JPH06348574A - Data managing method for mltiplexed file - Google Patents

Abstract

PURPOSE:To restore file data by the managing method for the multiplexed file that a computer utilizes when a temporary failure occurs in one file and restoration is done. CONSTITUTION:Plural files wherein the same contents should be registered are connected to one computer. Those files are called the multiplexed file. The managing method for the utilization state of the file is prepared and this managing method manages the utilization states of one multiplexed file with a matrix of two ros by (n) columns (n: multiplexity) when the management of the multiplexed file is specified. This managing method restores the file data when a temporary fault occurs in one file of the multiplexed file and restoration is performed, and consists of four procedures at the time of the start of application software, restoring procedure execution, shutdown execution, and file closing and resetting procedure as time points of the restoration.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a procedure for recovering a data unit in a file or an entire file when a failure is detected in any of the multiplexed files.

[0002]

2. Description of the Related Art As a known example of the prior art, Japanese Patent Laid-Open No. 62-2
60239 is known.

In this publicly known example, immediate recovery is performed by recovering a defective portion from the data of the file that can be normally processed by the input / output operation of the duplicated file, and the reliability of the duplicated file is deteriorated. It is to prevent it.

Regarding the prior art,
When an error occurs in a file being used, a method of copying a normal file to the file in which the error occurred and performing error recovery after the work using the file ends.

[0005]

SUMMARY OF THE INVENTION The above-mentioned conventional data management method for multiplexed files is disclosed in Japanese Patent Laid-Open No. 62-26023.
The method of 9 has a drawback that the file cannot deal with the multiplexing of more than triple, because only the duplicated file is considered.

Further, in the method of the prior art,
The recovery procedure for a failed file is when all file processing is completed and a recovery command is issued. Therefore, if the operating time of the computer system increases, the correct data in the file will decrease. Therefore, there is a drawback that the reliability of the system is deteriorated.

[0007]

SUMMARY OF THE INVENTION It is an object of the present invention to recover a file when the data in the multiplexed file is used in addition to the above-mentioned method when the mismatch of the multiplexed files occurs due to a temporary file failure. This is to improve the reliability of file data by performing a recovery process on the data accessed to reduce the inconsistency of the data in the file.

Therefore, a data management method for a multiplexed file according to the present invention is a multiplexing method for a computer system having two or more multiplexed files and an input / output control unit for controlling input / output of the multiplexed files. A method for managing data of an encrypted file, wherein an input / output control unit performs an input / output process on the multiplexed file and a recovery process when a failure occurs in any of the multiplexed files. And a matrix of 2 rows and n columns (n; the number of multiplexed files) is used as a criterion for determining the execution status of the input processing of the multiplexed file and the execution status of the restoration processing. There is.

[0009]

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

FIG. 1 is a diagram showing a matrix of 2 rows and n columns (n: multiplicity) used in an embodiment of a data management method of a multiplexed file according to the present invention, and FIGS. 2 to 8 show the matrix of FIG. It is a flowchart of the process performed using.

First, a matrix of 2 rows and n columns (n: multiplicity) used in this embodiment will be described with reference to FIG.

As shown in FIG. 1, in the 2nd row and the nth column (n: multiplicity), the first row shows the usage status of the file at the present time, 2
The line shows the usage status from the start of system operation.

The value in the first line is represented by ai, and this value represents either normal, fault, or occlusion. The value on the second line is represented by bi, and this value caused normal / temporary failure, but during this time no update occurred in the normal system.
Indicates that the normal system has been updated while the failure occurred.

Next, the case of activating an application will be described with reference to the drawings.

2 to 4 are flow charts showing the file management processing when the application is executed using the matrix of FIG.

When the application is started (step 21), if the file used by the application has a multiplexed file, or if all the multiplexed files can be accessed, the record READ in the file, all the multiplexed files WRITE is performed and the process ends (steps 231, 232, 233, 234). When some of the multiplexed files cannot be accessed, the matrix is rewritten by the records READ and WRITE in the file (steps 241, 242, 2421, 24).
22, 2423, 2424). If the management status of the multiplexed file to be used by this application is not 0 matrix at the time of starting the application, the procedure of FIGS. 3 and 4 is performed.

Next, the case of performing file restoration will be described with reference to the drawings.

FIGS. 5 and 6 are flow charts showing the recovery processing when a file failure occurs using the matrix of FIG.

A multiplexing file to be restored is designated by a job control language or the like (step 51), an argument i is set to 1 so that the first column of the management table can be checked (step 52), and the designated multiplexing is performed. The management table for the file is checked (step 53), and in the step 53, for the argument bi corresponding to the i-th file, bi ≠
When it is 0, file recovery described later is performed. After the file is restored, 0 is set in the argument of the i-th column (step 54),
Add 1 to the argument (step 55). Step 5 above
When the value of the 2nd row and i-th column (= bi) in 3 is not 0, 1 is set to the argument j for searching for a normal file (step 61), and the argument corresponding to the j-th file is set. , Bj = 0, the j-th file which is the normal system is copied to the i-th file which is the abnormal system (step 6).
3), when bj ≠ 0, 1 is added to the argument j (step 62). Step 64 sets 1 to the argument k when a set of multiplexed files has all been in a failure state regardless of whether or not there is a file update. For the argument bk corresponding to the kth file, b
If k = 1 (although there was a failure, but other normal files were not updated during this period), the kth file is changed to the i-th file which is the abnormal system. Copy (step 67), and if bk ≠ 1, argument k
Is added to 1 (step 65). In step 66, if all of the set of multiplexed files are abnormal, a check is performed for each multiplexed file on a record-by-record basis, and if there is a mismatch, all are output in a dump.
It also closes files.

Next, file recovery at the time of executing shutdown will be described with reference to the drawings.

FIG. 7 is a flow chart showing a recovery process at the time of executing shutdown using the matrix of FIG.

This is based on the restoration procedure of FIG. 5 and FIG.
Except for the specification of the multiplexed file for performing the recovery procedure in step 51, the recovery procedure is unconditionally performed for all the multiplexed files to be managed.

Next, the case of releasing the blocking of the multiplexed file blocked by the file sub-serving will be described with reference to the drawings.

FIG. 8 is a flowchart showing the block release processing when the file is in the block state using the matrix of FIG.

A job control language or the like is used to describe or specify a correct record-by-record multiplexed file to be restored (step 81), replace the specified record (step 82), and release the blockage. In the case, the records of the multiplexed file are checked one by one (step 83), and when all the multiplexed files match, the file blocking is released (step 8).
Along with 4), the management table T is also set to 0 (step 8)
5). If the records in the multiplexed file do not match, dump output is performed for these records (step 86).

[0026]

As described above, according to the data management method of the multiplexed file of the present invention, the utilization status of the multiplexed file is managed by using the matrix, so that the management of the failure status can be facilitated. Has the effect.

Further, in the method of managing data of a multiplexed file according to the present invention, when the matrix means that the files do not coincide with each other, the record accessed in the execution of the application using this file is replaced with the record of the normal system of the failure system. By copying the file, it is possible to improve the reliability of the file and simplify the failure recovery.

[Brief description of drawings]

FIG. 1 is a diagram showing a matrix of 2 rows and n columns (n: multiplicity) used in an embodiment of a data management method of a multiplexed file according to the present invention.

FIG. 2 is a flowchart showing file management processing when an application is executed using the matrix of FIG.

FIG. 3 is a flowchart showing file management processing when an application is executed using the matrix of FIG.

FIG. 4 is a flowchart showing a file management process when an application is executed using the matrix of FIG.

5 is a flow chart showing a recovery process when a file failure occurs using the matrix of FIG. 1. FIG.

FIG. 6 is a flowchart showing a recovery process when a file failure occurs using the matrix of FIG.

FIG. 7 is a flowchart showing a recovery process at the time of executing shutdown using the matrix of FIG.

FIG. 8 is a flow chart showing blocking release processing when a file is blocked using the matrix of FIG. 1;

Claims (3)

[Claims] 1. A method for managing data of a multiplexed file in a computer system, comprising: two or more multiplexed files; and an input / output control unit for controlling input / output of the multiplexed file. The control unit has means for performing input / output processing on the multiplexed file, and means for performing recovery processing when a failure occurs in any of the multiplexed files, A data management method for a multiplexed file, characterized in that a matrix of 2 rows and n columns (n; the number of multiplexed files) is used as a criterion for determining the execution status of the file input processing and the recovery status. . 2. The input / output control unit has means for managing the multiplexed file, and has a matrix of 2 rows and n columns (n: multiplicity) for each set of multiplexed files. The data management method for a multiplexed file according to claim 1, wherein the usage status of the file is managed by a matrix. 3. When any of the multiplexed files has a failure, the input / output control unit performs input / output processing except for this file, recognizes the failure status, and recognizes the failure status of the multiplexed file. When it is recognized that there may be inconsistency in any of the above, the procedure to copy each file that accessed the abnormal file from the normal file and the normal operation for the abnormal file after the work using the file is completed A recovery procedure for copying a system file, a procedure for performing a recovery process according to the recovery procedure at the time of shutting down the computer system, and a data recovery procedure for recognizing that there may be a mismatch for all the multiplexed files. The data management method for a multiplexed file according to claim 2, further comprising: