JPH0343841A - Fault information store system - Google Patents

Abstract

PURPOSE:To obtain the satisfactory fault information for a restoring process via the next switch even though the fault information is not obtained with a switch applied first by storing and taking out the fault information via a control means in accordance with the state of a flag means. CONSTITUTION:It is shown whether the fault information is stored in a store means 111 or not according to the state of a flag means 112. Thus the control is carried out so that the fault information is stored in the means 111 only when no fault information is stored in the means 111. As a result, the contents of the fault information stored once in the means 111 are never changed as long as they are not read to outside. Thus it is possible to obtain the fault information at occurrence of the first fault via the next switch means as long as the fault information is stored in the means 111 even though the information is not obtained with a switch means S applied first.

Description

[Detailed Description of the Invention] [Objective of the Invention] (Industrial Application Field) The present invention relates to a fault information storage method, and particularly to a fault information storage method for a computer system having a plurality of switches for instructing collection of fault information. (Prior Art) Generally, in a computer system, a system failure (stall) may occur, for example, in which transmission from a terminal becomes impossible. When such a system failure occurs, in order to execute recovery processing, it is necessary to recognize what kind of program status the system was in at the time of the failure. Therefore, when a failure occurs, processing is executed to collect information at the time of the failure.

In conventional systems, a plurality of manual switches for instructing collection of fault information are provided in the system body, and when a fault occurs, the first switch is first turned on by the operator. Then, failure information when the first switch is turned on is collected and stored in the internal memory. Thereafter, the fault information is retrieved from the internal memory and analyzed.

However, if a failure occurs again at the stage of retrieving the failure information from the internal memory, the failure information cannot be extracted and analysis of the failure information becomes impossible. In this case,
The second switch is turned on and failure information is newly collected. However, the fault information obtained by this new process is not the first fault information when the first switch is turned on, but a different fault when the second switch is turned on, that is, when the information collected by the first switch is retrieved to the outside. It is information.

In order to perform failure recovery processing, it is important to analyze the first failure information, and it is difficult to perform sufficient information analysis with the failure information obtained by the second switch. For this reason, in the past, if fault information was not obtained by the first switch turned on, there was a drawback that sufficient fault information for recovery processing could not be obtained.

(Problem 8 to be solved by the invention) Conventionally, even if multiple manual switches are provided to instruct the collection of fault information, if the fault information cannot be obtained by the first switch turned on, the recovery process cannot be completed. The drawback was that sufficient failure information could not be obtained.

This invention has been made in view of the above points, and it is possible to obtain fault information that is sufficient for recovery processing by the next switch even if fault information cannot be obtained by the first switch turned on. The purpose is to provide a preservation method.

[Structure of the Invention] (Means for Solving the Problems) The fault information storage method according to the present invention includes a plurality of switch means for instructing collection of fault information when a system fault occurs, and for storing the fault information. storage means for;
a flag means that is set when the fault information is stored in the storage means and reset when it is not stored;
information collecting means for collecting fault information and storing it in the storage means and setting the flag means in a set state; information retrieval means for reading information from the storage means to the outside and setting the flag means in a reset state; and information retrieval means for reading out information from the storage means and setting the flag means in a reset state; and determining whether the flag means is in a set state or a reset state when any one of the plurality of switch means is turned on. control means for determining and causing the processing by the information collecting means and the processing by the information retrieval means to be sequentially executed when the flag means is in a reset state, and for causing the processing by the information retrieval means to be executed when the flag means is in a set state; It is characterized by comprising:

(Operation) In this fault information storage method, the state of the flag means indicates whether or not fault information is stored in the storage means, so only when fault information is not stored is the fault information stored in the storage means. Can be controlled to store. Therefore, the contents of the failure information once stored in the storage means are not changed unless read out to the outside. Therefore, even if fault information is not obtained by the first switch turned on, as long as the fault information is stored in the storage means at that time, the next switch turned on will be able to detect the fault at the time of the first failure. You can get information.

(Example) Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 shows a system configuration for realizing a fault information storage method according to an embodiment of the present invention.

The host CPU II is in charge of controlling the entire system, and is connected to the terminal devices 12a and 12b via the bus.

12c. These terminal devices 12a.

12b and 12c each execute data processing and exchange data between the host computer 11 and these terminal devices. An external memory 13 is further connected to the host CPU II. This external memory 13 consists of, for example, a floppy disk device or a hard disk device.

The host CPU II is provided with three manual switches 5l-83 for instructing collection of failure information and recovery when a system failure occurs. Also,
The host CPU II has RAM for storing failure information.
ll1. and a flip-flop 112 for indicating whether fault information is stored in the RAMtti. The flip-flop 112 is controlled to be in a set state when fault information is stored in RAM III and to be in a reset state when fault information is not stored. Functions as a flag to indicate.

When one of the manual switches Sl to S3 is turned on, the host CPU II collects, for example, the program status at that time, the contents of a register inside the CPU, etc. as failure information, and stores it in the RAMI. in this case,
The host CPU II sets the flip-flop 112 to the set state. Then, the failure information is stored in RAM lit.
The data is read from the external memory 13 and written to the external memory 13, and the flip-flop 112 is set to the reset state.

The process of reading out failure information from the RAM 111 by the host CPU II is executed in a form that corresponds to the switch that is turned on. For example, when the switch Sl is turned on, the host CPU II reads failure information from the RAM ill using a program specific to the system while the failure has occurred. Furthermore, when the switch S2 is turned on, the host CPU II reads another dedicated program from the external memory 13,
The fault information is read from the RAM 111 using the program.

Further, when the switch S2 is turned on, the host CPU II reads the fault information from the RAM 111 while the system is restarted.

In this way, the storage process and the retrieval process of the failure information are executed, and whether or not the storage process of the failure information by the host CPU II is executed is controlled depending on the state of the flip-flop 11=8-. . That is, when the flip-flop 11 is in the set state, when the host CPU II is storing the fault information, the host CPU II executes both the storing process and the retrieving process of the fault information.

Next, with reference to the flowchart of FIG. 2, the procedure of processing operations performed by the host CPU II when the switch is turned on will be described.

For example, when a system failure occurs that makes it impossible to receive data from the terminal 12a, the operator first turns on the switch Sl. By turning on the switch Sl, the host CPU II executes interrupt processing for storing and retrieving fault information.

When transitioning to interrupt processing, first, the state of the free knob flop 112 is determined by the host CPU II (step AI). In the interrupt processing caused by turning on the first switch St, the fault information is not yet stored in the RAM 111, so the flip-flop 11 at this time
2 is a reset state. Therefore, the process branches to step A2, and the failure information at that time, such as the program status and the contents of the CPU's internal registers, is sent to the host cput.
i and stores it in RAM ill. In this case, flip-flop 112 is connected to host C
Set to set state by PUII.

Next, the host CPU II recognizes that the turned-on switch is the switch S1 (step A3), reads the fault information from the RAM 111 in a processing mode according to the switch St, and writes it to the external memory 13 (step A3). A4).

If the process of step A4 is executed normally, fault information will be obtained, but if an error occurs during the process of step A4, no fault information will be obtained. In this case, the second switch S2 is turned on by the operator. By turning on the switch S2, the host cptrtt returns to step AI again to execute interrupt processing.

First, the state of the flip-flop 112 is
It is determined by it (step AI), but since the flip-flop 112 is in the set state at this time, the process of step A2 is not executed. Then, it is recognized that the turned-on switch is switch S2 (step A
5) RAM in the processing mode according to the switch S2
The failure information is read from the external memory 111 and written to the external memory 13 (step Aft).

If the process of step A6 is executed normally, failure information is obtained. This obtained fault information is not the fault information at the stage of turning on the switch S2, but the fault information at the stage of turning on the first switch Sl.

Further, if the process of step A6 is not executed normally, switch S3 is turned on, and in the same way, step A
7. The process of A8 is executed. If the process of step A8 is executed normally, failure information at the initial turning-on stage of switch Sl is obtained.

In this way, in the fault information storage method of this embodiment,
The contents of the failure information once stored in RAM III will not be changed unless read out to the outside.

Therefore, even if the fault information is not obtained by the first switch turned on, the fault information will be sent to the RA at that time.
As long as the fault information is stored in Mill, the fault information at the time of the first fault occurrence can be obtained by the next switch turned on.

Note that here, the switches Sl, S2 . Although the case where the switches are turned on in the order of S3 has been described, it goes without saying that no matter what order the switches are turned on, the fault information at the stage when the switches are turned on first can be obtained.

[Effects of the Invention] As described above, according to the present invention, even if fault information cannot be obtained with the first switch turned on, sufficient fault information for recovery processing can be obtained with the next switch.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of a system that implements a failure information storage method according to an embodiment of the present invention, and FIG. 2 is a flowchart illustrating the failure information storage operation of the system shown in FIG. 1. 11--Host CPU, 12a, 12b,
12c one terminal, 13...external memory, txt...
・RAM, 112...Flip-flop, 5l-33
···switch.

Claims (1)

[Claims] a plurality of switch means for instructing collection of fault information when a system fault occurs; a storage means for storing the fault information; and a plurality of switch means for storing the fault information when it is stored in the storage means. a flag means that is reset when the flag is not present; an information collection means that collects fault information and stores it in the storage means and sets the flag means in a set state; and a switch that is turned on among the plurality of switch means. information retrieval means for reading out the failure information from the storage means to the outside in a processing form corresponding to the means, and setting the flag means to a reset state;
When any of the plurality of switch means is turned on, determining whether the flag means is in a set state or a reset state;
and control means for sequentially executing the processing by the information collecting means and the processing by the information retrieval means when the flag means is in a reset state, and for causing the processing by the information retrieval means to be executed when the flag means is in a set state. A fault information storage method featuring: