JP2004070458A - Program with self-diagnostic function, program supervising device and method, and program with program supervising function - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To restore a program to its normal state without depending on an operator's control nor having to initialize an entire system. <P>SOLUTION: A self-diagnostic function, comprising a process which waits for periodic event messages from a main thread by operating on a sub-thread different from the main thread on which the original process operates, and another process operating on the main thread to receive periodic timer messages produced by a timer operating on the main thread, to report events periodically to that process, is incorporated into the program to allow the program itself to detect errors and to terminate itself in the event of error. Then it is monitored whether or not the program has terminated because of error, and if it has terminated because of error, then it is determined whether the program terminated can be restarted. Based on the result, the program terminated because of error is restarted. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a program having a self-diagnosis function, a program monitoring apparatus and method for monitoring the program, and a program having a program monitoring function used to realize the program monitoring method.
[0002]
[Prior art]
Software such as a central monitoring device that monitors and controls plant and building equipment consists of a number of programs, including background processing.However, deadlock is caused by processing that exclusively accesses shared memory between programs. The operation cannot be performed at a rare timing such that the operation cannot be performed due to the occurrence of the error.
[0003]
Essentially, it is necessary to design to avoid such timing, but in a complicated software environment, it is actually difficult to completely realize this.
[0004]
On the other hand, if the deadlocked program is forcibly terminated and then restarted, the entire function is often restored.
[0005]
Therefore, in the related art, if there is a program that cannot be operated due to a deadlock, the user forcibly terminates the program by requesting the task manager.
[0006]
[Problems to be solved by the invention]
As described above, in the related art, when a program that cannot be operated due to deadlock occurs in a central monitoring device that monitors and controls equipment in a plant or a building, the user requests the task manager to execute the program. Forced termination.
[0007]
This forced termination operation is realized by a special key operation different from a normal operation.
[0008]
However, many operators who operate a central monitoring device that monitors and controls equipment in plants and buildings are not very familiar with computers. In many cases, special operations such as the above cannot be performed immediately.
[0009]
From now on, according to the conventional technology, when a program that deadlocks and becomes inoperable occurs in a central monitoring device that monitors and controls the equipment of plants and buildings, it takes time to recover the program, and the original A problem occurs that monitoring processing cannot be performed.
[0010]
On the other hand, the central monitoring device that monitors and controls plant and building equipment must operate continuously 24 hours a day, 365 days a year. A method of initializing the whole cannot be used.
[0011]
The present invention has been made in view of such circumstances, and automatically terminates a program that cannot be operated due to a deadlock or the like, and automatically restarts the self-terminated program, thereby reducing the operator's need. It is an object of the present invention to provide a new program technology capable of restoring a program to a normal state without initializing the entire system irrespective of an operation.
[0012]
[Means for Solving the Problems]
In order to achieve this object, the present invention realizes a program having a self-diagnosis function so that a program that cannot be operated due to deadlock or the like can be automatically terminated by itself, and the program can be monitored. By implementing a program monitoring technology that restarts the program when the program terminates by itself, it is possible to automatically restart a program that has abnormally terminated due to deadlock etc. I do.
[0013]
(1) A program having a self-diagnosis function realized by the present invention The program having a self-diagnosis function of the present invention operates on (1) a sub-thread and times out when an event is not notified for a specified time. Then, a process for generating a diagnostic program portion for performing a process of self-terminating the program (for example, when the program is started) and (2) running on the main thread in accordance with the generation of the diagnostic program portion Then, a process of starting a timer for performing a process of generating a timer message in accordance with a prescribed period, and (3) a process of operating on a main thread and notifying an event to a diagnostic program portion when a timer message occurs. Are executed by a computer.
[0014]
The program having the self-diagnosis function of the present invention configured as described above adopts a configuration in which a self-diagnosis function equivalent to WatchDogtimer is incorporated.
[0015]
This self-diagnosis function operates on a sub-thread different from the main thread on which the original processing operates, and waits for a periodic event notification from the main thread (executed by the diagnostic program part). The process comprises a process of receiving a timer message of a specified period generated by a timer operating on the thread and operating on the main thread, and notifying an event to the diagnostic program portion at a fixed period.
[0016]
A program operating on Windows (registered trademark) operates in an event manner by a message. However, when the main thread becomes inoperable, a timer message having a specified period cannot be generated due to this configuration. Event notifications can no longer be made.
[0017]
From now on, if an event notification is not received within a certain period of time, the diagnostic program part detects an abnormality by timeout and terminates the program from the sub-thread by itself.
[0018]
Here, CPU time is allocated to the diagnostic block part separately from the main thread, and since the diagnostic block part has a simple configuration that only waits for an event notification, the diagnostic block part becomes abnormal. The consequences can be considered negligible.
[0019]
In this way, the program having the self-diagnosis function realized by the present invention automatically detects a program that cannot be operated due to deadlock or the like, and forcibly terminates the program.
[0020]
(2) Program monitoring device realized by the present invention The program monitoring device of the present invention has a self-diagnosis function, and according to the self-diagnosis function, monitors a program that terminates processing when an abnormality occurs. (1) means for registering a program to be monitored, (2) means for monitoring abnormal termination of a registered program, and (3) detection of abnormal termination of a program by the monitoring. Means for determining whether the restart of the abnormally terminated program is appropriate and means for restarting the abnormally terminated program when determining the validity of the restart. To be configured.
[0021]
The program monitoring method of the present invention realized by each of the above processing means can be realized by a computer program, and the computer program can be provided by being recorded on a suitable recording medium such as a semiconductor memory.
[0022]
The thus configured program monitoring device of the present invention has a self-diagnosis function, and according to the self-diagnosis function, registers a program that terminates processing when an abnormality occurs as a monitoring target. Monitor the abnormal termination of the registered program.
[0023]
Then, when the abnormal termination of the registered program is detected by the monitoring, it is determined whether the restart of the abnormally terminated program is appropriate.
[0024]
For example, by determining whether a fixed time has elapsed since the previous startup, it is possible to determine whether the restart of the abnormally terminated program is appropriate, or whether the specified amount of memory capacity remains. By determining whether or not the restart of the abnormally terminated program is appropriate, it is determined.
[0025]
When the validity of the restart is determined by this determination processing, the program that abnormally ends is restarted.
[0026]
In this way, the program monitoring device realized by the present invention determines whether it is appropriate to restart a program that has terminated itself due to inoperability due to deadlock or the like, and restarts it. The processing is performed so that the self-terminated program in which the abnormality has occurred is restarted on condition that it is determined that the program is valid.
[0027]
As described above, according to the present invention, the program itself detects an abnormality, terminates the program if it is abnormal, and determines whether it is possible to restart the abnormally terminated program. Then, based on the result of the determination, the abnormally terminated program is restarted, so that the program can be restored to the normal state without initializing the entire system without depending on the operation of the operator. become.
[0028]
Then, it is determined whether or not restarting of the abnormally terminated program is appropriate, and on the condition that it is appropriate, the abnormally terminated program is restarted. Being able to do so can greatly improve the reliability of the system.
[0029]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, the present invention will be described in detail according to embodiments.
[0030]
FIG. 1 shows an example of an embodiment of a central monitoring device 1 provided with the present invention.
[0031]
The central monitoring device 1 according to this embodiment, which includes the present invention, performs a process of monitoring and controlling facility equipment 2 connected to a building via a communication network 3, and includes an operating system 10 and a building The system includes a plurality of application programs 11 for executing monitoring / control processing of the equipment 2 and a monitoring program 12 for executing restart processing of the abnormally terminated application program 11.
[0032]
The application program 11 performs, for example, a process of displaying a screen simulating a building and displaying state data of the equipment 2 installed at a building position selected from the screen. In the case of the configuration according to the present invention, a self-diagnosis function is provided, and when the occurrence of an abnormal state is detected according to the self-diagnosis function, the processing is forcibly terminated by itself.
[0033]
The application program 11 executes processing according to a processing unit called a thread, which is a unit of allocation to the CPU.
[0034]
On the other hand, when configured according to the present invention, the monitoring program 12 monitors the abnormal termination of the application program 11 notified from the operating system 10, and detects the abnormal termination of the application program 11 when detecting the abnormal termination. It is determined whether or not the restart of the terminated program is appropriate, and if it is determined that the restart is appropriate, processing is performed to restart the abnormally terminated application program 11. .
[0035]
FIG. 2 illustrates an embodiment of a program configuration of the application program 11 configured according to the present invention, and FIG. 3 illustrates an embodiment of a program configuration of the monitoring program 12 configured according to the present invention.
[0036]
First, a self-diagnosis process executed by the application program 11 configured according to the present invention will be described in detail with reference to FIG.
[0037]
The application program 11 includes a program portion that operates on a main thread and performs a process inherent to the application, and a program portion that operates on a sub thread and performs a self-diagnosis process.
[0038]
The program part that performs the processing of steps 1 to 6 shown in FIG. 2 is a program part that operates on the main thread, and the program part that performs the processing of steps 10 to 12 shown in FIG. This is the part of the program that does. Here, 110 shown in FIG. 2 is a timer that generates a timer message according to a prescribed cycle.
[0039]
When the application program 11 is started, as shown in FIG. 2, first, in step 1, it is determined whether or not a message has occurred. Here, when the application program 11 is started, it generates a start message first.
[0040]
From now on, when it is detected in step 1 that this start message has been generated, the process proceeds to step 2 to generate a notification event. That is, a path for event notification is generated.
[0041]
Subsequently, in step 3, the timer 110 that generates a timer message in accordance with the specified cycle is started.
[0042]
Subsequently, in step 4, a diagnostic program portion (diagnostic sub-thread) that operates on the sub-thread and performs the processes of steps 10 to 12 is generated, and then returns to step 1.
[0043]
On the other hand, if it is detected in step 1 that an event inherent to the application has occurred instead of a start message, the process proceeds to step 5 to execute the original process of the application indicated by the event that occurred, and then to step 1 Return to
[0044]
On the other hand, if it is detected in step 1 that the event of the timer message generated by the timer 110 has occurred instead of the event inherent in the application, the process proceeds to step 6 and the event notification path generated in step 2 is used. After the event is notified to the diagnostic program portion generated in step 4, the process returns to step 1.
[0045]
Next, the processing of the diagnostic program portion will be described.
[0046]
When the diagnostic program part is generated by the processing on the main thread, it operates on the sub thread. First, in step 10, it is determined whether or not a specified time has elapsed since the previous event notification. When it is determined that the specified time has not elapsed, the process proceeds to step 11, where it is determined whether an event has been notified from the main thread.
[0047]
When it is determined that an event has been notified from the main thread by this determination processing, the process proceeds to step 12, where the elapsed time from the previous event notification is cleared, and the process returns to step 10, where the event is not notified. When it is determined that the time has elapsed, the process immediately returns to step 10 without clearing the elapsed time.
[0048]
When it is determined in step 10 that the specified time has elapsed from the previous event notification, the process proceeds to step 13 in which execution of ExitProcess operating on the main thread is instructed to execute the processing of the application program 11 itself. The process is terminated by causing the sub thread to terminate itself.
[0049]
That is, when the operation of the main thread is stopped, the event notification is not performed because the timer message is not generated. In response to this, the sub-thread side is instructed to execute the ExitProcess by timeout of the event waiting, thereby , The application program 11 terminates itself.
[0050]
In this way, the application program 11 configured according to the present invention operates on the sub-thread and, when an event is not notified for a specified time, times out and terminates the program by itself. When an abnormality is detected by performing a self-diagnosis using the generated self-diagnosis, a process is performed so as to terminate the self-operation according to an instruction from the diagnosis program portion.
[0051]
When implementing the present invention, it is necessary to incorporate the logic described above into the application program 11, but Windows provides a library mechanism called OCX which is easily incorporated.
[0052]
From now on, when Windows is used, as shown in FIG. 2, if the diagnostic program part and the processing of steps 2, 3, 4, and 6 are incorporated in this OCX, the modification can be minimized. Become.
[0053]
Next, a program monitoring process executed by the monitoring program 12 according to the present invention will be described in detail with reference to FIG.
[0054]
When the monitoring program 12 is started, as shown in FIG. 3, first, at step 20, the monitoring program 12 interacts with the operator to, for example, execute the application program 11 which is not convenient if it is abnormally terminated. By acquiring the ProcessHandle as a registration target, the application program 11 to be monitored is registered.
[0055]
For example, when the application programs A, B, and C are to be monitored, the application programs A, B, and C are registered as monitoring targets by acquiring the ProcessHandles of the application programs A, B, and C.
[0056]
In response to this registration, the operating system 10 terminates any of the application programs A, B, and C for which the ProcessHandle has been acquired. Or abnormal termination for some other reason), the monitoring program 12 is notified of this.
[0057]
Subsequently, in step 21, by monitoring the notification from the operating system 10, the abnormal termination of the registered application program 11 is monitored.
[0058]
It is a waste of CPU time to always check whether the application program 11 to be monitored is running. On the other hand, in the case of Windows, by using an API called WaitForMultipleObject (), the application program 11 to be monitored can be monitored by the OS kernel without wasting CPU time.
[0059]
From now on, when Windows is used, the application program 11 to be monitored can be monitored without wasting CPU time by using the API of WaitForMultipleObject ().
[0060]
Subsequently, in step 22, it is determined whether the abnormal termination of the registered application program 11 has occurred by the monitoring process executed in step 21, and if it is determined that the abnormal termination has not occurred, the process proceeds to step 21. And the monitoring of the registered application program 11 for abnormal termination is continued.
[0061]
On the other hand, when it is determined in step 22 that the abnormal termination of the registered application program 11 has occurred, the process proceeds to step 23 to determine whether it is appropriate to restart the abnormally terminated application program 11. I do.
[0062]
For example, if a certain period of time has elapsed since the previous startup of the abnormally terminated application program 11, it is determined that it is appropriate to restart the application program. Processing such as judging that the possibility of abnormal termination is high again even if activated is determined, and that restarting is not appropriate is performed.
[0063]
Also, for example, if the memory usage of the system is below a certain threshold, it is judged that it is appropriate to restart because the memory does not run out, and conversely, if it is not below the threshold, the system is running out of memory. Is determined to be highly likely to end abnormally again even if it is restarted due to the unstable state, and processing such as determining that restarting is not appropriate is performed.
[0064]
When it is determined that restarting the abnormally terminated application program 11 is appropriate by the determination processing in step 23, the process proceeds to step 24, where the abnormally ended application program 11 is restarted. Returning to 21, the monitoring of the registered application program 11 for abnormal termination is continued.
[0065]
On the other hand, when it is determined that restarting the abnormally terminated application program 11 by the determination processing in step 23 is not appropriate, the process proceeds to step 25 and requests the operating system 10 to initialize the system. Then, the process ends.
[0066]
Here, depending on the application program 11 that has abnormally terminated, it is possible to perform other methods such as not performing initialization but performing any operation.
[0067]
In this way, the monitoring program 12 configured according to the present invention monitors the application program 11 that has terminated by the self-diagnosis function, and detects an abnormal termination of the application program 11, and detects the abnormal termination of the application program 11. It is determined whether restarting the program 11 is appropriate and restarting the abnormally terminated application program 11 on condition that it is determined that restarting is appropriate. Process it.
[0068]
Although the present invention has been described according to the illustrated embodiment, the present invention is not limited to this. For example, in the embodiment, the present invention has been described by taking the application to the central monitoring device 1 as a specific example, but the present invention is not limited to the application to the central monitoring device 1.
[0069]
【The invention's effect】
As described above, according to the present invention, the program itself detects an abnormality, terminates the program if it is abnormal, and determines whether it is possible to restart the abnormally terminated program. Then, based on the result of the determination, the abnormally terminated program is restarted, so that the program can be restored to the normal state without initializing the entire system without depending on the operation of the operator. become.
[0070]
Then, it is determined whether or not restarting of the abnormally terminated program is appropriate, and on the condition that it is appropriate, the abnormally terminated program is restarted. Being able to do so can greatly improve the reliability of the system.
[Brief description of the drawings]
FIG. 1 is an embodiment of a central monitoring device equipped with the present invention.
FIG. 2 is an example of a program configuration of an application program configured according to the present invention.
FIG. 3 is an embodiment of a program configuration of a monitoring program configured according to the present invention.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Central monitoring apparatus 2 Equipment 3 Communication network 10 Operating system 11 Application program 12 Monitoring program

Claims (7)

A process that runs on a sub-thread and generates a diagnostic program portion that performs a process of self-terminating when a time-out occurs when an event is not notified for a specified time;
Processing for starting a timer that operates on the main thread and performs processing for generating a timer message in accordance with a prescribed cycle, in accordance with the generation of the diagnostic program portion;
A program that runs on a main thread and has a self-diagnosis function for causing a computer to execute a process of notifying an event to the diagnostic program portion when the timer message occurs. Means for registering as a monitoring target a program having a self-diagnosis function and, according to the self-diagnosis function, self-terminating processing when an abnormal state occurs;
Means for monitoring abnormal termination of the registered program;
Means for determining whether restarting the abnormally terminated program is appropriate when detecting the abnormal termination,
Means for restarting the abnormally terminated program when determining the validity of the restart,
Characteristic program monitoring device. The program monitoring device according to claim 2,
The determining means determines whether or not the restart of the abnormally terminated program is appropriate by determining whether or not a predetermined time has elapsed since the previous startup.
Characteristic program monitoring device. The program monitoring device according to claim 2,
The determining means determines whether or not the restart of the abnormally terminated program is appropriate by determining whether or not the specified amount of memory capacity is left.
Characteristic program monitoring device. The program monitoring device according to any one of claims 2 to 4,
The program to be monitored operates on the sub-thread, generates a diagnostic program portion that performs a process of terminating the program by itself when the event is not notified for a specified time, and furthermore, At the same time as the generation of the diagnostic program portion, a timer that operates on the main thread and performs a process of generating a timer message according to a specified cycle is started, and further operates on the main thread to generate the timer message. Sometimes, the self-diagnosis function is realized by performing a process of notifying an event to the diagnostic program part.
Characteristic program monitoring device. Having a self-diagnosis function, according to the self-diagnosis function, registering a program for self-terminating processing when an abnormal state occurs as a monitoring target;
Monitoring the abnormal termination of the registered program;
When detecting the abnormal termination, determining whether restarting of the abnormally terminated program is appropriate; and
Restarting the abnormally terminated program when judging the validity of the restart,
Characteristic program monitoring method. A process of having a self-diagnosis function and registering a program for self-terminating the process when an abnormal condition occurs according to the self-diagnosis function as a monitoring target;
Monitoring the abnormal termination of the registered program;
When detecting the abnormal termination, a process of determining whether restarting the abnormally terminated program is appropriate, and
A program having a program monitoring function for causing a computer to execute a process of restarting an abnormally terminated program when determining the validity of the restart.

Images