JPH10187520A - File update processing method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To conform an updating process for a file group on a slave server side to that of a master server by updating files on the other information processor according to an updating process kind based upon logging data. SOLUTION: Log data based upon the update of a data base 14 on the master server 1 are stacked at a log stacking process part 11 and adjusted at a log editing process part 12 as to repeated or duplicate update data, etc., and then the data are sent from a communication part 13 and received by a communication part 23 of the slave server 2; and a data base update part 21 updates the data base S24 according to the log data. At this time, a file which is already updated is obtained by accessing the data base M14 on the master server as to some file whose update is specified with the log data to update the data base S24 on the slave server 2, and a file whose deletion is specified is processed directly in the data base S24 on the slave server 2.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for maintaining a group of files owned by a plurality of servers connected to a line and each server owning the same group of files, and more particularly to a method of updating files in one server. It relates to a file update reflection process for reflecting the status to other servers.

[0002]

2. Description of the Related Art In recent years, instead of centralized processing by a host information processing apparatus installed at a center, a plurality of servers are connected to a LAN (L
Local area networks or public lines are connected, and distributed processing in which each server individually has a database is widely used. Then, it is required that common assets and the like stored in the database when information processing is performed by the distributed servers are the same on all servers. For this reason, it is necessary to perform a process for periodically reflecting the contents of file updates in the database on the master server in the file configuration of the database in the slave server.

As an update method in the conventional reflection processing,
When one of the files in the master server has been updated or the directory has been updated, the method of copying all the files to the slave server has been applied.

[0004] Alternatively, as another update method in the reflection process, the date of the process is recorded at the time of update, and the update date of the file or directory is compared with the update date of the previous process, and the date is changed. The method of copying only objects was applied.

[0005]

In the former method, when the amount of files stored in the database becomes enormous, the processing time for copying all files increases in proportion thereto. In addition, since there is no change in the file to be copied, when the number of updated files is small, the time required for copying is wastefully wasted.

In the latter method, for example, when a file name is changed (renamed) or a file is deleted, the date is not recorded, so that the slave server is excluded from the update processing. was there.

The present invention is a novel invention created in view of such a point, and extracts only an updated file from a file group of a master server, and updates the file group on the slave server side. An object of the present invention is to provide a means for matching a master server.

[0008]

An object of the present invention is to provide an information processing apparatus linked by a line, which owns a file group having the same configuration, and performs update processing in a file group of one information processing apparatus with another information processing apparatus. A file update processing method that is also applied to a file group of a processing device, wherein at least one information processing device performs an update process on a file,
Set the identification information of the update process performed on the file according to the update process type, collectively notify the data logging the update process identification information, and receive the notified data in the other information processing device, By performing the update process on the file in the other information processing device according to the update process type based on the logging data, it is possible to achieve a method of matching the configuration of each file group between the linked information processing devices. .

[0009]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a block diagram illustrating a functional configuration of a master server and a slave server to which the present invention is applied.

In the figure, 1 is a master server,
Reference numeral 10 denotes a file processing unit, 11 denotes a log accumulation processing unit, 12 denotes a log editing processing unit, 13 denotes a communication unit of the master server, and 14 denotes a database M of the master server. 2 is a slave server, 21 is a database update unit, 23
Is a communication unit of the slave server, and 24 is a database S of the slave server.

The master server 1 and the slave server 2 are connected by a line, and a database M in the master server 1
Log data (to be described later) resulting from updating 14 is
After being accumulated in the log accumulation processing unit 11 and adjusted in the log editing processing unit 12 for repeated or duplicated update data, the data is transmitted from the communication unit 13 via a line, and received by the communication unit 23 of the slave server 2 and received. The database updating unit 21 updates the database S24 based on the log data thus obtained.

At this time, for a file having an update designation (a U symbol described later) according to the log data, the database M14 of the master server 1 is accessed to obtain the updated file, and the database S24 of the slave server 2 is updated. The file having a delete designation (to be described later: D symbol) is directly processed in the database S24 of the slave server 2.

The database M1 in the master server 1
4 has a tree-like directory configuration, and the database S24 in the slave server 2 has the same configuration as that of the database M14.

The update in the database M14 is performed on a file-by-file basis, and includes an update process for each file (a type described later) and an update process for a directory in which each file is stored (a type described later). .

A file format having a tree-like directory structure will be described with reference to FIG. As shown in FIG. 4, individual files are files (names A, B, C, D,...)
And directories (path names: P, P1, Q,...) To which the files (A, B, C, D) are connected.

The files (A, B, C, D,...) Are located at the end of the tree and are described by a path from a directory indicated by a capsule box. That is, file A
And file B are connected to “directory P” from “root directory”, so the paths are respectively described as “$ P @ A” and “$ P @ B”. In the case of file C, Since it is connected to “directory P1” from “directory P”, the path is “$ P @ P1”.
$ C ", and the file D is connected to" directory Q "from" root directory ", so the path is described as" $ Q @ D ".

FIG. 2 is a diagram for explaining the file update processing on the master server 1 side, and explains the type of update processing for the above file (for example, A). That is, the boxes indicated by the boxes 31 to 35 are the processes of the file processing unit 10 in FIG. 1, and the boxes indicated by the boxes 311 to 351 are the logs accumulation processing unit 1 in FIG.
This is the first process. In the box, "{P}
“U” in parentheses described as “A (U)” indicates that file A has been updated, and “$ P @ A (D)”
"D" in parentheses indicates that file A has been deleted.

If the log data in FIG. 2 is described individually, since file A has been updated in box 31 "update / create", "@P
\ A (U) "is output as log data.

In the box 32 "delete", since the file A is deleted, "{P}" shown in the box 321 is displayed.
A (D) "is output as log data. Box 33
In the “rename”, the name of the file A is changed to B, so that the file A is deleted and the file B is updated to the same path P. Therefore, "$ P @ A (D), $ P @ B (U)" shown in the box 331 is output as log data.

In the box 34 "Move", the file A is moved from the directory P to the directory Q. Therefore, the file A is deleted, and the file B is transferred to a different path Q.
Will be updated. So box 34
"$ P @ A (D), $ Q @ B (U)" shown in FIG. 1 is output as log data.

In box 35 "copy", file B having the same contents as file A is updated to a different path Q, so only "$ Q \ B (U)" shown in box 351 is output as log data.

FIG. 3 is an explanatory diagram of a file update process on the slave server 2 side, and illustrates an update process type for a file received as log data. That is, the received log data is always provided with the “U” symbol or the “D” symbol as the update type information as described above.

Accordingly, for the file with the "U" symbol, the file in the database M14 of the master server 1 is accessed to receive a copy of the file connected to the path, and the file in the database S24 of the slave server 2 is overwritten and modified. Will be performed. on the other hand,
For the file with the “D” symbol, the file in the database S24 of the slave server 2 is deleted.

As described above, even in the processing for one file, the log data of either the update “U” or the delete “D” is notified in “update / create”, “delete”, and “copy”. However, the log data of both the update “U” and the delete “D” are notified in “name change” and “move”.

In the embodiment of the present invention, two methods are adopted as a method of acquiring log data. First method: log data newly created, collected, edited and adjusted in the master server 1 is directly sent to the slave server 2. In this case, every time the file update operation is completed, the entire amount of log data is sent out to all the linked slave servers 2 sequentially or simultaneously.

Each slave server 2 receiving the log data completes updating its own database S24 before the next file update occurs in the master server 1. Second method: log data newly created, accumulated, edited and adjusted in the master server 1 is stored in a common memory (not shown) with a sequence number and the like, and the slave server 2 is communicated via the communication unit 13. Notify only that log data for which new update work has been completed is accumulated.

In this case, the slave server 2 accesses the common memory without delay and acquires the accumulated log data. The sequence number of the acquired log data is checked against the previous sequence number, and if continuity is guaranteed,
Only the file based on the log data is updated.

On the other hand, when the continuity with the previous sequence number is lost, the database M1 of the master server 1
Then, the process proceeds to a process of copying the entire amount of No. 4. In the embodiment of the present invention described above, each update target is an individual file. However, the present invention is not limited to this.
The same applies when the directory is updated.

That is, the update processing type of the directory is
For example: 1) Make Directory (MD: Make Directory)
ctry) 2) Delete directory (RD: Remove Di)
3) Moving a path (Move) 4) Copying a directory (Copy) FIG. 5 is an explanatory diagram of a directory update process on the master server 1 side.

The log data in FIG. 5 will be described individually. In box 31D "Create", a new directory P2 is created in directory P.
“$ P | P2 (MD)” shown in FIG. 1D is output as log data.

In the box 32D "delete", the directory Q is deleted.
Q (RD) "is output as log data. Box 3
In the 4D “move”, the directory P1 is moved to the directory P2, so the file C connected to the directory P1 is deleted and the file C is updated to the designated path P2. Then, "{P {P1} C (D), {P}" shown in box 341D
P2 @ C (U) "is output as log data.

In the box 35D "copy", the directory P1 is copied to the directory P2, so that the file C connected to the directory P1 is copied to the directory P2.
"$ P @ P2 @ C (U)" shown in D is output as log data.

On the other hand, the slave server 2 which has been notified of the log for the directory is connected to the file connected to the directory by the same linkage method (the first method or the second method) as the above-described file update processing. And updates its own database S24.

Each procedure program applied to the above-mentioned file update is provided from the outside of each processing device, that is, loading the program via a line, or a portable medium such as a floppy disk or CD.
-May be provided in ROM, MO, etc.

FIG. 6 is an explanatory diagram for loading this computer program into the processing device. In the figure, 900 is a processing device, 901 is a display provided in the processing device, 903 is a driver for reading a computer program recorded on a portable medium, and 902 is a processing device 9
Reference numeral 904 denotes a portable medium, such as a floppy disk or CD-ROM. Reference numeral 905 denotes another device that is a provider of the computer program, and reference numeral 906 denotes a line.

As shown in the figure, when the computer program is provided from another processing device 905 connected online such as a network, the computer program is installed in the processing device 900 via a line 906 and then loaded into the memory 902. Be executed. If the portable medium 904 is provided directly by the portable medium 904, the portable medium
After being set to 3 and installed in the processing device 900, it is loaded into the memory 902 and executed.

[0037]

As described in detail above, according to the present invention, only the part changed on the master server side is extracted without omission including "delete" and "rename" to the slave server. Since the notification can be made, and the reflection processing is performed only on the notified part in the slave server, it is possible to efficiently provide the same asset to a plurality of servers in a short time.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a file update process on the master server side.

FIG. 3 is an explanatory diagram of a file update process on the slave server side.

FIG. 4 is an explanatory diagram of a path and a file.

FIG. 5 is an explanatory diagram of a directory update process on the master server side.

FIG. 6 is an explanatory diagram for loading a program into a processing device.

[Explanation of symbols]

 Reference Signs List 1 master server 2 slave server 10 file processing unit 11 log accumulation processing unit 12 log editing processing unit 13, 23 communication unit 14 database M 21 database update unit 24 database S 900 processing device 901 display 902 memory 903 driver 904 portable medium 905 and others Equipment 906 lines

Claims (6)

[Claims] 1. An information processing apparatus linked by a line owns a file group having the same configuration, and an update process in a file group of one information processing apparatus is applied to a file group of another information processing apparatus. A file update processing method for performing update processing on a file in at least one information processing apparatus, setting identification information of the update processing performed on the file in accordance with an update processing type, and setting the update processing identification information to By collecting and notifying the logged data, receiving the notified data in the other information processing device, and performing the update process on the file in the other information processing device according to the update process type based on the logging data. File updating, wherein the configuration of each file group in the information processing apparatus linked by the line is matched. Management method. 2. The update processing type for a file includes, at least, new creation, modification of an existing file, deletion of an existing file, renaming of an existing file name, change of the path of an existing file, or copying of an existing file. 2. The file update processing method according to claim 1, wherein any one of the following is applied. 3. The logging method according to claim 1, wherein at the time of the logging, the previously processed logging information when the processing type of the already processed file is a new creation or a modification of an existing file is invalidated. File update processing method described. 4. The method according to claim 1, wherein the acquired logging data and the generated sequence are recorded and integrated in a common memory. 5. The file update processing method according to claim 1, wherein the update processing type is applied to processing for each directory in which a file is organized. 6. An information processing apparatus linked via a line owns a file group having the same configuration, and the update processing in one information processing apparatus file group is applied to another information processing apparatus file group. A computer program that performs an update process on a file to be updated is loaded into one information processing device to perform an update process on a file, and identification information of the update process performed on the file is set and logged according to the update process type. A computer program that receives data from the information processing device on the notification side, and that performs a process for updating a file in the information processing device on the notification side in accordance with an update process type based on the logging data; Medium.