KR20020077000A - Methode for fault management of intranet OSI layer - Google Patents

Abstract

PURPOSE: A method of managing a layer failure of intranet is provided to include a failure correlative analyzing function between each layer in a communication network, to understand a basic cause of a failure by using a layer correlation, and to analyze a trap generated in each layer, thereby understanding a system having a failure. CONSTITUTION: System, subsystem, and link failures(101-103) generated from a physical layer(100) are transmitted to a data link layer(200). A DL sub-network failure(201) and a DL link failure(202) generated in a data link(200) are transmitted to a network layer(300). The network layer(300) generates a network sub-network failure(301) and a network topology link failure(302), and transmits the failures(301,302) to service layers(4-7 layers). Traps are classified by the physical layer(100), the data link layer(200), the network layer(300), and the service layers(4-7 layers). Correlation between cause and effect is analyzed for the classified traps. A failure cause/effect graph for each layer and a failure cause/effect graph have to be included between layers. A corresponding failure is selected on the failure cause/effect graph. A failure-generated position is shown by repeating the above procedure.

Description

Intranet hierarchical fault management method {Methode for fault management of intranet OSI layer}

The present invention relates to a failure management method of an intranet, and more particularly, to analyze the correlation of failures occurring hierarchically in an intranet, to find the root cause of the failure, and to alert the administrator to take an action. Hierarchical failure management method.

As the network speeds up recently, large-scale data loss occurs in the event of short-term faults. In order to solve this problem, the problem management technology of communication network has been studied.

Conventional fault management methods of communication networks include such functions as detection of faults, localization of faults, recovery of faults, and root cause detection of faults.

For example, the present invention may be applied by a network management technology such as Simple Network Management Protocol (SNMP), Telecommunications Management Network (TMN), and Telecommunications Information Networking Architecture (TINA). Here, SNMP refers to the simplest network management technology as a simple network management protocol, and TINA refers to a network management technology applicable to a distributed communication network.

In case of equipment equipped with SNMP, trap is generated in case of failure, and in case of equipment equipped with TMN and TINA network management function, alarm is generated in case of failure.

Here, the trap refers to a message informing the administrator of the failure of the system in which the problem occurred when a problem occurs in the system.

To date, research has shown that there are not many functions for detecting the root cause of various obstacles in the network. SNMP-based network management programs such as Hewlett-Packard (HP) 's Node Manager and Openview include some features that correlate to failures.

Software such as Node Manager and OpenView associate similar traps in the event of a failure.

However, such a failure management method according to the related art is a layer 2 (data link layer), layer 3 when a failure occurs in one layer (physical layer) of the seven OSI (Open System Interconnection) layer for interconnecting heterogeneous computer systems. There was a problem that the root cause could not be found by analyzing duplicate traps occurring in the (network layer).

The present invention is to solve the above problems of the prior art, the purpose is to include a failure correlation analysis function between each layer in the communication network, to identify the root cause of the failure using the hierarchical correlation, and to transmit to the administrator The present invention provides a hierarchical failure management method of an intranet that analyzes traps generated at each layer of a communication network to identify a location of a failed system.

1 is a conceptual diagram for hierarchical failure management of an intranet according to an embodiment of the present invention;

2 is a failure cause-result display diagram for hierarchical failure management of an intranet according to an embodiment of the present invention;

3A, 3B are disability correlation diagrams in cause-effect of disability in FIG. 2,

Figures 4a, 4b is a failure position analysis according to the cause-effects of the disorder in FIG.

<Description of Symbols for Major Parts of Drawings>

100: physical layer 101: system failure

102: subsystem failure 103: link failure

200: data link layer 201: DL sub-network failure

202 DL link failure 211 Subnetwork 1.1

212: Subnetwork 1.2221: Subnetwork 1.1.1

222: subnetwork 1.2.1 300: network layer

301: network subnetwork failure 302: network topology link failure

400: service layer

In order to achieve the object of the present invention, a hierarchical failure management process of an intranet according to the present invention performs a network management function of detecting a failure generated in an OSI (Open Systems Interconnection) layer for communication of an intranet system and alerting the user of the failure. A method for alarming an intranet equipped with a protocol for performing the above, the first step of generating a corresponding trap when a failure occurs in at least one layer of the system, the second step of classifying the trap by layer, and the layer A third step of analyzing the correlation of traps classified by each other; a fourth step of calculating a cause / result relationship graph of the failure by the analyzed correlation; and a step of selecting a corresponding failure on the failure cause / result graph Step 5 and determine the location of the failure according to the correlation of the cause of the failure on the failure cause / result graph It is characterized by made of an sixth step.

The present invention made as described above will be described in detail with reference to the accompanying drawings.

1 is a correlation diagram of occurrence of failures in each layer for implementing a hierarchical failure management method of an intranet according to an exemplary embodiment of the present invention.

As shown here, the OSI layer for intranet communication includes a physical layer 100, a datalink layer (abbreviated as "DL") 200, a network layer ( 300) and a service layer (Service Layer) 400.

The failures occurring in the physical layer 100 include a system failure 101, a subsystem failure 102, and a link failure 103, and the data link layer 200 is a DL. There is a DL Subnetwork Failure 201 and a DL Link Failure 202, and the failures occurring at the Network Layer 300 are Network Subnetwork Failure 301 and Network Topology Link Failure. Link Failure) (302).

When described in more detail with reference to the accompanying drawings, the operation according to the embodiment of the present invention configured as described above are as follows.

1 is a correlation diagram of the occurrence of failures in each intranet layer according to an embodiment of the present invention, the failure in the communication network is divided hierarchically 1 layer, 2 layer, 3 layer and higher layer (4-7 layer) It can be classified as a disorder. Failures in each layer are correlated in each layer.

A failure occurring in the physical layer (layer 1) 100 necessarily causes a failure in the data link layer (layer 2), the network layer (layer 3), and the upper layer.

That is, the system, subsystem, and link failures 101 to 103 generated at the physical layer 100 are transmitted to the data link layer 200, and the DL subnetwork failure 201 and DL generated at the data link 200 are transmitted. The link failure 202 transmits the failure to the network layer 300.

In the network layer 300, a network subnetwork failure 301 and a network topology link failure 302 occur, and the failure is transmitted to a service layer (4-7 layers).

2 is a failure cause-result graph for hierarchical failure management of an intranet according to the present invention.

As such, a system equipped with SNMP includes a trap for fault management. Here, traps are classified by layer (physical layer, data link layer, network layer, service layer).

The physical layer 100 and the data link layer 200 are composed of an Ethernet-based network and an Asynchronous Transfer Mode (ATM) -based network. In addition, the data link layer 200 is classified into an 'EML, L-NML, M-NML, H-NML' layer.

The network layer (IP) 300 is similarly classified into 'EML, L-NML, M-NML, H-NML' layers, respectively.

In this way, the correlation between the cause / effect of traps classified by hierarchy is analyzed.

Every system contains different traps, so you should have a separate failure cause / effect graph for each system.

You must have a failure cause / effect graph for each layer and a failure cause / effect graph between each layer.

3A and 3B are failure cause analysis / correlation analysis diagrams using a failure cause / result graph for explaining a hierarchical failure management method of an intranet according to an exemplary embodiment of the present invention.

As shown therein, the SNW1.2 212 of the first link Link1, which connects the predetermined subnetwork 1.1 (SNW1.1) 211 and the subnetwork 1.2 (SNW1.2) 212, which is loaded with SNMP, is shown. When a link failure occurs at the port, the corresponding trap is generated to the subnetwork (SNW1) 210 of the upper layer.

That is, as shown in FIG. 3B, the SNW1.2 212 is transmitted to the SNW1 210 of the upper layer through a connection termination point (CTP) and a link termination point (LTP) of the upper layer.

The failure so generated is reported as any one or more causes of the failure cause / results graph.

Each cause of disability is transmitted to the upper level according to the priority of the cause / effect of the failure, and each cause is reported as an alarm representing the correlation of the higher cause according to the priority of the cause / effect.

As described above, the process of reporting the cause of failure occurring in each layer to the uppermost level is repeated to report the highest level.

4A and 4B are analysis diagrams for identifying a location of a failure using a failure cause / result graph for implementing a hierarchical failure management method of an intranet according to an exemplary embodiment of the present invention.

If a failure occurs in a system of any hierarchy, the failure management device selects the failure on the failure cause / result graph.

If a connection failure is reported to SNW1 210 as shown in FIG. 4A, the cause of the failure is searched for in the failure cause / result graph.

That is, Link 1 between SNW1.1 211 and SNW1.2 212 in the lower layer of SNW1 210 and each link 1.1 of SNW1.1.1 221 in the next lower layer. The faults of the respective links Link1.3 and Link1.2 of the link 1.3 and the SNW1.2.1 222 are searched for.

As a result of the search, it is determined that a failure has occurred in Link 1.1 of snw1.1.1 (221).

In addition, the search for the cause of the failure takes precedence from the cause of the alarm representing the correlation when the alarm is delivered as shown in FIG. 4B. That is, the SNW1.1 210 tracks and determines the link (Link1.1) of the lower layer after the SNW1.1 (211), the link (Link) and the link failure occurs in the lower layer.

In the absence of alerts representing correlations, all faults with a cause / effect relationship should be searched and examined on the fault cause / effect graph.

By repeating the above process to determine the location where the actual failure occurred.

Although the preferred embodiment according to the present invention has been described above, it can be applied to other network communication as well as Ethernet, ATM, and those skilled in the art can be various without departing from the claims of the present invention. It is understood that modifications and variations can be made.

As described above, the hierarchical failure management method of the intranet according to the present invention can be used to find the root cause for various failures occurring hierarchically in the intranet. Since all faults occurring in the physical layer, data link layer, and network layer are not independent, there is an effect that the exact cause and location of the fault can be identified by calculating the correlation of faults by layer.

Claims (6)

In the intranet failure warning method equipped with a protocol for performing a network management function that detects a failure occurred in the OSI (Open Systems Interconnection) layer for communication of the intranet system and alerts the user, Generating a corresponding trap when a failure occurs in at least one layer of the system; Classifying the trap by layer; A third step of analyzing correlations of traps classified by the layers; A fourth step of calculating a cause / effect relationship graph of the disorder based on the analyzed correlation; Selecting a corresponding disorder on the disorder cause / effect graph; And a sixth step of determining a location of a failure based on the correlation of the cause of the corresponding failure on the failure cause / result graph. The method of claim 1, wherein the failure is a trap occurring at a physical layer, a data link layer, a network layer, and a service layer. The method of claim 1 or 2, wherein the traps generated in the corresponding layer are transmitted to upper layers. The method of claim 1, wherein the fourth step comprises a failure cause / result graph for each layer and a failure cause / result graph for each layer in each layer. 2. The hierarchical failure management method of an intranet according to claim 1, wherein the cause of failure transmitted from each layer is transmitted as an alarm representing an upper correlation according to the priority of the cause / effect of the failure. The method of claim 1, wherein the sixth step comprises: searching for the cause of the alarm representing the correlation in the transmission of the failure occurrence alarm; Retrieving all the causes of the disorder having a cause / effect relationship on the cause / effect graph of the failure when there is no delivery of the alarm representative of the correlation; And And repeating the steps to determine the location of the actual failure.