JP2004164271A - Job execution monitoring method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a job execution monitoring method and a program, capable of immediately giving notice of generation of a system failure even when a job comes to an abnormal end. <P>SOLUTION: This operation monitoring program 5 operates when a job execution system 2 starts, monitors predetermined end standard time of each the job inside a monitoring information parameter file 3, determines that the job execution system 2 has abnormality when log information corresponding to the job inside a log file 4 is not recorded by a prescribed amount at point of time reaching the end standard time of the job or when actual end time of the job passes the end standard time, and starts a notification processing program 6 notifying the outside of it. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to a method and a program for monitoring the execution of a job in a computer system, and more particularly, to a method and a program for monitoring a job execution for detecting a system abnormality during job execution.
[0002]
[Prior art]
Due to the advanced use of computer systems, for example, offices of financial institutions operate 24 hours a day. During the day, an online business processing system mainly for dealing with customers operates, and at night, a batch business processing system is mainly operated as daily processing after the online business processing is closed.
[0003]
In recent years, automation in job operation has been advanced, and in a batch job processing system, a computer system is operating in a state close to unmanned operation. Therefore, it is required that the job operates normally and that the batch business processing is completed by the start of operation of the online business processing system the next morning.
[0004]
Therefore, if a failure due to hardware or software failure occurs in such a computer system and the recovery of the system is delayed, the subsequent operation of the business processing system will be greatly affected. Subsequent actions will be urgent.
[0005]
Further, when the processing capacity of the computer system is close to the limit and the data processing amount reaches a peak, a situation in which the batch job processing is not completed by a predetermined time in the next morning is also assumed.
[0006]
In the system described above, there is known a system that monitors the operation status of a job and notifies the occurrence of a system failure when a predetermined process is not completed by a predetermined reference time. (For example, see Patent Documents 1 and 2).
[0007]
[Patent Document 1]
JP-A-11-120033 (FIG. 2)
[Patent Document 2]
JP-A-2002-278804 (FIG. 1)
[0008]
[Problems to be solved by the invention]
However, such a system is designed on the assumption that a job is completed. For this reason, if a system failure occurs during job execution and the job ends abnormally, the job end time is not recorded and the scheduled end time (the latest execution end time in Patent Document 1 or the comparison table in Patent Document 2). End time) cannot be compared. In this case, there is a problem that an operation status cannot be monitored to detect an event in which a predetermined process has not been completed by a predetermined reference time, and a notice of occurrence of a system failure is greatly delayed.
[0009]
Therefore, an object of the present invention is to provide a job execution monitoring method and a program capable of promptly notifying the occurrence of a system failure even when a job ends abnormally.
[0010]
[Means for Solving the Problems]
To achieve the above object, according to a first aspect, a job execution monitoring method according to the present invention provides a method for monitoring a job corresponding to a job at a time when a predetermined end reference time at which the job ends is reached. It is a method of determining whether or not a desired value is recorded in the information.
[0011]
According to a second aspect, in the job execution monitoring method according to the present invention, a predetermined amount of log information corresponding to the job is recorded at a time when a predetermined end reference time at which the job ends is reached. If not, or if the actual end time of the job exceeds the end reference time, a method is provided for activating a reporting system that reports to the outside of the system that there is an abnormality in the system that executed the job. You.
[0012]
Further, according to a third aspect, the job execution monitoring method according to the present invention operates when the batch job processing starts, monitors a predetermined end reference time for each job in the batch job processing, and When a predetermined amount of log information corresponding to the job is not recorded at the time when the job end reference time is reached, or when the actual job end time exceeds the end reference time, the batch business process It is a method of activating a notification system that determines that there is an abnormality in the system and notifies the outside of the batch business processing system.
[0013]
In the present invention, preferably, the end reference time may be obtained by adding a predetermined time difference to a predetermined reference time.
[0014]
In the present invention, preferably, the notification system executes at least one of a message display on a monitoring terminal, an operator call by sounding a buzzer, and an automatic notification by telephone or e-mail to a system related person registered in advance. Is also good.
[0015]
According to a fourth aspect, a job execution monitoring program according to the present invention includes a monitoring information parameter table in which an end reference time predetermined for each job is recorded, and a log including an actual end time for each job. A log information table in which information is recorded; a step of determining whether a time has reached a reference end time in the monitoring information parameter table corresponding to a certain job; and a determination that the time has reached the reference end time. Determining whether a predetermined amount of log information corresponding to the job is recorded in the log information table; and determining whether the log information corresponding to the job is recorded in a predetermined amount. Determining whether or not the actual end time of the job within the job exceeds the reference end time of the job; and storing log information corresponding to the job in a predetermined amount. If not, or if the actual end time of the job exceeds the reference end time of the job, a notification system that notifies the outside of the system that there is an abnormality in the system executing the job. Activating the program.
[0016]
In the present invention, preferably, the job execution monitoring program may be started when the batch job processing system starts.
[0017]
BEST MODE FOR CARRYING OUT THE INVENTION
Embodiments of the present invention will be described below with reference to the accompanying drawings.
[0018]
FIG. 1 is a block diagram illustrating a configuration of a batch job processing system according to an embodiment of the present invention. The batch job processing system includes a job execution monitoring system 1, a job execution system 2, and a notification system 7. The job execution monitoring system 1 is a system for monitoring a job executed by the job execution system 2, and includes a monitoring information parameter file 3, a log file 4, an operation monitoring program 5, and a notification processing program 6. Be composed.
[0019]
The monitoring information parameter file 3 is a file in which a predefined execution end reference time for each processing unit of a job is tabulated and recorded. The log file 4 is a file in which generated event information, that is, log information of a unit of job processing during execution from the start to the end of the execution of the batch business process in the job execution system 2 is tabulated and recorded. The operation monitoring program 5 is a program for comparing and verifying the contents of the monitoring information parameter file 3 and the contents of the log file 4 for each processing unit. The notification processing program 6 is started when the operation monitoring program 5 detects that a job has abnormally ended, and is a program for starting a notification system 7 for automatically notifying the abnormal end of the job to the outside of the system.
[0020]
The job execution system 2 is a system for executing batch business processing. When started, the job execution system 2 starts execution of the operation monitoring program 5, processes a job according to the batch business, and logs information at a set checkpoint. Output to file 4. That is, log output 0, process 1, log output 1, process 2, log output 2, ..., process n, log output n, ..., process z, and log output z are executed, and finally, the operation monitoring program 5 is executed. Terminate. FIG. 1 shows a state where up to the log output n is written to the log file 4.
[0021]
In the above configuration, the operation monitoring program 5 starts operating when the job execution system 2 starts, and monitors the end reference time in the monitoring information parameter file 3 for each job in the batch job processing. If the desired value is not recorded in the log information in the log file 4 corresponding to the job at the point in time when the job end reference time is reached, for example, the log information is not recorded or the predetermined amount is insufficient. If the job has been completed, or if the actual end time exceeding the reference end time of the corresponding job is recorded, it is determined that the job has ended abnormally. When it is determined that the job has ended abnormally, the notification processing program 6 is started, and the notification processing program 6 starts the notification system 7, and the notification system 7 executes a predetermined notification process.
[0022]
Therefore, the job execution monitoring system 1 operates as described above. When the job execution monitoring system 1 detects an abnormal end of a job when the time reaches a reference end time of the job, the job execution monitoring system 1 immediately notifies the occurrence of a system failure. be able to.
[0023]
【Example】
In order to describe the above-described embodiment of the present invention in more detail, an embodiment of the present invention will be described with reference to the drawings.
[0024]
FIG. 2 is a diagram showing a data structure of a monitoring information parameter file according to the embodiment of the present invention. The job start reference time Ts0 and the end reference time Tsj of the process j are written in advance before the job execution system operates. Here, j is an integer from 1 to z.
[0025]
The job start reference time Ts0 and the end reference time Tsj of the process j, which are the criteria for verifying the job operation status, are determined by the system operation review committee or the like after the job configuration and the job execution order of the batch business processing system are determined. It is determined based on the unit execution time and the like.
[0026]
That is, the end point of each processing unit constituting a job is set as a check point, and each end reference time Tsj is registered in the monitoring information parameter file 3 in chronological order starting from the job operation start reference time. I have.
[0027]
It should be noted that the end reference time Tsj needs to be reviewed regularly or at any time according to a change in the processing data amount or a system change, and must be maintained so that an appropriate time is set.
[0028]
FIG. 3 is a diagram illustrating a data structure of a log file according to the embodiment of the present invention. When the job execution system 2 starts, the actual job start time Tr0 is written in the log file 4, and the log information j at the checkpoint of the processing j of the job j is sequentially written. However, FIG. 3 shows a case where j has passed from 1 to n.
[0029]
The log information j includes an item of the actual end time Trj of the process j. A large value that does not exist at the time may be written in the item of the actual end time Trj before executing the job. In this way, even if the job ends abnormally or the job does not end and the job end time is not written to the actual end time Trj, the actual end time of the job is set to the end reference time. It can be determined that it has exceeded.
[0030]
FIG. 4 is a diagram illustrating a processing flow of the operation monitoring program according to the embodiment of the present invention. When the job execution system 2 starts, the operation monitoring program 5 starts in step S0.
[0031]
In step S1, the log file 4 is initialized, and the log information is cleared or set to a predetermined value. Further, a counter i for counting the processing order is set to 0.
[0032]
In step S2, the current time is acquired and set to the actual job start time Tr0 in the log file 4.
[0033]
In step S3, a job start reference time Ts0 and an end reference time Tsi (i is 1 to z) of the process i in the monitoring information parameter file 3 are read.
[0034]
In step S4, a time difference Δ = Tr0−Ts0 is obtained. In a batch business processing system, generally, job operation is not always started at a time determined by the job start reference time Ts0. Therefore, a time difference Δ between the actual job start time Tr0 and the job start reference time Ts0 is calculated, and The end reference time Tsi in the step is corrected.
[0035]
In step S5, 1 is added to the counter i, and the count is incremented.
[0036]
In step S6, it is determined whether the counter i has exceeded the upper limit (z). If it is equal to or less than the upper limit, the process proceeds to Step 7 If the upper limit is exceeded, the end processing is executed in step S13 assuming that all the predetermined jobs have ended normally, and the operation monitoring program ends in step S15.
[0037]
In step S7, the current time T is obtained.
[0038]
Step S8 is a standby routine, and waits until the current time T reaches or exceeds the sum of the end reference time Tsi of the process i and the time difference Δ. That is, the current time T is acquired every moment, and if Tsi + Δ ≦ T, the process proceeds to step 9.
[0039]
In step S9, the log information i (including the actual end time Tri) existing in the log file 4 is read.
[0040]
In step S10, the presence or absence of log information i is determined. If there is no log information i or the log information is not enough, it is determined that a delay has occurred in the execution of the job due to some system failure, and the process proceeds to step S12. If there is a predetermined number of log information i, the process proceeds to step S11.
[0041]
In step S11, it is determined whether or not the sum of the end reference time Tsi of the process i and the time difference Δ is equal to or longer than the actual end time Tri of the process i. If Tsi + Δ ≧ Tri, it is assumed that the job has been executed within the reference time, and the process proceeds to step S5. If Tsi + Δ <Tri, it is determined that a delay has occurred in the execution of the job due to some system failure, and the process proceeds to step 12. Note that the positions of step S10 and step S11 may be interchanged.
[0042]
In step S12, the notification processing program 6 is started, and in step S14, the operation monitoring program ends.
[0043]
In the above steps, the operation monitoring program 5 starts operating when the job execution system 2 starts, and monitors the end reference time Tsi in the monitoring information parameter file 3 for each job in the batch job processing.
[0044]
In addition, when the time reaches the job end reference time Tsi, the operation monitoring program 5 does not record the log information i corresponding to the job i in the log file 4 or runs out of the predetermined number. In this case, or when the actual end time Tri of the job i exceeds the end reference time Tsi, it is determined that the job i has ended abnormally.
[0045]
That is, the case where the log information i is not recorded in the log file 4 or the case where the log information i is insufficient is a predetermined number. Abnormal termination due to a failure of the server or a hardware failure is assumed. Also, the case where the actual end time Tri of the job i exceeds the reference end time Tsi means that the processing is delayed due to a large amount of data processing or a limitation of hardware resources. Is done.
[0046]
As described above, when it is determined that a job has ended abnormally due to some system failure, the operation monitoring program 5 starts the notification processing program 6. The notification processing program 6 operates the notification system 7. The notification system 7 performs a message display on a monitoring terminal, an operator call by sounding a buzzer, an automatic notification by telephone or e-mail to a pre-registered system-related person, and the like, whereby a system failure occurs. Notify.
[0047]
【The invention's effect】
As described above, the job execution monitoring method and the program according to the present invention determine whether or not a job has ended normally at the reference time for ending the job. It can be notified outside.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a batch job processing system according to an embodiment of the present invention.
FIG. 2 is a diagram showing a data structure of a monitoring information parameter file according to the embodiment of the present invention.
FIG. 3 is a diagram illustrating a data structure of a log file according to the embodiment of the present invention.
FIG. 4 is a diagram showing a processing flow of an operation monitoring program according to the embodiment of the present invention.
[Explanation of symbols]
Reference Signs List 1 job execution monitoring system 2 job execution system 3 monitoring information parameter file 4 log file 5 operation monitoring program 6 report processing program 7 report system

Claims (8)

A job execution monitoring method, comprising: determining whether a desired value is recorded in log information corresponding to a job when a time reaches a predetermined end reference time at which the job ends. When a predetermined amount of log information corresponding to the job is not recorded at the time when a predetermined end reference time at which the job ends is reached, or the actual end time of the job exceeds the end reference time In this case, a job execution monitoring method characterized by activating a notification system for notifying the system that has executed the job that there is an abnormality to the outside of the system. It operates when batch business processing starts,
Monitor a predetermined end reference time for each job in the batch business process,
At the time when the time reaches the end reference time of the job, if a predetermined amount of log information corresponding to the job is not recorded, or if the actual end time of the job exceeds the end reference time, Judge that there is an error in the batch business processing system,
A job execution monitoring method, characterized by activating a notification system for notifying the outside of the batch job processing system. 4. The job execution monitoring method according to claim 1, wherein said end reference time is obtained by adding a predetermined time difference to a predetermined reference time. The system according to claim 1, wherein the notification system executes at least one of a message display on a monitoring terminal, an operator call by sounding a buzzer, and an automatic notification by telephone or e-mail to a system related person registered in advance. 4. The job execution monitoring method according to any one of the items 2 to 3. A monitoring information parameter table in which an end reference time predetermined for each job is recorded,
A log information table in which log information including the actual end time is recorded for each job;
Determining whether or not the time has reached an end reference time in the monitoring information parameter table corresponding to a certain job;
When it is determined that the time has reached the end reference time, determining whether a predetermined amount of log information corresponding to the job is recorded in the log information table;
Determining whether the actual job end time in the log information table exceeds the job end reference time when log information corresponding to the job is recorded by a predetermined amount;
If a predetermined amount of log information corresponding to the job is not recorded, or if the actual end time of the job exceeds the reference end time of the job, the system that is executing the job has an error. Activating a reporting system that reports to the outside of the system,
A job execution monitoring program characterized by including: The job execution monitoring program according to claim 6, wherein the end reference time is obtained by adding a predetermined time difference to a predetermined reference time. 7. The job execution monitoring program according to claim 6, wherein the job execution monitoring program is started when the batch job processing system starts.

Images