JPH09251402A - Computer fault detection system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a computer fault detection system which can reduce the load applied to every computer and a cable set between both computers and also can quickly and surely detect the faults of the computers. SOLUTION: A timer operation process 11 of a computer 1 repetitively starts a timer 13 from the prescribed count value to evade a time count-up state of the timer 13. If the timer 13 has a time count-up state, a fault notification part 12 produces a fault notification signal and transmits it to a computer 2 via a cable C. A fault detection part 21 of the computer 2 detects the fault notification signal received via the cable C and realizes the occurrence of a fault in the computer 1. At the same time, the part 21 gives an interruption to a fault process 22. The process 22 receives the interruption and transfers its control to a handier 23 which deals with the the fault.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer fault detection system, and more particularly to a computer fault detection system that detects a fault occurring in a predetermined computer and that another computer replaces the computer and provides a service to a customer. .

[0002]

2. Description of the Related Art Conventionally, when a computer (computer) is used to provide an information service to a customer, another computer replaces it so that the customer will not be inconvenienced even if a predetermined computer fails. The system has been adopted to increase the availability by providing the same service. In such a system, a cable is provided between computers to communicate information on the occurrence of a failure,
When a failure occurs in one computer, another computer can detect the occurrence of the failure. There are the following methods for detecting the occurrence of this failure.

FIG. 4 is a schematic diagram showing a configuration example of a conventional computer fault detection system. In this configuration example, the computer 102 makes a call to the computer 101 at regular time intervals in order to ask whether the computer 101 is operating normally. The computer 101 naturally responds to the call from the computer 102 if it operates normally without any failure. The computer 102 determines that the computer 101 is operating normally by detecting the response from the computer 101.

On the other hand, when a failure occurs in the computer 101, the computer 101 cannot respond to the call from the computer 102. Calculator 102 is Calculator 1
When there is no response from 01, it is determined that the computer 101 has failed.

FIG. 5 is a schematic diagram showing another configuration example of a conventional computer fault detection system. In this configuration example, the computer 101 sends to the computer 102 a signal called a heartbeat indicating that it is operating normally. The computer 102 determines that the computer 101 is operating normally by detecting the heartbeat from the computer 101.

[0006]

However, in the system shown in FIG. 4, it is necessary to always carry out both "call" and "response" communication between computers, which causes a load on the cable. In addition, the computer 102 must always make a “call” and always detect a “response” no matter how busy the work of the computer 102 is.
A large load is applied to the computer 102. In addition, the computer 101 must always make a “response” to the “call” no matter how busy the work of the computer 101 is, so that a great load is applied to the computer 102.

Even when the "response" from the computer 101 is not detected in the computer 102, it does not mean that the computer 101 is in trouble and is simply busy with other work, so it takes time to respond. There may be cases. In this case, the computer 102 mistakenly determines that a failure has occurred in the computer 101. Therefore, computer 1
It has been difficult for the computer 102 side to accurately determine whether or not 01 has fallen into a state of being unable to continue providing the service to the customer.

Further, in the system shown in FIG. 5, it is necessary to continuously perform "heartbeat" communication between computers, which causes a load on the cable. Also, calculator 1
01 must always send a "heartbeat" no matter how busy his job is.
Will be overwhelmed. Also, the computer 102 is always a “heartbeat” no matter how busy the job is.
Must be detected, which imposes a large load on the computer 102.

Even if the "heartbeat" from the computer 101 is not detected by the computer 102, it does not mean that the computer 101 has a failure and is busy with other work, so it takes time to respond. In some cases, Also in this case, the computer 102 erroneously determines that a failure has occurred in the computer 101. Therefore, it is difficult for the computer 102 side to accurately determine whether or not the computer 101 has fallen into an impossible state to the extent that it cannot continue providing services to the customer.

In order to reduce the load on each computer,
If the time intervals of “call” and “response” in the method of FIG. 4 and “heartbeat” in the method of FIG. 5 are set to be long, it takes time to detect a computer failure. Also in this case, the processing by the computer 102 when a failure occurs is delayed, and the service to be provided to the customer is hindered.

The present invention has been made in view of the above circumstances, and provides a computer fault detection system capable of reducing the load on each computer and a cable between computers and detecting a fault of a computer quickly and reliably. The purpose is to

[0012]

A computer fault detection system according to the present invention comprises a timer, a means for repeatedly starting the timer from a predetermined value before the timer times out, and a timer when the timer times out. A first computer having a means for generating a signal; and means for detecting a signal generated from the signal generating means and determining that a failure has occurred in the first computer. And a second computer.

Further, the computer fault detection system according to the present invention includes a timer, means for repeatedly starting the timer from a predetermined value so that the timer does not expire, and a signal when the timer expires. A first computer having means for generating the signal, a cable for transmitting the signal when the signal is generated from the signal generating means, and detecting the signal when the signal is transmitted through the cable. And a second computer having means for determining that a failure has occurred in the first computer.

In the computer fault detection system, the second computer may be the first computer by the judging means.
It may further be provided with means for executing measures for occurrence of a failure when it is determined that a failure has occurred in the computer. Further, the first computer may further include means for providing a service to a customer. Also,
It is preferable that the timer start unit is operating while the service providing unit is operating. Further, the treatment executing means may include means for providing a service to the customer on behalf of the first computer when the determining means determines that a failure has occurred in the first computer. . Further, the computer fault detection system,
The first and second computers may further comprise shared database means for use in providing services to customers.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a schematic diagram showing the configuration of a computer failure detection system according to an embodiment of the present invention.

The computer fault detection system is composed of a plurality of computers. For example, as shown in FIG. 1, it is composed of two computers 1 and 2. The computer 1 and the computer 2 are adapted to detect a fault occurring on the other side and to take action when the fault occurs. Less than,
For simplification of the description, it is assumed that the computer 2 detects a failure in the computer 1, and the functions used in this case will be mainly described.

The computer fault system sends various information to the user terminal 3 in order to provide the service to the customer. For example, when a failure occurs in the computer 1 when the computer 1 provides the service, the computer 2 performs the same service to the customer instead of the computer 1, and a system having high availability is provided. Has been built.

The user terminal 3 is used by the customer to receive the service. For example, the user terminal 3
When the computer 1 receives a service from the computer 1 and a failure occurs in the computer 1, the computer 2 is replaced by the computer 2 instead of the computer 1.
From the same service. In this case, the customer does not know that the service provider has changed from the computer 1 to the computer 2, and the service equivalent to the case where no failure occurs in the computer 1 is provided. .

In this computer failure system, a shared database (not shown) is provided for use by the computer 1 and the computer 2 when providing a service to a customer. The computer 1 or the computer 2 can retrieve the information requested by the customer from the shared database and send the predetermined information to the user terminal 3 of the customer.

A timer operation process 1 is provided inside the computer 1.
1 is provided. The timer operation process 11 is realized by software, for example. Timer operation process 11
Repeats the timer 13 from a predetermined value before the timer 13 times out. That is, this timer operation process 11 is operating normally unless a failure occurs in the computer 1, and when the timer 13 is pointing to the position of the remaining 5 seconds, the timer 13 is operated for the remaining 10 seconds, for example. The operation of setting the position is repeated. Even when the computer 1 cannot set the timer 13 when the timer 13 is pointing to the position of the remaining 5 seconds because the computer 1 is busy with its own work, the work is calm down and the timer 13 is stopped. If the time has not yet expired, the computer 1 sets the timer 13 at the position of the remaining 10 seconds.

A fault notifying section 12 is provided inside the computer 1.
Is provided. When the timer 13 times out (that is, when the remaining time runs out), the failure notification unit 12 regards the computer 1 as a failure and generates a signal to that effect. The timer 13 is provided inside the failure notification unit 12, for example. The generated signal is sent to the computer 2 as a fault occurrence notification through the cable C between the computer 1 and the computer 2. It should be noted that this cable C is used only when a fault is notified.

A fault detecting section 21 is provided in the computer 2. The failure detection unit 21 is a failure notification unit 1 in the computer 1.
When a fault notification signal is sent from 2 through the cable C, this is detected and it is determined that a fault has occurred in the computer 1. In this case, the failure detection unit 21 interrupts the failure processing process 22.

In the computer 2, the failure processing process 2
2 are provided. The fault handling process 22 is realized by software, for example. The fault handling process 22
In order to cause the handler 23 to take a concrete action upon occurrence of an interruption when the interruption is caused by the failure detection unit 21,
Control is passed to this handler 23.

When the control is passed by the fault handling process, the handler 23 takes a concrete action when the fault occurs, and returns the control to the fault handling process when the action is completed. This handler 23 handles
Instead of the computer 1, the computer 1 provides the customer with a service that cannot be provided to the customer due to the occurrence of a failure.
That is, the handler 23 records in advance all the contents of the service that the computer 1 should provide to the customer, so that the service that the computer 1 was performing immediately before the failure occurred can be continuously provided to the customer. It has become.

Next, the operation of the computer fault detection system will be described. FIG. 2 is a flowchart for explaining the operation of the computer 1 in the computer failure detection system. The operation in the computer 1 will be described below.

Now, it is assumed that the computer 1 is operating normally without any trouble. At some point, the timer operation process 11 sets the timer 13 at the position of the remaining 10 seconds (step A1). As a result, the timer 13 starts and the remaining time gradually decreases.

At the time when the remaining time of the timer 13 indicates 5 seconds (step A2), no trouble has occurred in the computer 1 and the timer is operated unless the work is abnormally busy. Since the process 11 has a sufficient margin to operate the timer 13, the timer 13 is reset to 1
Reset to the 0 second position. As a result, the timer 13 restarts from the time when the remaining 10 seconds have passed.

If, in step A2, a problem occurs and there is no room to operate the timer 13, the timer operation process 11 does not reset the timer 13 at this point.

Thereafter, before the timer 13 times out (step A3), if the problem that has occurred is solved and there is room to operate the timer 13, the timer operation process 11 restarts the timer 13 again. Reset to the 10 second position. As a result, the timer 13 restarts from the time when the remaining 10 seconds have passed.

If the timer operation process 11 cannot reset the timer 13 by the time immediately before the timer 13 times out, the timer 13 naturally times out. At this time, the fault notifying unit 12 considers that a fault has occurred in the computer 1, and sends a fault notification to the computer 1 through the cable C (step A4).

FIG. 2 is a flow chart for explaining the operation of the computer 1 in the computer fault detection system. Below, computer 2 in case of failure of computer 1
Will be described.

The fault detecting unit 21 detects a fault notification from the fault notifying unit 12 in the computer 1 through the cable C, and determines that a fault has occurred in the computer 1 (step B1). . At this time, the failure detection unit 21 interrupts the failure processing process 22 (step B2). As a result, the fault handling process 22 passes control to the handler 23 in order to make the handler 23 take action upon occurrence of a fault (step B3).

When the control is passed by the above fault handling process, the handler 23 carries out a concrete action upon occurrence of a fault (step B4). On the occasion of this failure occurrence, the handler 23 provides the service, which the computer 1 cannot provide to the customer due to the failure occurrence, to the customer instead of the computer 1. That is, the handler 23 records in advance all the contents of the service that the computer 1 should provide to the customer, and continues to provide the service that the computer 1 was performing immediately before the failure occurred to the customer until the end.

The handler 23 returns the control to the failure processing process when the above-mentioned processing upon occurrence of the failure is completed (step B5).
As described above, according to the above-described embodiment, the cable C is used only when the failure notification is performed, so that it is possible to reduce the communication load between the computers.

Further, since the computer 2 does not need to perform any process related to the fault detection until the computer 1 receives the fault notification from the computer 1, it is possible to reduce the load on the computer 2. Further, the reset operation of the timer 13 by the timer operation process 11 in the computer 1 is light in burden. Therefore, it becomes possible to perform the reset operation at a shorter time interval,
The failure that occurs in the computer 1 can be notified to the computer 2 faster, and the computer 2 can detect the failure occurrence of the computer 1 faster.

Further, by realizing such high-speed detection, since the computer 2 can immediately replace the service provided to the customer by the computer 1 and continue the service, the quality of the service to the customer is not deteriorated. .

[0037]

As described above, according to the present invention,
It is possible to provide a computer fault detection system that can reduce the load on each computer and a cable between computers and can detect a fault in a computer quickly and reliably.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing the configuration of a computer fault detection system according to an embodiment of the present invention.

2 is a flowchart for explaining the operation of the computer 1 shown in the computer fault detection system of FIG.

3 is a flowchart for explaining the operation of the computer 2 shown in the computer fault detection system of FIG.

FIG. 4 is a schematic diagram showing a configuration example of a conventional computer failure detection system.

FIG. 5 is a schematic diagram showing another configuration example of a conventional computer fault detection system.

[Explanation of symbols]

 1, 2 ... Computer, 3 ... User terminal, 11 ... Timer operation process, 12 ... Fault notification unit, 13 ... Timer, 21 ... Fault detection unit, 22 ... Fault processing process, 23 ... Handler.

Claims (7)

[Claims] 1. A first computer comprising a timer, means for repeatedly starting the timer from a predetermined value before the timer times out, and means for generating a signal when the timer times out. A second computer having means for detecting a signal generated from the signal generating means and determining that a failure has occurred in the first computer. Computer fault detection system. 2. A first means comprising: a timer; means for repeatedly starting the timer from a predetermined value so that the timer does not time out; and means for generating a signal if the timer times out. A computer, a cable for transmitting the signal when the signal is generated from the signal generating means, and detecting the signal when the signal is transmitted through the cable, and a failure occurs in the first computer. A computer failure detection system, comprising: a second computer having means for determining that the failure has occurred. 3. The second computer further comprises means for executing measures for occurrence of a failure when the determining means determines that a failure has occurred in the first computer. The computer fault detection system according to claim 2. 4. The computer fault detection system according to claim 3, wherein the first computer further comprises means for providing a service to a customer. 5. The computer fault detection system according to claim 4, wherein the timer start means is in operation while the service providing means is in operation. 6. The treatment executing means comprises means for providing a service to a customer on behalf of the first computer when the determining means determines that a failure has occurred in the first computer. 6. The computer fault detection system according to claim 5, wherein. 7. The computer fault detection system according to claim 6, further comprising a shared database means for use when the first and second computers provide services to customers.