JPH02159636A - Network fault diagnostic system - Google Patents

Abstract

PURPOSE:To effectively diagnose a fault generated in a network by hypothesizing the cause of a fault, verifying its hypothesis, hypothesizing a more detailed cause investigation and repeating a processing for instructing the inspection of information. CONSTITUTION:The diagnostic system is provided with an inference part 1 for searching the cause of those faults against a network fault, and also, hypothesizing the cause of the fault in order to be restored, instructing the inspection of network information to a state inspection part 3 in order to verify its hypothesis, hypothesizing a more detailed cause investigation, based on the result of inspection obtained by its state inspecting part 3 and repeating a processing for instructing the inspection of the information, determining the cause as its result, and notifying a restoring method corresponding to the cause. Also, this system is provided with the state inspection part 3 for deciding whether its collected information coincides with the using condition of each equipment determined in advance or not and sending the result of its decision to the inference part 1. In such a manner, the fault of the network can be diagnosed effectively.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a network fault diagnosis method, and particularly to a network fault diagnosis method for early detection, diagnosis, investigation of the cause, and quick recovery from a network fault.

[Conventional technology]

The use of knowledge processing is known as an effective method for reflecting human thinking in computer processing. Even when diagnosing network faults, the conventional methods used by maintenance personnel are highly specialized and follow rational procedures. Therefore, when converting such expert procedures into a device, it is necessary to utilize knowledge processing. Although there are some examples of devices in which knowledge processing is utilized for diagnosis in medical diagnostic systems, I have not yet seen one in which it is utilized for network fault diagnosis, which is the object of the present invention.

[Problem to be solved by the invention]

Recent trends in the construction of networks for computers and communication equipment include wide-area decentralization and large-scale expansion. Accordingly, when a failure occurs, the loss inevitably tends to increase. At the same time, network operation and maintenance become more complex and costs increase. Therefore, there is a need for a method to effectively diagnose network failures.

The purpose of the present invention is to effectively diagnose network failures in a large-scale, increasingly complex network system environment using knowledge processing in a manner similar to that of experts, so that users can use the system with peace of mind. An object of the present invention is to provide a network fault diagnosis method that can effectively diagnose faults.

[Means to solve the problem]

The configuration of the network fault diagnosis method of the present invention is such that in a network for the purpose of electrically transmitting information consisting of computers, terminal equipment, exchanges, and communication equipment, faults occur in the equipment and transmission paths that make up the network. Malfunction of network equipment due to physical failure, incorrect operation by the user of the network equipment, malfunction of the computer program created to use the network equipment, or failure of the equipment due to improper setting of parameter values for operating various equipment. In order to investigate the cause of network failures caused by malfunctions and perform recovery, we formulate a hypothesis of the cause of the failure.
In order to verify the hypothesis, the status inspection unit is instructed to inspect the network information, and based on the inspection results obtained by the status inspection unit, a process is performed to formulate a hypothesis for further detailed cause investigation and instruct the inspection of the information. Iteratively determines the cause as a result,
An inference section that posts recovery methods according to the cause, and a system that collects the network information according to the instructions of this inference section, determines whether the collected information matches the predetermined usage conditions of each device, and The present invention is characterized in that it includes a state checking section that sends a determination result to an inference section.

〔Example〕

Embodiments of the present invention will be described in detail using the drawings.

FIG. 1 is a diagram showing an embodiment of the network fault diagnosis method of the present invention. In FIG. 1, an inference unit 1 uses knowledge used in fault diagnosis in an inference knowledge base 2 to formulate a strategy for investigating the cause. The inference knowledge base 2 records knowledge regarding network failures, and the state in which a certain device has fallen, or the test result of network information obtained as a test result in the state inspection unit 3, is recorded as the fault state of each device. are associated. The failure states of each device are hierarchically related according to the type and level of detail of confirmed data and inspection results, and once a certain failure state is confirmed, more detailed possibilities may be developed to get closer to investigating the cause. Several types of failure conditions are posted.

In order to check possible fault conditions posted in the inference knowledge base 2, the inference section 1 instructs the state inspection section 3 on how to test the fault conditions by referring to the inference knowledge base 2;
Let it run. In other words, the inference knowledge base 2 can be thought of as a collection of data cards that define their failure states according to the content of collected information. These cards are arranged in a hierarchical manner from the rough stages of collected information to the detailed stages, and as the data changes, several failure states are related from one failure state to the next failure state. It is attached. When the cause of a failure is identified, it is associated with a recovery method.

The state inspection section 3 collects network information according to instructions from the inference section 1. Collection methods include asking users about their status, inspecting line traces, and checking parameters necessary for device usage, such as system configuration definitions. Each collected data is subject to the usage conditions recorded in the inspection knowledge base 4 (parameters, line trace logic regulations,
operation method, etc.), confirm whether it is correct or not, and send the result to the inference section 1.

FIG. 2 is a diagram showing the operation details of each function of the present invention.

In FIG. 2, in operation A, a report of the occurrence of a failure is sent to the reasoning unit 1 from the user 5- or the target network device 6 through the status inspection unit 3. In operation B, the inference section 1 instructs the state inspection section 3 to conduct an initial interview with the user 5 or the target network device 6. In operation C, the condition inspection unit 3 performs an initial interview according to the instructions. In operation, the user 5 and the target network device 6 send their answers to the questions through the status checking section 3 to the reasoning section '1. In operation E, the inference unit 1 hypothesizes the failure state and instructs the status inspection unit 3 to confirm the status.In operation F, the status inspection unit 3 hypothesizes the failure state Relevant data is collected from the network equipment 6 and inspected.

In operation G, the state inspection section 3 sends the inspection result to the inference section 1. In operation H, when the inference unit 1 finds the cause and confirms the recovery method, the status inspection unit 3 notifies the user 5 or the target network device 6 of the recovery method.

FIG. 3 shows a flowchart of the diagnostic procedure of the present invention.

The procedure is divided into an initial interview in which questions are asked to roughly confirm the status of the failure, cause investigation in which hypotheses are repeatedly generated and verified to determine the cause, and recovery.

In the initial interview, in the diagnostic procedures P1 and P2, the second
As explained in the diagram, information is collected from users and target network devices, and this information is used to roughly classify failures.

In the cause investigation, the cause is investigated by repeatedly generating and verifying hypotheses in diagnostic steps P3 to P6.

In the recovery, in the diagnosis step P7, the contents of the inference knowledge base used by the inference section are referred to, and a recovery method for the cause is posted to the user and the target network device.

As described in 2, the present invention provides network fault diagnosis that enables effective fault recovery by investigating the cause of faults that occur in a network using processing similar to the method that experts would take. It is a method.

〔Effect of the invention〕

As explained above, the network fault diagnosis method of the present invention has the advantage of automatically performing fault recovery.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a diagram showing the operation details of each function of the invention, and FIG. 3 is a flowchart showing the diagnostic procedure of the invention. A, B,..., H...Operation content, PL, P2.・・・
・P7...Diagnosis procedure, 1...Inference section, 2...Inference knowledge base, 3...Status inspection section, 4...Testing knowledge base, 5...User, 6... Target network equipment.

Claims (1)

[Claims] In a network for the purpose of electrically transmitting information consisting of computers, terminal equipment, switching equipment, and communication equipment, physical failures that occur in the equipment or transmission paths that make up the network, or caused by the users of the network equipment. Network failures resulting from malfunctions of network equipment due to incorrect operation, defects in computer programs created to use said network equipment, or malfunctions of equipment due to incorrect setting of parameter values for operating various equipment, In order to investigate the cause of a failure and perform recovery, we formulate a hypothesis of the cause of the failure, instruct the status inspection unit to inspect network information to verify that hypothesis, and use the inspection results obtained by the status inspection unit. Based on this, a reasoning unit repeats the process of forming a hypothesis for a more detailed cause investigation and instructing the inspection of information, and as a result, determines the cause and posts a recovery method according to the cause;
and a status inspection unit that collects the network information according to instructions from the inference unit, determines whether the collected information matches predetermined usage conditions for each device, and sends the determination result to the inference unit. A network fault diagnosis method characterized by: