JPH06110809A - Mounting system for network file system - Google Patents

Abstract

PURPOSE:To attain the use of any client computer with the mounting of the same position by deciding a fixed position to mount a server computer regardless of the types of client computers. CONSTITUTION:A file system 13 of a server computer 11 includes the link tables 14-17 which store the pairs of architecture identifiers of a client computer 19 and path names of the file systems mounted onto the computer 19. Then the computer 11 reads the architecture information on the computer 19 based on the mounting request received from the computer 19. Then the computer 11 refers to the tables 14-17 and mounts the directory of the path name corresponding to the designated architecture to a file system 21 of the computer 19 as a directory to be mounted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mount processing system for a server computer of a network file system in response to a mount request from a plurality of client computers having different architectures.

[0002]

2. Description of the Related Art Computers having different architectures cannot share files that depend on the architecture such as files of program execution modules, and require dedicated files for each.

When a plurality of client computers having different architectures mount the file system on the same file server computer and the server computer provides a file dependent on the architecture of each client computer, a file dependent on the architecture of each client computer You need to refer to the separate directories that contain.

FIG. 4 is a diagram showing an example of a tree structure of a file system directory and a mounting state when a conventional client computer having a different architecture mounts a file system on the same server computer.

The architecture of the client computer A is SPARC, the architecture of the client computer B is 68000, the architecture of the client computer C is i368, and the file dependent on SPARC in the server computer is / usr. 68000 for sparc
Files that depend on / usr. 68000, i36
The file that depends on 8 is / usr. It is located at i386.

Since the architecture of the client computer A is SPARC, / usr. Mount the sparc. Since the architecture of the client computer B is 68000, / usr68000 is mounted on / usr. Since the architecture of the client computer C is i368, mount / usri368 on / usr.

Normally, the directory structure is not changed for each client computer, and the program is placed in the path of / usr on any computer as described above. If you do not do this, the directory where the program exists will differ for each client computer you use, so the user should be aware of the architecture of the computer he is currently using and know where the program for that computer is. There must be. To avoid this, the client computer mounts the file system on the server computer so that the paths are the same.

As a result, the user does not need to be aware of the architecture of the client computer being used, but the system administrator must know which directory of the server computer should be mounted. For each architecture of the client computer, the administrator must know the mount position suitable for each client computer and set up each client computer.

As described above, in the conventional system, when a client computer having a different architecture mounts a file system on the same file server computer, the system administrator of the client computer has a server computer having a file suitable for the client computer. Each client computer must be set up so that it knows and mounts the directory on the file system. Multiple client computers can be connected to the same server computer. Also, it can be installed geographically at a position distant from the server computer.

In the conventional system, the system administrator must manage both the server computer and the client computer. Alternatively, when the system administrator of the server computer and the system administrator of the client computer are different, the system administrator of the server computer must inform the system administrator of the client computer of the position where the client computer should be mounted.

[0011]

As described above, in the conventional system, in managing the client computer, the system administrator should check the architecture, check the file position of the server computer suitable for it, and mount it at an appropriate position. Each computer must be set up for each client computer with a different architecture.
Therefore, there is a problem in that files and the like for storing the setting information necessary for the configuration of the client computer must be prepared and managed by the number of types of the architecture of the client computer.

The present invention has been made in view of the above situation,
The purpose is to fix the position where the server computer should be mounted in one place regardless of the type of client computer, so that all client computers only need to mount the same position, and mount it for each client computer. By eliminating the need to consider the power location, the management load on the client computer can be reduced, and the location of files dependent on the client computer can be managed only by the server computer. Since there is no need to notify the system administrator of the mounting position, the client computer and the server computer can be managed independently, providing a network file system mount method that facilitates management and saves labor. The purpose is to do.

[0013]

SUMMARY OF THE INVENTION The present invention is a system having a client computer for requesting mounting of a network file system, a server computer for mounting a network file system, and a secondary storage device for holding a file system. In the above, a link table for storing the identifier for identifying the architecture of the client computer in the file system of the server computer and the path name of the file system to be mounted on the client computer in the file system of the server computer as a pair , A communication path between the server computer and the client computer, and when the client computer issues a mount request to the server computer, the architecture information of the own client computer is added to the parameter to support the server. The server computer reads the architecture information of the client computer according to the mount request received from the client computer and refers to the link table to mount the directory with the path name corresponding to the specified architecture. It is characterized in that it is mounted on the file system of the client computer as the directory of.

[0014]

According to the present invention, when mounting a remote file through the network file system by mounting, the computer architecture information of the mount request side (client computer) is added as a parameter, and the architecture information is attached to the mount request side (server computer). By interpreting and adding information indicating the directory where the file actually suited to the architecture is placed for each architecture on the directory structure and replacing the directory mounted by the client with the directory indicated by that information, the client On the computer side, the directory on the server computer always seems to be in one place, but in reality, by referring to another directory on the server side and mounting it, there is one place from the client. To appear in only the mount position.

[0015]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing the configuration of an embodiment of a network file system mounting method according to the present invention. Figure 1
, 11 is a server computer and 12 is a server computer 1.
A secondary storage device 1 and a file system 13 are constructed on the secondary storage device 12.

Reference numerals 14 to 17 denote the file system 13
14 is a file system 13
Root directory, 15 is a directory represented by pathname / usr, and 16 is pathname / usr. directory represented by arc1, 17 is pathname / usr. arc2
It is a directory represented by.

Reference numeral 18 is a communication path between the server computer 11 and the client computer 19, 19 is a client computer,
Reference numeral 20 is a secondary storage device of the client computer 19, and 21 is a file system constructed on the secondary storage device 20.

Reference numerals 22 and 23 are components of the file system 21, 22 is a root directory of the file system 21, and 23 is a directory represented by a path name / usr.

In the file system 13, 1
01 is a file control table constituting the root directory 14, 106 is a file control table constituting the directory 15 represented by pathname / usr, 114 is pathname / usr. File control table (FCT-c) 117 that constitutes the directory 16 represented by arc1
Is the path name / usr. directory 1 represented by arc2
7 is a file control table (FCT-d), and the detailed structure of these file control tables is shown in FIG.

FIG. 2 is a file control table showing the contents of the directory structure in the file system 13 of the server computer 11 referred to in the mount process.

In FIG. 2, 101 is a server computer 11.
File control table (FCT-a) for the root directory 14 of the file system 12 of the above, 102 is a file name input field for entering the file name of a file included in the root directory 14, 103 is a file control for the file name of the file name input field 102 Pointer input field for inputting pointer to table, 104 is path name / u
entry for directory 15 represented by sr, 1
05 is a pointer that points to the file control table of the directory 15 represented by pathname / usr, and 106 is the file control table (FCT) pointed to by the pointer 105.
-B), 107 is a file control table (FCT-b)
A link table storing the link information of the file represented by 106, 108 is an identifier input field for entering the architecture identifier, 109 is a path name input field for entering the path name of the link destination, 110 is an entry for the architecture arc1, and 111 is The link destination path name in the case of the architecture of arc1, 112 is / usr. a
entry for rc1, 113 is / usr. arc1
To the file control table (FCT-c) of
114 is / usr. It is a file control table (FCT-c) for arc1. 115 is / usr. arc
2, an entry for / usr. Pointer to the file control table (FCT-d) of arc2, 11
7 is / usr. It is a file control table (FCT-d) for arc2.

FIG. 3 is a flow chart showing the procedure for mounting the network file system in the above embodiment, and shows the flow of processing when the client computer mounts a different file system for each architecture on the server computer. . The operation of the embodiment of the present invention will be described with reference to FIGS. Here, a process in which the client computer 19 mounts the file system of the server computer 11 will be described. The client computer 19 is / u of the file system 21.
Consider a case where the file system 13 of the server computer 11 is mounted on sr.

The architecture of the client computer 19 is arc1. In the file system 13 of the server computer 11, a file that depends on the architecture arc1 is / usr. A file that depends on the architecture arc2 is /usr.arc1. It is summarized in arc2. / U of the file system 13 of the server computer 11
The path name of the link destination is set in sr for each of the architectures arc1 and arc2.

When the client computer 19 sends a mount request to the server computer 11 to mount the directory / usr of the server computer on the directory / usr of the file system 21, the client computer 19 also sends information indicating its own architecture as a parameter. (Step S1 in FIG. 3). The server computer 11 receives the mount request and interprets the parameters (step S in FIG. 3).
11, S12).

Here, the server computer 11 checks the file control table of the mount request directory / usr. First, check the file control table (FCT-a) 101 in the root directory, find usr there,
Than / usr file control table (FCT-b
) 106 is found (steps S13 to S15 in FIG. 3).

Next, the file control table (FCT-b)
Find the entry 110 for the architecture arc1 of the mount request source in the link table 106,
The link destination path name is obtained (steps S16 to S18 in FIG. 3).

Here, the link destination is / usr. Since it is arc1, the file control table of the root directory is checked, and usr. Find the entry for acr1. With this search / usr. Find the entry 112 for arc1. Than that / usr. Get the file control table 114 for arc1.

In this way, / usr. When the file control table for arc1 is found, a normal network file system mount process (NFS mount process) is performed thereafter for that directory as a mount process target (step S19 in FIG. 3).

As described above, when the client computer 19 issues a mount request to the server computer 11, the architecture information of the own client computer is added to the parameter to issue a request to the server computer 11, and the server computer 1
1 reads the architecture information of the client computer according to the mount request received from the client computer 19, refers to the link table, and sets the directory having the path name corresponding to the specified architecture as the mount target directory in the file system of the client computer 19. By mounting, it is possible to fix the mounting position of the server computer in one place regardless of the type of client computer, and it is not necessary to consider the mounting position for each client computer,
It reduces the management load on the client computer, and the location of files that depend on the client computer can be managed only by the server computer, eliminating the need to inform the system administrator of the client computer of the mount location. Therefore, the client computer and the server computer can be managed independently, which facilitates the management and saves labor.

In the above-described embodiment, when the client computer issues a mount request, the architecture information of the client computer is added to the parameter to notify the server computer of the architecture of the client computer which is the mount request source. However, if the client computer does not add the architecture information to the parameter at the time of mount request, the server computer receiving the mount request queries the client computer of the mount request source for the architecture and obtains the architecture information of the client computer of the mount request source. Is also possible.

[0031]

As described above, according to the present invention, regardless of the type of the client computer, by fixing the mount position of the file system on the server computer to one place, the same position can be obtained on any client computer. Mount it and you're done. As a result, when the client computer mounts the file system on the server computer, it is not necessary to consider the position to be mounted for each client computer, and therefore the management of the client computer can be reduced. Further, since it is possible to manage the position of files depending on the client computer only by the server computer, the file system of the server computer can be managed independently of the client computer.
In addition, in the case of a mechanism that automatically mounts a network file system when there is access, such as automount, prepare a file that indicates which directory on which server computer should be mounted when accessed. In this case, it is necessary to mount a different directory for different architectures in the past, so this file had to be prepared for each architecture. Since it can be limited to, only one file is required. As described above, according to the present invention, the management of the server computer and the client computer can be facilitated and the labor can be saved.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration of a system configured by applying a mount method of a network file system according to an embodiment of the present invention.

FIG. 2 is a diagram showing the configuration and relationship of a file control table in the system of the above embodiment.

FIG. 3 is a flowchart showing a procedure for mounting a network file system in the system of the above embodiment.

FIG. 4 is a diagram showing a relationship of mounts in a conventional general system.

[Explanation of symbols]

11 ... Server computer, 12 ... Secondary storage device of server computer 11, 13 ... File system constructed on secondary storage device 12, 14 ... Root directory of file system 13, 15 ... Represented by path name / usr Directory, 16 ... Path name / usr. directory represented by arc1, 17 ... Path name / usr. directory represented by arc2, 18 ... communication path, 19 ... client computer, 20 ... secondary storage device of client computer 19, 2
1 ... A file system constructed on the secondary storage device 22, 22 ... A root directory of the file system 21, 23 ... A directory represented by a path name / usr, 1
01 ... File control table (FCT-a) constituting the root directory 14, 102 ... File name input field for entering file names of files included in the root directory 14, 103 ... File control table for file name in file name input field 102 Pointer input field for inputting pointer to 104, entry for directory 15 represented by path name / usr, 105 ... path name / u
A pointer to the file control table of the directory 15 represented by sr, 106 ... A file control table (F that constitutes the directory 15 represented by pathname / usr
CT-b), 107 ... File control table (FCT-)
b) A link table storing the link information of the file represented by 106, 108 ... an identifier input field for entering the architecture identifier, 109 ... a path name input field for entering the link destination path name, 110 ... architecture arc1
, 111 ... Link destination path name of architecture arc1, 112 ... / usr. entry for arc1, 113 ... / Usr. Pointer to the file control table (FCT-c) of arc1, 114 ... Path name / usr. File control table (FCT-c) constituting the directory 16 represented by arc1, 115 ... /
usr. entry for arc2, 116 ... / us
r. Pointer to the file control table (FCT-d) of arc2, 117 ... Path name / usr. A file control table (FCT-d) that constitutes the directory 17 represented by arc2.

Claims (2)

[Claims] 1. A client computer requesting a mount of a network file system, a server computer performing a mount process of a network file system, a secondary storage device holding a file system, and a file system of the server computer, wherein: A link table for storing an identifier for identifying the architecture of the client computer and a path name of a file system to be mounted on the client computer; and a communication path between the server computer and the client computer, the client computer When the mount request is issued to the server computer, the architecture information of the client computer itself is added to the parameter to issue a request to the server computer, and the server computer receives the request received from the client computer. A network characterized by reading the architecture information of the client computer in accordance with a request and mounting a directory having a path name corresponding to the architecture specified by referring to the link table as a mount target directory in the file system of the client computer. File system mount method. 2. A client computer requesting a mount of a network file system, a server computer performing a mount process of a network file system, a secondary storage device holding a file system, and a file system of the server computer, wherein: A link table for storing an identifier for identifying the architecture of the client computer and a path name of a file system to be mounted on the client computer; and a communication path between the server computer and the client computer, the server computer Receives a mount request from the client computer, acquires architecture information from the client computer that is the source of the mount request, refers to the link table, and acquires the path name directory corresponding to the acquired architecture. Mounting scheme of the network file system, characterized in that to mount the file system of the client computer as a mounted object directory directory.