JP2001229033A - Device for re-executing job net in file failure - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automate file failure restoration processing by making it unnecessary to interpose an operator in a batch processing system using a job net, and to shorten a failure restoring time by quickly realizing the failure restoration processing. SOLUTION: This job net re-executing device is provided with a re-execution job deciding means for deciding a necessary re-execution job, a job re-executing means for re-executing a job, an execution JCL library in which an execution job control sentence is stored, an access history file in which file information processed in the job is stored, and a re-execution job management file in which the name of the job whose re-execution is necessary is stored.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a job net re-execution method in a batch processing system using a job net.

[0002]

2. Description of the Related Art In a batch processing system for processing a plurality of files in a job, when an error occurs in an external storage device in which a file is stored and a file failure occurs, an apparatus in which the error occurs without the intervention of an operator. Means for restoring an internal failure file and re-executing an abnormally terminated job is disclosed in Japanese Patent Application Laid-Open No. 7-56793.

[0003] In the above-mentioned prior art, a method of acquiring failure information when a file failure occurs, executing a recovery job corresponding to the failure type, recovering the failure file, and re-executing the job that ended abnormally due to the file failure is performed automatically. I'm going. In addition, data is restored from a previously saved file to a recovered file, and the operator need only grasp the occurrence of a failure and the recovery status.

[0004]

The above-mentioned prior art is effective in a batch processing system that executes a single job, but the following problems are raised in the case of a batch processing system using a job net.

[0005] As a problem of the prior art, when a save file used in the event of a file failure cannot be used due to a failure or the like,
Since it is necessary to start all over from the previously executed jobs, there is a case where operator intervention is required. This involves operations such as judgment and operation due to human intervention, and the recovery time greatly increases.

[0006] The present invention has solved the above problems,
An object of the present invention is to automate a file failure recovery process which does not require the intervention of an operator in a batch processing system using a job net, and to shorten a failure recovery time by a quick failure recovery process.

[0007]

In order to solve the above-mentioned problems,
Re-execution job determination means for determining a required re-execution job,
Job re-execution means for job re-execution, execution JCL library storing execution job control statements, access history file storing file information processed in the job, and job name requiring re-execution are stored Provide a re-execution job management file.

[0008]

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

FIG. 1 is an overall configuration diagram showing the relationship between the present invention and the prior art. First, a configuration of a file failure automatic recovery processing apparatus according to the related art will be described. 1 is a file information management database, 11 is a failure recovery procedure file,
21 is a basic JCL image file, 100 is a failure occurrence recognizing means, 110 is a failure recovery procedure assembling means, 120 is a failure recovery job control statement assembling means, and 130 is a failure automatic recovery executing means.

In these methods, when a file failure occurs, failure information is acquired, a recovery job corresponding to the failure type is executed, the failed file is recovered, and a job that has been abnormally terminated due to a file failure is re-executed.

Next, the configuration of the job net re-executing device in the event of a file failure according to the present invention will be described. 31 is an access history file, 41 is an execution JCL library, 51 is a re-execution job management file, 100 is a failure occurrence recognition unit, 20
Reference numeral 0 denotes a re-execution job determination unit, and 210 denotes a job re-execution unit. Among them, the fault occurrence recognizing means 100 uses the conventional technology, and is called when the backup file becomes unusable for some reason when the fault automatic recovery executing means 130 is executed as an interface. The re-execution job determining unit 200 inputs the access history file 31 and determines a necessary re-execution job. The job re-execution unit 210 inputs the re-execution job management file 51 and the execution JCL library 41, and starts the re-execution job set by the re-execution job determination unit 200. The access history file 31 stores file information processed in the job. Execution JC
The L library 41 stores business job control statements used in the batch processing system. The re-execution job management file 51 stores job names that need to be re-executed.

FIG. 2 shows a processing flow when the present invention is applied to a batch processing system using a job net.

Job A71, Job B72, Job C7
Reference numeral 3 denotes a job group constituting the job net, a file 61,
File 62 and file 63 are job A71 and job B
72 is an input or output file used in the job C73. Needless to say, the number of jobs and files is actually larger. Now, job B7
If a failure occurs in the input file 62 and the saved file cannot be used for some reason, the failure occurrence recognition means 100 detects the failure occurrence message, the re-execution job determination means 200 is started, and the access history file 3
Enter 1 to search for a job that creates the file 62. The job to be re-executed is the re-executed job management file 5
1 is stored. The job re-execution unit 210 inputs the execution JCL library 41 and the re-execution job management file 51, and starts the job determined by the re-execution job determination unit 200.

FIG. 3 is a processing flow of the re-execution job determining means and the job re-execution means.

The re-execution job determining means 200 searches the access history file 31 and checks whether there is a job for creating a failure file (step 202). If there is no job for creating the failure file, the process ends because the file cannot be recovered. If there is a job for creating a failure file, the failure file creation job is set as a re-execution target (step 203), and the abnormally ended job is set as a re-execution target (step 204). Finally, the re-execution target job is stored in the re-execution job management file 51 (step 205). The job re-executing means 210 reads the re-executed job management file 51 (step 21).
1) The same job control statement is read from the execution JCL library 41 (step 212) and executed (step 21).
3). It is determined whether the job has been completed normally (step 21).
4) In the case of normal termination, it is determined whether or not all the jobs have been re-executed (step 215). If there is a job that has not been started, the next job is executed. This is repeated for all the jobs in the re-execution job management file 51. If the job did not end normally, the process ends.

FIG. 4 shows the internal structure of an input / output file used in the present invention.

The execution JCL library 41 stores job control statements used in the batch processing system as a job name table 42. The access history file 32 contains file information (job name, file name,
(Save information) is stored as an access history table. The re-execution job management file 51 stores job names required for re-execution as a re-execution job name table.

[0018]

As described above, according to the present invention, in a batch processing system using a job net, it is possible to automate file failure recovery processing without requiring operator intervention and to shorten the failure recovery time by rapid failure recovery processing. It is possible.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram showing a relationship between the present invention and a conventional technique.

FIG. 2 is a flowchart of processing when the present invention is applied to a batch processing system using a job net.

FIG. 3 is a processing flowchart of a re-execution job determining unit and a job re-execution unit.

FIG. 4 is a diagram showing an internal configuration of an input / output file used in the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... File information management database, 11 ... Disaster recovery procedure file, 21 ... Basic JCL image file, 31 ...
Access history file, 32 ... access history table,
41: execution JCL library, 42: job name table,
51: re-execution job management file, 52: re-execution job name table, 100: failure occurrence recognition means, 110: failure recovery procedure assembling means, 120: failure recovery job control statement assembling means, 130: failure automatic recovery execution means, 200 ...
Re-executing job determining means, 210... Job re-executing means.

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Yoshihiro Tanaka 890 Kashimada, Saiwai-ku, Kawasaki-shi, Kanagawa F-term in Information Systems Division, Hitachi, Ltd. 5B027 AA04 BB01 BB05 CC04 5B098 AA03 DD02 GA03 GC01 GD02 GD14 JJ02

Claims (1)

[Claims] In a batch processing system using a job net, means for determining a job for creating a failure file when a backup file becomes a failure, means for executing the job determined by the left means, and There is an access history file that stores processed file information, an execution JCL library that stores execution job control statements used in the batch processing system, and a re-execution job management file that stores the names of jobs that need to be re-executed. A job net re-executing device at the time of a file failure, which enables re-execution from a previous job.