JPH09244907A - Data restoration system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently perform a data restoration processing in the information processor of a small scale whose storage area is relatively small and processing speed is slow. SOLUTION: When the restoration of a processing by the control of an OS 4 becomes impossible by a fault, data inside a real memory 2 are dumped to an external storage device 3. At the time of reactivation, the management information of a page directory and a page table or the like used for managing the respective data inside the real memory 2 in sequential relation is read from a memory dump file 31, the respective data are sequentially called by relating data units from the memory dump file 31 based on the management information and rearranged and an aligned file 32 is generated. A data group (user data) processed by an APP 5 until the generation of the fault is extracted from the aligned file 32 and delivered to the reactivated APP 5 and the user data are edited to a file form corresponding to the present APP 5 and restored by the APP 5.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for recovering data (user data) edited on an actual memory of an application program when a failure occurs in an information processing apparatus, and more particularly to a personal computer, a workstation or the like. The present invention relates to a data recovery system suitable for efficiently recovering user data in a small-scale information processing device.

[0002]

2. Description of the Related Art In an information processing apparatus, when the operation of an operating system is stopped due to a failure, an application program cannot store the data (user data) edited on the memory until then in an external storage device. As conventional techniques for dealing with such a problem and recovering user data edited until immediately before the occurrence of a failure, for example, the following three techniques are disclosed as described in Japanese Patent Laid-Open No. 3-252825. It's being used.

First technique: The contents of the body of the data before the update are temporarily saved during processing, the information is taken out from the log before the update, and the body of the data is restored. Second technique: The updated data is temporarily saved during the processing, and the updated data is used to recover the data body again. Third technology: Instead of saving the pre-update log / post-update log in an area unrelated to the data to be updated as in the first and second technologies, the data area is pair-structured and managed by a pointer. When the update is interrupted, the data is recovered only by updating the pointer.

However, in any of these conventional techniques, it is necessary to save the pre-update log or the post-update log in another storage device or work area or in a pair structure with the update target data in advance. Therefore, an extra area for saving these log data is required in addition to the update target data. Further, in the first and second techniques, in the recovery process after the update is interrupted, the log data has to be read and the data area must be updated again, which requires a long time for the update process and also during the recovery process. It takes a lot of time because it needs to read and write data. For this reason,
In particular, it cannot be efficiently applied to a small-scale information processing device such as a personal computer or a workstation.

[0005]

The problems to be solved by the prior art are that the conventional technique requires an extra area for saving the pre-update log or the post-update log in advance, and that after the update is interrupted. In the recovery process, the log data must be read and the data area must be updated again, which requires a lot of time because the update process takes time and data must be read and written during the recovery process. It is a point.
An object of the present invention is to solve these problems of the prior art and provide a data recovery system capable of efficiently performing data recovery processing in a small-scale information processing device having a relatively small storage area and a slow processing speed. That is.

[0006]

In order to achieve the above object, a data recovery system of the present invention is a data recovery system which is edited on an actual memory of an application program when a failure occurs in an information processing apparatus. The memory dump processing unit 7 that dumps the data in the real memory 2 to the external storage device 3 to generate the memory dump file 31 when a failure occurs, and the memory dump file 31 when restarting.
A management information reading unit 8 for reading the management information (page directory 21, page table 22, chain information) used for sequentially managing the data in the real memory 2 in association with each other, and the management information reading unit 8 On the basis of the management information read by, the memory dump file 31 sequentially calls and rearranges each data in a related data unit and rearranges them to generate an alignment file 32. From the alignment file 32,
A cutout unit 1 that extracts a data group (user data) that has been edited by the application program 5 up to the occurrence of a failure and passes it to the restarted application program 5.
2 is included, and the restarted application program 5 edits the data into a file format corresponding to the own application program 5 and recovers the data.

[0007]

BEST MODE FOR CARRYING OUT THE INVENTION In a small-scale information processing apparatus, when a system cannot be recovered due to a failure due to control of an operating system and the system is stopped, an application program is edited on a memory until the system is stopped. The transient data cannot be stored in the external storage device. In the present invention, by adopting the configuration as shown in FIG. 1, the memory area of the stopped system is dumped to the external storage device as a memory dump (a routine for outputting the stored contents in the actual memory), and the memory dump file is stored. To recover the excessive data (user data) edited by the application program. 1 to 3 below
Will be used to provide a detailed description of the present invention.

FIG. 1 is a block diagram showing a position embodiment of the configuration according to the present invention of the data recovery system of the present invention.
In FIG. 1, 1 is a CPU (Central Processing Uni
t, a central processing unit) for performing various processes according to a storage program method.
A real memory such as an ess memory), 3 is an HDD (Hard D)
is an external storage device such as an isk Drive).

The processing device 1 is an operating system (hereinafter referred to as OS) that controls the operation of the entire system.
4, an application program (hereinafter referred to as APP) 5 that performs specific processing control on the OS 4, and a data recovery processing unit 6 that performs data recovery processing according to the present invention.
Is provided. In addition, the data recovery processing unit 6 includes
Memory dump processing unit 7, management information reading unit 8, alignment processing unit 9, compression unit 10, compression / decompression unit 11, cutout unit 12
The data recovery processing unit 6 is provided below.
The data recovery process according to the present invention will be described based on the description of each of the constituent elements.

OS4 and APP5 are stopped due to a failure, and
Due to the control of S4, it becomes impossible to recover the processing.
When P5 can no longer store the user data edited in the real memory 2, the memory dump processing unit 7 dumps all the data in the real memory 2 to the external storage device 3 and stores the data in the external storage device 3. The dump file 31 is generated. Originally, the data including one file to be processed by APP5, including the APP5 main body, is not physically stored in the memory 2 in a physically sequential manner, so the memory dump file 31 is also the same. It is a storage state of various data.

That is, each APP 5 is executed by multitasking, and in order to operate in the real memory 2 having a limited capacity, the OS 4 manages the page directory 21 and the page table 22 shown in FIG. By the paging function using information, the APP5 main body and processing data (user data) are stored on the real memory 2 in many pages (a1 to a1).
a4, b1 to b4, c1 to c4, ...). And, if necessary, APP in page units
Move 5 to make room for the page.

If the page cannot be freed even if the OS 4 is moved, the OS 4 pages out the page a1, for example, to the external storage device 3. When the program or data of the page a1 is required, the OS 4 pages in from the external storage device 3 and passes it to the APP 5. In this way, the OS
4 sequentially manages a set of related data stored in the real memory 2 in a distributed manner by generating predetermined management information, and sorts such non-sequential data as if it were sequential processing. do it,
Passed to APP5.

Since the memory dump file 31 also stores the management information used to sequentially manage the respective data in the real memory 2 in this manner, the management information reading unit 8 has a failure. At the time of restarting after being released, the management information (page directory 21 and page table 22 in FIG. 2) used to manage each data in the real memory 2 in this order from the memory dump file 31 in a sequential manner is managed. (Chain information of) is read.

Based on the management information read by the management information reading unit 8, the alignment processing unit 9 sequentially calls and rearranges each data from the memory dump file 31 in related data units to generate the alignment file 32. To do. For example, the alignment processing unit 9 performs
Based on the chain information of the page table 22 in FIG. 2, from the memory dump file 31, the memory space in which APP5 is operating, which is divided into pages on the real memory 2,
That is, the APP area having the user data to be recovered is rearranged in a sequential manner, and the alignment file 32 is generated.

That is, the alignment processing section 9 restarts A
The PP name of the user data to be recovered and the data format on the memory shown in FIG. 3 are received from PP5, and the memory dump file 31 is a character string based on the chain information of the page table 22 in the order indicated by the page directory 21 of FIG. It searches and finds the page directory 23 of the target APP5 of the received APP name, for example, “b”. Then, the area of the page directory "b" found is
Chain information “b1, b2, b” of the page table 22
3, b4 ”, and rearranged sequentially, and stored in the external storage device 3 as the aligned file 32.

The compression unit 10 compresses the alignment file 32 to generate a compression alignment file 32a, and the compression / decompression unit 11 decompresses the compressed compression alignment file 32a to restore the alignment file 32. The compression processing by the compression unit 10 and the decompression processing by the compression / decompression unit 11 take into consideration the efficiency of data movement and the efficiency of media storage when transferring a file to another device in failure analysis or the like. , To prevent a large capacity.

The alignment file 32 contains not only the user data but also the APP5 main body, and the cutout unit 1
2 is APP5 from the alignment file 32 to the occurrence of a failure.
APP5 restarted after extracting the data group processed by
Pass to. That is, the cutout unit 12 uses the restarted AP.
From P5, the APP name of the user data to be recovered, and FIG.
The data format on the real memory shown in
From the name and the data format on the real memory 2 shown in FIG. 3, a character string search is performed for the expanded file in the external storage device 5, the beginning of the user data section is found, and the subsequent data group is APP.
Pass to 5. Then, the restarted APP5 makes the received data group correspond to its own APP5, that is, its own APP5.
3 is stored in the external storage device 3 as a recovery data file 33.

As described above, in the data recovery system of this embodiment, the log data is not always saved as in the conventional data recovery technique, but the APP5 is stored in the memory only when the APP5 cannot store the log data when the system is stopped. The user data edited above is dumped in memory and saved in the external storage device 3. Then, after restarting the system, from the memory dump file 31 saved in the external storage device 3,
By retrieving the user data area that was being edited and passing the data equivalent to the user data section that was on the real memory 2 when the system stopped to APP5, the recovery data can be externally stored in a file format that APP5 can handle. It is stored in the storage device 3.

In the prior art, since the pre-update log / post-update log is always saved in an area irrelevant to the update target data or in a paired structure with the data area, a save area is required separately from the target data. However, as described with reference to FIGS. 1 to 3, in the present embodiment, the data being updated is externally transferred from the real memory 2 only when the user data being edited cannot be stored due to a failure. To save to the storage device 3,
The limited resources of a small-scale information processing device can be released to the user instead of being used for maintenance.

Further, in the prior art, at the time of restoration after the interruption of updating, it was necessary to read the log, rewrite the data to be updated, and update the pointer. However, in the data recovery system of this embodiment, the editing is performed. The user data is saved in a transient state, and after the system is restarted, the saved data is passed to APP5 to continue the interrupted process and recover the user data, so that the recovery process is facilitated.

The present invention is not limited to the embodiments described with reference to FIGS. 1 to 3, and various modifications can be made without departing from the scope of the invention. For example, the configuration may be such that the compression processing by the compression unit 10 and the decompression processing by the compression / decompression unit 11 in consideration of the efficiency of data movement and the efficiency of media storage are not used.

[0022]

According to the present invention, when the operation of the application program is stopped due to a failure and the user data being edited cannot be updated or stored, the data in the memory being updated is temporarily saved and restarted. To recover the transient state immediately before the failure expected by the user, that is, to recover the user data, by passing the saved memory data to the saved application program and continuing the interrupted application program. As a result, the data recovery process can be efficiently performed in a small-scale information processing device having a relatively small storage area and a slow processing speed.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a configuration of a data recovery system of the present invention according to the present invention.

FIG. 2 is an explanatory diagram showing an example of a mechanism of a paging function by the operating system in FIG.

FIG. 3 is an explanatory diagram showing an example of each data format on a real memory in FIG.

[Explanation of symbols]

1: Processing device, 2: Real memory, 3: External storage device, 4:
Operating system, 5: application program, 6: data recovery processing unit, 7: memory dump processing unit, 8: management information reading unit, 9: alignment processing unit, 10:
Compression unit, 11: compression / decompression unit, 12: cutout unit, 21:
Page directory, 22: Page table, 22: Page directory, 31: Memory dump file, 3
2: alignment file, 32a: compression alignment file, 33:
Recovery data file, a1 to a4, b1 to b4, c1
c4: page.

Claims (1)

[Claims] 1. A system for recovering data edited in an actual memory of an application program when a failure occurs in an information processing apparatus, wherein the data in the actual memory is dumped to an external storage device when the failure occurs. Means for generating a memory dump file, a means for reading management information used for sequentially managing each data in the real memory from the memory dump file at the time of restart, and the read management A means for sequentially calling and rearranging the respective data in the related data units from the memory dump file based on the information and rearranging them, and generating an alignment file, and editing the application file from the alignment file to the occurrence of the failure. Extracted the data that was used and restarted the above application program And a means for transferring the data group extracted from the sorting file to the file format corresponding to the restarted application program and restored by the restarted application program. Data recovery system.