JPS6344254A - Holding and managing system for history of generation of fault - Google Patents

Abstract

PURPOSE:To exactly grasp the generation of a fault, by performing the preservation and the management of the history of the generation of the fault, corresponding to contents stored in an accumulation counter, a fluctuation counter, and a device manufacturing number holding register. CONSTITUTION:The accumulation counter 4 which counts the number of fault, and the fluctuation counter 4, are provided in a processor 1n having the device manufacturing number holding register 3 in which a device manufacturing number is stored. For example, when the fault is generated in a processor 1a, a monitoring device 2 adds 1 on the value of the accumulation counter 4 and that of the fluctuation counter 5, and reads and keeps the renewed values of the accumulation counter 4, and the fluctuation counter 5, and the device manufacturing number of the processor 1a stored in the device manufacturing number holding register 3.

Description

[Detailed Description of the Invention] [1 Already Required] The number of failures that have occurred since the processing devices 13 to 13 started operating is counted in each processing device 1a''In having a device manufacturing number holding register 3 that stores the device manufacturing number. A cumulative counter 4 and a fluctuation counter 5 that counts the number of failures that occur after the component parts of the processing devices 1a to 1n are replaced.
are provided, respectively, and each of the processing devices 1a to 1n is connected to the monitoring device 2.

The monitoring device 2 that has received the failure occurrence signal of the processing devices 1a to 1n increments the counts of the cumulative counter 4 and the fluctuation counter 5 by 1, and retains the updated value of the cumulative counter 4, the value of the fluctuation counter 5, and the device serial number. Read and save the device serial number stored in register 3.

[Industrial application field]

The present invention relates to a method for maintaining and managing a failure history of an information processing device.

[Conventional technology]

When maintaining information processing equipment, maintaining and managing its failure history is extremely necessary to ensure good operation.

Recently, the reliability of information processing devices has become higher and higher as the quality of electronic components has improved, and as a result, the majority of failures have shifted from being fixed to being intermittent. That is, it has become extremely difficult to reproduce failures that occur during operation at a test site.

For this reason, effective maintenance management measures are needed.

[Problem that the invention seeks to solve]

However, since the conventional failure occurrence history maintenance and management method is performed manually, there is a drawback that the number of failure occurrences may not be accurately known.

[Means for solving the problem] As shown in the principle diagram of FIG. A cumulative counter 4 that is reset before 1n starts operating and counts the number of failures that will occur thereafter, and a variable counter that is reset when the component parts of the processing devices 1a to 1n are replaced and counts the number of failures that will occur thereafter. 5, and the processing apparatuses 1a to 1a.
1n to the monitoring device 2, respectively.

[Effect]

According to the present invention, for example, when a failure occurs in the processing device 1a, the monitoring device 2 increments the value of the cumulative counter 4 and the value of the fluctuation counter 5 by 1, and the updated value of the cumulative counter 4 and the value of the fluctuation counter 5. , and the device serial number of the processing device 1a stored in the device serial number holding register 3 are read and stored, making it easy to conduct a follow-up investigation for each processing device 1a to ln using the built-in data processing function of the monitoring bag W2. becomes possible.

〔Example〕

FIG. 2 is a diagram showing an embodiment of a failure occurrence history retention/management method according to the present invention.

In the figure, 1a to 1n are processing units, 2 is a monitoring device, 3 is a device serial number holding register, 4 is an accumulation counter, 5 is a fluctuation counter, 6 is an external auxiliary recording device, 7 is a system console, and 8 is a communication line. , 9 is a remote support center device.

The devices to be maintained and managed now are the processing devices 1a to 1n, and a monitoring device 2 is provided for the processing devices 13 to 1n in order to facilitate the maintenance and management of these processing devices 1a to In.

As is well known, recently, with the advent of high-density LSIs, one device is often composed of one board.

There are multiple sets of systems consisting of a plurality of processing devices 1a to 1n and monitoring devices 2 like this, and each of these systems is installed in a distributed manner throughout the country, and the monitoring devices 2 of each system are It is connected to a remote support center device 9 that serves as a maintenance center via a communication line vA8.

In addition, in each of the processing devices 1a to ln, a device serial number holding register 3, a cumulative counter 4,
and a fluctuation counter 5 are provided.

The device serial number holding register 3 is the processing device 1a to which it belongs.
The device serial number of In is stored, and this device serial number is output to the monitoring device 2 in response to a command from the monitoring device 2.

Further, the cumulative counter 4 is a counter that counts the number of failures that have occurred in the processing devices 1a to 1in, and is initialized when the processing device 1 is shipped.

Furthermore, the fluctuation counter 5 is a component of the processing devices 1a to 1n (
It is initialized when the LSI (mainly LSI) is replaced.

It is assumed that the contents of the device serial number holding register 3, the cumulative counter 4, and the fluctuation counter 5 are retained even if the processing devices 13 to 1n are powered off. For this reason, for example, a mechanical storage medium is used for the device serial number holding register 3, and for example, a dry cell-powered RAM or the like is used for the cumulative counter 4 and the fluctuation counter 5.

According to the present invention, if a fault occurs in the processing device 1a, the processing device 1a issues a fault occurrence signal to the monitoring device 2.

The monitoring device 2 that received this failure occurrence signal is the processing device 1
Add 1 to the values of the cumulative counter 4 and fluctuation rank 5 of a, and add the updated value of the cumulative counter 4, the updated value of the fluctuation counter 5, and the device serial number holding register 3.
The value of is sent to the monitoring device 2, and this data is stored in the external auxiliary recording device 6.

Furthermore, when the value of the cumulative counter 4 written in the external auxiliary recording device 6 and the value of the fluctuation counter 5 reach the specified values, an alarm is issued and displayed on the system console 7. The maintenance personnel can also manually write data related to the occurrence of a failure into the external auxiliary recording device 6 using the keyboard of the system console 7.

In addition, the monitoring device 2 remotely records the value of the cumulative counter 4, the value of the fluctuation counter 5, fault information, etc. written in the external auxiliary recording device 6, based on a request from the remote support center device 9 or voluntarily. Since the support center device 9 can be notified, the remote support center device 9 can perform real-time management.

Note that the cumulative counter 4 is reset before the processing devices 1a to 1n start operating, and the fluctuation counter 5 is reset to the processing devices 1a to 1n.
Although it is reset when the component 1n is replaced, it is also possible to perform this reset by the monitoring device 2.

〔Effect of the invention〕

As described in detail above, according to the present invention, the monitoring device 2 uses the built-in data processing function to calculate the value of the cumulative counter 4 and the value of the fluctuation counter 5 of the processing devices 1a to 1n under its control, using the device serial number as an index. , failure information, etc. can be organized and maintained, and the tracking of the processing devices 1a to 1n can be easily conducted.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the principle of the present invention. FIG. 2 is a diagram showing an embodiment of a fault occurrence signal retention and management method according to the present invention. In the figure, 1a to 1n are processing units, 2 is a monitoring device, 3 is a device serial number holding register, 4 is an accumulation counter, 5 is a fluctuation counter, 6 is an external auxiliary recording device, 7 is a system console, and 8 is a communication line. , 9 is a remote support center device. Kyorō no Chikikiri ■ Te 1 San

Claims (1)

[Claims] Device manufacturing number holding register (3
) to each processing device (1a to 1n) having the processing device (
A cumulative counter (4) that is reset before the processing device (1a to 1n) starts operating and counts the number of failures that occur thereafter; and a cumulative counter (4) that is reset when the component parts of the processing device (1a to 1n) are replaced and A variable counter (5) is provided to count the number of failures that occur, and each of the processing devices (1a to 1n) is connected to a monitoring device (2), and each of the processing devices (1a to 1n) is connected to a monitoring device (2) for counting the number of failures. The monitoring device (2) adds 1 to the count values of the cumulative counter (4) and the fluctuation counter (5), and updates the updated value of the cumulative counter (4), the updated value of the fluctuation counter (5), and the device. A fault occurrence history retention/management method characterized by reading and storing the device serial number stored in the serial number holding register (3).