JPS6279531A - Job step restarting method - Google Patents

Abstract

PURPOSE:To execute job step restarting at the time of the reexecuting of the job step by using the contents of a second memory area on a secondary memory device at the time of the processing of the restarting after a system down. CONSTITUTION:A restarting processing part 109 refers to the identifying information of the job which is executed at the time of generating the system down and selects the job which comes to be the object of the restarting processing. Further, the pointer to show the presence of a job control language file 104 of the job is referred to, the job control language file 104 is read, it is decided whether or not in the information, the job step restarting prohibiting displaying exists, when the displaying is not present, from the read information, the contents of a job control information backup file 106 are copied to a job control information file 105, and the information of the resources succeeded from the prior job step is reloaded. Thus, without wasting many computer resources at the time of the usual operation, the job step can be re-executed.

Description

Detailed Description of the Invention [Industrial Application Field] The present invention stores job control information that is updated during a job processing process on a secondary storage device, and when a system failure occurs, the job control information is stored in the secondary storage device. The present invention relates to a job step restart method for re-executing a job step using job control information stored on an apparatus.

[Conventional technology]

If the system goes down for some reason while the system is in use, the job restart function is generally called the job restart function, which automatically re-executes the jobs whose execution was interrupted due to the system down after the system is restored. In this case, one job consists of several job steps that are executed sequentially, so it is preferable to be able to re-execute the job step that actually experienced a system down from the beginning. It's called a function.

The information required by the system when processing a job is called job control information, and is created using the job control language that describes the job as input. One job step consists of an initiation processing section, a processing program section, and a termination processing section. Job control information is updated in the initiation processing section and termination processing section, and the processing program section also updates the job control information. Job control information may be updated by a job control macro instruction, a file control macro instruction, or the like used by a processing program to be executed.

In order to carry over the job control information updated in this way over sequentially executed job steps, the job control information is usually created in an area on a secondary storage device.

The conventional job step restart method was to restore job control information at the time of system failure and re-execute job steps based on job control information created in an area of secondary storage. .

Another example of the conventional job step restart method is to duplicate the job control information on the secondary storage device on the X and Y sides of the storage medium, and update the job control information in the initiation processing section and the termination processing section. When job control information is updated by a job control macro command or a file control macro command used by a processing program executed in the processing program section, the update is first performed on the X plane, and then the X plane is updated. By copying from
JP-A-58-1H444).

[Problem that the invention seeks to solve]

Among the conventional job step restart methods mentioned above,
In the case of the former method, the job control information is updated when the job control information is updated in the initiate processing section and the termination processing section, or by job control macro instructions, file control macro instructions, etc. executed in the processing program section. When a system failure occurs, the job control information on the secondary storage device becomes uncertain, making it impossible to restart the job step.

In general, the frequency of updating job control information during the processing of one job is quite high, so this method has the disadvantage that there is a high possibility that job step restart will not be possible.

In the case of the latter method, access to the duplicated secondary storage device always occurs when job control information is updated, resulting in a large overhead in job processing. The problem is that it is necessary to separately store information about which side the screen is in on a secondary storage device, which makes management unnecessarily complicated, and job control macros used by processing programs executed in the processing program section. Since the job control information on the secondary storage device is also updated by commands, file control macro instructions, etc., there is a drawback that a program checkpoint of the processing program must be taken every time the job control information is updated. That is, this method has the disadvantage that a large amount of computer resources must be wasted during normal operation in preparation for a system failure that generally occurs only rarely.

An object of the present invention is to store job control information that is updated during a job processing process on a secondary storage device, and to use the job control information stored on the secondary storage device when a system down occurs. An object of the present invention is to provide a job step restart method for re-executing a job step.

Another aspect of the present invention is to significantly reduce the possibility that a job step cannot be re-executed because the job control information in the secondary storage device is uncertain. The purpose of the present invention is to provide a job step restart method.

Yet another object of the present invention is to make it possible to re-execute job steps in a timely manner without wasting a lot of computer resources during normal operation, in preparation for a system down that generally occurs only in the morning. How to restart job step〃,
Our goal is to provide the following.

[Failure to solve the problem]

The job step restart method of the present invention includes the steps of: storing job control information updated during initiator 1/processing of a job step in a first storage area on a secondary storage device; Secondary storage device
storing the job control language in the first storage area of the job step in the area of the main storage device-L allocated to the job step; and during the execution and termination process of the job step. Using the job control information stored in the area on the main storage device, when termination processing is completed, the job control information on the main storage device that was changed during the execution of the job step and during the termination processing is stored in the secondary storage. storing in a first storage area on the device; and then copying the contents of the first storage area on the secondary storage device to a second storage area on the secondary storage device, It is characterized in that the contents of the second storage area of the secondary storage device -1- are used during restart processing after a downtime.

In this way, writing to the job control information backup file that stores job control information required for re-execution of a job step is reduced to once during termination processing, that is, once per job step. As a result, when a system failure occurs, the possibility that a job step cannot be re-executed because the job control information on the secondary storage device is uncertain is greatly reduced, and further, By performing all access to job control information during processing program execution on the main storage device, not writing to job control information files and job control information backup files, and not requiring program checkpoints, etc. In preparation for system downtime, which generally occurs rarely, job steps can be re-executed without wasting much computer resources during normal operation.

〔Example〕

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

FIG. 1 is a block diagram showing control and data flow in an embodiment of a job step processing process and a restart processing process to which the job step restart method of the present invention is applied. In the figure, "←" indicates the flow of control, and "伽" indicates the flow of data.

The job step processing process in this embodiment includes an initiation processing section 101, a processing program section 102, and a termination processing section 103. In addition, as a file on the secondary storage device used during the job step processing process, a job control language file 104 that stores the job control language that describes the job is stored in the first area on the secondary storage device. The job control information file +05, which is the job control information that was created, and the second job control information file on the secondary storage
A job control information backup file 106, which is job control information stored in an area, exists for each job. Here, the job control language file 104 and the job control information file 105 are the same type of files that are generally found in systems that perform job step restart processing.

The initiation processing unit 101 executes the job control language file 104 at the beginning of the job step initiation processing.
The job control language corresponding to the job step is read from the information contained in the job control language that describes the job step to be processed and the job control information inherited from the preceding job step. Create information and update the job control information file 105. Thereafter, resource allocation is performed, and the updated job control information is also sequentially written to the job control information file +05. When the job step initiation process is completed, the job control table 107 is created from the contents of the job control information file +05, and is stored in the - area of the main storage device 1- reserved for the execution of the job step. Processing program section 1
Control is passed to 02.

In the processing program section 102, job control table +0
7, the processing program 108 to be executed is determined, loaded into the secured area on the main storage device, and control is transferred. Resources may be dynamically allocated or released by job control macro instructions, file control macro instructions, etc. used by the processing program 108, and job control information may be updated. In this case, changes are made only to the job control table 107, and the contents of the job control information file 105 are not changed. When the execution of the processing program for 10 days is completed, control is passed to the termination processing 1, 1 unit 103.

The termination processing unit 103 executes the job control table updated by the processing program 108 in order to perform post-processing of various resources specified by the job control language after normal job step termination processing commonly found in conventional systems. Based on the contents of 107, job control information at the time of completion of that job step is created. In order to carry over this job control information to subsequent job steps, the job control information file 105 is updated using the created job control information. Thereafter, the contents of the job control information file 105 are copied to the job control information backup file 106. After waiting for the completion of this copying, control is passed to the initiation processing unit 101 in order to process the subsequent job step.

Further, referring to FIG. 1, the restart processing process in this embodiment involves re-execution of a job step after a system down occurs using a restart processing unit 109, a job control language file 104, a job control 2 control information file +05 and job control information backup file 1 (1θ).

The restart processing unit 109 uses the job control language file +04 of the job that was being executed when the system went down.
The job step is re-executed by reading the job control information backup file 10B and restoring the job control information of the last completed job step to the job control information file 105 based on the job.

The following describes the initiation processing unit 101 in the embodiment shown in FIG. The operations of the termination processing section 103 and the restart processing section 108 will be explained in more detail.

FIG. 2 is a flowchart showing the operation of the initiation processing section 101. However, FIG. 2 mainly shows the parts related to the present invention, and does not show the details of the initiation processing part generally seen in conventional systems. Referring to FIG. 2, the initiation processing unit 101 first reads the job control language file 104 of the job to be processed.
is read to obtain information regarding a job step to be processed and a job control language that describes the job step (step 201). It is determined whether the job step to be processed is the first job step of the job (step 202), and if it is the first job step, an area for the job control information file 105 and the job control information backup file 10B is saved. After securing the area on the secondary storage device and initializing each area, pointers for each area are written in the job control language file 104 (step 203). If it is not the first job step, the job control information file 105 already exists, so the job control information file 105 is read using the pointer to the job control information file 105 contained in the information read from the job control language file 104. (step 204). Step 203 or Step 2
04, if the job control information file 105 is read, based on its contents and the job control language that describes the job step read from the job control language file 104, A. Create job control information. (Step 205). In this case, the content read from the job control information file 105 means information such as resources inherited from the preceding job step. Furthermore, the content obtained by interpreting the job control language means information such as the type and location of resources newly required in the job step, and post-processing after use.

Next, the job control information file 105 is updated using the job control information created in this way at the time of starting the resource allocation process (step 206). Thereafter, an initiation process commonly found in conventional systems is performed (step 207). In the process of this initiation process, allocation of various resources required by the job step is attempted by referring to information regarding required resources and allocated resources indicated by job control information written in the job control information file 105. , if the allocation is successful, information to that effect is stored in the job control information file 105.
write back to. When all the necessary resources are finally allocated by repeating the process 1-1, the job control information file 105 contains job control information indicating the allocation state of all the resources required by the job step, i.e. This means that the job control information at the time when job step execution starts has been created. At the end of these initiation processes, an area of main storage is reserved for execution of the job step. Finally, the job control information necessary for the execution of the processing program section 102 is read from the job control information file 105, and the job control table 107 is created in an area of the main storage device -L in -[-.
Step 208).

FIG. 3 is a wave diagram showing the operation of the termination processing section 103. However, FIG. 3 mainly shows the parts related to the present invention, and does not show the details of the termination process commonly seen in conventional systems. Referring to FIG. 3, the termination processing unit 103 first reads the job control information file 105 to obtain job control information at the time when job step execution starts (step 301). This information and job control table 10? Data for billing/statistics can be obtained by comparing the contents of the data. Thereafter, termination processing commonly found in conventional systems is performed (step 302). In the process of this termination process, the job control table 107
Referring to the information regarding the post-processing of resources after use indicated by the job control information of -1=, the resources to be carried over to the subsequent job step are left in the allocated state,
Performs release processing for resources that will not be inherited by subsequent job steps. 2. When the post-processing of all resources is finally completed by repeating the process described in 2 above, job control information is created in the job control table 1071-, indicating the allocation state of resources to be carried over to the subsequent job step. There will be. Therefore, the job control information file 105 is created using the job control information created in this way.
is updated (step 303). Thereafter, it is determined whether the job step being processed is the final job step of the job (step 304), and if it is not the final job step, the part corresponding to the job step in the job control language file 104 is A job step restart prohibition display is written (step 305). After waiting for the completion of step 305, the contents of the job control information file 105 are copied to the job control information backup file 10B (step 306). Step 30
After waiting for the completion of step 6, the job step restart prohibition display written in step 305 is erased (step 307), and the processing of the termination processing section is ended.

If the determination in step 304 is that it is the final job step, the area for the job control information backup file 10B and the job control information file 105 used by the job is released (step 308), and the processing of the termination processing section is ended.

FIG. 4 is a flow chart showing the operation of the restart processing section 108. However, FIG. 4 mainly shows the parts related to the present invention, and does not show the details of the restart processing part commonly seen in conventional systems. In a system that performs job step restart, the identification information of the job that was being executed when the system went down,
A pointer indicating the location of the job control language file +04 of the job is retained even after the system is down, and job step restart processing is performed with reference to them. Referring to FIG. 4, the restart processing unit 109 first selects a job to be restarted by referring to the identification information of the job that was being executed when the system went down as described in 1- above. (Step 401). Furthermore, referring to the pointer indicating the location of the job control 8-word file 104 of the job, the job control word J^ file +04 is read (step 402). It is determined whether or not the job step restart prohibition 11-display written in step 305 of FIG. Obtain pointers to the job control information file 105 and job control information backup file l0ff from the information, copy the contents of the job control information backup file 106 to the job control information file 105, and copy the job that was being executed when the system went down. The step restores information such as resources inherited from the preceding job step (Stella 7' 404). Thereafter, job step restart processing commonly found in conventional systems is performed (step 405). In the process of job step restart processing, the job control of the job is transferred to the word file 104 and the restored job control information file 10.
5 can be used to rerun the job step in much the same way as normal initiation processing. If the job step restart prohibition display exists in the test settings in step 403, the contents of the job control information backup file 106 are uncertain, so job step restart processing is not performed, and job step restart processing is not performed.
04 to re-execute the entire job (step 408). (2) Repeat the process described above, and when all the jobs targeted for restart processing are finally finished, the restart processing unit 109
processing is completed.

〔Effect of the invention〕

As described below, the present invention allows the job control information backup file storing 17 job control information required for re-execution of a job step to be occupied once during termination processing, that is, By reducing the number of times per job step to once, there is a possibility that when a system failure occurs, the job step cannot be re-executed because the job control information on the secondary storage device is uncertain. significantly lowers
For example, all access to job control information during processing program execution is performed on the main memory, no writing is done to the job control information file or job control information backup file, and program checkpoints are not required. This has the effect of making it possible to re-execute job steps without wasting a lot of computer resources during normal continuous use, in preparation for a system failure that generally occurs only in the morning.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the control and data flow of job step processing and restart processing in one embodiment of the job step restart force U of the present invention, and FIG. FIG. 3 is a flow chart showing the operation of the initiation processing section 101 in the embodiment of FIG. 1. FIG. 4 is a flow chart showing the operation of the termination processing section 103 in the embodiment of FIG. 1. 10 is a flowchart showing the operation of the section 1013. +01... Initiate processing section, 102... Processing program section, 103... Termination processing section, 104... Job control language file, 105... Job control information file, 108... Job control information backup File, 107...Job control table, +08...Processing program, 109...Restart processing unit.

Claims (1)

[Claims] Job control information that is updated during the job processing process is stored on a secondary storage device, and when a system down occurs, the job control information stored on the secondary storage device is used. A job step restart method for re-executing a job step includes: storing job control information updated during job step initiation processing in a first storage area on a secondary storage device; and upon completion of the initiation processing; the first on the secondary storage device;
storing the job control information in the storage area of the main storage device in an area on the main storage device allocated to the job step; Using the job control information stored in the area, when the termination process is completed, the job control information on the main storage device that was changed during execution of the job step and during the termination process is stored in the first memory area on the secondary storage device. and then copying the contents of the first storage area on the secondary storage device to a second storage area on the secondary storage device, A job step restart method, characterized in that, during start processing, contents of a second storage area on the secondary storage device are used.