KR20020065188A - Method for managing fault in computer system - Google Patents

Abstract

PURPOSE: A method for managing a fault of a computer system is provided to monitor the fault occurrence of the computer system and report the data for the occurred fault to an upper system. CONSTITUTION: An application program requests a system service to an operating system and receives the return value. The application program calls a fault management program by confirming the return value. The application program transfers the kind of the fault and the related value to the fault management program. The fault management program confirms the kinds of the fault and collects the fault data(S210,S220). After completing the data collection, the fault management program configures a local database in a preset fault management storing region(S230) and reports the fault occurrence to the upper system by transmitting the fault data(S240). The fault data is stored in the local DB(S250) and the fault management program carries out the preset fault recovery operation(S260).

Description

METHODS FOR MANAGING FAULT IN COMPUTER SYSTEM}

The present invention relates to a computer system, and more particularly to a method for managing a failure occurring in a computer system.

As is known, a computer system consists of a plurality of processor boards for performing unique functions, and the processor boards are operated by an operating system (OS) program and an operating system program that controls the operation of the corresponding processor board. It has at least one application program and additional programs that perform various functions.

1 shows a program data storage area of a typical computer system. Referring to FIG. 1, the operating system program 110 executes at least one application program 120, a middleware program 130, a device driver 140, a hardware controller 150, and the like.

When an error occurs on the basis of an error occurrence condition predetermined externally or internally in the computer system configured as described above, corresponding programs determine that a failure is detected and perform a predetermined failure management operation. Referring to the program-specific failure management operation according to the prior art as follows.

When a failure occurs in the application program 120, the operating system program 110 sets the return value to the application program to a value indicating the failure. If the return value received from the operating system program 110 is a value indicating the failure, the application program 120 processes itself according to the type of the failure.

The middleware program 130 also checks the return value received from the operating system program 110 and, if a failure is detected, prints out the contents of the corresponding failure. For example, if the send message is not sent due to insufficient buffer capacity, the printer prints out the error message "Tx Error (Buffer Full)" and increases the transmission error count.

Like the middleware program 130, the device driver 140 detects a failure by a return value and prints out the contents of the corresponding failure by itself. If the hardware controller 150 detects a failure in the corresponding hardware, the hardware controller 150 manages the failure. If an interrupt occurs as an exception, the operating system program 110 stores the register value and the contents of the stack.

As described above, the prior art does not process or database a failure generated in a computer system, and the corresponding failure data cannot be confirmed in a higher system. Therefore, it is impossible to take action on various errors and failures that occur in the online situation, and it is impossible to solve them when a problem occurs. In other words, it is difficult to identify and solve the cause of the problem because the information on various failures occurring in the computer system is not reported to the upper system. Therefore, there was a problem that the contents of the obstacles could not be stored and their types could not be stored systematically.

Accordingly, an object of the present invention, which was devised to solve the problems of the prior art operating as described above, provides a method of monitoring occurrence of a failure in a computer system and reporting data on the generated failure to a higher system. will be.

Embodiments of the present invention, which are invented to achieve the above object, are provided in an error management method of a computer system including an operating system program, an application program executed by the operating system program, a middleware program, a device driver, and a hardware controller. ,

When a failure occurs in any one of an application program, a middleware program, a device driver, and a hardware controller, calling a failure management program from a corresponding program and transmitting information on the generated failure to the failure management program;

Confirming the type of the generated failure in the failure management program;

Collecting failure data corresponding to the generated failure;

Configuring a local database in a pre-allocated fault data storage area;

Reporting the fault data to a higher system;

Storing the fault data in the local database;

And performing a failover operation corresponding to the generated failure.

1 is a diagram showing a program data storage area of a conventional computer system.

2 illustrates a program storage area of a computer system according to the present invention.

Figure 3 is a flow chart illustrating a failure management operation in the software field according to the present invention.

Figure 4 is a flow chart showing the operation of the problem management program according to the present invention.

Hereinafter, with reference to the accompanying drawings will be described in detail the operating principle of the preferred embodiment of the present invention. Like reference numerals are used to designate like elements even though they are shown in different drawings, and detailed descriptions of related well-known functions or configurations are not required to describe the present invention. If it is determined that it can be blurred, the detailed description thereof will be omitted. Terms to be described later are terms defined in consideration of functions in the present invention, and may be changed according to intentions or customs of users or operators. Therefore, the definition should be made based on the contents throughout the specification.

2 illustrates a program storage area of a computer system according to the present invention. Referring to FIG. 2, the operating system program 210 executes an application program 220, a middleware program 230, a device driver 240, a hardware controller 250, and a problem management program 260. In particular, the application program 220, the middleware program 230, the device driver 240, the hardware controller 250 may call the failure management program 260 and the failure management program 260 is assigned for failure management. Fault data storage can be accessed.

When an error occurs on the basis of an error occurrence condition predetermined externally or internally in the computer system configured as described above, the programs determine that the failure is detected, call the failure management program, and then cause the failure caused by the failure management program. It will pass information about. The failure management operation for each program according to the present invention will be described in detail as follows.

The application program 220, the middleware program 230, and the device driver 240, which are software fields, check the return value received from the operating system program 210 and, if the return value indicates a fault, execute the fault management program 260. It is programmed to call.

3 is a flowchart illustrating a failure management operation in a software field according to the present invention. Referring to FIG. 3, in operation S110, an application program 220 (or a middleware program or a device program) may provide a system service to an operating system program 210. Request and receive a return value in response. In operation S120, the application program 220 (or the middleware program or the device program) checks whether the received return value is a failure notification value, and if it is a failure notification value, calls the failure management program in operation S130. . In this case, the application program 220 (or the middleware program or the device program) transfers the type of failure and its value to the failure management program.

A case of the middleware program 230 in the software field operating as described above will be described below. The resources of the system processed by the middleware, especially when there are no available resources in the IPC (Inter-Processor Communication) receiving and sending buffers, timers, redundancy, etc. In this case, the operating system program 210 sets the return value to a value indicating the failure. For example, if the send message could not be sent due to insufficient buffer capacity, the return value is -1 and the type of failure is set to 10. When the middleware program 230 determines that a failure has occurred by checking the return value, the middleware program 230 calls the failure management program 260 and transmits the type of the corresponding failure and the value of the failure to the failure management program 260.

In addition, the device driver 140 checks the return value in the same manner as the middleware program 130 and calls the failure management program 260 when detecting a failure. In this case, in order for the operating system program 210 to set a return value related to the device driver 140, a failure type related to an error such as an I / O needs to be added with respect to the device driver 140.

Unlike the software field, in the hardware field, the hardware controller 150 determines that a failure has occurred if the value exceeds the predetermined range or the read / write operation is not performed normally as a result of measuring the output range of the corresponding hardware. The management program 260 is called and the location of the corresponding failure and the difference value are transmitted to the failure management program 260.

When an interrupt occurs as an exception, the operating system program 110 stores the contents of the program counter and the fault address register and notifies the fault management program 260 of its location.

When the failure management program 260 is called as described above, the failure management program 260 performs a failure management operation according to the present invention.

4 is a flowchart illustrating the operation of the failure management program 260 according to the present invention. Referring to FIG. 4, the failure management program 260 confirms the type of failure based on the information received from each program in step S210, and collects corresponding failure data in step S220. Fault data is a fault management requirement that is predetermined by the operator, such as the location of the faulty software or hardware and the values of registers and stacks. For example, in the case of a fault in the software field, the fault data is the ID, program count, error value, etc. For a fault in the hardware field, the fault data is the task ID, program count, error type, error value, CPU register values. The fault data is generated in a predetermined format and can be output by a debug command.

When the collection of the failure data is completed in step S230, the failure management program 260 configures a local database in a failure management storage area pre-allocated for failure management. The failure management storage area is allocated at system initialization and cannot be accessed by other programs.

In step S240, the failure management program 260 transmits the failure data to an upper system to report that a failure has occurred. Then, the host system stores the fault data, and the operator analyzes the fault data stored in the host system to immediately identify the cause of the fault and solve the problem.

In step S250, the fault data is stored in a local database configured in step S230. Then, in step S260, the failure management program 260 performs a corresponding predetermined failure recovery operation. To recover from the failure, an operation such as restarting or rebooting the system or deleting a corresponding task may be performed.

Meanwhile, in the detailed description of the present invention, specific embodiments have been described, but various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be defined not only by the scope of the following claims, but also by those equivalent to the scope of the claims.

In the present invention operating as described in detail above, the effects obtained by the representative ones of the disclosed inventions will be briefly described as follows.

In the present invention, in the case of a system down or abnormal operation occurring in an operating system, the problem may be easily determined by using fault data reported to a higher system of the corresponding system and fault data stored in a corresponding processor board. It can shorten the resolution time and solve the problem fundamentally. In addition, even if a failure occurs seriously and needs immediate action, there is an effect that immediate action is possible in the field depending on the type of failure.

Claims (4)

In the failure management method of a computer system comprising an operating system program, an application program executed by the operating system program, a middleware program, a device driver and a hardware controller, If a failure occurs in any one of an application program, a middleware program, a device driver, and a hardware controller, calling a failure management program from a corresponding program and transferring information on the generated failure to the failure management program; Confirming the type of the generated failure in the failure management program; Collecting failure data corresponding to the generated failure; Configuring a local database in a pre-allocated fault data storage area; Reporting the fault data to a higher system; Storing the fault data in the local database; And performing a failover operation corresponding to the generated failure. The method of claim 1, wherein the failure data includes a task ID, a program count, a type of failure, an error value, a register value, and a stack value. The method of claim 1, wherein the failover operation is a restart, reboot, or deletion of a failed task. The method of claim 1, wherein the application program, the middleware program, and the device driver identify a return value received from the operating system program in response to a system service request, and determine that a failure has occurred if the return value indicates a failure. How to.