JP2001005694A - System monitoring device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a system monitoring device capable of monitoring the wholesomeness of each process in a computer system, specifying an abnormal process and monitoring also the operation state or the like of a process to be monitored. SOLUTION: An operating system(OS) for driving plural processes is provided with a monitoring process 13, a process to be monitored 14 and a semaphore 11 to be used for synchronizing both the processes 13, 14. The process 1 executes each process by locking the semaphore 11 and the process 13 judges whether the semaphore is locked within a prescribed monitoring time or not, and when it can not be locked within the time, judges the occurrence of a process abnormality.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for monitoring a computer system in which a plurality of processes (a process is an execution form of a program, which is used in the same sense in this application).

[0002]

2. Description of the Related Art Conventionally, various methods have been proposed as methods for detecting an abnormality in a computer system. For example, a special hardware (watchdog timer) is prepared in a system, and a system program periodically executes the above hardware. Some systems detect the health of the system by resetting the system.

FIG. 11 is a block diagram showing a system monitoring apparatus using a watchdog timer (WDT). In the figure, a watchdog timer (WDT) 100 is hardware provided in a computer system, and a system process 103 is software that operates in the computer system. The counter 101 is a system process 103
, The initial value is set, and the watchdog timer (WD
T) is subtracted by 100. The timer 102 starts the system process 103 periodically, and is used by the system process 103 to set a counter value in the counter 101 to an initial value.

Next, the conventional watch dog timer (W
The operation of the computer system using DT) will be described. The system process 103 stores the initial value in the counter 101
And watchdog timer (WDT) 1
00 is instructed to start. Watchdog timer (WD
T) 100 decrements the counter value in the counter 101 at regular intervals. On the other hand, the system process 103 resets the counter value in the counter 101 to the initial value at the cycle of the timer 102.

If the operation of the system process 103 (set of initial values) is interrupted due to an abnormality in the computer system, the counter value becomes zero. In this case, the watchdog timer (WDT) 100 determines that an abnormality has occurred in the computer system, and generates an abnormal interrupt in the computer system.

Thus, a watchdog timer (WDT)
In the system provided with 100, the system process 103 periodically resets the above initial values to monitor the health of the entire system.

[0007]

Since the conventional system monitoring apparatus is configured as described above, there are the following problems.

First, it is not possible to detect abnormalities of individual processes constituting the computer system (for example, hang-up of individual processes).

[0009] Next, there is no means for investigating the cause of which process caused the abnormal state.

Further, the set value of the priority value of the system process (priority in the computer at the time of executing the process) allows the system process to operate even if a process below the priority value of the system process runs away. , Cannot detect abnormalities.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and can monitor the health of each process in a computer system, identify an abnormal process, and specify a monitoring target. It is an object of the present invention to obtain a system monitoring device capable of monitoring the operation status of a process.

[0012]

According to a first aspect of the present invention, there is provided a system monitoring apparatus, comprising: a monitoring process, a monitored process, a monitoring process, and a monitored process on an operating system (OS) on which a plurality of processes operate. A semaphore used for synchronization with the process is provided. The monitored process locks the semaphore and executes each process, and the monitoring process determines whether or not the semaphore is locked within a predetermined monitoring time. If the lock cannot be performed within the time, it is determined that a process abnormality has occurred.

According to a second aspect of the present invention, there is provided a system monitoring apparatus comprising a plurality of monitoring threads which are sequentially activated for monitoring targets from a monitoring process, and wherein the monitored process is monitored by the plurality of monitoring threads. I do.

According to a third aspect of the present invention, there is provided a system monitoring apparatus comprising means for setting a variable-length sleep time on the monitoring process side, and the monitoring process keeps the semaphore locked during the sleep time. And

According to a fourth aspect of the present invention, there is provided a system monitoring apparatus for temporarily suspending a lock time of a semaphore by a monitoring process, and for investigating a state of the monitored process during that time, in particular, executing a dump process of the monitored process. Features.

[0016]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram showing a system monitoring device according to a first embodiment of the present invention, and FIG. 2 is a flowchart showing an operation of the first embodiment.

In the computer system of FIG. 1, a monitoring process 13 and a monitored process 14 exist in parallel, and the semaphore 1 is used as a means for monitoring the health of the process.
1 is provided between the monitoring process 13 and the monitored process 14. The timer 12 monitors the lock time of the semaphore 11. Here, the semaphore means a system variable (shared variable) used for synchronizing the parallel process processing, and indicates whether a certain action has been completed or whether a certain event has occurred.

Next, the operation of the first embodiment will be described with reference to the flowchart of FIG. Monitored process 14
Locks the semaphore 11 (S100) and performs processing in each process (S101). Thereafter, the semaphore 11 is unlocked (S102).

On the other hand, in the monitoring process 13, the timer 12
Is set (S110) to monitor the lock waiting time. Then, a semaphore lock used for synchronization with the monitored process 14 is attempted (S111). In S112, it is determined whether or not the semaphore lock is successful between the settings of the timer 12. This judgment depends on the implementation of the semaphore system service call of the computer system. If the lock cannot be performed, an error may be returned immediately, or a wait may occur in a system call until the lock can be performed. In some cases, the waiting time limit can be specified. If it becomes possible to lock within the monitoring time, the monitoring process 13 assumes that the monitored process 14 is operating normally, unlocks the semaphore (S113), and returns to S1.
The monitoring operation from 10 is continued.

In step S112, if the semaphore 11 cannot be locked in time, the monitoring process 13
It is assumed that some abnormality has occurred in the
114), for example, forcibly terminating the monitored process 14 and outputting abnormal information.

In this embodiment, the monitoring process 1
It is also possible to remove the monitoring mechanism by providing a flag or the like between the monitoring process 3 and the monitored process 14, transmitting the necessity of monitoring by the flag or the like, and further providing a means for determining the necessity of monitoring. is there.

As described above, according to the first embodiment, a plurality of monitoring processes can be performed without using special hardware by using a semaphore that can be used in almost all computer systems and a general-purpose software timer. Since it is configured to be able to monitor, there is an effect that the system monitoring device can be constructed at low cost. In addition, since an abnormality of an individual process can be detected and an abnormal process can be easily specified, maintenance time from the occurrence of a trouble to the investigation of the cause can be reduced.

Embodiment 2 FIG. Hereinafter, a second embodiment of the present invention will be described with reference to the drawings. FIG. 3 is a functional block diagram showing a system monitoring device according to a second embodiment of the present invention, and FIG. 4 is a flowchart showing an operation of the second embodiment.

In FIG. 3, as in the first embodiment (FIG. 1), a monitoring process 13 and a monitored process 14 exist in parallel, and a semaphore 11 is used as a means for monitoring the health of the process. 13 and the monitored process 1
4 are provided. And in this embodiment,
A monitoring thread 23 that is sequentially activated for the monitoring target from the monitoring process 13 is provided. The timer 12 monitors the lock time of the semaphore 11.

Next, the operation of the second embodiment will be described with reference to the flowchart of FIG. Monitored process 14
Locks the semaphore 11 (S200) and performs processing in each process (S201). Thereafter, the semaphore 11 is unlocked (S202).

On the other hand, in the monitoring process 13, threads are activated for the number of monitoring processes (S210). The following is an individual process by the monitoring thread 23. Monitoring thread 23
Then, the timer is set (S221), and the lock waiting time is monitored. Then, a semaphore lock used for synchronization with the monitored process 14 is attempted (S222).

In S223, it is determined whether or not the lock is successful within the monitoring time. This judgment depends on the implementation of the semaphore system service call of the computer system.
If the lock cannot be performed, an error may be returned immediately, or a wait may occur in a system call until the lock can be performed. In some cases, the waiting time limit can be specified.

If the lock is possible within the monitoring time, the monitored process 13 regards the monitored process 14 as operating normally, unlocks the semaphore 11 (S213), and performs the monitoring operation again from S221. To continue.

If the semaphore 11 cannot be locked in time in S223, it is considered that some abnormality has occurred in the monitoring process 13, and the abnormality processing (S225), for example, the forced termination of the monitored process 14, the abnormality information Output.

In the second embodiment, one monitored area is set in the monitored process, one semaphore is set for the monitored area, and one monitoring thread is associated with one monitored area. It is also possible to set a plurality of monitoring target areas for one monitored process. At this time, a plurality of monitoring areas in the monitored process may be monitored by the same semaphore, or another semaphore may be prepared for each monitoring area. Further, a plurality of monitoring threads may be prepared for each of a plurality of monitoring areas of the monitored process. According to this means, a plurality of processes and a plurality of monitoring areas can be monitored.

Embodiment 3 Hereinafter, a third embodiment of the present invention will be described with reference to the drawings. FIG. 5 is a functional block diagram showing a system monitoring device according to a third embodiment of the present invention, and FIG. 6 is a flowchart showing an operation of the third embodiment.

In FIG. 5, as in the first embodiment (FIG. 1), a monitoring process 13 and a monitored process 14 exist in parallel, and the semaphore 11 is used as a means for monitoring the soundness of the process. 13 and the monitored process 1
4 are provided. The timer 12 monitors the lock time of the semaphore 11. In the present embodiment, the monitoring process 13 sets the semaphore 1
A timer 15 for sleeping while locking 1 is provided, and an area 16 for setting a variable-length sleep time (time during which no command is executed) is provided.

Next, the operation of the third embodiment will be described with reference to the flowchart of FIG. Monitored process 14
Locks the semaphore 11 (S300) and performs processing in each process (S301). Then, the semaphore 11 is unlocked (S302).

On the other hand, in the monitoring process 13, the timer 12
Is set (S321), and the lock wait time is monitored. Then, a semaphore lock used for synchronization with the monitored process 14 is attempted (S322). In S323, it is determined whether or not the lock is successful. This judgment depends on the implementation of the semaphore system service call of the computer system. If the lock cannot be performed, an error may be returned immediately, or a wait may occur in a system call until the lock can be performed. In some cases, the waiting time limit can be specified.

When the lock becomes possible within the monitoring time, the computer sleeps for a certain time according to the sleep time set in the sleep time setting area 16. Thereafter, the semaphore 11 is unlocked (S325), and the monitoring operation is continued (S321). By this means, it is possible to give an arbitrary stop time to the monitored process.

If the semaphore 11 cannot be locked in time in S323, it is regarded that some abnormality has occurred in the monitoring process 13 and the abnormality processing (S32
6) For example, forced termination of the monitored process 14 and output of abnormal information are performed.

As described above, according to the third embodiment, an arbitrary delay time can be given to a process to be monitored, and provision is made for provisional measures for avoiding a system failure and effective means for debugging and the like. can do.

Embodiment 4 FIG. Hereinafter, a fourth embodiment of the present invention will be described with reference to the drawings. FIG. 7 is a functional block diagram showing a system monitoring device according to a fourth embodiment of the present invention, and FIG. 8 is a flowchart showing an operation of the fourth embodiment.

In FIG. 7, as in the first embodiment (FIG. 1), a monitoring process 13 and a monitored process 14 exist in parallel, and the semaphore 11 is used as a means for monitoring the health of the process. 13 and the monitored process 1
4 are provided. The timer 12 monitors the lock time of the semaphore 11. In the present embodiment, a stop flag setting area 17 in which a stop flag indicating that the monitoring process 13 is stopped while being locked is provided, and a processing device 18 for dumping the state of the monitored process 14 is provided. And a control program 19 for turning on the stop flag.

Next, the operation of the fourth embodiment will be described with reference to the flowchart of FIG. Monitored process 14
Locks the semaphore 11 (S400) and performs processing in each process (S401). Then, the semaphore 11 is unlocked (S402).

On the other hand, in the monitoring process 13, the timer 12
Is set (S421), and the lock wait time is monitored. Then, a semaphore lock used for synchronization with the monitored process 14 is attempted (S422). By S423,
It is determined whether the lock is successful within the monitoring time.
This judgment depends on the implementation of the semaphore system service call of the computer system. If the lock cannot be performed, an error may be returned immediately, or a wait may occur in a system call until the lock can be performed. In some cases, the waiting time limit can be specified.

If the lock is possible within the monitoring time, it is determined whether or not the stop flag set by the control program 19 in the stop flag setting area 17 is on (S424). If the stop flag is on, the stop flag is turned off (S425), and the dump processing unit 18 is requested to perform a dump process of the monitored process 14 (S42).
6) Wait for the end of the dump (S427).

If the stop flag is off in S424, the monitoring operation is continued (S421).

In the dump processing unit 18, the monitoring process 13
Receives the dump instruction (S426) from the server, and dumps the monitored process 14 (S430). When the dump is completed, an end notification is sent to the monitoring process 13 (S431).

If the semaphore 11 cannot be locked in time in S423, it is regarded that some abnormality has occurred in the monitoring process 13, and the abnormality processing (S42) is performed.
8) Forcibly terminate the monitored process 14 or output abnormal information, for example.

As described above, according to the fourth embodiment, it is possible to obtain dump information at any time of a monitored process, so that it is possible to obtain internal information of an abnormal process. become.

Embodiment 5 Hereinafter, a fifth embodiment of the present invention will be described with reference to the drawings. FIG. 9 is a functional block diagram illustrating a system monitoring apparatus according to a fifth embodiment of the present invention, and FIG. 10 is a flowchart illustrating the operation of the fifth embodiment.

In FIG. 9, as in the first embodiment (FIG. 1), the monitoring process 13 and the monitored process 14 exist in parallel, and the semaphore 11 is used as a means for monitoring the health of the process. 13 and the monitored process 1
4 are provided. The timer 12 monitors the lock time of the semaphore 11. In the present embodiment, a control program 19 for instructing the number of measurements in the statistics table and a statistics table 20 for recording the operation time of the monitoring process 13 are provided, and a measurement for measuring the lock time is provided. A timer 21 is provided.

Next, the operation of the fifth embodiment will be described with reference to FIG.
This will be described with reference to the flowchart of FIG. Monitored process 1
4 locks the semaphore 11 (S500) and performs processing in each process (S501). And semaphore 11
Is unlocked (S502).

On the other hand, in the monitoring process 13, the timer 12
Is set (S521), and the lock wait time is monitored. Next, the timer 21 for measuring the execution time of the monitored process 14 is started. (S522) Then, a semaphore lock used for synchronization with the monitored process 14 is attempted (S523). Then, in S524, it is determined whether or not the lock is successful within the monitoring time. This judgment depends on the implementation of the semaphore system service call of the computer system. If the lock cannot be performed, an error may be returned immediately, or a wait may occur in a system call until the lock can be performed. In some cases, the waiting time limit can be specified.

If the lock becomes possible within the monitoring time, the elapsed time is obtained from the measurement timer 21, and the recording of the processing time and the number of times of measurement are recorded in the statistics table 20 (S52).
5). After that, the monitoring semaphore 11 is unlocked (S
526), the monitoring / measuring operation is continued (S521). The number of measurement instructions in the statistical table 20 is determined by the control program 1
Set by 9.

If the semaphore 11 cannot be locked in time in S524, it is considered that some abnormality has occurred in the monitoring process 13, and the abnormality process (S52).
7) For example, the monitored process 14 is forcibly terminated or abnormal information is output.

As described above, according to the fifth embodiment, detailed execution time performance can be obtained for the performance of a process to be monitored or for an individual area of the process to be monitored.

[0054]

As described above, according to the first aspect of the present invention, a plurality of monitoring processes are performed by using a semaphore which can be used in almost all computer systems and a general-purpose software timer. Since the system can be monitored without providing hardware, the system monitoring device can be constructed at low cost. In addition, since an abnormality of an individual process can be detected and an abnormal process can be easily specified, maintenance time from the occurrence of a trouble to the investigation of the cause can be reduced.

According to the second aspect of the present invention, a plurality of processes can be monitored, and a plurality of areas can be monitored even in the same process.

Further, according to the third aspect of the present invention, it is possible to give an arbitrary delay time to a process to be monitored, thereby providing a provisional workaround for a system failure,
It is possible to provide an effective means for debugging or the like.

According to the fourth aspect of the present invention, it is possible to obtain dump information at any time of a process to be monitored. It is possible to obtain the internal information of the process that has become.

According to the fifth aspect of the present invention, it is possible to acquire the performance of a process to be monitored or a detailed execution time performance for an individual area of the process to be monitored.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a system monitoring device according to a first embodiment of the present invention.

FIG. 2 is a flowchart illustrating an operation according to the first embodiment of the present invention.

FIG. 3 is a block diagram showing a system monitoring device according to a second embodiment of the present invention.

FIG. 4 is a flowchart illustrating an operation according to the second embodiment of the present invention.

FIG. 5 is a block diagram showing a system monitoring device according to a third embodiment of the present invention.

FIG. 6 is a flowchart illustrating an operation according to the third embodiment of the present invention.

FIG. 7 is a block diagram showing a system monitoring device according to a fourth embodiment of the present invention.

FIG. 8 is a flowchart illustrating an operation according to the fourth embodiment of the present invention.

FIG. 9 is a block diagram showing a system monitoring device according to a fifth embodiment of the present invention.

FIG. 10 is a flowchart illustrating an operation according to the fifth embodiment of the present invention.

FIG. 11 is a block diagram showing a conventional system monitoring device using a watchdog timer (WDT).

[Explanation of symbols]

11 semaphore, 12 timer (for semaphore lock monitoring), 13 monitoring process, 14 monitored process, 1
5 timer (for sleep time), 16 sleep time setting area, 17 stop flag setting area, 18 dump processing unit, 19 control program, 20 statistics table,
21 timer (for measurement), 23 monitoring thread, 100
WDT (watchdog timer), 101 counter, 102 timer (for periodic activation), 103 system program.

Claims (5)

[Claims] An operating system (OS) on which a plurality of processes operate includes a monitoring process, a monitored process, and a semaphore used for synchronizing the monitoring process and the monitored process. The semaphore is locked and each process is executed, and the monitoring process determines whether or not the semaphore is locked within a predetermined monitoring time. If the semaphore cannot be locked within the monitoring time, it is assumed that a process error has occurred. A system monitoring device characterized by the above-mentioned. 2. The system monitoring apparatus according to claim 1, further comprising a plurality of monitoring threads that are sequentially activated for the monitoring targets from the monitoring process, and wherein the monitored processes are monitored by the plurality of monitoring threads. 3. The monitoring process according to claim 1, further comprising means for setting a variable-length sleep time in the monitoring process, wherein the monitoring process keeps the semaphore locked during the sleep time. A system monitoring device according to claim 1. 4. The system monitoring according to claim 1, wherein a lock time of the semaphore by the monitoring process is temporarily stopped so that a state of the monitored process can be checked. apparatus. 5. The method according to claim 1, wherein statistics from execution of the monitored process are acquired by collecting statistics from a semaphore lock request by the monitoring process to completion of the semaphore lock. The system monitoring device according to the paragraph.