JPH05313977A - File restoring method - Google Patents

Abstract

PURPOSE:To provide a file restoring method capable of quickly performing the restoration of a file and evading the lowering of throughput. CONSTITUTION:A main memory device 3a which stores access history information T while a system is being operated, and restoration utility UT which performs the analysis of the access history information T on the main memory device 3a when a fault occurs are provided. The restoration utility UT specifies updated file data out of file data in a file group Fd based on the content of the access history information T in processing when the fault occurs, and returns the file data to the one before update.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a file recovery method capable of automatically recovering a file prepared as a resource of a computer system when a failure occurs.

[0002]

2. Description of the Related Art As a resource of a computer system, for example, a file that is referenced or updated during processing can be cited. This file is composed of file data stored in a secondary storage device such as a magnetic disk device.

FIG. 2 shows a block diagram of a general computer system. Computer system 1 shown
The processor 2 includes a processor 2, a main memory 3, a secondary memory 4, a communication controller 5, a keyboard 6, a display 7, and a printer 8 connected to the processor 2.

The processor 2 refers to an operating system and an application program to carry out predetermined arithmetic processing and control of each part constituting the computer system 1 in accordance with this arithmetic processing. The main memory 3 is an RA for storing programs and data referenced by the processor 2, and further calculation results of the processor 2 and the like.
It is composed of a memory element such as M. The secondary storage device 4 is composed of a magnetic storage device or the like that stores an operating system, application programs, and the like referenced by the processor 2.

The communication control device 5 carries out protocol processing for realizing communication with a host computer (not shown). The keyboard 6 is the processor 2
It is an input device equipped with character and numerical keys for inputting commands and data to. The display 7 is a display device that displays commands and input data input by the keyboard 6, and further displays a calculation result by the processor 2. The printer 8 is a printing device that prints the calculation result of the processor 2 and the like.

In the computer system 1 having the above-mentioned configuration, for example, a database file is stored in the secondary storage device 4, and the processor 2 executes arithmetic processing for updating the file data constituting the file as needed. It shall be. As an example of a database, assuming that a financial institution manages customer's transaction details, files are provided for each customer, and file data are transaction details and balances.

For example, a transfer transaction can be mentioned as one of transactions in a financial institution. Here, the transfer transaction will be described with reference to FIG. FIG. 3 is an explanatory diagram of a transfer transaction. The figure shows a file group F (files F1, F2, F3) stored in the secondary storage device 4.

When performing a transfer transaction of 300,000 yen from the file F1 to the file F3, the processor 2 first performs a subtraction process (payment transaction) of 300,000 yen on the file F1. Further, when the subtraction process on the file F1 is completed, an addition process (deposit transaction) of 300,000 yen is executed on the file F3. When the addition process for the file F3 is completed, the processor 2 determines that the transfer transaction is completed, and executes another new transaction process, for example.

By the way, it is assumed that a failure occurs in the operation process of the computer system 1 and a restart (recovery) must be performed. For example, it is assumed that the processor 2 is executing a transfer transaction and a failure occurs at the time when the subtraction process for the file F1 is completed. If the system is restarted in this case, the processor 2 will carry out a new transaction process without recognizing that a failure has occurred during execution of the transfer transaction.

That is, a new operation is started in a state where the addition process (deposit transaction) for the file F3 is not executed and only the subtraction process (payment transaction) for the file F1 is executed. As described above, if a failure occurs before the transaction processing is completed, the reliability of the file is significantly reduced. Therefore, when the failure occurs, a measure is taken to restore the file group F to the state before the failure occurred.

FIG. 4 shows a first explanatory view of a conventional file restoration method. The figure shows the processing relating to the secondary storage device 4 (file group F) in the operation process of the computer system 1 (processor 2). The processor 2 backs up the storage contents of the secondary storage device 4 to, for example, the magnetic tape TP before actually executing the transaction processing. After that, the operation of the transaction process involving the process of updating the file data of the file group F and generating a new file group Fa is started. If a failure occurs during operation and the system goes down, the processor 2 transfers the contents of the magnetic tape TP to the secondary storage device 4 and restores the file group F before restarting the operation. The processor 2 will construct a new file group Fb based on the restored file group F and carry out re-operation. As described above, the method of recovering the file group F by making a backup on the magnetic tape TP and preparing for a failure is applied when the update frequency of the file group F is low.

Next, a method of recovering a file when the frequency of updating the file group F is high, that is, when the computer system 1 is installed in a financial institution or the like as described in FIGS. 2 and 3, will be described. FIG. 5 is a second explanatory diagram related to the conventional file recovery method. Similar to FIG. 4, the figure shows a process regarding the secondary storage device 4 (file group F) in the operation process of the computer system 1 (processor 2).

When the processor 2 recognizes the start of the transaction processing accompanied by the update of the file data of the file group F, the processor 2 indicates the processing content of the access processing executed to update the file group F with the transaction processing. The specified log data D is generated and written in the log file Fc1. For example, in the case of the subtraction process of subtracting 300,000 yen from the contents of the file F1 as described in FIG. 3, the log data D indicates the beginning of a series of processes and indicates that the access target is the file F1. It is composed of one item D1, a second item D2 indicating the file data of the file F1 before updating, and a third item D3 indicating the file data of the file F1 after updating. When updating the file data, the processor 2 once reads the file data to be updated and grasps the contents before the update.

When the log data D is written in the log file Fc1, the processor 2 updates the file F1 of the file group F and constructs a new file group Fc.
If a failure occurs during operation and the system goes down, the processor 2 checks the log file Fc1 before restarting operation.
Of the access (processing of the log data D in which the first item D1 indicating the end of the processing does not exist) that was being processed at the time of occurrence of the failure, To do. That is, if it has already been updated, it is returned to the pre-update data. As a result, the transaction F is not completed and the file F1 whose file data has been updated can be returned to the state before being updated.

[0015]

DISCLOSURE OF THE INVENTION Magnetic tape TP
In case of backing up and restoring to a secondary storage device 4 and the like, the storage capacity of the secondary storage device 4 is increased, and the backup time and the time to transfer to the secondary storage device 4 for restoration are very long and the restoration is quick. There was a problem that it was difficult to do. Further, when the log file Fc1 is constructed during operation, the processing time becomes longer than that in the case where only the file is simply updated because the secondary storage device 4 is additionally accessed. Fc
As compared with the case where 1 is not constructed, there is a problem that the processing capacity is significantly reduced. The present invention has been made in view of the above points, and an object of the present invention is to provide a file recovery method that can quickly recover a file and can avoid a decrease in processing capacity.

[0016]

A file recovery method of the present invention stores access history information indicating a history of access processing including processing for updating file data forming a file in a main storage device, and a failure occurs. At this time, the access history information is extracted from the storage content of the main storage device, the updated file data is specified based on the access history information, and the updated file data is set to the content before update. Restore.

[0017]

In this method, when the failure occurs, the access history information stored in the main storage device and indicating the access contents is recognized along with the file access processing. Based on the access history information, if there is file data that was related to the process that was being processed at the time of the failure and was already updated, restore this file data to the contents before the update based on the access history information. Restore the file. The access history information is information that is always stored in the main storage device with file access. Therefore, it is possible to easily realize file recovery in the event of a failure without performing special processing in advance assuming file recovery.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is realized by the computer system 1 described above with reference to FIG. Therefore, the computer system 1 will be taken as an example here. FIG. 1 shows an explanatory diagram relating to the file restoration method of the present invention. When updating the file data of the file group F, the processor 2 reads the file data before the update, then writes new contents as the file data, and completes the update of the file data.

In general, the processor 2 includes a secondary storage device 4
The data read from and the data to be written are read in a predetermined area (for example, a cache area) of the main storage device 3 once with the task information T1 indicating the identification and access type (read or write) of the accessed file. Access history information composed of a pair of write data (file data T2) is written.

This access history information is erased at a fixed timing, for example, when the area is full or when a series of processing is completed. Further, the contents indicating the beginning and the end are added to the access type so that the beginning and the end of the series of processes can be analyzed.

When a failure occurs during operation and the system goes down, the processor 2 writes the contents of the main memory 3a (main memory 3 in which the access history information T is written) before restarting the operation. The process of transferring to the secondary storage device is performed.
Here, the list of data stored in the main storage device 3a is called a post-mortem dump PD, and the process of transferring this post-mortem dump PD to the secondary storage device 4 is referred to as acquisition of the post-mortem dump PD. I will call it.

In the processor 2, a recovery utility UT that analyzes the post-mortem dump PD is activated,
Post-mortem dump PD stored in the secondary storage device 4
From this, the access history information T is recognized. After that, the recovery utility UT specifies the file data that has been updated without completing the processing based on the task information T1 of the access history information T, and updates the file data on the file group Fd at the time of the failure before the update. File group Fe that has been restored to the contents and has no updated file data due to incomplete processing
To build.

FIG. 6 shows a flowchart according to the present invention. The figure illustrates the processing when a failure occurs. When the recovery utility of the processor 2 acquires the post-mortem dump (step S1), it extracts the access history information PD (step S2). After that, the file data of the secondary storage device 4 indicated by the access history information PD and related to the processing being executed at the time of failure is specified (step S3). After that, the specified file data is restored to the contents before the update (step S4), and the process is ended.

The present invention is not limited to the above embodiments. In the embodiment, the case where the post-mortem dump PD is transferred to the secondary storage device 4 has been described as an example, but for example, the access history information T is directly accessed from the main storage device 3 without performing the transfer.
You may recognize.

[0025]

As described above, according to the file restoration method of the present invention, it is not necessary to transfer a large amount of data before the start of operation and at the time of restoration for file restoration.
Furthermore, since it is not necessary to collect special information for file recovery during operation, rapid recovery can be realized, and it is possible to avoid a situation where processing capacity is lowered due to the assumption of recovery.

[Brief description of drawings]

FIG. 1 is an explanatory diagram relating to a file restoration method of the present invention.

FIG. 2 is a block diagram of a general computer system.

FIG. 3 is an explanatory diagram of a transfer transaction.

FIG. 4 is a first explanatory diagram relating to a conventional file restoration method.

FIG. 5 is a second explanatory diagram related to a conventional file restoration method.

FIG. 6 is a flowchart according to the present invention.

[Explanation of symbols]

 2 processor 3, 3a main memory F, Fd, Fe file group T access history information

Claims (1)

[Claims] 1. Access history information indicating a history of access processing including processing for updating file data constituting a file is stored in a main storage device, and when a failure occurs, the access contents are stored in the main storage device according to A file recovery method comprising extracting access history information, identifying the updated file data based on the access history information, and recovering the updated file data to the contents before the update.