JPS6362007B2 - - Google Patents

Description

[Detailed Description of the Invention] Purpose of the Invention Industrial Application Field The present invention is an automatic system that automatically performs recovery processing without human intervention such as an operator when a software failure occurs in a business processing job of a computer system. Regarding recovery processing method.

Conventional technology Conventionally, the method of recovery processing when a software failure occurs in a computer system is to identify the cause of the software failure by a support system engineer (programmer) or operator from a list obtained from a memory dump, file dump, trace dump, etc. It is common to find them and process them using one of the following methods.

That is, in the first method, in order to eliminate the cause of the failure and avoid future failures, the program, etc. is first corrected, then the file that is being updated is restored, the corrected program is activated, and reprocessing is performed.

The second method is to delete the data that caused the failure and restore the file that is being updated, then restart the program that caused the failure and perform reprocessing. Please note that modifications to the program and processing of deleted data will be made at a later date after thorough consideration.

Problems to be Solved by the Invention The above-mentioned conventional software failure recovery processing requires support system engineers and operators to search for the cause of the failure from a huge list, resulting in recovery failures due to omissions in the search, and long search times. There is a problem that the system operation rate decreases because it takes time.

In addition, since lists from memory dumps, file dumps, etc. are converted into hexadecimal numbers and displayed, the person searching for the cause of the failure must have sufficient knowledge to interpret the contents of the list. There is also the problem that the number of workers who can be involved in the recovery process is limited.

Means for Solving the Problems of Structure of the Invention The automatic recovery processing system of the present invention that solves the problems of the prior art described above provides a method for deleting fault-inducing data in response to the occurrence of a software fault in a business processing job. A recovery control information file that stores recovery control information including how to restore and restart updated files, and input/output requests issued by business processing jobs, file access sequence numbers, input/output files being accessed, and records. An interface control means that records addresses and converts them into operating system input/output macro instructions while managing them by an interface control information recording means that records addresses, and a recovery control information file and The system is configured to refer to the interface control information recording means to delete failure-inducing data and restore update files, and to restart the business processing job in a method specified by the recovery control information. ing.

Hereinafter, the effects of the present invention will be explained in detail by way of examples.

Embodiment FIG. 1 is a functional block diagram showing an example of the configuration of an automatic recovery processing system of the present invention.

The business processing job 1 that is automatically restored by this system includes an input file 2 that stores data input to the business processing job 1, an update file 3 that is updated by the business processing job 1,
The business processing job is executed using the output file 4 that stores the output data of the business processing job 1.

The interface control unit 10 configuring the automatic recovery processing system for the business processing job 1 includes:
It includes an interface control information table 100 that stores file access sequence numbers (the number of input/output requests made by business processing job 1) and addresses on the storage medium of input/output records, and handles input/output requests made by business processing job 1. , while referring to the interface control information table 100 and converting the input/output requests into input/output macro commands of the operating system (hereinafter referred to as "OS") and passing them to the OS.

Reference numeral 20 denotes a job control unit, which stores recovery control information (data information to be deleted as failure-inducing data,
Recovery control information file 20 that records information on how to restore updated files, how to restart business processing jobs, etc.
0, and when a software failure occurs, the failure-inducing data is determined by referring to the recovery control information file 200 and the interface control information table 100, and the failure-occurring business processing job is restarted after recovery.

Reference numeral 30 denotes an automatic operation control unit that refers to the recovery control information file 200 and requests recovery processing while securing resources for recovery processing, and also changes resources and handles failure-occurring operations after recovery is completed. Requests restart of processing job.

Reference numerals 40 and 50 denote a data deletion processing unit and a file restoration processing unit, respectively, which delete failure-inducing data and restore files that are being updated, respectively, in response to the above restoration processing request from the automatic operation control unit.

A recovery control information file 200 for recording in advance recovery control information set by an operator according to the business type before executing a business processing job in order to enable automatic recovery processing when a software failure occurs.
will be provided. 2 to 4 illustrate three types of recovery control information recorded in the control information file 200.

FIG. 2 is a conceptual diagram showing an example of the structure of job control information 201, which is created for each business processing job. The job control information 201 includes (1) the business processing job name, (2) the business processing program name used in the business processing job, (3) the format for restarting the failed business processing program after recovery, and (4) the restart format. Storage location of checkpoint information that is expanded when is a checkpoint restart (5) Checkpoint number to be restarted (6) Processing form of data that can be deleted when a software failure occurs, and (7) Contains information for each type of data deletion processing.

The data deletion processing format (6) is set by the operator in the recovery control information file 200 before starting a business processing job, from among the following four types, depending on the operation of the business.

Form 1: Data is not deleted.

Type 2: Delete records of recently accessed input files.

Form 3: When there are multiple input files, a record with the smallest key value is searched among the records currently being processed, and records with the same key from the input file are deleted.

Form 4: When there are multiple input files, specify the deletion processing file name and deletion key value, search for the deletion compatible file name and deletion key in the record currently being processed, and search for the same key value in the file. Delete the record you have.

FIG. 3 shows the recovery control information file 20 in FIG.
2 is a conceptual diagram showing an example of the configuration of used file information 202 recorded in 0. This used file information 202 is management information for all files used by each business processing job, and includes (1) business processing job name, (2) number of used files used by the business processing job, and (3) information on the above-mentioned files used. Internal file name (logical name) (4) External file name of the file used above (actual file) (5) Distinction whether the file used above is an input file, output file or update file, and (6) Used above Contains information necessary for file allocation. Note that the above information (3) to (6) is repeated for the number of files used.

FIG. 4 shows the recovery control information file 20 in FIG.
2 is a conceptual diagram showing an example of the configuration of file restoration processing information 203 recorded in 0. This is created for each updated file. This file restoration processing information 203 includes: (1) External file name of the updated file (2) Business process job name that uses the above updated file (3) Restoration processing type of the updated file (4) File restoration for restoring the updated file Job name (5) Name of the file used by the file restoration job to perform restoration processing (6) Media type of the updated file (disk, drum, tape, etc.) (7) Distinction of whether the updated file is a duplicated file and (8) Contains information necessary for allocating update file area.

In this embodiment, one of the following four types of update file restoration processing modes is used to restore the restoration control information file 20 before starting a business processing job.
Set to 0.

a. Reload the previously saved save file 5 onto the failed update file 3.

b. Repaint the update file that caused the problem with the pre-update information from the journal file.

However, the history of journal files is managed by the OS.

c Load the saved file into the updated file that caused the failure and repaint it with the updated information stored in the journal file 4.

d. Redo the file creation process.

FIG. 5 is a conceptual diagram showing an example of the configuration of the interface control information table 100 shown in FIG.
The interface control information table 100 is created by the interface control unit 10 for each business processing job.

As shown in FIG. 5, these interface control information tables include interface control processing control information 101 that records the number of inputs and outputs requested by the relevant business processing job, and information input by each task of the relevant business processing job. File access-parameter table (hereinafter referred to as "FAP table") 101-1, 102-2, ... 102-n that records the current access point of the file to be output.
It consists of

Furthermore, as shown in Figure 6, the above FAP table contains file access parameter information (hereinafter referred to as "FAP") for each file used by each task.
103-1, 103-2,...
It consists of 103-m.

FIG. 7 is a conceptual diagram showing an example of the configuration of the interface control processing control information 101 shown in FIG. This interface control processing control information 101
(1) File access sequence number (This is the number of OS input/output macro instructions issued by the interface control unit 10 to the OS based on the input/output request of business processing job 1, and a sequence number is assigned for each job. (2) Number of tasks currently being executed for business processing job 1 (=
(number of FAP tables) and (3) address list (list of starting addresses) of the FAP table that manages files used by each task being executed. The above address list is repeated as many times as there are tasks.

FIG. 8 shows the FAP table 102 shown in FIG.
FIG. 2 is a conceptual diagram showing an example of the configuration of FAP information 103 that manages . This FAP information 103 includes: (1) Internal file name of the file used by the task (business processing program) (2) External file name of the above file (3) Storage location within the buffer of the record currently being processed (latest access) (4) ) Record address on the storage medium currently being processed and (5) File access sequence number FIG. 9 is a flowchart showing an example of interface control processing by the interface control unit 10 shown in FIG.

When the interface control unit 10 receives an input/output request from the business processing job 1, it starts interface control processing. The interface control unit 10 first determines whether the parameters of the input/output request from the business processing job 1 are appropriate in step 11, and if they are inappropriate, sets an error code in step 18 and returns to the business processing job 1. If the parameters are correct, the process advances to step 12, where an OS input/output macro command corresponding to the above input/output request is issued.

The interface control unit 10 performs the next step 1.
3, it is determined whether the return code returned from the OS after issuing an input/output macro instruction to the OS is normal. If the return code is not normal (input/output has failed), proceed to step 18, set an error code, and return to business processing job 1.

If the return code is normal (input/output completed), the interface control unit 10 executes step 1.
4 and 15, the record storage position (3) and record address (4) are recorded in the FAP information 103, respectively. Next, in step 16, the interface control unit 10 counts up the file access serial number. Interface control unit 10
In the next step 17, the counted up file access sequence number is stored in the FAP information 103.
(5) and interface control processing control information 101
(1) and end the interface control processing.

FIG. 10 is a flowchart showing the entire process up to restart of the failed business processing job when a failure occurs in the business processing job 1 in the system of FIG.

Deletion data determination processing Deletion data determination unit 21 of job control unit 20
When a failure occurs in a business processing job, the system refers to the information in the recovery control information file 200 and the interface control information table 100 described above to determine data to be deleted.

Activation Process of Recovery Process The recovery process requesting unit 31 of the automatic operation control unit 30 receives the recovery process request from the job control unit 20.
Performs preprocessing necessary to start the recovery process, such as securing resources necessary for deleting data and restoring files.

Data deletion processing and. File restoration processing data deletion processing section 40 and file restoration processing section 50
Each of these performs data deletion processing and file restoration processing upon activation from the automatic operation control unit 30.

Restart request processing The restart request unit 32 of the automatic operation control unit 30
Upon receiving notification of the end of data deletion processing and file restoration processing, processing such as securing resources necessary for restarting business processing job 1 is performed.

Restart Process When the restart unit 22 of the job control unit 20 receives a restart command from the restart request unit 32 of the automatic operation control unit 30, it refers to the recovery control information file 200 and restarts the business processing job in the specified format. Restart 1.

Hereinafter, each part will be explained in detail using a flowchart according to the process flow shown in FIG.

Deletion Data Determination Process FIG. 11 is a flowchart showing an example of deletion data determination processing in the deletion data determination unit 21 of the job control unit 20 when a software failure occurs.

When the deletion data determination unit 21 receives a notification of abnormal termination by a business processing program or a notification of a program exception by the OS when a business processing program exception occurs, the deletion data determination unit 21 starts deletion data determination processing.

When the deletion data determination unit 21 starts the deletion data determination process, first in step 210 it analyzes the content of the notification regarding the corresponding business processing program and checks whether there is a software failure (logically inconsistent program exception). If the content of the notification is a software failure, the process advances to step 211 and all other business processing programs in this business processing job are forcibly terminated.

The deletion data determining unit 21 performs the next step 212.
At , the data deletion processing type information (6) is read from the job control information 201 of the recovery control information file 200, and the presence or absence of deleted data (whether it is type 1 or not) is checked. If there is no deleted data (Form 1)
), the process proceeds to step 215 without performing recovery processing. When the deletion data determination unit 21 determines in step 212 that deletion data exists (in the case of any of forms 2, 3, or 4), it proceeds to step 213, where it performs the following processing.

(b) Data deletion processing type 2 (deleting the latest access record in the latest access input file)
In the case of: The deletion data determining unit 21 determines the latest access input file and its latest access record by referring to the interface control processing control information 101, FAP information 103, and used file information 202,
Obtain the following information as deleted data information.

(1) External file name (actual file name) (2) Record address on the medium of the applicable record (for example, in the case of a disk, cylinder address,
track address, relative byte position within the block, etc.) (3) Job name (4) Data deletion processing format (b) In the case of data processing format 3 (deleting the input record with the minimum key value): Deletion data determination unit 21 is job control information 20
1 and used file information 202 and FAP information 10
3, determine the input record with the minimum key value among the records currently being processed in multiple input files, and delete data information consisting of (1) to (4) similar to the case (a) above. get.

(c) In the case of data deletion processing form 4 (deleting the record of the input file with the specified key value of the specified file): The deletion data determination unit 21 deletes the job control information 20
1 and used file information 202 and FAP information 10
3, determine the record with the specified key value among the records currently being processed for one or more files to be deleted, and perform steps (1) to (1) similar to information (a) and (b).
(4) Obtain the deleted data information consisting of:

However, if there is no record with the specified key value,
Abnormally terminate the business processing job as a job not subject to automatic reprocessing.

When the deletion data determining unit 21 determines the deletion data as described above, in the next step 214, it attaches the determined deletion data information and instructs the recovery activation unit 31 of the automatic operation control unit 30 to start the recovery process. request.

Furthermore, the deletion data determining unit 21 performs step 21.
In step 5, all business processing jobs including the business processing program in which the software failure has occurred are abnormally terminated, and all deletion data determination processing is terminated.

Activation Process of Recovery Process FIG. 12 is a flowchart showing an example of the operation of the recovery process activation unit 31 within the automatic operation control unit 30.

When the recovery process starting unit 31 receives a recovery process request from the deleted data determining unit 21 of the job control unit 20, it starts the process of starting the recovery process. That is, the recovery processing starting unit 31 first performs step 310.
In this step, resources necessary for the data deletion process are secured based on the deletion data information passed from the deletion data determining unit 21 and the file restoration processing information 203 in the restoration control information file 200. Next, in step 31, the recovery processing starting unit 31 adds the above-mentioned deleted data information to the data deletion processing unit 4.
0 to request data deletion processing.

Furthermore, the recovery processing starting unit 31 performs step 312.
At this point, one or more updated files that require restoration are determined by referring to the used file information 202 and file restoration processing information 203.

Next, the recovery processing starting unit 31 performs step 313.
In step 314, after securing the resources necessary for file restoration processing for one of the updated files to be restored, a request is made to the file restoration processing unit 50 to perform file restoration processing, with information regarding file restoration attached.

In step 315, the restoration processing starting unit 31 determines whether the issuance of file restoration processing requests has been completed for all updated files to be restored, and if it has not been completed, the process returns to step 313.
Repeat the above process for the next update file.

Data Deletion Processing FIG. 13 is a flowchart showing an example of data deletion processing in the data deletion processing section 40. The data deletion processing section 40 includes the recovery processing starting section 3
When the data deletion processing request from 1 is received, the data deletion processing is started.

The data deletion processing unit 40 performs steps 41 and 42.
In accordance with the deletion data information received from the recovery process requesting section 31, deletion flags are set for all of the relevant data in the input file, thereby deleting all of the relevant data.

File Restoration Processing FIG. 14 is a flowchart showing an example of updated file restoration processing in the file restoration processing section 50.

The file restoration processing section 50 is a restoration processing starting section 3.
When a file restoration request is received from 1, file restoration processing is started.

In step 51, the file restoration process 50 performs any of the steps a to d already explained with respect to the file restoration process information 203 in FIG. In this manner, the restoration process for the updated file that caused the failure is performed.

Restart Request Processing FIG. 15 is a flowchart showing an example of the operation of the restart requesting section 32 of the automatic operation control section 30.

When the restart requesting unit 32 receives notification of the end of processing from the data deletion processing unit 40 and the file restoration processing unit 50, it starts restart request processing. In step 320, the restart request unit 32 changes the resources used upon completion of the data deletion processing and file restoration processing, and in the next step 321, restarts the business processing job that has caused the software failure. A request is made to the restart unit 22 of the control unit 20.

Restart Processing FIG. 16 shows the restart unit 22 of the job control unit 20.
2 is a flowchart illustrating an example of restart processing for a faulty business processing job according to the method of FIG.

When the restart unit 22 receives a restart processing request from the restart request unit 32 of the automatic operation control unit 30,
Start the reboot process. That is, the restart unit 22
First, in step 220, the job reprocessing mode (3) of the business processing job is determined by referring to the job control information 201. If this reprocessing mode is not a checkpoint restart but a rerun, the business processing job is restarted as normal processing in step 221.

If the reprocessing mode is checkpoint restart, the restart unit 22 proceeds to step 222;
The checkpoint information storage area name (4) and checkpoint serial number (5) of the job control information 201 in the recovery control information file 200 are obtained and expanded.

Next, in step 223, the restart unit 22 restarts the corresponding business processing job from the corresponding checkpoint point.

In this way, all automatic recovery processing in the event of a software failure is completed.

For business processing jobs that terminate abnormally when the corresponding record does not exist in the case of data deletion processing type 1 or data deletion processing type 4, the operator must use an appropriate method to determine the cause of the failure and remove it. Then, the operator restarts it.

Automatic operation control unit 30, data deletion processing unit 40,
The file restoration processing section 50 may be of job format.

Effects of the Invention As explained in detail above, the automatic recovery processing system of the present invention deals with the occurrence of software failure in a business processing job, and sets recovery control information suitable for the business processing type in advance. Since the configuration is such that data to be deleted is automatically determined by managing the file usage status of the business processing job in the interface control unit, it is possible to automatically perform recovery processing in response to a failure.

For this reason, as in the conventional example, support SE
This eliminates the need for operators and operators to search for the cause of failures from a huge list, eliminates the problems of failures in recovery due to omissions in the search, and the long time it takes to search, which greatly improves system availability.
The load on the operator can be reduced.

[Brief explanation of drawings]

FIG. 1 is a functional block diagram showing the configuration of an embodiment of the present invention, FIGS. 2 to 4 are conceptual diagrams showing an example of the configuration of various control information in the control information file 200 of FIG. 1, and FIG. 8 are conceptual diagrams showing examples of the configuration of various information in the interface control information table 100 shown in FIG. 1, and FIG. 9 is a flowchart illustrating an example of the operation of the interface control unit 10 shown in FIG. 10 is a flowchart illustrating an example of the operation of the entire system in FIG. 1, FIG. 11 is a flowchart illustrating an example of the operation of the deletion data determination block 21 in FIG. 1, and FIG. 1. FIG. 13 is a flowchart for explaining an example of the operation of the data deletion processing section 40 in FIG. 1. FIG. 14 is a flowchart for explaining an example of the operation of the data deletion processing section 40 in FIG. 15 is a flowchart for explaining an example of the operation of the file restoration processing section 50 shown in FIG. 1, and FIG.
FIG. 16 is a flowchart for explaining an example of the operation of the restart unit 22 from FIG. 1. DESCRIPTION OF SYMBOLS 10...Interface control part, 20...Job control part, 30...Automatic operation control part, 40...Data deletion processing part, 50...File restoration processing part, 100...Interface control information table, 200...Recovery control information file.

Claims (1)

[Scope of Claims] 1. Recovery control in which recovery control information including a method for deleting failure-inducing data, a method for restoring updated files, and a restart method is stored in advance in response to the occurrence of a software failure in a business processing job. The information file and the input/output requests issued by the business processing job are managed by an interface control information recording means that records the file access sequence number, the input/output file being accessed, and the record address, while the operating system an interface control means that converts the data into input/output macro commands, and when a software failure occurs in a business processing job, refers to the recovery control information file and the interface control information recording means to delete the failure-inducing data and update the update file. Along with the restoration,
An automatic recovery processing system for a software failure, comprising means for restarting the business processing job in a method specified by the recovery control information.