JPH11143848A - Maintaining method of distribued information processing system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce communication load owing to the dispersion of traffic by the other information processor switching processing information at the present point of time to new processing information after the other information processor receives the new processing information from a specified information processor on condition of receiving a command. SOLUTION: An operation of an operator of a reversing device executes an inquiry operation of maintenance execution (601), an inquiry of a version is requested to a management device, response is received from the management device and version information is acquired (602 and 603). When version information of a self-device is compared with version information of the management device and there is difference, new processing information is carried out (604 to 606). Further, the new processing information is sent to the management device (608) and the reversing device side receives the new processing information (609). Consequently, it is possible to reduce communication load owing to the dispersion of traffic.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information processing method for a distributed information processing system, and more particularly to a distributed information processing system suitable for exchanging programs and the like in a plurality of distributed information processing apparatuses. Maintenance method.

[0002]

2. Description of the Related Art In recent years, techniques for decentralizing computer systems and the like have been developed due to higher performance and lower cost of information processing systems such as computers and further development of network technology.

In distributing the computers in such an information processing system, it is necessary to modify or add programs in order to cope with program defects and program function expansion. I have.

When the program of each device is modified, the modification work is developed one by one, and the program or the like is
There has been a method of carrying the FD by storing it in a storage medium such as a (floppy disk).

In recent years, there has been a method using a client server system in which one machine in the system is used as a common information server to reduce the number of maintenance steps of common information for all machines. In other words, when the common information changes, the server issues a request to acquire the common information to all machines, and each client compares the time obtained by the server with the time obtained last time. There is a known method for capturing data from a server.

[0006]

According to the prior art, when an attempt is made to perform maintenance in a distributed system, as the size of the system increases, it is difficult to rotate one place at a time.
take time. Even if it is desired to switch all at once, it is necessary to send a person to each device for switching, which is a problem.

In the maintenance method using a client-server system, when the number of devices to be replaced increases, the network load increases due to simultaneous data acquisition requests and transfer processing to the server, and the processing of the server computer increases. There was a possibility that the response would be delayed due to exceeding the ability and the replacement could not be performed all at once.

Further, when the contents of correction are large-scale, it takes a considerable amount of time to transfer only the contents of correction, and there is a problem in that the time for replacement is limited.

Further, in the data transfer processing time zone and the like, since the operation of each computer is not a common time zone, the use of the operator is limited due to a decrease in responsiveness and the like, causing a problem in usability. There was a possibility.

SUMMARY OF THE INVENTION An object of the present invention is to provide an environment in which data transfer and replacement are performed by operation and judgment of a computer operator in a time period during which processing of each computer is not performed in maintenance of a plurality of distributed computers, thereby facilitating the use of the computer. It is an object of the present invention to provide a maintenance method for a distributed information processing system which aims to create a system and can recover even if the replacement fails.

[0011]

The following means are conceivable to solve the above-mentioned problems.

According to the present invention, a specified information processing apparatus among a plurality of distributed information processing apparatuses is queried from another information processing apparatus for a version having a current processing information, and as a result of the inquiry, To the operator of the information processing device, and instructs the switching to the new processing information according to the instruction of the operator based on the result of the inquiry, and on receipt of this instruction, the other information processing device After receiving new processing information from an information processing apparatus, another information processing apparatus employs a maintenance method for a distributed information processing system in which current processing information is switched to new processing information.

Even if the contents of the maintenance are distributed from the designated information processing apparatus, the switching is not performed until the switching instruction is issued from the instruction apparatus such as the management of the operation version of each apparatus and the start-up. It is desirable to carry out replacement.

[0014] When a switching instruction is received, it is desirable to perform the replacement when the next apparatus is started. Also,
This switching can be simultaneously switched to the new processing information at a certain timing when the current processing information and the new processing information are managed in two planes.

Further, it is desirable that the distributed maintenance program and data are saved in each device before the replacement in order to prepare for a failure, and can be used at the time of restoration.

It is desirable that a record at the time of maintenance be kept in each device and referred to by the management device as needed.

That is, according to the present invention, a specified information processing device among a plurality of information processing devices distributed is inquired from another information processing device about the version of the current processing information that should be present. As a result, it notifies the operator of the other information processing apparatus, instructs the switching to the new processing information according to the instruction of the operator based on the result of the inquiry, and the other information processing apparatus is conditioned on receiving this instruction. After receiving the new processing information from the designated information processing apparatus, another information processing apparatus switches the current processing information to the new processing information, so that the communication amount can be minimized. In addition, since this method allows the operator to select a time period that limits the use of the information processing device,
The burden on the operator can be minimized.

Further, even if the replacement fails, the original operating program or data can be recovered by the recovery processing without transferring the data again.

[0019]

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

FIG. 3 shows a network configuration example for explaining one embodiment of the present invention.

301, 305, 306, 307, 30
9, 310 represents an apparatus. Loops 302 and 304, which are networks, are connected via a coupling device 303 such as a router, and data transmission is reliably performed. Here, the device 301 is an information management device, and the devices 305 and 306,
Let 307, 309, and 310 be maintenance target devices for which replacement is desired.

FIG. 1 shows a functional block diagram of the device to be replaced. In FIG. 1, computers are connected via a network. The network can be realized by, for example, a LAN (local area network) using an optical fiber. The computer includes a system management unit, a version management table, a running file, a switching file, a save file, a user program, and a management information file.

These are realized by, for example, a main storage unit in a computer and a program stored therein.

The user program is, for example, various application programs, and is a maintenance target program to be replaced. The system management unit has a data transmission / reception processing function for exchanging programs and data for exchange with other devices, a version management function for managing stored programs and data, an exchange function for actually exchanging, starting and stopping the computer itself. And a user notification function that prompts the operator to perform an operation. The data transmission is performed according to a predetermined communication protocol.

The version management table is a means for storing whether or not the above-mentioned programs have been replaced.
The content is started before the switching is performed, is referred to when the computer is restarted, and is updated after the replacement is performed.

The running file indicates a file before loading of the user program and data, for example, an object module and a command file for storage. The switching file indicates a user program to be replaced and a file before data loading, for example, an object module or a command file for storage.

Similarly, the backup file is a file prior to loading of the user program and data being executed, for example, when the replacement is performed in preparation for failure of the actual replacement operation.
The object module and the command file for storage are saved.

The management information file is a file in which the version of the operating program and the record at the time of replacement are recorded.
For example, it is realized by a UNIX file.

Similarly, FIG. 2 shows a functional block diagram of a management device for distributing a program to be replaced.

The network can be realized, for example, by a LAN (local area network) using an optical fiber. The computer includes a system management unit, a version management table, a switching file, a management information file, and a user program.

These are realized, for example, by a main storage unit in a computer and a program stored therein.

The user program is, for example, various application programs, and is a work program corresponding to a maintenance target to be replaced.

The system management unit has a data transmission / reception processing function for exchanging exchange programs and data with other devices, a version management function for managing distributed programs and data, and a recording / editing processing function for recording distributed results and exchanges. Have. The data transmission is performed according to a predetermined communication protocol.

The switching file indicates a user program to be replaced and a file before data loading, for example, an object module or a command file for storage.

The management information file is a file in which the version of the operating program and the record at the time of replacement are recorded.
For example, it is realized by a UNIX file.

Now, let us consider a case where it is desired to maintain the device 306. First, FIG. 5 shows registration of a switching file to be maintained by the management device. In step 501, a switching file to be maintained is registered. The registration file includes a switching program and a switching procedure file indicating a loading method at the time of switching.

Next, in step 502, the version of the switching file is confirmed.

If an incorrect switching file is registered,
It can be canceled by the cancel instruction in step 503, and the file registration can be started again. After that, the switching version registered is responded to the inquiry of the version from each device by 504.

Next, FIG. 6 shows a processing procedure up to the reception of the switching file in the replacing apparatus.

In step 601, an operation for inquiring of maintenance is accepted by the operation of the operator of the replacement apparatus. The inquiry operation is performed by a screen operation on the device or an input from a console screen. At this time, it is determined by the operation of the operator whether to reflect the new processing information immediately or to store the new processing information by the two-sided management, and store it in the version management table.

Thereafter, at step 602, a version inquiry request is made to the management device.

In this embodiment, the data transmission "P
As a protocol (communication protocol) using the "OINT TO POINT line", TCP / IP (Transmission Con
trolProtocol / Internet Protocol). This protocol is a transmission protocol that has high reliability and provides one-to-one communication.

A configuration having a transmission error check function, an error retransmission function, and the like may be used.

FIG. 4 shows an image diagram of transmission using the “POINT TO POINT line”.

In step 603, a response is received from the management device, and version information is obtained.

At this time, the version information held in the own device is compared with the version information acquired from the management device, and
If there is no difference, the fact that there is no information update is displayed and notified to the operator (step 604).

Further, the version information held in the own apparatus is compared with the version information obtained from the management apparatus, and if there is, a message as to whether or not to acquire new processing information is output, and the judgment is left to the operator. (Step 60
5).

The operator selects the execution if the operation of the apparatus does not affect his own operation. If there is an influence, the acquisition of the new process information is canceled by a cancel instruction (steps 606 and 607).

If there is no cancel instruction, step 60
In step 8, a request for acquiring new processing information is transmitted to the management device. If there is a plurality of new process information, step 6
In step 04, a request is sent for only information that has a difference when compared.

The management apparatus transmits the new processing information of the version requested in step 505 as data. When the transmission is completed, the management apparatus updates the version management table of the apparatus and records that the new processing information has been transmitted as maintenance information.

The replacement apparatus receives the new process information in step 609.

The new processing information is stored as a switching file. The procedure for the actual replacement work will be described with reference to the flowcharts of FIGS. First, at step 701, the switching-side apparatus checks whether the version of the switching file is the version to be stored as new processing information. At that time, the user can refer to the version management table to determine whether the switching instruction is for two-sided management or whether the user wants to update immediately.

In this embodiment, a case will be described in which the switching is performed at the next start-up.

On the replacement apparatus side, the restart is performed in step 702 by the configuration control function. Step 7
In step 03, the apparatus checks whether the startup has been performed normally. At this time, the replacement device outputs the message completed by the user notification process and notifies the operator. If there is a problem, the recovery processing in step 705 is performed, and the intended maintenance is performed again. If successful, normal operation is performed in step 704.

The details of the processing procedure on the replacement device side when the system is restarted will be described with reference to FIG.

At step 801, an instruction to restart the CPU is accepted. At this time, it is determined whether or not there is a switching instruction. If not, the normal start-up process of step 802 is performed, but if there is a switching instruction, replacement is performed. In step 803, the file being executed is saved to a save file in case of replacement failure.

Next, in step 804, the switching file is saved in the execution file.

In step 805, the switching file is loaded into the real machine environment.

In step 806, the loading result is stored in the replacement result file, and it is determined whether the loading process has been completed normally.

If the processing has been completed normally, the version management table is updated at the processing end in step 807, and the version is displayed. Thereafter, in step 808, a message indicating the completion of startup is displayed. If the loading has failed, in step 809, the version management table is updated with an abnormal end, and the version is displayed.

Thereafter, an abnormal termination message is output in step 810, and a recovery process is awaited.

The normal processing of the actual replacement has been described above.

Next, a method for recovering from a failed replacement will be described with reference to FIGS. 9 and 10. FIG.

First, as a preliminary preparation, the switching-side apparatus checks in step 901 whether the new processing information has failed to switch in the corresponding version to be maintained. After that, the actual recovery processing is performed. First, in step 902, the replacement side apparatus issues a version restoration instruction to the replacement target apparatus.

At step 903, the version management table is updated, and the fact that the recovery processing will be performed at the next start-up is stored.

In step 904, the replacement device is restarted. In step 905, the apparatus checks with the replacement apparatus whether the startup has been performed normally. The recovery process only recovers the programs and data that were running at that time, and if maintenance is actually desired, the switching file is reviewed and the process is performed again.

The details of the recovery processing procedure on the replacement device side when the system is restarted will be described with reference to FIG.

In step 1001, an instruction to restart the CPU is received. At this time, it is determined whether or not there is a switching instruction. If not, the normal start-up process of step 1002 is performed, but if there is a switching instruction, replacement is performed.

In step 1003, the saved file is saved in the executable file and restored to the original file. Then step 10
04 loads the save file to the actual device environment. In step 1005, the version management table is updated at the end of the process, and the version is displayed. Thereafter, in step 1008, a message indicating the completion of startup is displayed. In this way, even if the replacement fails, it can be recovered.

In the present embodiment, the maintenance is performed by the device 406, but it can be dealt with by repeating the same operation in the other devices 407, 409, 410 and the like. Further, in the present embodiment, the management device and the replacement device are separately separated.
It is possible, though only one, to implement both functions in one device.

In this embodiment, a case has been described in which it is not necessary to be aware that the device 403 exists as a network connection device by, for example, a router.
If the device 403 is configured by a gateway or the like and needs to be considered as a relay device, the device 401 and the device 4
05, the fact that there is a relay device 403 in each data transmission / reception process is stored, and the device 403 may have the functions of the distribution device and the status monitoring / switching instruction device as one device, and perform the same process.

Further, in the present embodiment, the switching by restarting is described, but the switching may be performed automatically on condition that the switching data is received without restarting. Here, the operation according to the present embodiment described above and the operation according to the above-described conventional technique will be compared.

As described in the prior art, the method of maintaining one device at a time is not efficient in terms of work efficiency, and when there is a time constraint for maintenance or when a plurality of devices are targeted, simultaneous maintenance is performed manually. , Or step by step while selecting the maintenance target device.

Further, the maintenance function using the client server function is provided when the target device is in a form of a plurality of wide-area distribution.
It is conceivable that an information acquisition request is issued all at once, which causes a problem in processing on communication or exceeds the processing capacity of the computer of the server.

Further, when the usability of the operator is reduced, it is conceivable that it is impossible to determine whether or not the maintenance is an influence.

On the other hand, according to the present embodiment, an inquiry is made from another information processing apparatus to a specified information processing apparatus among a plurality of information processing apparatuses which are distributed and arranged to determine the version of the current processing information. As a result, it notifies the operator of the other information processing apparatus, instructs the switching to the new processing information according to the instruction of the operator based on the result of the inquiry, and the other information processing apparatus is conditioned on receiving this instruction. After receiving the new processing information from the designated information processing apparatus, another information processing apparatus switches the current processing information to the new processing information, so that the communication amount can be minimized. In addition, since this method allows the operator to select a time period that limits the use of the information processing device,
Since the processing can be executed at the convenience of the operator, the burden can be minimized. Further, even if the replacement fails, the original running program or data can be recovered by the recovery process without transferring the data again.

[0077]

As described above, according to the present invention, the switching to the new information can be performed in a time zone convenient for the operator, so that the burden on the operator can be reduced and the communication amount can be dispersed. Therefore, it is possible to provide a method capable of reducing the communication load and recovering even if the replacement fails.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of a replacement device to be maintained when the present invention is implemented.

FIG. 2 is a functional block diagram of a management device when the present invention is implemented.

FIG. 3 is a configuration diagram of an entire system for explaining an embodiment of the present invention.

FIG. 4 is an explanatory diagram of POINT TO POINT transmission.

FIG. 5 is a flowchart illustrating a processing procedure on the distribution device side according to an embodiment of the present invention.

FIG. 6 is a flowchart illustrating a switching processing procedure (replacement work preparation) on the switching device side according to an embodiment of the present invention.

FIG. 7 is a flowchart illustrating a switching processing procedure (outline) of the switching device according to an embodiment of the present invention.

FIG. 8 is a flowchart showing a switching processing procedure (details) on the switching device side in one embodiment of the present invention.

FIG. 9 is a flowchart illustrating a recovery processing procedure (replacement work preparation) on the replacement apparatus side in one embodiment of the present invention.

FIG. 10 is a flowchart showing a restoration processing procedure on the replacement device side in one embodiment of the present invention.

[Explanation of symbols]

401, 405, 406, 407, 409, 410: device, 402: loop, 403: router, etc., 404: loop.

Claims (8)

[Claims] An information processing apparatus inquires a specified information processing apparatus among a plurality of information processing apparatuses arranged in a distributed manner about a current version of processing information from the other information processing apparatus. A command to switch to the new processing information is issued by an instruction of the apparatus operator, and on receipt of this command, another information processing apparatus receives new processing information from the designated information processing apparatus, and then executes other information processing. A maintenance method for a distributed information processing system in which a processing device switches current processing information to new processing information. 2. The maintenance method according to claim 1, wherein the maintenance is performed by inquiring of one management device, comparing operating versions of the devices, notifying a comparison result to an operator of the device, and starting the operation by the operator. A maintenance method for a distributed information processing system, characterized in that replacement is not performed until a switching instruction is given, and replacement is performed according to a switching instruction. 3. The maintenance method for a distributed information processing system according to claim 1, wherein when a switching instruction is received, the replacement is performed at the time of starting the next apparatus. 4. A maintenance method for a distributed information processing system according to claim 1, wherein said switching information is managed by each of the devices that perform replacement. 5. The distributed information processing method according to claim 1, wherein the distributed processing information for each device is saved in each device before the replacement in preparation for a failure. How to maintain the system. 6. The maintenance of a distributed information processing system according to claim 1, wherein a record at the time of maintenance is left in each device and can be referred to by a management device as needed. Method. 7. A maintenance method for a distributed information processing system according to claim 1, wherein maintenance is simultaneously performed when configuration control such as start-up and stop of each device is completed so that replacement can be performed simultaneously. 8. A maintenance method for a distributed information processing system according to claim 1, further comprising means for recovering when the maintenance has failed.