JPH07230401A - Method and system for file maintenance - Google Patents

Abstract

PURPOSE:To apply the file maintenance method and system to an information processor which does not have a recording means for update information of each update operation by securing the restoration of a file without adding a device for gathering restoration information to a client. CONSTITUTION:An update backup process and a file recovery process are performed. The former is started with an execution indication or periodically and has a step wherein the client 101 detects an updated file and its update information and a step wherein a server 102 gathers the restoration information on the file and stores it in the storage device 104 of the server 102. The latter has a step wherein the server 102 generates a directory information table 106 at the time of the indication on the basis of directory information and the update information, a step wherein the server 102 restructures the contents of the file indicated at the time of the indication on the basis of the file contents of the restoration information, and a step wherein the client 101 stores the restructured file contents in a restoration object file.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method of maintaining a file held in an external storage device of an information processing apparatus, and a file maintenance system using the method.

[0002]

2. Description of the Related Art An information processing apparatus usually holds data to be processed in an external storage device or the like. However, the data held in these external storage devices or the like may be destroyed due to a failure of the storage devices or the like, or an insufficient situation such as an operation error. Therefore, it is necessary to maintain the data by guaranteeing the restoration of these data. For this reason,
Generally, for these data, in preparation for an unexpected situation such as data destruction due to device failure, in order to restore the data to be processed, information for the restoration of the data is collected, and another external storage device is used. Restoration information collection processing held on a magnetic tape device or the like, so-called backup processing, and restoration processing for restoring data using the collected and held restoration information, so-called recovery processing is performed.

The simplest backup processing method is to copy all data to an external storage device on a regular basis. According to this method, even if the data is destroyed, the state at the time of making the copy can be restored. However, with this method, the results of file addition, deletion, update processing, etc. performed between the time when the copy is created and the time when the data is destroyed are lost. Therefore, in order to minimize the loss of the data processing result, the copying interval must be shortened. However, copying is generally a time-consuming process, and there are limits to how often copying can be performed.

Therefore, as another backup method, once a copy of all data is created, only the contents of the copy are successively saved when the data is updated, such as written or deleted. There is a way. Saving the update processing content can be executed in less time than copying all the data. When the user wants to restore the backed up data, it can be restored by copying the first collected copy to the data and reproducing the stored update processing contents for the copied data. According to this method, even if the data is destroyed, the data can always be restored to the latest state.

As a specific example of the backup method, there is a method disclosed in Japanese Patent Laid-Open No. 3-71351. This method stores data to be backed up, duplication of the data, update processing contents of the data on separate storage devices, and a high-speed storage device for storing update processing contents that are frequently accessed. It is intended to realize reliable and high-performance backup processing by using.

[0006]

In the backup method of the above-mentioned conventional example, the latest state can always be restored when data is destroyed even if the frequency of time-consuming copying of all data is reduced, and it is possible to save the contents of data update processing that is normally performed. This is an excellent method in that the overhead is also reduced. However, in the above conventional example, it is necessary to add a storage device dedicated to backup for holding the restoration information to each information processing device. Therefore, in the above-mentioned conventional example, in a network system in which a large number of information processing devices having information that needs to be guaranteed to be restored are connected, if this method is applied to each information processing device, the cost becomes high. There was a problem of being lost.

Further, in the above-mentioned conventional method, since the backup is held by the combination of the duplication of the data at a certain time and the contents of the data updating operation after that time, the contents are recorded every time the data updating operation processing is performed. There is a problem that it is difficult to apply it to an information processing device for which no means is prepared.

In view of the above, according to the present invention, an information processing apparatus connected to a network retains a copy of data held by a plurality of other information processing apparatuses on the network at a certain time and data operation contents after the time. It is a first object of the present invention to provide a data maintenance method and a data maintenance system that can be realized at low cost without adding a special additional device to each information processing device.

Further, the present invention provides a data maintenance method and a data maintenance system applicable to an information processing apparatus which is not provided with a means for successively recording the contents of a data update operation each time an update operation is executed. The second purpose is to provide.

[0010]

In order to achieve the above-mentioned first object, in the present invention, at least one server having a storage device and an arithmetic processing unit is connected to the server via a communication line. At least one client, the client stores at least one file, an arithmetic processing unit having means for creating, updating, and deleting the file, and an input for receiving an input from the outside. In a method for maintaining a file in a network system having a device, an input receiving process for receiving an instruction for restoration, an instruction for a file to be restored, and an instruction regarding a point of time for restoration through an input device, and a file restoration. The update backup process that collects the information for The first file maintenance method for performing at least the file recovery process to restore the intended time it is instructed through is provided. Furthermore, a file maintenance system using this file maintenance method is also provided.

The update backup process in the first file maintenance method is at least one of a start command issued at a predetermined cycle and a backup process execution instruction accepted via the input device. Is started by the client, and the processing unit of the client detects a file that has been created, updated, or deleted, and update information indicating the content of the change, and identifies the file. A changed file detection step of transferring the update information of the file to the server, the arithmetic processing unit of the server that has received the transfer, directory information indicating the configuration of the directory, the content of the detected file, and the update of the file. Recovery information including information for recovering the file is collected, and the recovery information is collected. The and an update information storing step of storing in the server storage device.

Further, the file recovery processing in the first file maintenance method is activated by a file restoration instruction accepted by the arithmetic processing unit of the client via the input unit, and the arithmetic processing unit of the client operates the input unit. The step of transferring the instruction at the time of restoration received via the server to the server, and the arithmetic processing unit of the server receiving the transfer of the instruction at the time point are instructed based on the directory information and the update information. A step of creating a recovery directory information table representing a directory at a time point, a step of causing the arithmetic processing unit of the client to transfer an instruction of a file to be restored received via the input device to the server, and a step of transferring the file instruction. The processing unit of the server that accepted the above-mentioned recovery directory information Based a table, a file contents to be restored contained in the restoration information, to reconstruct the contents of the instruction file in the time of the above indicated,
Transferring the content to the client, and storing the content of the reconstructed file in the file to be restored by the arithmetic processing unit of the client that has received the transfer of the reconstructed file content. Have.

In order to achieve the above-mentioned second object, according to the present invention, a storage device holding at least one file, an arithmetic processing unit having means for creating, updating, and deleting the file, and from the outside In a method of maintaining a file in an information processing apparatus having an input device that receives an input of a process, a process of receiving an instruction of restoration, an instruction of a file to be restored, and an instruction about a time of restoration through the input device. , At least executing an update backup process for collecting information for file restoration and a file recovery process for restoring the above-mentioned file instructed via the input device to the one at the time instructed via the input device A second file maintenance method is provided. A file maintenance system using this file maintenance method is also provided.

The update backup process in the second file maintenance method is at least one of a start command issued at a predetermined cycle and a backup process execution instruction accepted via the input device. And a change file detection step of detecting a file that has been changed by any one of creation, update, and deletion, and update information indicating the contents of the change, directory information indicating a directory structure, and An update information storage step is included that includes the detected file content and update information for the file, collects restore information that is information for restoring the file, and stores it in a storage device.

The file recovery process in the second file maintenance method is started by a file restoration instruction received through the input device, and is instructed based on the directory information and the update information. Content of the instructed file at the instructed time based on the step of creating a recovery directory information table representing the directory of And a step of storing the contents of the reconstructed file in the file to be restored.

[0016]

In the first file maintenance method and the file maintenance system using the method, since the restoration information is held in the storage device of the server, it is necessary to provide each client with a storage device for holding the restoration information. There is no. Also,
According to the present method, even if the processor of the client does not have the function of collecting the restoration information each time the modification operation is performed in order to detect the modified file, the modification is performed according to the present method. It is possible to collect the restoration information of only the file.

Further, in the second file maintenance method and the file maintenance system using the method, the files changed up to the present time since the restoration information was last collected periodically or by an instruction from the outside. Is detected and the restoration information of the file is collected, the restoration information can be collected even if the information processing device does not have the function of collecting the restoration information every time the change operation is executed. Restoration can be guaranteed.

[0018]

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

(Embodiment 1) FIG. 1 shows a block diagram of an information processing apparatus system to which this embodiment is applied. The system to which this embodiment is applied is via the network line 105,
At least one information processing device (hereinafter, referred to as a client) 101 having local data for which data restoration should be guaranteed, and an information processing device holding information (restoration information) for restoring local data held by the client 101 A network system (hereinafter referred to as a server) 102 is connected. The server 102
May be a dedicated device for restoring local data, or may be an information processing device having other functions such as a function as a file server. Also, the server 10
The computer 2 may be a personal computer, a workstation, or a host computer as long as it has a sufficient storage capacity.
The server 102 of this embodiment is a dedicated workstation for restoring local data.

The server 102 includes an arithmetic processing unit 121,
And an external storage device 104. The external storage device 104 of the server 102 stores a directory information file 109 that stores the oldest directory information of the local files 117 and a data file 110 that is a file that stores the contents of the local file 117 itself.
An update information file 111 that records the contents of updates made to the local file 117 after collecting the oldest restoration information, a backup registration file 116 in which a backup identifier that is the identifier of the client that collects the restoration information is registered, and a backup An update history file 118 for recording the time when the processing is performed. The directory information file 109, the data file 110, the update information file 111, and the update history file 118 are provided for each client. Two or more backup identifiers are assigned to one client 101, and the four files 109, 110,
By providing 111 and 118 for each backup identifier, dividing into a set of a plurality of local files held in the external storage device 103 of the same client 101, and assigning a backup identifier to each set, the backup identifiers are separately provided. The restoration information can be collected at the timing of.

The arithmetic processing unit 121 of the server 102 has, as a storage area, the latest directory for holding the directory information of the local file 117 when the server 102 executes a process (backup process) for collecting information for restoration. The information table 106 and the checkpoint counter 11 that increases the value each time one backup process is executed and records the number of times the backup process is executed.
9 and. A recovery directory table (not shown) is provided in the storage area temporarily secured in the arithmetic processing unit 121 when the local file restoration processing is executed. Note that, in this way, the table in which the area is temporarily reserved and used is not shown in FIG.

Further, the arithmetic processing unit 12 of the server 102
1 is a server-side backup execution unit 107 that performs processing necessary for restoring the local file 117 on the server 102 side.
And a backup registration acceptance unit 120 that accepts a registration request for restoration information collection from the client 101, and a communication processing unit 108 that performs communication performed via the network line 105.

The client 101 also includes a processor 122, an external storage device 103, and an input / output device 125.
Have and. External storage device 103 of client 101
Holds a local file 117 in which local data to be restored is stored.

The arithmetic processing unit 122 of the client 101
Is the old directory information table 11 for holding the directory information of the local file 117 when the client 101 executes the backup process as a storage area.
4 is provided. Further, when the restoration information is collected, the arithmetic processing unit 1
A current directory information table (not shown) is in the storage area temporarily reserved in 22. Further, the processor 122 includes a flag storage area (not shown) for storing the file restoration instruction.

Further, the arithmetic processing unit 122 of the client 101 has a communication processing unit 112 for performing communication performed via the network line 105, and the client 10.
The client-side backup execution unit 113 performs processing necessary for restoring the local file 117 on the one side, and the local file use means 115 that refers to and updates the local file 117.

The arithmetic processing units 121 and 122 include
A main memory (not shown) and a central processing unit (not shown) are provided. Processors 121 and 122
Each of the above units is realized by the central processing unit executing instructions stored in the main memory in advance.

First, an outline of the operation of the data restoration method of this embodiment and the structures of the above-mentioned tables and files will be described. In the present embodiment, when the client 101 is started up, restoration is registered in the server 102 and each restoration information is initialized. After that, in a fixed cycle,
Alternatively, the restoration information is collected and accumulated in the server 102 according to the user's instruction. When there is an instruction to restore the local file 117 from the user, the local file for which restoration is instructed by the initial restoration information and the restoration information accumulated thereafter is in the state at the time when the restoration information was last collected. Restored to.

FIG. 8 shows a process for restoring local data.
Shown in. First, when the client 101 is activated, the client-side backup execution unit 113 performs restoration information initialization processing (step 801). As a result, the fact that the restoration information should be collected is registered in the server 102, and the restoration information is further initialized. Note that in the restoration information initialization process (step 801), the client-side backup execution unit 113 causes the server-side backup execution unit 10 to execute.
7 is also started, and the server 10 in the restoration information initialization process
The processing on the second side is executed.

Next, the client side backup execution unit 113 checks whether a predetermined fixed time has passed (step 802), and if it has passed,
The process proceeds to step 804 to perform the update backup process. Also, the client side backup execution unit 1
If the predetermined time has not elapsed, 13 checks whether or not an instruction to execute the update backup process has been given (step 803). If so, the process proceeds to step 804 to update. Perform backup processing. Update backup process (step 80)
In 4), the update contents of the local file 117 by the local file using means 115 are updated to the update information file 11
1 and the data file 110 is additionally written.

Update backup process (step 804)
Is completed, the client side backup execution unit 11
3 advances the processing to step 805. Also, in step 803, even if the update backup processing instruction has not been issued, the client side backup execution unit 113 advances the processing to step 805. Step 8
In 05, the client side backup execution unit 11
3 checks whether or not a file restoration is instructed (step 805). The client-side backup execution unit 113 executes the file recovery processing if restoration is instructed (step 806), and if not instructed, returns to step 802 and repeats the above processing.
Details of the update backup process and the file recovery process will be described later.

In the present embodiment, the update backup process (step 804) is executed when it is detected in step 802 that a certain period of time has elapsed, so it is periodically activated. It is desirable that the cycle of execution of this update backup process (step 804) is determined according to the importance of the data to be restored. That is, restoration information is collected frequently for data that requires high reliability, and to reduce the load on the arithmetic processing unit 122 of the client 101 and the server 102 for data that does not require high reliability. In addition, it is desirable to lengthen the cycle of the update backup process (step 804). Further, the client 101 of this embodiment includes the input / output device 125, and the input / output device 125
When a user's instruction to collect the restoration information is received via, a value indicating the file restoration instruction is stored in a flag storage area (not shown) held in the arithmetic processing unit 122. In step 805, the client side backup execution unit 113 refers to this flag storage area, and based on the value held therein, determines whether or not the restoration information collection instruction has already been accepted.

In this embodiment, since the arithmetic processing unit 122 of the server 102 is a time sharing system, the server side backup execution unit 107 has as many as the number of clients 101 that have accepted the registration of restoration information collection. It is activated and each operates in parallel.

Next, the restoration information initialization process (step 8)
The flow of (01) is shown in FIG. First, the client 10
Client side backup execution unit 11
3, the client-side backup execution unit 113 requests the server 102 to register restoration information collection (step 2001). This registration request includes
A predetermined backup identifier is added to each client 101 so that the client 101 that issued the request can be identified.

The registration request is sent via the communication processing unit 112, the network line 105, and the communication processing unit 108.
The backup registration acceptance unit 120 of the server 102 is reached. It was started by receiving this registration request (Step 2
002) The backup registration receiving unit 120 checks whether the backup identifier added to the registration request is already registered in the backup registration file 116 (step 2003).

If it is not registered in the backup registration file 116, the backup registration receiving unit 120 newly registers the backup identifier of the client 101 that issued the registration in the backup registration file 116,
Directory information file 109, data file 110, update information file 1 relating to the client 101
11. The names of the update history file 118 are determined, and these file names are also registered in the backup registration file 116 (step 2004). Then, the backup registration acceptance requesting unit 120, the server-side backup executing unit 10
7 is started (step 2005) and the server side backup execution unit 107 is requested to execute the initial backup process. The server-side backup execution unit 10 that has been started
7 is the four files 109, 110, 111, 11
In order to perform the initialization of No. 8, the initial backup process is executed (step 2006), and the restoration information initialization process ends. Details of the initial backup process will be described later.

On the other hand, in step 2003, when the backup identifier added to the registration has already been registered in the backup registration file 116, the backup registration acceptance section 120 activates the server side backup execution section 107 (step 2007). , And requests the server-side backup execution unit 107 to start rebuilding the latest directory information table 106. The server-side backup execution unit 107 uses the backup registration file 116.
A directory information file 109, a data file 110, which is held in the client 101,
By referring to the names of the update information file 111 and the update history file 118, the latest directory information table 106 is created again using the contents of the directory information file 109 and the update information file 111 (step 2008).
Then, the server-side backup execution unit 107 transmits the contents of the reconstructed latest directory information table 106 to the client-side backup execution unit 113 via the communication processing unit 108. The client-side backup execution unit 11 that has received the contents of the latest directory information table 106 reconstructed via the communication processing unit 112
3 stores the contents in the old directory information table 114 (step 2009) and finishes the restoration information initialization processing.

The internal structure of the directory information file 109 is shown in FIG. The directory information file 109 is
For each local file 117 that guarantees restoration, information such as the file name, size, attribute, and last update date,
This is a file for storing the position in the directory. The directory information file 109 is shown in FIG.
As shown in (a), it has an array structure in which one or a plurality of directory information storage areas 200 holding directory information for managing information on one file or subdirectory are arranged. A unique index number is assigned to each of the directory information storage areas 200.

As shown in FIG. 2B, each directory information storage area 200 has a file name storage area 2 for storing a file or subdirectory name.
01, a file size storage area 202 for storing the size of a file, an attribute storage area 203 for storing an attribute indicating whether a file is a subdirectory, whether the file is write-protected, etc. The last update date / time storage area 204 to be stored, the data start position storage area 205 for storing the data start position indicating the position where the contents of the file in the data file 110 are stored, and this directory information is related to the sub-directory. In this case, the subdirectory index storage area 206 for storing the index number of the head information of the directory information storage area 200 of the file / subdirectory belonging to the subdirectory, and the index number of the next directory information are stored. With directory information index storage area 207 for lifting.

Directory information index storage area 2
The “index number of the next directory information” stored in 07 is the index number of the directory information 200 of the next file or subdirectory belonging to the same directory as the file to be restored by this directory information 200. That is. By the index number held in the directory information index storage area 207, the directory information storage area 200 of files and sub-directories belonging to the same directory constitutes a linear list. Further, the subdirectory index 206 stores the index number of the first directory information storage area 200 of the linear list of the directory information of files or subdirectories belonging to the subdirectory when the directory information is related to the subdirectory. It is an area for The storage areas 206 and 207 for these indexes store the position of the file in the directory to which it belongs.

The internal structure of the data file 110 is shown in FIG. As shown in FIG. 3A, the data file 110 has a structure including one or more local file information storage areas 305. As shown in FIG. 3B, each local file information storage area 305 has a header storage area 300 for storing attributes of a local file that guarantees restoration and the local file 11
7 and a storage area 304 for storing the contents of 7. At this time, the redundant part of the above-mentioned contents is, for example, LZW (Lempel-Zi).
If the data file 110 can be reduced by compression using an algorithm such as v-Welch method,
You may perform compression.

Each header part storage area 300 is shown in FIG.
As shown in (c), a compression flag storage area 301 holding a compression flag indicating whether or not the above compression has been performed, and an original data length indicating the size of the content before compression, respectively, are stored. It further includes an original data length storage area 302 and a post-compression data length storage area 303 for storing a post-compression data length indicating the size of the compressed content. If compression has not been performed, the values of the original data length and the compressed data length are the same. The respective storage positions of the local files 117 are included in the data start position storage area 205 of the directory information storage area 200 stored in the directory information file 109 or the update information stored in the update information file 111 described next. The data is stored at the beginning position.

The internal structure of the update information file 111 is shown in FIG.
Shown in. The update information file 111 is a file that stores file update contents. Update information file 1
The update content of the file stored in 11 is the content of the update performed from the state of the reference to the current state with the state of the oldest recoverable file as the reference. The information of the oldest recoverable file is the oldest information of the directory to which the file belongs, which is held in the directory information file 109, and the contents of the local file 117 at that time, which is held in the data file 110. A state of a file that can be reconstructed by.

As shown in FIG. 4A, the update information file 111 has a structure in which one or more update information storage areas 400 showing the contents of file update are provided in an array format. Each update information storage area 400,
As shown in FIG. 4B, a checkpoint number storage area 401 for storing the checkpoint number which is the value of the checkpoint counter 119 when the update information is stored, and the name of the update target file or subdirectory. Target name storage area 402 for storing
An update command storage area 403 for storing the type of update content, an updated file size storage area 404 for storing the updated size of the file, and an updated attribute of the file. An updated file attribute storage area 405, an updated last update date / time storage area 406 for storing the updated last update date / time of the file, and a position where the updated file content is stored in the data file 110. An updated data head position storage area 407 for storing the data head position is provided. In addition,
In the present embodiment, there are five types of updates stored in the update command storage area 403: “file creation”, “file update”, “file deletion”, “directory creation”, and “directory deletion”.

The internal structure of the update history file 118 is shown in FIG.
Shown in. The update history file 118 is a file for recording the time at which either the initial backup process or the update backup process is performed. The update history file 118 includes a storage area 501 of an oldest pointer that points to a storage area of an element related to the oldest recoverable restoration information and a storage area of a latest pointer that points to a storage area next to the storage area of an element related to the newest restoration information. 502 and at least one update history storage area 505. Each update history storage area 505 stores a checkpoint number storage area 503 which is the value of the checkpoint counter 119 at the time when the initial backup processing or the update backup processing is performed, and an update time when the processing is performed. A date / time storage area 504 is provided.

The oldest pointer 501 is a pointer pointing to the storage area 505 of the element relating to the oldest recoverable restoration information, and the latest pointer 502 is the storage area 50 next to the storage area of the element relating to the newest restoration information.
5, that is, a pointer indicating a portion to write information when the update backup is performed next time. In the example shown in FIG. 5, the oldest update backup process is 1
It was held at 19:30 on July 29, 993. Therefore, the oldest pointer 501 points to the update history storage area 505 of checkpoint number 7.

The internal structure of the backup registration file 116 is shown in FIG. The backup registration file 116 is
This is for enabling the server 102 to identify and manage the restoration information of the data held by the plurality of clients 101. The backup registration file 116 is shown in FIG.
As shown in (a), the storage area 600 of the registration information is provided for each client 101 in an array format. The registration information storage area 600 is shown in FIG.
, A backup identifier storage area 601 for storing a backup identifier for identifying which client 101 is related to the restoration information, and a directory information file name storage for storing a directory information file name for the client Area 602
A data file name storage area 603 for storing a data file name, an update information storage area 604 for storing an update information file name, and an update history file name storage area 605 for storing an update history file name. Have.

The structure of the latest directory information table 106 is shown in FIG. Latest directory information table 106
Is a table showing the directory status of the local file 117 at the end of the update backup process. The latest directory information table 106 has one or more directory information storage areas 700.

Each directory information storage area 700 stores a file name storage area 701 for storing a file or subdirectory name, a file size storage area 702 for storing a file size, and a file attribute. For storing the file attribute storage area 703 for storing the file, the last update date and time storage area 704 for storing the last update date and time of the file, and the data head position indicating the position where the file content is stored in the data file 110. Data head position storage area 705 and a file information storage area 700 of a file and subdirectory belonging to the subdirectory in the case of a subdirectory
A sub-directory address storage area 706 for storing the address of the first one of them and a directory information address storage area 707 for storing the address of the directory information 700 of the next file or sub-directory belonging to the same directory are included.

Subdirectory address storage area 706
Is stored in the directory information storage area 70 of the file / subdirectory belonging to the subdirectory.
The address is 0. For example, the sub-directory address storage area 706 of the directory information storage area 700a
Is stored in the directory information storage area 700a.
Is a directory information storage area 700b of a file belonging to a subdirectory in which information is stored. In this way, when the information held in the directory information storage area 700 is related to the subdirectory, the address of the first one on the linear list in the files belonging to the subdirectory and the file information storage area 700 of the subdirectory is By being held in the subdirectory address storage area 706, the directory information storage area 700 of this subdirectory is referenced,
It is possible to detect the storage location of the directory information of the file / subdirectory belonging to this subdirectory and search the linear list.

Also stored in the directory information address storage area 707 is the address of the directory information storage area 700 of the next file / subdirectory. For example, the directory information address storage area 707 of the directory information storage area 700a stores the next file / sub directory belonging to the directory to which the file / sub directory represented by the information held in the directory information storage area 700a belongs. This is the address of the directory information storage area 700c of the directory. In this way, by holding the address of the directory information storage area 700 of the next file / subdirectory, files / subdirectories belonging to the same directory
Directory information of subdirectories can be connected to form a linear list structure.

As described above, in the latest directory information table 106 of this embodiment, a linear list (herein referred to as an in-directory linear list) composed of directory information storage areas 700 of files / subdirectories belonging to the same directory, A linear list including directory information storage areas 700 of files / subdirectories belonging to the same subdirectory (herein referred to as an in-directory linear list) is formed by the values held in the directory information address storage area. A directory information storage area 700 of a subdirectory located in the linear list and an in-subdirectory linear list of files / subdirectories belonging to the subdirectory are held in a subdirectory address storage area 706. It is connected by an address that is. As a result, in the latest directory information table 106,
Similar to the directory information file 109, the directory information 700 is connected in the linear list structure so that the directory information 700 has a linear list structure corresponding to the structure of the directory to which the local file 117 belongs.

Although the old directory information table 114 on the client 101 side has the same structure as the latest directory information table 106, in the old directory information table 114, the data head position 705 has no meaning.

Next, an initial backup process (step 2)
006), update backup processing (step 804),
Details of each of the file recovery processes (step 806) will be described.

First, the initial backup process (step 2)
The contents of (006) will be described with reference to FIG. Figure 9
FIG. 8 is a sequence diagram showing the contents of the initial backup process (step 2006).

The server-side backup execution unit 107 started in step 2005 has the communication processing unit 108,
An initial backup process start frame 900 is transmitted to the client side backup execution unit 113 via the network 105 and the communication processing unit 112. Upon receiving the initial backup start frame 900, the client-side backup execution unit 113 sends an acknowledgment frame (hereinafter referred to as an ACK frame) 901, which means an affirmative response, to the server-side backup execution unit 107.

Upon receiving this ACK frame 901, the server side backup execution section 107 receives the latest directory information table 106 and checkpoint counter 11
9 is initialized, the directory information file 109, the data file 110, the update information file 111, and the update history file 118 are generated (step 907), and the initialization completion frame 902 is transmitted to the client side backup execution unit 113. Request to start sending the file to be backed up. This initialization completion frame 902
Client side backup execution unit 113 that received the
Transmits the directory information and the file content of one of the local files 117 determined in advance as the target for collecting the restoration information to the server side backup execution unit 107 (frame 903).

The server-side backup execution unit 107 creates directory information 700 regarding the file from the received directory information and file contents (frame 903) for one file, and determines the directory to which the file belongs in the latest directory information table 106. Connect to the list (step 908). The directory to which the file belongs is the latest directory information table 1
If it does not exist in the directory 06, the server-side backup execution unit 107 determines the directory information 7
00 is created and inserted into the latest directory information table 106 first, and then the directory information of the file to be processed is linked to the directory list.

Further, the server side backup execution unit 10
7 indicates the file contents of the file as a data file 11
0 is additionally written, and the write start position is stored in the data start position 705 of the directory information 700 of the file. At this time, the file contents may be compressed if necessary (step 909). When the above process ends normally, the server-side backup execution unit 107 sends the ACK frame 904 to the client-side backup execution unit 113.
Send to.

Upon receiving this ACK frame 904, the client side backup execution unit 113 sends the directory information and file contents of the next local file 117 to the server side backup execution unit 107. Upon receiving the transmission, the server-side backup execution unit 107 performs the processing of steps 908 and 909, and the ACK frame 904.
Is transmitted to the client side backup execution unit 113. In this way, the processing from the transmission of the frame 903 to the transmission of the frame 904 is repeated until the contents of all the local files that guarantee the restoration are transmitted to the server 102. I received an ACK frame 904,
When there is no local file 117 to be transmitted, the client side backup execution unit 113 transmits the initial backup end frame 905. Upon receiving this initial backup end frame 905, the server side backup execution unit 107 stores the contents of the latest directory information table 106 in the directory information file 109 (step 910). At that time, the subdirectory address storage area 7 which is a value in the memory address format
06 and the value held in the directory information address storage area 707 are respectively converted to the index number format, and then the sub directory index storage area 20
6, stored in the directory information index storage area 207. When the process of step 910 ends normally, the server-side backup execution unit 107 sends an ACK frame 906 and notifies the client-side backup processing unit 113 that the initial backup process has ended normally. The above is the contents of the initial backup process.

Next, an update backup process (step 8)
The contents of 04) will be described with reference to FIG. Figure 1
0 is a sequence diagram showing the contents of the update backup process (step 804).

Client side backup execution unit 113
First, the directory information of the current local file 117 is read, and using this directory information, a current directory information table (not shown) having the same structure as the old directory information table 114 is created (step 1001).

Next, the client side backup execution unit 113 compares the contents of the old directory information table 114 with the contents of the above-mentioned current directory information table, and for the local file 117 having a difference, sends that information to the server 102. Notify (step 1002).
For ease of comparison, the contents of the old directory information table 114 and the current directory information table are pre-sorted before the comparison, using the file or subdirectory name as a key. In this embodiment, it is assumed that they are sorted in ascending order. The comparison is performed by sequentially comparing the contents of the nodes of the old directory information table 114 and the current directory information table from the head node according to the list structure. Here, the node refers to an individual directory information storage area forming each table.

When a file change (creation, deletion, or update) is detected as a result of the comparison, the client side backup execution section 113 sends the update information and the contents of the file to the server side backup execution section 107. .
Here, the update information refers to data stored in each area of the update information storage area 400 that constitutes the update information file 111. That is, the update information is the checkpoint number, the update target file (or subdirectory)
Name, update command information (file creation, file update,
(File deletion, directory creation, or directory deletion), size after update, attribute after update, last update date and time after update, and data start position.

As a result of the comparison, if the file change (creation, deletion, or update) is not detected, the client side backup execution unit 113 does not perform this transmission process (step 1003). The client side backup execution unit 113 repeats the above-described comparison of directory information until the comparison of all directory information of the old directory information table 114 and the current directory information table is completed. This comparison process will be described later in detail.

When this comparison processing is completed, the client side backup execution unit 113 copies the contents of the latest directory information table 106 held by the server 102 to the old directory information table 114, and ends the update backup processing (step). 1005). The contents of this copying process will be described later in detail with reference to FIG.

On the other hand, the server side backup execution unit 107 that has received the update information and the contents of the local file 117 transmitted in step 1003 operates as follows. First, the received update information is added to the update information file 111, and the contents of the local file 117 are additionally written to the data file 110 (step 100
6). Next, the server-side backup execution unit 107 updates the latest directory information table 106 according to the content of the received update information (step 1007). That is, the server-side backup execution unit 107 adds a node related to the file if the update information indicates creation of the file, and deletes a node related to the file if the update information indicates deletion of the file. If the update information indicates the modification of the file, the server-side backup execution unit 107 uses the modification content and the modification date and time,
The information held in the directory information storage area 700 of the file is modified.

If the size of the update information file 111 or the data file 110 is equal to or smaller than the predetermined size, the server side backup execution unit 107 ends the processing as it is, but if it becomes large (step 1008). ), The following processing is performed in order to delete old update information and make it within a predetermined size.

First, the server side backup execution unit 107
Updates the contents of the directory information file 109 according to the contents of the update information having the oldest checkpoint number among the update information in the update information file 111 (step 1009). The oldest checkpoint number is written in the element indicated by the oldest pointer 501 among the elements in the update history file 118.

Next, the server side backup execution unit 10
7 deletes the data that is no longer needed in step 1009 (step 1010). That is, the server-side backup execution unit 107 deletes the local file information held in the local file information storage area 305 of the data file 110. The position of the storage area 305 of the local file information to be deleted in the data file 110 is set to the data head position storage area 407 of the update information storage area used for updating the contents of the directory information file 109 in step 1009. It is detected by referring to it. The size of the data to be deleted is the local file information storage area 305 indicated by the value held in the data head position storage area 407.
Of the compressed data length storage area 3 in the header section 300 of
It is detected by referring to 03. Furthermore, the server-side backup execution unit 107 uses the update information file 11
Of the update information stored in the first update information storage area 400, the value stored in the checkpoint number storage area 401 is the oldest number (used in step 1009 to update the contents of the directory information file 109). Deleted).

Finally, the server side backup execution unit 10
7 is due to the deletion of the data in step 1010,
Directory information file 109 and update information file 1
Since the data head positions 205 and 407 stored in 11 may be changed, they are also updated, and the oldest pointer 501 is updated to point to the next element, and the process ends. (Step 1011).

Next, the update backup process (step 8)
The comparison process (step 1002) of the current directory information table and the old directory information table 114 performed in 04) will be described.

FIG. 1 shows the initial state of the two tables and the pointer for comparison when the comparison processing of the current directory information table and the old directory information table 114 is performed.
3 shows. Client side backup execution unit 113
In step 1001, secures an area in the main memory of the arithmetic processing unit 122, creates a current directory information table, and, for comparison processing, directory information which is a processing target node of the old directory information table 114. A pointer p for indicating the storage area 700
Storage area 1302 of the current directory information table 1301 and the storage area 13 of the pointer q for indicating the directory information storage area 700n which is the node to be processed in the current directory information table 1301.
03 and secure.

In this state, the client side backup execution unit 113 compares the old directory information table 114 with the current directory information table 1301 according to the procedure shown in the flowchart of FIG.

First, the client side backup execution unit 113 initializes the pointers p1302 and q1303 (step 1401). That is, the client-side backup execution unit 113 stores the address of the leading directory information storage area 700 of the old directory information table 114 and the directory information storage area 700n of the current directory information table 1301 in the pointer storage areas 1302 and 1303, respectively. It stores the address and. Thus, in the initial state, the two pointers p and q respectively point to the head node of the old directory information table 114 and the head node of the current directory information table 1301, as shown in FIG.

When the initialization of the pointers p and q is completed, the client side backup execution unit 113
The file name held in the node pointed to by is compared with the file name held in the node pointed to by pointer q (step 1402). At this time, if the pointer does not point to any node, it is considered that the pointer points to a node having an infinitely large file name, and comparison is performed.

If the result of the comparison in step 1402 is the same, the client side backup execution unit 113 further compares the file size, the attribute and the last update date and time held in the respective nodes 700 and 700n. Perform (step 1403). If a difference is detected in step 1403, it means that the content of the file indicated by the node pointed to by the pointer p1302 has been updated. Therefore, the client-side backup execution unit 113 updates the update information whose update command information is “file update”. The created update information and the content of the file indicated by the processing target directory information are transmitted to the server-side backup execution unit 107 (step 1404). After the transmission, the client side backup execution unit 113 displays the pointer p storage area 13
02, the value held in the pointer q storage area 1303 is updated to point to the next node (step 1405), and the process proceeds to step 1410. In addition,
If there is no node, a value that does not point to any node is stored in the pointer storage area.

If the file name stored in the node pointed to by pointer p is larger than the file name stored in the node pointed by pointer q in step 1402, the directory information has already been sorted using the file name as a key. , The file indicated by the content of the node 700n pointed to by the pointer q is newly created. Therefore, in this case, the client-side backup execution unit 113 creates update information whose update command information is “file creation”, and writes the update information and the contents of the file indicated by the directory information pointed to by the pointer q to the server side. Send to the backup execution unit 107 (step 14)
06). Next, the value held in the pointer q storage area 1303 is updated to point to the next node (step 14
07), and the process proceeds to step 1410. If there is no node, a value that does not point to any node is stored in the pointer storage area.

In step 1402, the file name stored in the node pointed by the pointer p1302 is the pointer q1.
If the file name is smaller than the file name stored in the node pointed to by 303, the directory information has already been sorted using the file name as a key, indicating that the file pointed to by the node pointed to by pointer p1302 has already been deleted. . Therefore, in this case, the client-side backup execution unit 113 creates update information whose update command information is “file deletion” and sends it to the server-side backup execution unit 107 (step 1408). Next, the client side backup execution unit 113 updates the pointer p1302 to point to the next node (step 1409), and advances the processing to step 1410. If there is no node, the pointer value is a value that does not point to any node.

After performing the above processing, step 1410
In the client side backup execution unit 113
Checks the pointer p and the pointer q,
As long as at least one of q points to the node, the processing is returned to step 1402, and step 1402
The processing of 1410 is repeated. When both of the pointers p and q do not point to the node, the client side backup execution unit 113 ends the comparison process (step 1410).

If the node to be compared points to a subdirectory, this comparison process (step 1002) is recursively executed for the subdirectory as well. If the subdirectory has been created or deleted, the processing is performed assuming that all files or subdirectories belonging to the subdirectory have also been created or deleted.

Next, the content copying process (step 1005) from the latest directory information table 106 to the old directory information table 114, which is performed in the update backup process (step 804), will be described. FIG. 11 is a sequence diagram showing the contents of the directory information table copying process performed at the end of the update backup process. In this process, first, the client-side backup execution unit 113 sends the latest directory information transmission request frame 1101 to the server-side backup execution unit 107.
Send to. When the server side backup execution unit 107 receives the latest directory information request frame 1101, the server side backup execution unit 107 writes the value of the checkpoint counter 119 and the date and time at that point in the portion pointed to by the latest pointer 502 of the update history table 118 (step 1103). After updating 502 to point to the next storage area 505 (step 1104), the contents of the latest directory table 106 are transmitted to the client side backup execution unit 113 (frame 1102). The client side backup execution unit 113 stores the received contents of the latest directory information table 106 in the old directory information table 114 (step 1105).

Finally, the contents of the file recovery process (step 806) will be described with reference to FIG. FIG. 12 is a sequence diagram showing the contents of the file recovery process (step 806). In the file recovery process (step 806), the one or more local files 117 designated by the checkpoint number and the file name are transferred to the external storage device 10 of the client 101.
It is a process of reconstructing within 3.

File recovery processing (step 806)
First, the client side backup execution unit 113
Of the recovery start request frame 1201 to the server 102.
Start by sending. When the server side backup processing unit 107 receives the recovery start request frame 1201, the client 1 that issued the recovery start request
All update history information held in the update history file 118 corresponding to 01 is transmitted (frame 120
2). The update history information is stored in the update history storage area 5
The information is held in 05 and has a checkpoint number and an update date and time. When the client side backup execution unit 113 receives the update history information,
Output from the input / output device 125 (step 1210), the input of the checkpoint number indicating which update time point the output of the recovery processing target file is returned is accepted via the input / output device 125, The checkpoint number (frame 1206) is transmitted to the server side backup execution unit 107 (step 1211).

The server-side backup execution unit 107 updates the directory information and directory structure held in the directory information file 109 with the update information 40 up to the time designated by the received checkpoint number.
The update processing according to 0 is performed to newly create a recovery directory information table (not shown) (step 1212). The recovery directory information table has the same structure as the contents of the directory information file.

The contents of the recovery directory information table creation processing in step 1212 are shown in FIG.
Shown in. In this recovery directory information table creation processing, the server-side backup execution unit 107
First, an area of the recovery directory information table is secured in the arithmetic processing unit 121, and the contents of the directory information file 109 are copied there (step 2101).
The directory information file 109 holds the oldest recoverable state of the directory to which the file to be restored belongs.

Next, the server side backup execution unit 10
Reference numeral 7 detects the update information of the recovery target file from the update information file 111 corresponding to the client 101 that requested the recovery, and reflects the update information. That is, the server-side backup execution unit 107 sets the reference target as the top update information storage area 400 in the update information file 111 corresponding to the client 101 that requested recovery, and the update target name storage area of this storage area 400. It is determined whether or not the data held in 402 matches the file name of the recovery target file (step 2102). If they do not match, the process proceeds directly to step 2104. If they match, the update information storage area 400 The contents of the update information stored in the table are reflected in the recovery directory information table created in step 2101 (step 2103), and the process proceeds to step 2104. This reflection means that, according to the contents of the update process (“creation”, “update”, or “delete”) held in the update command storage area 403, the same correction as the directory correction associated with the update process is applied to the directory information table. To do to.

Since the update information is sequentially stored in the storage area 400 next to the already stored update information storage area 400, the update information is arranged in the oldest order, and the first one is the oldest. Therefore, the server-side backup execution unit 107 sets the reference target to the update information storage area 4 next.
00 (step 2104), and it is checked whether or not the value held in the checkpoint number storage area 401 of the update information storage area 400 of the reference target is less than or equal to the checkpoint number specified by the input (step 210).
5) If the number is less than or equal to the designated number, the process returns to step 2102 and the reflection of update information is repeated. If the number is greater than the specified number, it means that the recovery directory information table has already been obtained in which all the recovery information collected before the specified recovery time is reflected in chronological order. The backup execution unit 107 ends the process of creating the recovery directory information table.

The server-side backup execution unit 1 that has created the recovery directory information table as described above
07 transmits the contents of the table (frame 1207) to the client side backup execution unit 113 (step 1213). The structure of the recovery directory information table is the same as the structure of the latest directory information table 106.

Client side backup execution unit 113
Outputs the contents of the received recovery directory information table from the input / output device 125 (step 121
4) The input for designating which file is to be recovered is accepted via the input / output device 125 (step 1215). In this embodiment, a plurality of files to be recovered can be designated. When the recovery target file is specified, the client-side backup execution unit 113 specifies the file name (frame name) including the name of one of the files specified as the recovery target, including the directory name in which the file exists. 1203) to the server side backup execution unit 107 (step 1
216).

Upon receiving the frame 1203, the server-side backup execution unit 107 reconstructs the contents of the file designated as the restoration target based on the contents of the data file 110 (step 1217), and the client-side backup execution unit 113. To (step 121
8). The client-side backup execution unit 113 stores the received file contents as the file 117 in the external storage device 103, and thereby the local file 1
17 is restored (step 1219).

The process of reconstructing the file contents in step 1217 is shown in FIG. In step 1217, the server side backup execution unit 107 refers to the leading directory information in the recovery directory information table created in step 1212 (step 22
01), it is determined whether or not the recovery target name storage area of the directory information (the area corresponding to the recovery target name storage area 201 in the directory information file 109) holds the file name of the file specified as the recovery target (step 2202). If the file names are different, the server side backup execution unit 107 sets the reference target as the next directory information (step 2203) and returns the process to step 2202. When the file names match, the server side backup execution unit 107 detects the data head position held in the directory information (held in the area corresponding to the data head position storage area 205 in the directory information file 109) ( Step 2204). Server side backup execution unit 107
Reads out the information held in the local file information storage area 305 starting from the data head position detected as described above in the data file 110 corresponding to the client 101 requesting the recovery processing (step 2205). The compression flag (held in the compression flag storage area 301) in the header of the local file information is inspected (step 2206), and if the compression has been performed, it is held in the file content storage area 304. The contents of the compressed file are returned to the original state, and the contents of the file designated as the restoration target are set (step 2207). If the file has not been compressed, the server-side backup execution unit 107 sets the content held in the file content storage area 304 as the content of the file designated as the restore target. The contents of the file to be restored (frame 1208) reconstructed as described above are stored in step 1218 as described above.
At, the server-side backup execution unit 107 sends the client-side backup execution unit 113.

Thereafter, until the transfer of all the files designated in step 1215 is completed, the file names (frame 1
The processing from the transmission of (203) (step 1216) to the restoration of the local file 117 in the external storage device 103 (step 1219) is repeated (step 122).
0). When the transfer of all the files to be recovered is completed, the client side backup execution unit 113
Issues a recovery end notification (frame 1205) to the server-side backup execution unit 107 (step 122
1) The file recovery process is completed.

As described above, according to this embodiment,
It is possible to collect the restoration information of the local data of the information processing device (client) connected to the network and restore the contents at the designated time point without providing a special additional device. Furthermore, according to the present embodiment, even if the information processing apparatus on the client side is not provided with means for directly detecting the content of the data operation performed by the application program, the update of the file is detected and restored at appropriate intervals. Information can be collected and a file can be restored with the updated information.

(Embodiment 2) Next, another embodiment of the present invention will be described. In the present embodiment, an external storage device for storing restoration information is added to an information processing device not connected to a network, and the information processing device having a local file that guarantees restoration performs restoration information collection processing by itself. Here is an example.

A block diagram of the information processing apparatus according to this embodiment is shown in FIG. The information processing apparatus 150 according to the present embodiment includes an arithmetic processing unit 152, an external storage device 103, a restoration information external storage device 151, and an input / output device 125.

The restoration information external storage device 151 includes a directory information file 109 and a data file 110.
, Update information file 111, and update history file 11
8 and. In the present embodiment, the directory information file 109, the data file 110, the update information file 111, and the update history file 118 are provided one by one. It may be provided for each identifier. In this case, since the collection of the restoration information is registered, the backup registration file 11 is registered as in the first embodiment.
6 must be provided in the restoration information external mechanism device 151. The external storage device 103 also holds the local file 117 that is the target of the restoration process. In this embodiment,
The external storage device 151 for restoration information is stored in the local file 1
It is provided separately from the external storage device 103 holding 17
An area holding the restoration information and the local file may be provided in the same external storage device.

The arithmetic processing unit 152 uses, as a storage area,
The latest directory information table 106 for holding the directory information of the local file 117 and the checkpoint counter 119 are held. The latest directory information table 106 is a table for storing the directory information of the last collected local file 117. Although the storage area temporarily reserved and used is not shown in FIG. 15, the current directory information is used as the storage area temporarily reserved in the arithmetic processing unit 152 when the restoration information is collected. There is a table (not shown), and further, as a storage area temporarily secured in the arithmetic processing unit 152 at the time of file restoration processing,
There is a recovery directory information table (not shown).

Further, the arithmetic processing unit 152 includes a backup execution unit 154 for collecting restoration information and restoring a local file, and a local file using unit 115 for referring to and updating the local file 117. In the second embodiment, as in the first embodiment, the arithmetic processing unit 152 includes a main storage device (not shown) and a central arithmetic processing unit (not shown). The respective units 154 and 115 of the arithmetic processing unit 152 are realized by the central processing unit executing instructions stored in the main storage device in advance.

FIG. 16 shows the procedure for restoring local data in the second embodiment. When the information processing device 150 is activated, the backup execution unit 154 first checks whether or not the update history file 118 exists in the restoration information external storage device 151 (step 1601). If the update history file 118 does not exist, it means that the backup process has not been performed until now, so the backup execution unit 154 executes the initial backup process (step 1602). Details of the initial backup process will be described later. When the update history file 118 already exists, the backup execution unit 154 obtains the directory information of the newest restoration information from the contents of the directory information file 109 and the update information file 111, and rebuilds the latest directory information table 106 ( Step 1603).

As described above, steps 1601 to 1603 are executed as part of the initialization process when the information processing device 150 is started.

When the above steps are completed, the backup execution unit 154 determines whether the predetermined time has elapsed (step 1604) or the input / output device 12
When an instruction to collect restoration information is received from the user via step 5 (step 1605), update backup processing is performed (step 1606). Further, upon receiving a file restoration instruction from the user via the input / output device 125 (step 1607), the backup execution unit 154 executes file recovery processing (step 1608). Details of the update backup process and the file recovery process in the second embodiment will be described later.

Initial backup processing (step 160)
The procedure of 2) is shown in FIG. In the initial backup process of this embodiment, the backup execution unit 154 first initializes the latest directory information table 106 and the checkpoint counter 119, and the directory information file 1
09, data file 110, update information file 11
1 and generation of the update history file 118 (step 1701).

Next, the backup execution unit 154 creates the directory information 700 of the local file 117 that guarantees the restoration, stores it in the directory information storage area 700, and stores the directory information storage area 700 in the latest directory information table 106. The file is linked to the list of directories to which it belongs (step 1702). If the directory information of the directory is not yet in the latest directory information table 106, the backup execution unit 154 creates the directory information about the directory and stores it in the directory information storage area 700 in the latest directory information table 106. Is first connected to the list of parent directories of the directory, and then the directory information of the local file 117 is stored.

As described above, when the directory information among the restoration information of the local file 117 for which the restoration information is to be collected is collected in the latest directory information table 106, the backup execution unit 154 reads the file contents of the file 117, Additional writing is performed in the data file 110, and the write start position is stored in the data start position 705 of the directory information 700 of the file (step 1703). When writing the file contents to the data file 110, data compression of the file contents may be performed if necessary. When the data has been compressed, the compression flag storage area 301 stores a flag indicating that the data has been compressed.

The backup executor 154 repeats the above-described steps 1702 and 1703 until all the local files 117 predetermined to collect the restoration information are executed (step 1704). When the above restoration information collection processing (steps 1702 and 1703) is performed for all local files for which all restoration information is collected, finally,
The backup execution unit 154 sets the contents of the latest directory information table 106 to the directory information file 109.
To end the initial backup process (step 1705). At the time of this writing, the backup execution unit 154 determines that the sub-directory address 706 and the directory information address 70 represented by the address on the memory.
7 is converted into a sub directory index 206 and a directory information index 207 which are respectively index number formats.

Next, the update backup process (step 1604) of this embodiment will be described. The procedure of the update backup process (step 1604) of this embodiment is shown in FIG.
8 shows. The backup execution unit 154 first reads the current directory information of the local file 117, secures an area in the arithmetic processing unit 152, stores the read directory information therein, and stores the current directory information table (not shown). ) Is created (step 18)
01). The structure of this current directory information table is the same as the structure of the latest directory information table 106.

Next, in step 1801, the backup execution unit 154 secures an area in the main memory of the arithmetic processing unit 121, creates a current directory information table, and prepares the latest directory information table for comparison processing. The storage area of the pointer p for indicating the directory information storage area which is the processing target node of 106 and the storage area of the pointer q for indicating the directory information storage area which is the processing target node of the current directory information table are secured. To do. Further, the backup execution unit 154 stores the address of the leading directory information storage area of the latest directory information table 106 and the address of the directory information storage area of the current directory information table in the storage areas of the pointer p and pointer q, respectively. By doing so, p and q are initialized (step 1802).

Next, the backup execution unit 154 compares the contents of the latest directory information table 106 with the contents of the current directory information table created in step 1801, creates update information for the changed one, and restores it. Collect information.

First, the backup execution unit 154 detects a file in which any one of creation, update, and deletion is changed (step 1820), and update information and file contents of the file in which the change is detected are detected. Update information file 111 and data file 110, respectively
It is additionally stored in and (step 1821).

The process of adding restoration information in steps 1820 to 1821 (referred to as restoration information adding process) is repeated as long as at least one of the pointers p and q points to a node (step 1816). If the node to be compared points to a subdirectory, the restoration information adding process (steps 1820 to 1821) described above is recursively executed for the subdirectory as well. If the subdirectory has been created or deleted, the processing is performed assuming that all files or subdirectories belonging to the subdirectory have also been created or deleted.

When neither of the pointers p and q points to a node, the backup execution unit 154 copies the current directory information table created as described above to the latest directory information table 106.
As a result, since the collection of the restoration information is completed, the backup execution unit 154 ends the update backup process (step 1604).

The above change file detection process (step 1
As shown in FIG. 23, information 820) indicates the information held in the current directory table (showing the directory structure at the time of executing the update backup process (step 1604)) and the latest directory table 106.
The file that has been changed since the latest directory table information was collected by comparing it with the information (which shows the directory structure at the time when the most recent restoration information was collected). And the content of the change.

The backup execution unit 154 first compares the file name held in the node pointed to by the pointer p with the file name held in the node pointed to by the pointer q (step 1803). If the pointer p or q does not point to any node, it is considered that the pointer points to a node having an infinitely large file name, and comparison is performed.

If the comparison result in step 1803 is the same, the backup execution unit 154 further determines the size, attribute, and last update date / time of the files held in the respective directory information storage areas indicated by p and q. A comparison is made (step 1804). If a difference is detected in step 1804, the backup execution unit 154 creates update information whose update command information is "file update" (step 1807), and sets pointers p and q respectively.
Update to point to the next node (step 180
8), the changed file detection processing (step 1820) is terminated, and the processing is updated information storage processing (step 1821).
Proceed to. Note that steps 1806, 1808, 181
Since we have already compared all the nodes at 0,
When there is no node to be used for comparison, a value that does not point to any node is stored in the pointer storage area.

If the file name stored in the node pointed to by pointer p is larger than the file name stored in the node pointed by pointer q in step 1803, the directory information has already been sorted using the file name as a key. The file indicated by the contents of the node pointed to by the pointer q is newly created. Therefore, the backup execution unit 154 creates update information whose update command information is “file creation” (step 1
805), the pointer q is updated to point to the next node (step 1806), the changed file detection process (step 1820) is terminated, and the process proceeds to the update information storage process (step 1821).

If the file name stored in the node pointed to by pointer p is smaller than the file name stored in the node pointed to by pointer q in step 1803, the file pointed to by the node pointed to by pointer p has already been deleted. ing. Therefore, in this case, the backup execution unit 154
Creates update information in which the update command information is "delete file" (step 1809), updates the pointer p to point to the next node (step 1810), and ends the changed file detection process (step 1820). hand,
The process proceeds to the update information storage process (step 1821).

In the update information storage processing (step 1821), the backup execution unit 154 first, as shown in FIG. 24, first updates the local file 117 from which the update information is to be collected.
Is additionally written in the data file 110, and the update information created in any of steps 1805, 1807, and 1809 is additionally written in the update information file 111 (step 1811). By this processing, the data file 110 or the update information file 11
If the size of 1 does not exceed the predetermined file size (step 1812), the backup execution unit 154 ends the update information storage process (step 1821) and advances the process to step 1816 described above.

On the other hand, when the size of the data file 110 or the update information file 111 exceeds the predetermined file size, the old update information is deleted to keep the file size within the predetermined limit value. The backup execution unit 154 performs the following processing.

First, the backup execution section 154 updates the update information having the oldest checkpoint number among the update information in the update information file 111 (oldest pointer 50
The contents of the directory information file 109 are updated in accordance with the contents of (pointed by 1) (step 1813).

Next, the backup execution unit 154 deletes the data held in the data file 110 and the update information file 111 that is no longer needed by the process of step 1807 (step 181).
4). At this time, the unnecessary data is determined in the same manner as in step 1010 described above. Further, since the data head positions 205 and 407 stored in the directory information file 109 and the update information file 111 may change due to the deletion of data, the backup execution unit 154 also corrects them. Finally, the backup execution unit 154 updates the oldest pointer 501 to point to the storage area 505 located next to the update history storage area 505 currently pointed to (step 1815). Through the processing of steps 1813 to 1815, the data file 110 and the update information file 11
The size of 1 is reduced. When step 1815 is executed, the backup execution unit 154 ends the update information storage process (step 1821) and proceeds to step 1816 described above, as in the case where the file size does not exceed a predetermined value.

Next, the procedure of the file recovery process (step 1606) in the second embodiment will be described with reference to FIG. File recovery process (step 1606)
Uses the restoration information accumulated in the restoration information external storage device 151 and the contents of the latest directory information table 106 to specify one or more local files 1
17 is a process of reconstructing 17 to the contents at the time corresponding to the specified checkpoint number.

When the backup execution unit 154 receives a file recovery processing request from the user via the input / output device 125, it first displays the update history information stored in the update history file 118 on the input / output device 125. Output to the device (step 1901). This is so that the user can confirm the display and determine the checkpoint number to be instructed. The update history information displayed at this time is a set of the checkpoint number and the update date and time.

Next, the backup execution unit 154 accepts the input of the checkpoint number designating the point of time when the file is restored via the input / output device 125 (step 1902), and the point of time designated by the designated checkpoint number. A recovery directory information table (not shown) that has been updated according to the above update information 400 is created (step 1903). This step 1903
The method of creating the recovery directory information table in is similar to the creating method (illustrated in FIG. 21) in the case of step 1212 (FIG. 12) described above.

Next, the backup execution unit 154 outputs the contents of the created recovery directory information table to the display device of the input / output device (step 1904).
This is so that the user can confirm the directory information at the time point corresponding to the checkpoint number specified and determine the file to be restored. Further, the backup execution unit 154 receives the designation of the file name to be restored via the input / output device 125 (step 1905). In this embodiment, a plurality of file names can be accepted.

The backup execution unit 154 that has received the designation of the file name restores the contents of one of the designated files from the recovery directory information table and the data file 110, and stores it as the file 117 in the external storage device 103. By storing, the specified file is restored to the contents at the specified time (step 19).
06). The method of reconstructing the contents of the local file 117 is the same as the method (illustrated in FIG. 22) in the case of step 1217 (FIG. 12) described above. The file restoration in step 1906 is repeated until the restoration of all the files designated in step 1905 is completed (step 1907).

According to the present embodiment, even if the information processing apparatus is not provided with means for sequentially recording the contents of file update operation, by applying the file restoring method of the present embodiment, Restoration can be guaranteed.

[0127]

As described above, according to the present invention,
Even if a special backup device is not added to the client connected to the server via the network, the restoration of the file held in the client's storage device can be guaranteed, and the data can be preserved at low cost. The effect is that it becomes possible.

Further, according to the present invention, the information processing apparatus is provided with
Even if a means for sequentially detecting and recording the content of the update operation of the data stored in the storage device is not provided, the update content of the file can be detected at regular intervals.
There is an effect that it can be applied to an inexpensive personal computer system in which the above means is not provided.

[Brief description of drawings]

FIG. 1 is a configuration diagram of an information processing device system according to a first embodiment to which the present invention is applied.

FIG. 2 is an internal structure diagram of a directory information file according to the first embodiment.

FIG. 3 is an internal structure diagram of a data file according to the first embodiment.

FIG. 4 is an internal structure diagram of an update information file according to the first embodiment.

FIG. 5 is an internal structure diagram of an update history file according to the first embodiment.

FIG. 6 is an internal structure diagram of a backup registration file according to the first embodiment.

FIG. 7 is a structural diagram of a latest directory information table according to the first embodiment.

FIG. 8 is a flowchart showing a data backup procedure in the first embodiment.

FIG. 9 is a sequence diagram illustrating contents of initial backup processing according to the first embodiment.

FIG. 10 is a sequence diagram illustrating contents of update backup processing according to the first embodiment.

FIG. 11 is a sequence diagram showing the contents of directory information table copying processing in the first embodiment.

FIG. 12 is a sequence diagram illustrating contents of file recovery processing according to the first embodiment.

FIG. 13 is a diagram showing a table initial state when comparing directory information tables in the first embodiment.

FIG. 14 is a flowchart illustrating a directory information table comparison method according to the first embodiment.

FIG. 15 is a configuration diagram of an information processing apparatus of a second embodiment to which the present invention is applied.

FIG. 16 is a flowchart showing a data backup procedure in the second embodiment.

FIG. 17 is a flowchart illustrating a procedure of initial backup processing according to the second embodiment.

FIG. 18 is a flowchart illustrating a procedure of update backup processing according to the second embodiment.

FIG. 19 is a flowchart illustrating a procedure of file recovery processing according to the second embodiment.

FIG. 20 is a flowchart showing a procedure of restoration information initialization processing.

FIG. 21 is a flowchart showing a procedure of a recovery directory information table creation process.

FIG. 22 is a flowchart showing the procedure of a file content detection process used for restoration.

FIG. 23 is a flowchart showing a procedure of file change detection processing.

FIG. 24 is a flowchart showing a procedure of update information storage processing.

[Explanation of symbols]

101 ... Client, 102 ... Server, 103,
104 ... External storage device, 106 ... Latest directory information table, 107 ... Server side backup execution unit,
108, 112 ... Communication processing unit, 109 ... Directory information file, 110 ... Data file, 111
... update information file, 113 ... client side backup execution unit, 114 ... old directory information table, 115 ... local file using means, 116 ...
Backup registration file, 117 ... Local file, 118 ... Update history file, 119 ... Checkpoint counter, 120 ... Backup registration file, 121, 122, 152 ... Arithmetic processing unit, 12
5 ... I / O device, 154 ... Backup execution unit, 1
51 ... External storage device for restoration information, 200, 700 ... Directory information storage area, 305 ... Local file information storage area, 400 ... Update information storage area, 600 ...
Registration information storage area.

 ─────────────────────────────────────────────────── ─── Continued Front Page (72) Inventor Wataru Kosuge 5-2-1, Omika-cho, Hitachi-shi, Ibaraki Hitachi Ltd. Omika Plant

Claims (8)

[Claims] 1. At least one server having a storage device and an arithmetic processing unit, and at least one client connected to the server via a communication line, the client holding at least one file. In a method for maintaining a file in a network system having a storage device for storing, an arithmetic processing unit having means for creating, updating, and deleting the file, and an input device for accepting an input from the outside, a restoration instruction, An input reception process for receiving an instruction of a file to be restored and an instruction about a time to restore through the input device, an update backup process for collecting information for restoring the file, and the input device. Restores the above-specified file to the one at the time specified via the input device. At least run the file recovery process activation command the updating backup processing, issued by a predetermined cycle,
And a file that is started by at least one of the backup processing execution instructions received via the input device, and the arithmetic processing unit of the client has made any change in creation, update, and deletion. And a change file detection step of detecting the update information indicating the content of the change and transferring the information specifying the file and the update information of the file to the server, and the arithmetic processing unit of the server that accepted the transfer. Collects restoration information, which is information for restoring the file, including directory information indicating a directory structure, the content of the detected file, and update information of the file, and collects the restoration information from the server. And a step of storing update information to be stored in the storage device of A step of being activated by a restore instruction of the file received by the arithmetic processing unit of the client, the arithmetic processing unit of the client transferring to the server the instruction at the time of restoration received via the input device; The processing unit of the server, which has received the transfer of the instruction at the time point, creates a recovery directory information table representing the directory at the instructed time point based on the directory information and the update information, and the client. And a step of transferring the instruction of the file to be restored received through the input device to the server, and the arithmetic processing device of the server receiving the transfer of the file instruction is configured to perform the recovery directory information table. And the contents of the file to be restored included in the above restoration information A step of reconstructing the contents of the instructed file at the instructed time point based on the above, and transferring the contents to the client, and an operation of the client that accepted the transfer of the reconstructed file contents. A processing device stores the content of the reconstructed file in the file to be restored, the file maintenance method. 2. The change file detection step according to claim 1, wherein the changed file and its change content are detected by comparing the state of the directory at the time of executing the step and the directory information collected last. A file maintenance method characterized by being carried out. 3. The client according to claim 1, further comprising an output device that outputs information, wherein the instruction processing of the file to be restored in the input receiving process is performed by the arithmetic processing device of the server. The recovery directory information table is transferred to the client, and the arithmetic processing unit of the client that has received the transfer of the recovery directory information table displays the contents of the recovery directory information table on the output device and displays the contents. The file maintenance method is performed by receiving an instruction of a file included in the created directory information. 4. An information processing apparatus comprising: a storage device that holds at least one file; an arithmetic processing device that has means for creating, updating, and deleting the file; and an input device that receives an input from the outside. In the file maintenance method, a process of receiving an instruction for restoration, an instruction of a file to be restored, and an instruction regarding a point of time of restoration via the input device, and collecting information for restoring the file At least executing an update backup process and a file recovery process for restoring the file instructed via the input device to a file at the time instructed via the input device, and the update backup process is predetermined. Start-up command issued by the specified cycle,
And a file that has been started by at least one of the backup processing execution instructions received via the input device and that has been modified by any of creation, update, and deletion, and the content of the modification. A change file detection step of detecting update information, directory information indicating a directory structure, the content of the detected file, and update information of the file, and restore information that is information for restoring the file. And a step of storing the update information in the storage device, wherein the file recovery processing is started by a restore instruction of the file received through the input device, and the directory information and the update are stored. The recovery directory information table that indicates the directory at the time pointed above based on the information Creating a file, the recovery directory information table, and the content of the file to be restored included in the restoration information, and the step of reconstructing the content of the designated file at the designated time point. And a step of storing the content of the reconstructed file in the file to be restored, the file maintenance method. 5. The change file detection step according to claim 4, wherein the changed file and the changed contents are detected by comparing the state of the directory at the time of executing the step and the directory information collected last. A file maintenance method characterized by being carried out. 6. The processing system according to claim 4, wherein the client further includes an output device that outputs information, and an instruction of a file to be restored in the input acceptance process is accepted by the arithmetic processing device of the server. The recovery directory information table is transferred to the client, and the arithmetic processing unit of the client that has received the transfer of the recovery directory information table displays the contents of the recovery directory information table on the output device and displays the contents. The file maintenance method is performed by receiving an instruction of a file included in the created directory information. 7. At least one server having a storage device and an arithmetic processing unit, and at least one client connected to the server via a communication line, the client holding at least one file. In the maintenance system of the file in a network system having a storage device for storing, an arithmetic processing device having means for creating, updating, and deleting the file, and an input device for receiving an input from the outside, the arithmetic processing of the client. The device has a start command issued at a predetermined cycle,
And a file which is activated by at least one of the execution instructions of the backup process received through the input device and which has been changed by any one of creation, update, and deletion, and the contents of the change. Change file detecting means for detecting update information and transferring the information for specifying the file and the update information to a server; and receiving an instruction to restore the file via the input device, Means for receiving an instruction about a time point to restore and transferring information indicating the time point to the server, and means for receiving an instruction of a file to be restored via the input device and transferring an instruction of the file to the server When the transfer of the reconstructed file contents of the server is accepted, the file contents are stored in the file to be restored. When the processing unit of the server receives the transfer of the information indicating the changed file and the update information of the file, the processing unit of the server receives the directory information indicating the configuration of the directory, the content of the file, and Update information storage means for collecting restore information, which is information for restoring the file, including the update information of the file, and storing the restore information in the storage device of the server, and transfer of information indicating the time point Means for creating a recovery directory information table that represents the directory at the time point instructed based on the directory information and the update information; and when receiving the transfer of the information that specifies the file, the recovery When the above instruction is given based on the directory information table for restore and the file contents of the restore target included in the above restore information File maintenance system to reconstruct the contents of the above instruction, the contents and having a means for transferring to the client in. 8. An information processing device comprising: a storage device for holding at least one file; an arithmetic processing device having means for creating, updating, and deleting the file; and an input device for receiving an input from the outside. In the file maintenance system, the arithmetic processing device is instructed via the input device to update backup processing means for collecting information for restoring the file, and the file instructed via the input device. And a file recovery processing means for restoring to a point in time when the update backup processing means is a start command issued in a predetermined cycle,
And a file that has been started by at least one of the backup processing execution instructions received via the input device and that has been modified by any of creation, update, and deletion, and the content of the modification. A change file detecting unit that detects update information, directory information indicating a directory structure, the content of the detected file, and the update information of the file, and is restoration information that is information for restoring the file. Update information storage means for collecting information and storing the restoration information in the storage device, wherein the file recovery processing means is started by a restoration instruction of the file received through the input device, Based on the directory information and the update information, the recovery directory indicating the directory at the point of time is instructed. Based on the means for creating a bird information table, the recovery directory information table, and the contents of the file to be restored included in the restoration information, and reconstructs the contents of the instructed file at the instructed time point. And a means for storing the contents of the reconstructed file in the file to be restored, the file maintenance system.