JPH03144831A - System recovery method - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] Industrial Field of Application The present invention (Dai) relates to a system recovery method for automatic recovery in the event of a system failure.
In many cases, restoration work was carried out by the specialist based on the specialist's own knowledge. However, with this method, whenever a failure occurs, at least one specialist is always on hand to carry out restoration work. In addition, if there is no one on duty, there is the problem of having to wait for recovery work until an expert arrives.So now, with the development of computers, failure diagnosis methods using computers are becoming more popular. An expert system is used for fault diagnosis using the system, which stores an expert's knowledge base on a computer in advance, and when a fault occurs in the system, this knowledge base is used to determine the cause of the fault from the results obtained. This is a method in which the user recovers from a failure according to the system's instructions.

Problems to be Solved by the Invention In this conventional recovery work method using a failure diagnosis expert system, the system asks the user questions as necessary, so the user can freely input the observed situation. In addition, the order of questions depends on the order in which the rules were written, so it is not necessarily in a natural order, and in this situation human labor becomes a burden on the user. The present invention has been devised in view of these points, and has a simple configuration. Although the purpose of the present invention is to provide a system recovery method that can automatically perform failure recovery work without putting a burden on the user, the present invention is a method for solving the problem. As empirical rules, there is a knowledge base section that stores recovery work procedures in the event of a failure as recovery rules, an operation observation section that constantly observes the operating status of the system, an empirical rule that is stored in the knowledge base section, and A system comprising a failure determination unit that compares the results observed by the operation observation unit to determine whether there is a contradiction, and a recovery unit that performs failure recovery work using recovery rules stored in the knowledge base unit. This is a recovery method.

Function: With the above-described configuration, the present invention analyzes the cause of failure and automatically performs recovery work without human intervention, thereby facilitating system maintenance.

EXAMPLES Below, the present invention will be described in detail with reference to the drawings.
In the functional block diagram of the system recovery method of the present invention, knowledge base section IHL stores system-dependent observations as experience rules 12 (knowledge), that is, the normal operation of the system as experience.

The knowledge base unit 11 (top) stores the recovery procedure in the event of a failure as the recovery rule 13. The operation observation unit 14 constantly observes the operation, and the observed operation is stored as the experience rule 12 of the knowledge base unit 11. , the failure determination unit 15 successively compares and recognizes whether or not there is an abnormal operation (failure).If it is recognized as a failure, ('L)
Based on this, the recovery unit 16 connected to the operation observation unit 22 through a dedicated connector performs recovery work. FIG. 2 is a configuration diagram showing a system recovery method in an embodiment of the present invention.

Operation observation unit 22 (upper) User 21 who uses the system
This is the part that observes the Q situation and behavior when using the system.

The observation 26 obtained by the operation observation unit 22 (Y) is compared with the operation principle 27 of the sequential knowledge base unit 33 by the failure determination unit 25, and if the observation 26 does not match the operation principle 27,
The operation is transferred to the recovery unit 24.

For example, when the user 21 is using the system, the operation observation unit 22 constantly observes the system status and operation. In observation 26, we obtained a result A for the observation item X and a result D for the observation item Y. Then, in the operating principle 27 of the knowledge base section 23 (above) for the item X of the operating principle 27. The knowledge that A-C is normal is obtained, and the knowledge that Z is obtained for item Y.Then, when this observation result is compared with the knowledge in the operating principle 27 in the failure judgment unit 25, The result A for observation item
Observation result D for observation item Y (Operating principle 2
7 is inconsistent with the knowledge that item Y is Z (because of this event, the operation regarding observation item Y is recognized as a failure operation. According to Rule 28, if the result Z for Y is not obtained (Y), the recovery unit 24 is requested to perform operation a and perform recovery work, but when the recovery work is completed (reluctant), the user 21 is notified of the completion of recovery. A specific example of the operation of the LSI mask layout editor system will be explained using FIG. 3. When the user 33 starts up the layout editor system, the operating principle 341 of the knowledge base unit 35; Although the four processes registered as processes at system startup are generated, these processes (a) remain in existence until the layout editor system is terminated.
The operating status of these processes is observed over a certain period of time. This operation observation result 38 is compared with the operating principle 34 of the system during normal operation stored in the knowledge base unit in a constant failure determination unit 39 to determine whether the current operating status of the system is normal or abnormal. If the operation observation unit 36 observes that the four processes have terminated abnormally due to some reason, the operation regarding this layout editor system is recognized as a failure operation, and the operation observation unit 36 observes that the four processes have terminated abnormally. Another knowledge of the knowledge base unit 35 is how to restore the process that has been restored.Recovery rules 37 are determined in order to restore the system to normality.The recovery unit 38 automatically performs the recovery work in accordance with the recovery rules 37.

Effects of the Invention As described above, the present invention provides an automatic recovery method that can automatically recover from failures by interfacing the system used by the user with the knowledge base. extremely useful

[Brief explanation of the drawing]

Figure 1 shows functional block blocks of the system recovery method of the present invention.
FIG. 2 is a configuration diagram showing a system recovery method in an embodiment of the present invention. FIG. 3 is a configuration diagram of a specific example of the present invention.

Claims (2)

[Claims] (1) A knowledge base unit that stores the operation of the system during normal times as empirical rules and recovery work procedures in the event of a failure as recovery rules, an operation observation unit that constantly observes the operating status of the system, and the knowledge base unit a failure determination unit that compares the empirical rules stored in the memory with the results observed by the operation observation unit and determines whether there is a contradiction; and a failure determination unit that performs failure recovery using the recovery rules stored in the knowledge base unit. A system recovery method comprising a recovery section that performs the work. (2) The system recovery method according to claim 1, characterized in that the operation observation section and the failure determination section are connected by a dedicated connector.