JP2002278766A - Center loading method of information processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To update data by collectively and simultaneously center loading update objects to a plurality of information processors designated through the Internet, etc. SOLUTION: The centerload (hereafter FR) A of a server 1 starts the FRB of a server 2 and centerloads transmission data, and the FRB of the server 2 receives the transmission data, subsequently erases its own IP addresses in addresses t2 to tn (204) and starts the FRB of a server 3 having the next address. FRBs after the server 2 also have an FRA function. The FRB of the server 2 uploads data to its own server and performs a completion report to the FRA of the server 1 (203). The operation of the server 3 is also the same as that of the server 2. The FRB of the server 3 starts the FRB of a server 4, but a completion report is not performed to the FRA of the server 1 to turn a timeout because the server 4 is turned off, and the FRB of the server 1 starts center loading to the FRB of a server 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an environment where a plurality of users coexist with an information processing apparatus such as a server of various manufacturers coupled to an open platform communication path such as the Internet. The present invention relates to a center information processing method for selecting and updating a target information processing apparatus and a target object. In particular, the information processing apparatus that performs center loading is a short-term processing of a plurality of types of target objects (model, OS, user program, etc.) for the information processing apparatus that is center-loaded.
And that the center load can be simultaneously performed on one or more information processing devices at low cost. Here, the open platform refers to the hardware that constitutes each terminal connected to the network,
This refers to a state in which the specifications of software such as an OS are released and existing resources in each terminal can be used when newly constructing a system operating on this network. Also, for example, requesting necessary information from a terminal device, for example, to a host computer, and sending the requested information to the terminal device is referred to as downloading. Unilaterally, for example, sending information to a terminal device.

[0002]

2. Description of the Related Art Japanese Patent Laid-Open No.
No. 571, "How to remotely upgrade software on a network", and Japanese Patent Application No. 9-11 / 1990.
No. 4671, "Remote program change method" is available. The method disclosed in Japanese Patent Laid-Open No. Hei 10-198571 is a method of upgrading information stored in a client processing system which is coupled to a server processing system on a network and communicates with the server processing system. If so, the download request is sent to the server processing system on the network, the information is downloaded from the server processing system to the client processing system via the network, and the downloaded information is stored in the flash memory of the client processing system. Upgrade existing information written and stored in this flash memory. Further, the method disclosed in Japanese Patent Application No. Hei 9-114671 rewrites a current control program operating on a network device to a new control program which is a revised version, and operates the network device with the new control program. When operation by the revised new control program becomes necessary, the new control program is transmitted from the monitoring control device to the network device via the network, and the network device executes the remote processing program and causes the communication control unit to receive the new control program. Rewrite the current control program with the new control program and reset the network device. As a result, the network device executes the startup process again, and is controlled by the rewritten new control program.

[0003]

DISCLOSURE OF THE INVENTION Problems to be Solved by the Invention
The method described in Japanese Patent Application No. 8571 relates to a download between a server processing system and a client processing system, and relates to a point-to-point processing. This is a method of remotely updating a control program of a network device via a network. However, the above-mentioned prior art does not consider the following points. 1. The target update object is centrally and collectively loaded into a plurality of designated information processing devices via a communication path such as the Internet, and updates the designated plurality of information processing devices. 2. The information processing apparatus that executes the center load transfers the program that executes the center load to another information processing apparatus having each environment (model, OS), so that the information processing apparatus that executes the center load is connected to the Internet. The center load is performed without being affected by the environment (model, OS) of a plurality of information processing apparatuses that operate while being connected to a communication path such as. Here, the porting of the program means that the program is loaded and the loaded program is changed so as to match the loaded OS. 3. At the center load, target / non-target information processing devices are collectively selected from among various information processing devices connected to the open platform, such as the company, other companies, and other areas. 4. Execution of center load The spread speed is designated, and the center load is executed for each district and area where the information processing device is installed. 5. The center load can be executed according to the global date and time for each group of information processing devices performing the center load, and for each district and area where the information processing devices are installed,
In addition, the center load can be executed from anywhere (for example, from the server 1 connected to the net A in Japan or from the server s connected to the net S in Singapore). An object of the present invention is to solve and enable the above points.

[0004]

In order to achieve the above object, the present invention provides an information processing apparatus which is connected to a communication path such as the Internet and is provided with a unique address which is connected to the communication path such as the Internet. A center loading method of an information processing apparatus for center loading data into one or more other information processing apparatuses, wherein the information processing apparatus for center loading specifies transmission information including destination information and an update object in the destination information. One or more other information processing apparatuses are collectively and centrally loaded via a communication path, and receive completion information indicating completion of center loading from each of the center-loaded information processing apparatuses via the communication path. To manage the progress of the center load. The corresponding object is updated by the update object, and are to be transmitted to the information processing apparatus to the center load via a communication path completion information indicating the center load completion. Also, the information processing apparatus that performs center loading collectively and simultaneously transmits transmission data including the diffusion designation information, the destination information, and the update object to one or more other information processing apparatuses designated by the destination information via a communication path. Center load,
In addition, the progress of the center load is managed by receiving completion information indicating the completion of the center load from each of the center-loaded information processing devices via a communication path, and each of the center-loaded information processing devices receives the completion information. The object corresponding to the update object is updated by the update object, and completion information indicating the completion of center loading is transmitted to the information processing apparatus for center loading via a communication path. Therefore, the transmission data is centrally and collectively loaded to one or more other information processing apparatuses via the communication path. Also, the information processing apparatus that performs center loading collectively and simultaneously loads the transmission data including the destination information and the update object to one or more other information processing apparatuses specified by the destination information via a communication path. A first program for realizing the function of managing the progress of the center load by receiving completion information indicating the completion of the center load from each of the center-loaded information processing apparatuses via a communication path; The information processing apparatus has a function of updating an object corresponding to the update object received by the center load with the update object, and transmitting completion information indicating completion of the center load to the information processing apparatus that has performed the center load via the communication path. Second to be realized
The information processing apparatus executes the first program when center-loaded to another information processing apparatus, and executes the second program when center-loaded to another information processing apparatus, By porting the first and second programs to an information processing device that does not store the first and second programs in the storage device, the information processing device can be center-loaded.

[0005]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS.
This will be described in detail with reference to FIG. FIG. 1 is a device connection diagram showing connection of each device in the open platform. FIG. 2 is a diagram illustrating a control relationship of each device in the upload method. FIG. 3 is a diagram showing an outline of a method of loading data to a plurality of devices in the simultaneous upload method.
FIG. 4 is a diagram for explaining a simultaneous upload path showing load development to a plurality of devices. FIG. 5 is a diagram illustrating a protocol operation between the transmitting server and the receiving server.
FIG. 6 is a diagram showing a data format of transmission data and a completion code. FIG. 7 is a diagram showing a data transmission block for explaining a data verification and recovery method. FIG.
FIG. 7 is a diagram showing a SUM check generation flow. FIG.
FIG. 8 is a diagram showing a center load A operation flow showing an outline of an operation of a transmission side program. FIG. 10 is a diagram showing a center load B operation flow showing an outline of the operation of the receiving program.

The device connection diagram of FIG. 1 shows a schematic connection of each device in an open platform including an embodiment of the present invention. Open network is Net A
(9), a net B (8), and a net C (10). The net A (9) has a server 1 (1) of a maker A and a server 2 of a maker B for an open platform.
(2) The server 4 (5) of the maker C, the server 6 (6) of the maker A in the net B (8), and the maker A in the net C (10).
Server 3 (3), maker A's server 5 (7), and maker D's server 7 (4). Each server of manufacturer A is under the operating system (OS) of manufacturer A, each server of manufacturer B is under the operating system (OS) of manufacturer B, and each server of manufacturer C is under the operating system (OS) of manufacturer C. , And each server of the maker D operates under the operating system (OS) of the maker D. Also, the server 1 (1), the server 3 (3), and the server 4 connected to the network
(5), Server 5 (7), Server 6 (6), Server 7 (4)
The server 2 (2) operated by the same company and connected to the network is operated by another company. The above one company cannot operate the server 2 (2) operated by the above other company. In this center load, a target server for center load is selected from the various servers connected to the above-mentioned open network by designating the OS of the specified manufacturer and the IP address of the server.

In FIG. 2, server 1 (IP # 1) (10
0), server 3 (IP # 3) (101), server 4 (IP # 3)
# 4) (116), server 6 (IP # 6) (117), server 7
(IP # 7) An outline of an upload method for uploading performed on (118) will be described. Note that the above IP # 1
Is an IP address (Internet Protocol address),
Points to an IP address having an address value of # 1. IP #
3, IP # 4,... Server 1 (IP #
1) The program structure of (100) is OSI (Open System In
terconnection) reference model application layer (10
4), hereinafter referred to as a presentation layer, a session layer, a transport layer, a network layer, a data link layer, and a physical layer, and a center load A (102) which is a center load transmission side program of the present invention is a user program.
As (106), it is located above the application layer (104), below the presentation layer, session layer, transport layer, network layer, data link layer, and physical layer. For this reason, the O which constitutes an open platform
It can operate without being affected by the SI reference model. The center load A (102), which is a transmission side program, outputs information on the center load execution status to a user (system administrator or the like) executing the center load, and inputs center load execution conditions from the user. It manages the execution of the center load, such as an interface to be accepted, and includes a job having a function as an operation manual and an operation procedure manual for the center load program, and a center load program. The center load program in the center load A (102) is the server 1 (IP # 1).
On the (100) OS, under the root directory / flas
h. A (107).

Similarly, the program structure of the server 3 (IP # 3) (101) also includes an application layer (122) of the OSI reference model, a presentation layer, a session layer,
Transport layer, network layer, data link layer,
According to the physical layer, the center load B (109), which is the center load receiving program of the present invention, also includes a user program (123) as an application layer (122), hereinafter a presentation layer, a session layer, a transport layer, and a network. Layer, data link layer, and physical layer. Therefore, the operation can be performed without being affected by the OSI reference model constituting the open platform. The center load B (109), which is a receiving program, also includes a job having a function as an operation manual and an operation procedure manual for the center load program, and a center load program.
The center load program in the center load B (109) is located on the OS of the server 3 (IP # 3) (101).
Under the root directory / flash. B (110). '/' Means a root directory.

[0009] From server 1 (IP # 1) (100) to server 3
(IP # 3) Session to (101) (Logical connection between source and destination of data in data communication)
/ Center is a program for center load under the root directory in response to activation of center load.
sh. A (107) is called (105), and passes through each layer under the center load A (102) as a user program (12).
4), the application layer (104), the presentation layer, the session layer, the transport layer, the network layer, the data link layer, and the user data area (112) after the network various headers (111) configured by the physical layer. The IP address of the target server used by the load
(113), IP address of other target server (114), flash center load object (115),
Through the physical layer, data link layer, network layer, transport layer, session layer, and presentation layer of the server 3 (IP # 3) (101) (108), and via the center load B (109). B (110).
/ Flash. A (107), the data format shown in FIG.
Received the start of transmission data center loading by MAT / flash. B (110) changes the completion code to / flash. Report to A (107). The target server IP address (113) used for the center load, and the IP addresses (11
4), the flash center load object (115), and the completion code will be described later using the data format of FIG. Similarly, server 4 (IP # 4) (116)
, A session (120) to the server 6 (IP # 6) (117), and a session (121) to the server 7 (IP # 7) (118).

The simultaneous upload method shown in FIG. 3 shows a basic operation of the simultaneous access method. As for the execution of the center load, transmission data according to a data format shown in FIG. 6 described later is created as advance preparation. In the upload, when the spreading number (511) is specified in the transmission data according to the data format of FIG. 6, the activation mode of the target servers 1 to n for the center load B changes according to the spreading number.

Here, the spreading number will be described with an example. As described in this example, the manner of center loading changes depending on the number of spreads, but the manner of center loading depending on the number of spreads is not limited to that shown in this example. For example, from server A to target servers 1 to 12
A case will be described as an example where transmission data is center-loaded to servers (abbreviated here as servers 1 to 12). The diffusion number is n
, And when n = 0, the server A to the servers 1 to 12
, The transmission data is center-loaded in a one-to-one relationship. If n = 2, server A to server 1,
Server 3, Server 5, Server 7, Server 9, Server 11
, The transmission data is center-loaded in a one-to-one relationship, and then server 1 is to server 2, server 3 is to server 4, server 5 is to server 6, and server 7 is to server 8
To the server 9 to the server 10, the server 11 to the server 12
Center load to When n = 3, the transmission data is center-loaded from the server A to each of the server 1, the server 4, the server 7, the server 10, the server 9, and the server 11 in a one-to-one relationship. 2
, The server 2 loads the center to the server 3 and the server 4
To the server 5, the server 5 to the server 6, the server 7 to the server 8, the server 8 to the server 9, the server 10 to the server 11, the server 11 to the server 11,
Is center-loaded to the server 12. Hereinafter, the same applies to the case where n = 4 or more. In this embodiment, the spreading number n is up to 100.

The simultaneous upload method shown in FIG. 3 is based on a principle which does not particularly consider the spreading number. The center load is performed from the server 1 (IP # 1) to the target servers 1 to n specified in the data format of FIG. In FIG. 3, target servers 1 to n
Are set as t2 to tn (202). Addresses t2 to tn are IP addresses, and t2
Is IP # 2, and tn is IP # n. At the start of the center load, the server 1 (IP # 1) sends the target servers 1 to 1 designated in the data format of FIG.
The center load B (201) of the server 2 having the head address t2 of the target server addresses t2 to tn (202) is started for the center load B for n. Server 2 (IP #
After receiving the transmission data according to the data format of FIG. 6, the center load B (201) of 2) deletes (204) its own IP address in the received target server addresses t2 to tn, and then executes the next target server. Address t3
The server 3 (IP # 3) having the above is activated (206). In addition,
Servers below server 2 (that is, server 2, server 3)
..) Also has the function of the center load A.

The server 2 (IP # 2) performs a center load on its own server, adds a completion code, and stores a center load A of the server 1 (IP # 1) (master server) specified in the transmission data. A completion report (203) is made to (200).
The center load A (200) of the server 1 (IP # 1) can know that t2 in the target server addresses t2 to tn (202) of the server performing the center load has been completed (205). Similarly, after receiving the transmission data, the center load B (209) of the server 3 (IP # 3) deletes (210) its own IP address among the received target server addresses t3 to tn, and then proceeds to the next target server. Address t4
The server 4 (IP # 4) having the above is activated (213). The server 3 (IP # 3) executes the center load to its own server, adds a completion code to the center load A (200) of the server 1 (IP # 1) (master server) specified in the user data. A completion report (208) is made. Server 1 (IP # 1)
The center load A (200) can know that t3 in the target server addresses t2 to tn (202) of the server performing the center load has been completed (207).

Further, the server 4 is not turned on.
When (IP # 4) cannot operate, the center load B (2
14) is the center load B (209) of server 3 (IP # 3)
Server (IP # 4) could not recognize the activation (213) from
The center load B (214) cannot receive necessary data, deletes its own IP address in the received target server addresses t4 to tn, and sends it to the server 5 (IP # 5) having the next target server address t5. Cannot start (215). Since the server 5 (IP # 5) cannot be started (215), the chain of starting stops. Server 4
(IP # 4) cannot execute the center load to its own server and cannot report the completion (208) to the center load A (200) of the server 1 (IP # 1) (master server). The time is detected by the time monitoring (211) of the center load A (200) of the server 1 (IP # 1), and the center load A (200) of the server 1 (IP # 1) determines the necessity of continuing the center load. , Target server addresses t2 to tn (2
16) having the remaining server address t5 of the server 5 (IP #
5) Start the center load B (217).

Center load B of server 5 (IP # 5)
(217) deletes the own IP address in the received target server addresses t5 to tn after receiving the transmission data, and deletes the server 6 having the next target server address t6.
(IP # 6) is started (220). Server 5 (IP # 5)
Executes the center load to the own server, adds a completion code, and specifies the server 1 (IP #
1) A completion report (218) is made to the center load A (200) of (master server). The center load A (200) of the server 1 (IP # 1) completes t5 in the target server addresses t2 to tn (202) of the server performing the center load (21).
9) Know what you did. Hereinafter, similarly, the center load is performed. As described above, if the server 4 (IP # 4) cannot operate because the power is not turned on and the chain is stopped, the center load B (217) at the end of the server 1 (IP # 1) changes the center load. The intended operation can be continued by detecting the expected completion time by the time monitoring (211) of the designated center load A (200). The above explanation is a principle explanation without considering the spreading number. However, when the spreading number is designated in the data format of FIG. 6, the simultaneous upload is performed in the manner described in the explanation of the spreading number.

The simultaneous upload route shown in FIG. 4 shows a method for implementing the method over a wide area and a method for implementing the method according to an area.
The server A (300) is a server B (301), a server F (305), a server G (312) that uses the OS of the A maker installed in the Tokyo headquarters area, and an OS of the B maker installed in the Osaka headquarters area.
C (302), D (303), and E (3
04), a center load command without a designated spread (that is, a spread number n in the data format shown in FIG. 6).
= 0), and each server returns a completion report to the server A (300) according to the simultaneous upload method. Also, the server A (300) uses the OSs of a plurality of manufacturers to specify the spread designation and the execution date and time in the data format of the transmission data shown in FIG. A center load command is issued to H311. Server H311 that received an instruction from server A (300)
Performs the above-described diffusion center load command to the server I, the server J, the server K, and the server L. Similarly, the server I, which has received the command from the server H (311), performs the above-described diffusion center load command to the server N. . The server L, which has received the command from the server H, issues the aforementioned diffusion center load command to the servers P and O. The server K that has received the command from the server H issues the above-described diffusion center load command to the server V, the server U, the server W, and the server M. The server M that has received the command from the server K issues the above-described diffusion center load command to the server T and the server Q. The server T that has received the command from the server M issues the aforementioned diffusion center load command to the servers X, S, and R. Each server H, I, J, K, L, I, N, L, P, O, V, U,
W, M, T, Q, X, S, and R execute the center load under the respective conditions, and report the completion of the diffusion center load to the server A (300).

For example, if the server to be center-loaded is 100
When the number of 00-spread number 100 is specified, the addresses of the target server 1 (507) to the target server n (508) in the transmission data according to the data format of FIG. Issue a center load instruction. That is, this corresponds to the case where n = 100 in the above description of the spreading number. At this time, the above 100
The center load B of each of the target servers receives a server address in units of 100 and data having a spreading number of 100 in accordance with the data format of FIG. Center load B of all servers that received the start of center load
Reports the completion code to the center load A of the master server # (501) specified in the data format at the time of the end of the center load regardless of the diffusion route. The method of center loading is as described in the description of the diffusion number, but the following center loading method may be adopted. That is, the above 1
The center load B of each of the 00 servers deletes its own address in the 100 target servers in the received data, and since the number of target servers 99 has become smaller than the spread number 100, the spread number is set to 0, and 1 The center load instruction is issued to all 99 target servers by designating the target server addresses.

FIG. 5 is a diagram for explaining the protocol operation between the transmitting server and the receiving server. Server A (400)
Prepares the following preparations as advance preparations for executing center loading, and creates a data file (403) in the data format of FIG. The data file (403) includes an evaluated target program or data (509) for performing center load, a designated date and time (503) that can be implemented in operation, an OS (504) of the implementation target server, a size (512), REV (505), V
The ER (506) is determined, and the simultaneous processing speed is taken into account in the diffusion number (511). Target server B (416), target server N
If the center load B programs (410) and (412) are not installed in (413), or if there is a report to update with the completion code, FTP (File Transfer Protocol) in the server A (400) The center load B program (405) is fetched by (431) (406), and the FTP (4
31) loads the center load B program (405) into the target server B (416) (408), and loads the target server B (4).
Center load B program loaded in 16)
(405) is changed to match the OS of the target server B (416), that is, ported to become the center load B program (410). Similarly, the FTP (431) loads the center load B program (405) into the target server B (413) (407), and the center load B program (405) loaded on the target server B (413) is Changed to match the OS of server B (413), ie
It is ported and becomes the center load B program (412).
Upon receiving the completion response (414) from each OS, the FTP (431) issues a restart instruction after reset (415) to link with the existing OS of the server, and executes the center load B program (410). , (412) can be accessed from the existing OS. The center load B program (410) can be arbitrarily and asynchronously executed independently of the execution of the center load. The center load A program may be transplanted in addition to the center load B program, and the center load A program and the center load B program may be transplanted.
Further, a center load B having the function of the center load A may be transplanted.

The center load A of the server A (400)
(417) sends S to the center load B (419) of server B (416).
The UM file is transmitted (418), and the center load B (419) of the server B (416) becomes the center load B (4) of the server N (413).
The SUM file is transmitted (426) in (27). Where SU
The M file is a data file including a vertical SUM, a horizontal SUM, and a vertical and horizontal SUM. Further, the center load B (419) of the server B (416) makes a completion report (424) to the center load A (417) of the server A (400), and the center load B (427) of the server N (413) is A completion report (428) is sent to the center load A (417) of the server A (400). FIG. 6 is a diagram showing the data format of the transmission data and the completion code. The data format for specifying the conditions when the center load is performed will be described with reference to FIG. Server 1 (IP # 1) (100) to Server 3 (IP #
3) The session to (101) is based on the control flag (500) of the user data, the master server IP # (501), and the SUB server IP.
# (502), date and time designation (502), OS designation (504), spreading number (511),
The transmission data format includes a size (512), a REV designation (505), a VER designation (506), a target server 1 (507), a target server n (508), data (509), and a SUM check code (510).

The control flag ((500)) can be specified as follows. bit00; 0: There is a SUB server. 1: No SUB server. Center load B consists of master server IP # (501) and S
A completion report is sent to the server designated as the UB server IP # (502). This is effective when the master server stops and the center load A (200) stops for some reason, and a SUB server is used instead of the master server. bit01; 0: No spreading 1: With spreading The flash load B that received the session specified with spreading in bit01 is specified as the spreading number ((511)) of the target server 1 (507) and the target server n (508). The session to the specified number of servers. For details, see the description of the spreading number described above. bit02; 0: Immediate start 1: Data format '
Start with the value of 'Time designation area'. This is particularly effective when there is a time difference, such as different uses / operations for each area. For example, this is effective when there are a plurality of areas such as an area in Singapore and an area in Hawaii as the area for center loading. bit03; 0: Depends on end mode 1: Immediately effective Specifies the effect when center load is completed. bit04; 0: Completion at shutdown 1: Re-start after reset Describe the termination mode. bit05; unused bit06; 0: center load B is not loaded 1:
The center load B is loaded and executed, that is, the center load B program of the target server is unconditionally updated. 00: Unconditional execution 01: Specified rev> Executed when target file rev. 10: Execute when specified rev = rev of target file. 11: Execute when specified rev ≧ rev of target file. Flash B of the target server stores the rev of the target file in the target server specified in the data (509).
(Revision) and transmission data format REV designation (50
The specified rev specified in 5) is collated. 00: Unconditional execution 01: Specified ver> Executed when target file ver. 10: Execute when specified ver = target file ver. 11: Execute when specified ver ≧ target file ver. Flash B of the target server contains the version of the target file in the target server specified in the data (509).
(Version) and VER designation of transmission data format (506)
The specified ver specified by is collated. bit12 to 13; 00: unconditional execution 10: executed when specified OS = target file OS. The flash B of the target server checks the OS of the target file in the target server specified in the data (509) against the specified OS specified in the OS specification (504) of the transmission data format.

The data format following the control flag (500) specifies the following. Master server IP #; IP # of the server that instructs the center load, all servers that have executed the center load
The result is reported to this IP #. SUB server IP #; 00 bit of control flag is'
1: Effective when reporting to the SUB server. This is effective when the master server fails to receive the result report for some reason. Date and time designation; 02 bit of control flag is'
1: Execution by the value of the "time specification area" of the data format. Executes when the target server reaches the specified value. OS specification; 12 to 13 bits of control flag
Is other than '00, the OS in the OS specification area of the data format is compared with the OS of the target file. The specified OS is Windows
(Registered trademark), UNIX (registered trademark), MS-DOS
(Registered trademark). Number of diffusion; 01 bit of control flag is
1: Diffusion Specify the number of diffusion for nest expansion. Size: Free space of target server required to store data to be center-loaded REV designation; 07 to 08 bits of control flag
Is other than '00, the rev of the REV designation area and the rev of the target file are conditionally collated. V ER specification; 10 to 11 bits of control flag
Is not '00, the ver of the VER designation area and the ver of the target file are conditionally collated. Target server: Designates the IP # (IP address) of the server that executes the center load. That is, the designation of the destination information of the center load. '00' is invalid. Data: Specify the file to be updated and its members (independent data groups that make up the file). SUM check code; automatic check of user data
/ Used for recovery.

The center load B of each target server returns a completion code to the center load A. The completion code is the data format for executing center load A.
Master server IP # specified in control code in mat
(501), master / SUB server address 52 instructing the center load of each 8 bytes of SUB server IP # (502)
0, the target server address 521 that performed the 8-byte center load, the 1-byte completion code 522,
It consists of a 2-byte additional code unique to the server to supplement the completion code. The meaning of the completion code is described below. 0x: Normal 00: Shutdown after normal end. 01: After normal termination, reset and restart. 1 ×: Control code 10: The receiving target server IP # does not have its own server IP # designation. Specification exception 11: The content of the spreading specification (01 bit of the control flag) in the received user data does not match the spreading number (the value of the spreading number (511)). 12: SUB server IP # is invalid with SUB server specified in control flag. 2x: Communication error 20: Data reception error was successfully repaired by SUM block. 21: Recovery of data reception error in SUM block failed. 3 ×: Device error 30: File creation failed due to device factors such as hardware failure / data invalid. 31: In the operation before renaming, the center load operation was interrupted by some factor such as hardware failure, data error, illegal operation, etc., and the old system started up. 4 ×: File specification exception 5 ×: Condition mismatch 50: The specified REV # in the received user data does not match the specified REV # in the own server. 51: The specified VER # in the received user data does not match the predetermined VER # in the own server. 52: There is no free area larger than the specified size in the received user data. 53: The model / OS in the receiving user data does not match the model / OS in the own server. 6 × to XX: Reserved F ×: Unrecoverable Note that “×” indicates that any numerical value may be used (however, in this case, a hexadecimal numerical value).

Next, the data transmission block shown in FIG.
The generation of the horizontal SUM data, the vertical SUM data, and the vertical and horizontal SUM data, and the recovery of the record data from the receiver using the SUM data will be described. Server 1 (IP # 1) (100) to Server 3 (IP # 3) (101)
Is divided into record data 1 (600) to record data 32 (601) having a length of 256 bytes, and the exclusive OR of each record is performed to generate a horizontal SUM1 (602) having a length of 256 bytes. . Similarly, horizontal SUM2 to horizontal SUMm (604) are generated for all data to be transmitted for every 32 record data. Block 1 (608): 32 in each of block n (603)
Exclusive OR is performed on the record in the unit of distance of the record, and a vertical SUM1 (605) having a length of 256 bytes is generated. Similarly, vertical SUM2 to vertical SUM32 (606) are generated. Exclusive OR is performed on each record of horizontal SUM1, horizontal SUM2,..., Horizontal SUMm, and vertical and horizontal SUM (607)
Generate The data generated in this manner is transmitted from the center load A of the master server IP # (501) to the center loads B of all the target servers.

The data received at the center load B of the target server can be recovered even if one record data in one block is lost.
For example, when the record data 2 in the block 1 (600) is lost, the receiving server removes the record 2 except the record 1, the record 3 to the record 32 (601), and the horizontal SUM1 (60).
By performing an exclusive OR on 2), record 2
Data can be recovered. The check of the recovered record data 2 is exclusive of the recovered record data 2 of the block 1, the block 2 record 34 having a distance of 32 record data,..., The record X-31 of the block n (603) and the vertical SUM2. By performing the logical OR, when the result becomes all “0”, it can be performed by assuming that the data can be correctly recovered. Thus, in this case, one record of data loss per block is:
Since recovery can be checked by using the horizontal SUM and recovery can be checked by using the vertical SUM, even if data of one record is lost in each block, data recovery of the record is possible.

Further, even when a plurality of records in one block are lost, data recovery of the records is possible. For example, if all record data of record 1 to record 32 in block 1 (600) is lost, record 1
Can be recovered by performing an exclusive OR operation on the record 33 of the block 2, the record X-32, and the vertical SUM1 (605). Similarly, the record data 2 to the record data 32 are the same as the vertical SUM2 to the vertical SUM32. , These vertical SU
Recovery can be performed by performing an exclusive OR operation on records other than the lost record used for obtaining M. The records 1 (600) to 32 (601) thus recovered include these records and the horizontal SUM1 (60).
The exclusive OR of 2) is performed, and the result becomes all “0”, so that it can be checked that all record data has been correctly recovered. In this case, continuous record data can be recovered up to a maximum of 32 records.

The SUM check generation flow of FIG. 8 will be described below. When the user data is transmitted from the server 1 (IP # 1) (100) to the server 3 (IP # 3) (101) (700), the user data to be transmitted is divided into 256-byte record data units. (701) Then, n record data (n = 32 in the illustrated example) are generated as one data block (step 20). The value of the record number is set to "0" (702), exclusive OR of 0 to n, that is, 32 record data is obtained, and the horizontal SUM data for the first data block is calculated (704).
0 to n, that is, 32 record data and its horizontal SUM data are transmitted (steps 21 and 22). Similarly, for blocks 1 to m, the horizontal SUM of the data block to be transmitted is calculated (705). The calculation is performed up to the last block (706). Assuming that the block number is “1” (707), the first vertical S
UM data is calculated (708), and the value of the control counter is set to +
While calculating 1, m, that is, 32 pieces of vertical SAM data are calculated, the calculated vertical SUM data is recorded (709), and vertical SUM data up to the last record is calculated (710).

The data transmission block shown in FIG.
According to the SUM check generation flow, the data transmission operation with high reliability and high quality as described above can be performed, and the error can be recovered on the receiving side. The embodiment described above is
Although the exclusive OR operation is used to calculate the horizontal SUM data, the vertical SUM data, and the vertical and horizontal SUM data, other logical operations or arithmetic operations such as addition, subtraction, multiplication, and division may be used. In this case, a similar effect can be obtained.

The operation of the center load transmission side program will be described with reference to the center load A operation flow shown in FIG. Server 1 that starts center load (IP # 1)
In the center load A (800) of (100), when the control flag of the user data to be transmitted includes a diffusion designation, the target servers are grouped in units of the number of diffusions, and the transmission is performed at every diffusion number in the designated target server. Settings (801) to be performed. For example, if the above-mentioned center load target server is 10
In the case where 000 units−spreading number 100 is specified, the addresses of the target servers 1 (507) to n (508) in the transmission data according to the data format of FIG. Address, that is, 1st, 101st, 201st,.
The address of each server of the unit and the spreading number 100 are set. If there is no spreading designation, the spreading number is set to 0 (804), but since it is transmitted to all the specified target servers,
It is necessary to consider that the load on the server 1 (IP # 1) (100) will increase. Send according to the above settings (81
5) do. At this time, the transmission does not involve the t0 mark and does not have the t0 mark. Note that the spreading number shown in the transmission (815) = 4,
The spreading number = 0 indicates an example of the spreading number.

The center row of the transmitted target server
In step 807, a time value until receiving the completion code and a timeout value Δt1 are set. Timeout value Δ
t1 is determined in consideration of the operation. For example, immediately execute
If the number of diffusions (511) to the target server (507) is set to end within the time, Δt1 = 1 hour (3
600 SEC). Also, if you specify the date and time (502) and execute the center load 3 days before the specified date,
Δt1 = 3 days + specified date = 4 days (345600SEC)
Set. I of the target server that received the completion code
The P # is recorded (808). When the completion codes are reported from all target servers that have instructed the center load, the execution result is displayed / printed (813). The operation of the center load A (800) of the server 1 (IP # 1) (100) instructing the center load is completed (814).

If the timeout has occurred before receiving the completion code reports from all target servers that have instructed the center load, the IP addresses in the target servers that have been managed and whose completion has been reported are deleted, and the completion report has been completed. Instead, the t0 mark is added to the first server whose center load is not completed (806). Then, the process returns to step (815). The first server whose center load has not been completed is the server that caused the timeout and is inoperable for some reason, so that the t0 mark is added to this server. The t0 mark is a mark indicating the first time-out, which is given to a target server that has not reported the completion of the center load to the server that instructed the center load. In the step (815) after the time-out, the target server (inoperable) marked with t0 indicating the first time-out is transmitted to the next designated IP #, and t
The 0 mark is updated to an abort mark. Next, the processing of step (807), step (808), and step (810) is performed. In the process of step (810), the server with the abort mark is excluded from the check target of the completion report. ab
When there is a completion report from a server other than the server with the ort mark, the execution result is displayed / printed (813).
Server 1 (IP # 1) that instructed center load (100)
The operation of the center load A (800) is completed (814). In one of the diffusion number units in the target server to manage,
When the t0 mark indicating the first time-out is added, if the next target server IP # remains after the mark, the diffusion chain is broken. In this case, an instruction is issued again in the initial diffusion unit from the next target server IP # after the mark, and the chains are connected.

The operation of the center load transmission side program will be described with reference to the center load B operation flow shown in FIG. The center load B of each target server that has received the center load instruction from the center load A (900) of the server 1 (IP # 1) (100) transmits data to be collated using a command provided by the OS of the own server. read out. When a center load instruction is received from the center load A (900) of the server 1 (IP # 1) (100), the server I (IP # 1) (100)
Is P # read and the target server in the received user data has IP # data of its own server? Collation (903)
I do. It should be noted that a server that does not perform center loading does not normally receive packets (111 and 112 in FIG. 1). If the collation does not match, the completion code '10' is created (902). If there is a match in the collation result, the process shifts to the check of free area (904), the free area of the own server is read, and is there a free area larger than the size in the received user data? Collate. If the collation does not match, the completion code '52' is created (905).

If the verification conditions are satisfied, REV verification (914)
Is made. When a center load instruction is received from the center load A (900), the rev value of the target file in the target server specified in the data (509) is read and specified by the REV specification (505) of the transmission data format. The value is compared (914) with the designated value of the specified rev. If the conditions do not match, the completion code '50' is created (912). For example, the revision designation collation area 0 of the control flag (500)
If '11' is specified in 7,08 bits, the rev value of the target file in the target server specified in the data (509) is changed to the specified rev specified in the REV specification (505) of the transmission data format. 'Smaller' than the value of
Is executed when the value matches or matches, and returns a completion code '0x'. If it is "large", a completion code "50" is returned.

When a center load instruction is received from the center load A (900), the ver value of the target file in the target server specified in the data (509) is read, and the VER specification (506) of the transmission data format is performed. ) Is collated (916) with the value of the specified ver specified by (). If the conditions do not match, the completion code '51' is created (913). For example, version designation collation area 1 of control flag (500)
When “10” is specified in 0 and 11 bits, the value of the ver of the target file in the target server specified in the data (509) is changed to the specified ver specified in the VER specification (506) of the transmission data format. Is executed only when the value matches with the value of, and returns a completion code '0x'. Disagreement '
In this case, a completion code '51' is returned.

When a center load instruction is received from the center load A (900), a predetermined model / O
S is read and collated (917) with the model / OS in the received user data. If the conditions do not match, the completion code is created (915). When the above condition is met, the user data from the center load A (900) is received and the designated file is stored in the empty area. If the file creation fails for some reason during reception, such as a hardware failure or data injustice, the completion code is created (907). If the file is created normally, check the next spreading number (911)
I do. Is diffusion specified in the received user data?
Check (911). If not specified, the current file and the updated file are renamed (919). If specified, the own target server # in the received user data is deleted and transferred to the next target server # (910). When the next target server # is no longer specified, the current file and the updated file are renamed (919). The current file is deleted, and files such as the update file, OS, and patch data created in the empty area are renamed so as to be recognized as the current file. As a result, by restarting the target server, the startup in the new environment after the center load is guaranteed. In addition, the operation of the center load B of each target server that has received the center load instruction from the center load A of the server that performs the center load has been interrupted due to a hardware failure, improper data operation, improper operation, etc. In this case, the center load operation is stopped and restarted, so that the startup in the old environment before the center load is guaranteed.

The server that has received the center load reports the completion code to the master server in the received user data (925). If the control flag specifies the SUB server, the completion code is also reported to the SUB server (925). The end processing (921) is performed according to the end mode of the control flag in the received user data. If shutdown is specified, shutdown (920) is performed.If restart after reset is specified, restart (923) after reset is performed.

[0036]

As described above in detail, according to the present invention, the server of each manufacturer, the target file in the server, the OS, the installation area of the server, and the update date and time can be arbitrarily set in the open platform environment. You can select and upload inexpensively all at once. In addition, the center can be efficiently loaded to the target server using an arbitrary server as a center.

[Brief description of the drawings]

FIG. 1 is a device connection diagram showing connection of each device within an open platform.

FIG. 2 is a diagram illustrating a control relationship of each device in an upload method.

FIG. 3 is a diagram showing an outline of a method of loading to a plurality of devices in a simultaneous upload method.

FIG. 4 is a diagram illustrating a simultaneous upload path showing load development to a plurality of devices.

FIG. 5 is a diagram illustrating a protocol operation between a transmitting server and a receiving server.

FIG. 6 is a diagram showing a data format of transmission data and a completion code.

FIG. 7 is a diagram showing a data transmission block for explaining a data verification and recovery method.

FIG. 8 is a diagram showing a SUM check generation flow.

FIG. 9 is a diagram showing a center load A operation flow showing an outline of the operation of the transmission side program.

FIG. 10 is a diagram showing a center load B operation flow showing an outline of the operation of the reception side program.

[Explanation of symbols]

 1 Server 1 (Manufacturer A) 2 Server 2 (Manufacturer B) 3 Server 3 (Manufacturer A) 4 Server 7 (Manufacturer D) 5 Server 4 (Manufacturer C) 6 Server 6 (Manufacturer A) 7 Server 5 (Manufacturer A) 8 Net A 9 Net B 10 Net C

Claims (3)

[Claims] 1. Information for center-loading data from an information processing apparatus coupled to a communication path such as the Internet to one or more other information processing apparatuses coupled to a communication path such as the Internet and assigned a unique address. A center loading method of a processing device, wherein the information processing device that performs center loading transmits transmission data including destination information and an update object to one or more other information processing devices designated by the destination information via a communication path.
The center load is collectively and simultaneously loaded, and the progress of the center load is managed by receiving completion information indicating the completion of the center load from each of the information processing devices loaded with the center via a communication path. Each information processing apparatus updates an object corresponding to the received update object with the update object, and transmits completion information indicating completion of center loading to the information processing apparatus performing center loading via a communication path. Center loading method of the information processing apparatus. 2. Information for center-loading data from an information processing apparatus coupled to a communication path such as the Internet to one or more other information processing apparatuses coupled to a communication path such as the Internet and assigned a unique address. A center loading method for a processing device, wherein the information processing device that performs center loading transmits transmission data including diffusion designation information, destination information, and an update object to one or more other information processing devices designated by the destination information. Through the center load all at once, and
The progress of the center load is managed by receiving completion information indicating the completion of the center load from each of the center-loaded information processing devices via a communication path, and each of the center-loaded information processing devices receives the update object. Is updated by the update object, and the completion information indicating the completion of the center load is transmitted to the information processing apparatus that performs the center load via a communication path. A center loading method for an information processing apparatus, wherein the transmission data is collectively and simultaneously loaded to one or more other information processing apparatuses via a communication path. 3. Information for center-loading data from an information processing apparatus coupled to a communication path such as the Internet to one or more other information processing apparatuses coupled to a communication path such as the Internet and assigned a unique address. A center loading method of a processing device, wherein the information processing device that performs center loading transmits transmission data including destination information and an update object to one or more other information processing devices designated by the destination information via a communication path.
The information processing apparatus performs a function of managing the progress of the center load by collectively and centrally loading the center load and receiving completion information indicating completion of the center load from each of the information processing apparatuses that have been center-loaded via the communication path. To the information processing apparatus that updates the object corresponding to the update object received by the center load with the update object, and transmits the completion information indicating the completion of the center load via the communication path to the center. A second program for causing the information processing apparatus to realize the function of transmitting is stored in the storage device, and the information processing apparatus executes the first program when center-loading the information processing apparatus into another information processing apparatus. Executes the second program when the first and second programs are executed. Center loading of the information processing apparatus by porting the first and second programs to an information processing apparatus that does not store the program in the storage device. Method.