine the connecting link having the lowest value of QoS connection links to the failed link.

Because of the network there are multiple paths, different paths between the presence of the cross and shared links, therefore, the present embodiment of the invention, through the network deployment of measurement points, monitoring through the measurement point of each path RTP stream, and each RTP stream quality of service QoS parameter values ​​are calculated, the initial identification of a failed path; then continue on the failed path of each hop connection link QoS parameter value calculation results were analyzed can be in a failed path excluded part of a healthy share link fail-ranges gradually narrowing, and ultimately failed link positioning.

In order to facilitate full understanding of the technical solutions of the present invention, the present invention will now be combined with the implementation of the accompanying drawings, were clear examples of technical solutions of the present invention, a full description.

Embodiment of the invention, the first to establish a flat network model IP network, the network model including a plurality of solid model (Entity Model), represent the real IP network entities. Solid model consists of two, respectively router solid model (Router Entity Model, REM) and link entity model (Link Entity Model, LEM). Each REM represents the IP network a physical router, each LEM represents a real IP network one link, LEM is recorded with REM's Connection between relationships and record links the latest status (including: R, and fault status, status updates, time and other information).

Both solid model can simulate the real physical properties and actions, and can from it represents an entity to read the latest property value to update their corresponding attributes. For example: REM through SNMP (Simple Network Management Protocol, tube Network Management Protocol), etc., from which they represent routers read the latest routing table, and simulate the routing table lookup and route selection operation.

Through the establishment of the two kinds of solid models, the flat network model can achieve the following two functions: 1, the simulation each network entity properties and actions; 2, hop analog router operation and deduce any pair of IP addresses between the network path .

Embodiment of the invention, a specific method between the IP address of the network path deduction may be: Find all REM routing table to find the distance between the source IP address of the nearest REM (from the REM routing table can be seen that the source IP where IP subnets network and the REM is directly connected to the REM recorded as REM1), then from REM1 start, in its routing table to find the destination IP next hop REM (referred to as REM2), until you find the next hop REM far to find the process of forming a REM sequence (REMl, REM2, REM3 ... REMn). Then find all LEM model, according with the REM connection relationship, in turn find REM1 and REM2 between the link (referred to as LEM1), REM2 and REM3 between the link (referred to as LEM2), until REMn-1 and REMn between the link (referred to as LEMn-1). These links form a new sequence (LEM1, LEM2, LEM3 ... LEMn-1), the sequence means that the network path from the source IP to the destination IP.

Network model, it is ready at preset router node deployment RTP flow measurement points through by RTP stream traffic analysis means, after the measuring point for all VoIP session quality of service (R value) is calculated; via the network path deduction , to identify these VoIP session routing paths, and in accordance with VoIP session QoS parameters, calculate the session belongs to the network path QoS parameters accordingly discovery service quality degradation network path. It should be noted that the quality of service here (QoS) is the R value of RTP streams, R values ​​ranging 0-100, R value, the higher the route quality, for example, may be specified R value of 70 points or more qualified, 70 points or less indicates encounter congestion and other failures. Calculated R-value reference ITU-T standard formula is as follows:

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 one-way transmission delay, p is the packet loss rate, these two parameters are available through known measurement Obtained.

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

HW = 0 (x <0)

H (x) = l (x≥ 0) of the measuring point deployment, may be provided on a shared link multiple paths or traffic data stream flow maximum link, through the measurement point on via the shared link or multiple paths traffic data stream flow maximum link in the RTP traffic packet capture, and the RTP traffic data packets QoS monitoring; for a determined target network can be measured point is set at the height of a fault is suspected link link, or the quality of service requirements are relatively high link, or LAN with an external network of interconnected on. This evening Bu, when said deployment of measurement points on the link a plurality of router node, you can be the measuring point is set at either a router node.

When it is determined fault path after, we can continue the failed link judgment. 2, the present invention is determined based on the fault path implementation failed link provided in the embodiment, comprising:

Step 201, acquiring the failed path with other paths occur superimposed connection link, the superposition link determined not fault connection link;

Step 202, for the fault path to remove the non-faulty connection link remaining after each hop connection link respectively QoS parameter estimation, the remaining hops connection link having the lowest QoS value chain of connections road identified as the failed link.

In the aforementioned measurement process has been assigned to QoS parameters in different network paths are calculated, since there be overlap between the network path, and there may be multiple shared link different network path, therefore, the plurality of network paths quality superimposed consider obtaining a connection link of the failed path superimposed with other paths occur superimposed link determined not fault connection links, you can exclude high-quality path (can be considered normal path) all along the link is the possibility of failure, thereby removing these links in the fault path. Finally, the fault path has not been excluded links respectively QoS parameter estimation, will have a minimum value of QoS connection link is determined as the failed link. This completes the network fault location.

Among them, the path quality superposition process is the path R value analysis process, follow the "on high" principle, namely: if two paths share the same link, wherein a channel quality higher (R value is high), while the other a path quality is poor (R value is low), the shared link quality is still judged to be high. For example: two paths (A and B) share the same link L, path A R value 100, the path B R value of 50 min., The L value of R takes a higher value of 100 points.

Below through a specific network scenarios, the embodiments described above technical scheme cylinder single explanation. For example, in Fig IP network 3, there are four routing node eight, B, C and D, there are two VoIP sessions in the network, two VoIP sessions end paths were ABC and DBC. In this case, ABC session quality 4 Gen good, but DBC session slightly lower quality, therefore, need to locate the fault.

Depending on the deployment mode of the measuring points on shared links BC point C deploy measurement point on a VoIP session ABC and DBC path monitored by RTP flow measurement, calculated on the assumption that the ABC path VoIP session R value 100, while DBC path VoIP session R value of 50, whereby the failed link initially identified in DBC path that DB link and BC link are likely to fail, after deduction, found two VoIP session share links to the BC, since the quality ABC session, you can exclude the possibility of BC link failure, the failed link to navigate to the DB link.

Visible, embodiments of the invention, by using VoIP session RTP stream R value as a link evaluation basis, you can quickly auto-discover IP network fault path, the faulty link in a larger impact of specific business before the problem occurred in a timely manner given warning, so as to enhance network stability and user satisfaction; in fault location process, without human intervention or active testing, the established IP network flat network model can automatically track network entity date and attributes, unattended precise fault location. Corresponding to the above-described method embodiments, the present invention also provides a IP network fault location device shown in Figure 4, may include:

Network modeling module 401 for establishing network model IP network, access traffic data stream in the path of the IP network transmission, wherein the network model comprising: a plurality among a plurality of router node and router nodes Article connecting link;

Fault path determining module 402 for monitoring through real-time transport protocol of each path preset router nodes RTP stream, determine the quality of service (QoS) parameters for each RTP stream, the path having the lowest QoS value is determined as the fault path; Faulty link determination module 403, for the fault path includes each hop connection link QoS parameter values ​​are estimated, determined to have the lowest value of QoS connection links to the failed link.

Due to network multiple paths cross and shared links exist between different paths, therefore, the present embodiment of the invention, through the deployment of measurement points in the network, by the fault path determining module monitoring through the various paths of the measuring point of RTP stream, and each RTP stream quality of service QoS parameter values ​​are calculated, the initial identification of a failed path; then, use the failed link determination module continues the failed path of each hop connection link QoS parameter value calculation results were analyzed, You may be in a failed path excluded part of a healthy share link fail-ranges gradually narrowing, and ultimately failed link positioning.

For network modeling module, which you can obtain an IP network each router node carries routing table information, according to the routing table information, access to any service data stream transmission path IP address between the pair. Specific methods for: Find all REM routing table, find the distance between the source IP address of the nearest REM (from the REM routing table can be seen that the source IP where IP subnets directly connected to the REM, the REM recorded as REM1), then from REM1 start, find the next hop REM target IP in its routing table (referred to as REM2), until no next hop REM far, the search process to form a REM sequence (REM1, REM2, REM3 ... REMn). Then find all LEM model, according with the REM connection relationship, in turn find REM1 and REM2 between the link (referred to as LEM1), REM2 and REM3 between the link (referred to as LEM2), until REMn-1 and REMn the link between (referred to as LEMn-l). These links form a new sequence (LEM1, LEM2, LEM3 ... LEMn-1), the sequence that is expressed from the source IP to a destination IP network path.

When the network model, can be deployed in a preset router nodes RTP flow measurement points through by RTP stream traffic analysis means, through the measurement point quality of service to all VoIP sessions (R value) is calculated; via the network path deduction, to identify these VoIP session routing paths, and according VoIP session QoS parameters, calculate the session belongs to the network path QoS parameters accordingly discovery service quality degradation network path.

Embodiment of the invention, the deployment of the measuring points can be set on a shared link multiple paths or traffic data stream flow maximum link through which the measuring point on the shared link or traffic data stream flows via multiple paths biggest link in the RTP traffic packet capture, and the RTP traffic data packets QoS monitoring; for a determined target network can be measured point is set at a height of suspected faulty link, or quality of service requirements are relatively high link, or LAN and external networks Interconnecting links. In addition, when the link above to deploy measuring point has multiple router nodes can be measured point is set at either a router node.

In the present embodiment provides a IP network fault localization apparatus shown in Figure 5, the failed link determination module 403, may include:

Superimposing link acquisition unit 501 for acquiring the connection link to the failed path superimposed with other paths occur superimposed link determined not fault connection link;

QoS estimation unit 502, for the failed path to remove the non-fault connection link remaining hops connecting links respectively QoS parameter estimation;

Results unit 503 for connecting link of the remaining hops connection link having the lowest QoS value is the failed link.

Determining module 402 through the fault path can be achieved on QoS parameters for different network path is calculated separately; because of overlapping and duplication between the network path, and there may be multiple shared links different network path, therefore, by the above superposition link Get unit 501, a plurality of network paths quality overlay consider acquiring the failed path superimposed connecting link with other normal paths will occur superimposed link determine fault connection link non, you can exclude high-quality path (available considered the possibility of a normal path) all along the link fails, thereby removing these links in the fault path. Finally, QoS estimation unit 502, the fault path has not been excluded links respectively QoS parameter value is estimated to have a connection link minimum QoS value is determined as the failed link. This completes the network fault location. For the IP network transmission systems, it can be applied above embodiment the IP network fault location device, in order to achieve IP network fault location. Typically, IP network transmission system typically includes a plurality of router node and router nodes between multiple connections link the specific implementation, you can preset the router node on which the above-described IP network fault location device so that the router nodes have fault location function, in order to achieve the final network fault location by the router nodes.

Here should be noted that the pre-router nodes into a plurality of paths shared links or traffic data stream flow maximum link on the deployment of any router nodes. For detailed network fault location principle and embodiments of the invention are no longer Chan said. For the apparatus and system embodiments, due to its substantially corresponding to the method of Example, it is described as Section compares the single-cylinder, relevant, see method embodiments described can be. The above described apparatus embodiments are merely illustrative, wherein the module as a separate member described may or may not be physically separated, as a component of the display module may or may not be a physical block, i.e., it may be located a place, or it can be distributed to multiple network modules. You can select some or all of the modules to achieve the object of the present embodiment of implementation of the program according to the actual needs. Those of ordinary skill without creative efforts that can be understood and implemented.

The above description of the disclosed embodiments, so that the skilled in the art to make or use the invention. These various modifications to the field of professional and technical personnel embodiments will be apparent, the general principles defined herein may be without departing square embodiment of the invention the spirit or scope of the cases, and in other embodiments achieve. Accordingly, embodiments of the invention will not be limited to these embodiments shown herein, but rather to meet consistent with the principles disclosed herein, and novel features of the widest range.