JPH06124214A - Continuation control system for fault hot job - Google Patents

Abstract

PURPOSE:To automatically continue the job of a fault occurring host at the time of fault occurrence by using other hosts. CONSTITUTION:Schedule execution means 2 and 2-1 execute the job corresponding to a schedule definition recorded in a schedule file 1 and also execute the later job specified by receiving a job restarting instruction corresponding to a job schedule. A job operating history write means 4 records information relating to the starting and stoppage of the job by the schedule execution means obtained through a job starting/stoppage recognition means 3 in a job operating history recording file 5 as a shared file. A job operating history read means 6 reads the information relating to the starting and stopping operations of the job of a system interrupting host obtained through an other host fault recognition means 7 from the job operating history recording file. A job operating state discrimination means 8 outputs the restarting instruction of the job at the time of system interruption occurrence from the information relating to the starting and stopping operations of the job of the system interrupting host.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a continuous control method for a job of a failed host when a plurality of hosts execute schedule definitions of jobs prepared in a shared file by a plurality of time series. Regarding

[0002]

2. Description of the Related Art Conventionally, a failed host has occurred while multiple hosts were executing business schedule definitions along multiple time series recorded in a schedule file prepared on a shared volume that can be read and written by multiple hosts. In this case, the job series in progress for the failed host has been restarted on another host according to the operator's judgment.

[0003]

In the above-mentioned conventional continuous control method for a failed host job, the operator determines a job in progress in the failed host and manually starts the job, so that it takes time to restart the job. There is a drawback that it takes time and leads to a decrease in the performance of the entire system.

It is an object of the present invention to provide a continuation control method for a failed host job that allows another host to automatically continue a job when a failure occurs in the failed host.

[0005]

SUMMARY OF THE INVENTION A first aspect of the present invention relates to a failed host job when a plurality of hosts execute job schedules defined in a plurality of time series prepared on a schedule file which is a shared file. In the continuous control method, each host executes the job according to the schedule definition recorded in the schedule file, receives a job restart instruction, and executes the jobs after the job specified by the instruction according to the job schedule. Means, a job start / stop recognition means for recognizing the start and stop of the job by the schedule execution means, and information on the start and stop of the job obtained by the job start / stop recognition means, which is a shared file of a plurality of hosts. The data recorded in the operation history record file Operation history writing means, another host failure recognition means for recognizing a system interruption due to a failure during system operation of another host other than itself, and activation of the job of the system interruption host from the job operation history recording file And job operation history reading means for reading information regarding stop operation, and job operation state determining means for outputting the restart instruction of the job at the time of system interruption from information regarding start and stop operation of the job of the system interruption host. It is characterized by having.

[0006]

Embodiments of the present invention will now be described with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of the present invention.

[0008] The schedule file 1 is a file in which the schedule definition of the work according to the time series set by the user is recorded, and is a file prepared on a shared volume readable and writable by a plurality of hosts, and the schedule execution means 2 Executes the schedule according to the schedule definition recorded in the schedule file 1. The job start / stop recognition unit 3 is a unit for recognizing the start and stop of the job by the schedule execution unit 2, and the job operation history writing unit 4 is for the job start / stop.
Information about the start and stop of the job obtained by the stop recognition unit 3 is recorded in a file. The job operation history recording file 5 is a file in which information about job start and stop operations is recorded by the job operation history writing unit 3 and is prepared on a shared volume that can be read and written by a plurality of hosts. The job operation history reading means 6 reads information on the job start and stop operations from the job operation history recording file 5. The job operating state determination means 8 determines the operating state of the job when the system is interrupted from the information about the job start and stop operations, and passes the operating state of the job to the schedule executing means 2-1. The other host failure recognizing means 7 recognizes a system interruption due to a failure while the other host is operating the system.
The schedule executing means 2-1 executes the schedule in accordance with the schedule definition recorded in the schedule file 1, and executes the job determined by the job operation status determining means 8 in the faulty host at the time of the fault occurrence. Starts without human intervention. still,
The respective means of the host A are also present in the host B and vice versa, but they are omitted in this embodiment.

FIG. 2 is a flow chart of an operation of executing a schedule before a system interruption occurs due to a failure during system operation, and an operation of recognizing a job start or stop operation and recording it in a file.

First, the schedule executing means 2 reads the schedule definition of the work in accordance with the time series set by the user from the schedule file 1 and executes the schedule (step 21). Next, when the job is started as a result of the execution of the schedule, the shovel start / stop recognition unit 3 recognizes the start operation of this job. This recognition is
It is also performed when the job is stopped as a result of the execution of the schedule (step 22). Information regarding the start and stop operations of the job recognized in step 22 is stored in the job operation record storage file 5 by the job operation history writing means 4.
(Step 23). The series of operations from step 21 to step 23 is repeated.

FIG. 3 is an operation flow chart relating to the continuous control method for a failed host job of the present invention when a system interruption occurs due to a failure during system operation.

First, the other host failure recognizing means 7 recognizes the system interruption of the host 9 in which a failure occurs during system operation (step 31). Next, the job operation history reading means 6 determines whether or not the job operation history recording file 5 has information regarding job start and stop operations (step 32). When it is determined that there is information by the operation of step 32, the job operation history reading unit 6
Information regarding job start and stop operations is read from the job operation history recording file 5 (step 33).

Using the information read out by the operation of step 33 as input information, the job operation state judging means 8 recognizes the job which was operating at the time of system interruption in the system interruption occurrence host 9 (step 34). Next, the schedule executing means 2-1 activates the job that was running when the system interruption occurred, which was recognized in step 34 (step 35).

By performing the series of operations shown in FIG. 2 after the series of operations shown in FIG. 3 in the host B, it becomes possible to continue the job operation without interruption.

[0015]

As described above, the continuous control method for a failed host job according to the present invention enables the other host to automatically continue the job at the time of the failure of the failed host. This has the effect of improving performance.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a flowchart of an operation of executing a schedule before a system interruption occurs and an operation of recognizing a job start or stop operation and recording the same in a file according to the present embodiment.

FIG. 3 is an operation flow diagram relating to a continuous control method for a failed host job when a system interruption occurs in the present embodiment.

[Explanation of symbols]

 1 Schedule File 2 Schedule Execution Means 2-1 Schedule Execution Means 3 Job Start / Stop Recognition Means 4 Job Operation History Writing Means 5 Job Operation History Recording Files 6 Job Operation History Reading Means 7 Other Host Failure Recognition Means 8 Job Operation Status Judging Means 9 Host A 10 Host B

Claims (1)

[Claims] 1. In a continuation control method for a failed host job when a plurality of hosts execute job schedules defined in a plurality of time series prepared on a schedule file which is a shared file, each host is A schedule executing unit that executes the job according to the schedule definition recorded in the schedule file and receives a job restart instruction, and executes subsequent jobs designated by the instruction according to the job schedule; Job start / stop recognition means for recognizing start and stop, and job operation history for recording information on job start and stop obtained by the job start / stop recognition means in a job operation history recording file which is a shared file of a plurality of hosts Writing means and self Other host failure recognizing means for recognizing a system interruption due to a failure during system operation of another host, which is a host other than the above, and a job for reading information on the start and stop operations of the job of the system interrupted host from the job operation history record file A faulty host comprising: an operation history reading unit; and a job operation state determination unit that outputs the restart instruction of the job when a system interruption occurs based on information about the start and stop operations of the job of the system interruption host. Job continuation control method.